fgam guide 25th mar 2015 it will change your life forever

.,
,
~ _ ”
-
.“”-
-.._ _--.
-..,.I)
“
l
i
l
.....
.._._,
,
... .,.
.
..
UNIVERSIDAD AUTONOMA METROPOLITANA
PLANTEL IZTAPALAPA
,
MEDICINA I V
//;íCANCER CERVICQ UTERINO
ALUMNOS 2
.,&ANA
ROC10 VICTORIA ROMERO
MIGUEL ANGEL RODRIGUEZ CAMPA
PROF.-
86228042
86333373
ALFONSO MARTINEZ O R T I Z
Juba 90.
MARCO NEDICQ
UNIVERSIDAD AUTONOMA METROPOLITANA
PLANTEL IZTAPALAPA
MEDICINA I V
CANCSR CERVICO UTERINO
ALUMNOS :
DIANA ROC10 VICTORIA ROMERO
86226042
MIGUSL ANGEL RODRIGUEZ CAMPA 86333373
POFRe-
ALFONSO MARTINEZ ORTXZ
.
., ~ , ,
,
,,,
, ,.,.
;.
......
1~~
_
"
,
.
e
-
".-s. .
.
,..,
,
. . .. ,
.,
r
I N P I C E
PAG
...
-.
r
- ..
--.
- ..
...
.
i.
..
~
.-.
...
~
..
..
I
..
...
..
.
L
"
.
E l cáncer cervicouterine constituye un problema de salud d e
importante magnitud en nuestro pals, que se manifiesta por
ui)
na tendencia ascendente como causa de mortalidad y de gran
trascendencia social a l afectar a l a s mujerss en su etapa mds
productiva y relevante dentre del núcleo familiar.
-
Por fertuna, l a acceskbilidad del útero y l a s caracterfstic
as bioldgicas de l a evoiuci6n de esta patología, a s í cemo e l
-
desarrollo de técnicas diagndsticas y terape6ticas, hacen vu1
nerable e l padecimienta s i &stas meaplican oportunamente en
sus etapas incipientes o preciínkas.
-
Es indudable e l valer de l a citelogía exfeliativa en e l dia
gnostico del cáncer cervical. Su a l t o grado de sensfbilidad y
,,.
.
especifidad, aunad@ a su sencillez y bajo costo relativo originan que en l a actualidad sea censiderado e l método más efi-
caz en e l dian6stico temprano del cáncer invasor del cérvix,
a s í como de sus precursores: e l carcinema
ill
'11
y las displ
asias.
Las autores de numeresos estudios en diversos
países coinc
iden en afirmar que l a implantacih de programas masives de J
d e t e c c i h con e l método citoibgice, que aseguren coberturas
impertantes de l a p e b l a c i h en riesgo, y se apeyen en un sic4
tema adecuado de tratamiento y contrel de l a patología cervi-
r e d u c c i b de las tasas de mer
-
ca1,contribuyen censiderablemente a l control del cáncer
ceuterino, que se r e f l e j a en l a
talidad per esta causa.
ervi
De l a s muchas responsabilidades asumidas p o r e l ginecblogo moderno, q u i z á ninguna es tan importante como l a detección
de l a s n e o p l a s i a s l o mas temprano y osortuno posible.
2h e s t e t r a b a j o se c r i t i c a n v a r i o s métodos diagndsticos
corrientemerite empleados; s i s e comprenden todas sus l i m i t a
-
c i o n e s y sus v e n t a j a s y s e u t i i i z q n apropiadamente, podría
s e r p o s i b l e aproximarse a l a meta de aue "nadie debería m o r i P
de carcinoma de c é r v i x .
La introducción de l a c i t o l o g i a en l a p r á c t i c a ginecoló-
-
g i c a ha cambiado s i g n i f i c a t i v a m e n t e l a deteccibn de cáncer c e r
vical.
Este cáncer, actualmente s e diagnostica en e s t a d i o s
c l í n i c o s mucho más tempranos.
Además l a c i t o l o g i a es capav de
d e t e c t a r a l o s precursores d e l c b c e r invasor:
e l carainoma i n
s i t u y l a d i s p l a s i a ; por l o tanto e s indudable su v a l o r en l a
-
deteccidn c i t o l 6 g ~ c a ; s i n embargo, e s t a sentencia no s e ha cum
p l i d o en poblaciones que en su t o t a l i d a d han s i d o examinadas
p o r e s t e método una v e z a l ano.
Bryans y colaboradores (4)
en
un extenso programa de detec,ción c i t o l ó g i c a en Columbia B r i t & - n i c a , mostraron que l a i n c i d e n c i a d e l Carcinoma c e r v i c a l invas o r f u e de 3.5
X 100,000 mujieres estudiadas, comparado con l a
t a s a de 24.1 X 180,000 en poblaciones de mujeres no estudiadas
Unil~garainomac e r v i c a l no s e d e s a r r o l l a en un.,año; p o r l o tanto
l o s cánceres c l í n i c o s que se presentaron en l a población estudiada previamenze p o r c i t o i o g í a , i n d i c a n l a t a s a de f a l s o s neg a t i v o s de e s t e método.
A causa de e s t a c i f r a , l a i n c i d e n c i a
de carcinoma invasor d e l c e r v i x nunca puede b a j a r a cero.
DATOS ANATOMICOS DEL TRACT0 GENITAL FEMENINO
-
El tract0 genital femenino astd compuesto por l a vulva, l a va
gins, e l dterc, l a s trompas de falapio y l e s evarios.
.
RECTO
VAGINA
, .
.
Lp
vulva, que representa la parte externa, es una estructura e
compleja formada por los siguientes e1ementes:l) mente de venus
, 2) labies mayores,
..
..
3 ) labies Rencores, 4 ) c l f t c r i s , 5)vestfbu-
Is, 6 ) meate urinario, 7 ) orificia, vaginal, 8 ) hiaicn (en l a s
vfrgenes) y 9 ) glándulas vulvwaginales.
.
<
MONTE DE VENUS
ES un mntfculo de g
ma, cubicrte ,
:
el
y situ d
F r e -
ima de l a sfnfhsis pubiana, en l a parte más inferier de l a pare
cd abdeminal anterior.
-3
-
LABIOS MAYORES
Sen dos pliegues lengitudinales salientes, fermados por tejid@
a d i p s a recubierto de pie&
-
que en sus partes externas se encue
n t r a n cubiertos de vello. Durante l a pubertad se desarrollan c?
-
nsiderablemente y despugs de! esta etapa se extienden hacia atrá
~
. ..
s, en direccidn a l perinee; s i se separan en l a parte pasterior
-
se puede observar una comisura ligeramente saliente, llamada ha
r qu i l l a.
LABIOS MENORES
Son dos pliegues Be direccidn anteroposterier que se observan
i
)
-
a l separar les labios mayores. En l a parte anterior se subdivid
en y cubren con una horquilla e l glande del C l f t Q r i S , farmando
-
una cubierta llamada capuchbn, mientras que l a otra h o j i l l a pas
a p o r debajo del glande para formar, con l a del lado epueste, e l
-
I
f r e n i l l e del c l í t e r i s . La p i e l que cubre los labios menores est*
” .
<.
.
S provista de follculos piloses, pero es muy r i c a en glándulas
sebáceas.
CLITORIS
Es un pequeña brgane, eriktil de forma cilindrica, que está for-
-4-
-
mads por un glande, un cuerps y dos raíces; e l glande de aproxi
madamente ti a 8 nun de didmetro, es l a única porcidin v i s i b l e de
6rgano
desde e l e x t e r i w . E l c l í t e r i s está formado por t e j i d s
erkctil en e l quo abundan canales venesos rodeados de gran cantidad de fibras musculares lisas.
VESTrBULQ
SE llama vestíbule a l a excavacib navicular que se observa a l
separar los labios de l a vulva; en é l se encuentra e l s r i f i c i o
-
vaginal y per delante de este e l meato urinario. En l a mujer v i
rgen, e1 ~ r b f i c i evaginal se encuentra parcialmente
9ClUidQ
per
e l himen, que es una membrana farmada per tejido caneetivo f i r -
-
me y cubierta por un e p i t e l i o plano estratificado que frccuntem
-
ente tiene forma lunar o semilunar, pero tambian puede ser c r i b
-
iforme; en casos anermales puede permanecer imperforado y ocasi
m a r retencibn del f l u j o menstrual.
MEATO U R I N A R I O
-
Es e l pequen@ o r i f i c i o externo de l a uretra, de forma triangul
ar
6
-
de hendedura. Frecuentemente se e-bserva a cada lado del me
ato una pequefia depresi6n en l a que se encuentran l a s llamadas
- 5-
su per
-
glándulas meneres del vestfbdla. La uretra, en su posición prox
imal, se encuentra revestida por epitelio de transicibn;
cidn d i s t a l está cubierta per e p i t e l i o plano estratificado no
Q
c$rneo.
GLANDULAS VULVOVAGINALEC O DE BARTHOLIN
-
Son glándulas arracimadas, dispuuestas en lbbulos que se encuen
tran a cada lado del o r i f i c i o vaginal, especialmente durante e l
coito, su funci6n censiste en secretar mucasidad para lubricar
e l s r i f i c i o y e l conducto vaginal. El [email protected] de l a
glándula está cubierts, por e p i t e l i o de transicinn.
TROMPA DE FALOPIO
&[email protected] UTERO
CAVIDAD ENDOMETRIAL
FIM5RIAS
CUERPO UTERINO
CERVIX
VAGINA
VAGINA
La vagina es un conducto musculomembranoso que une l a vulva a
a l btero y se relaciona anathicamente con l a vejiga por delani
e y cen e l recta hacia atrás ; mide entre 9 y 10 cm de longitud
.I
y se d i r i g e hacia arriba y hacia atrás desde su extremidad vul-
-
.
.
,,
var hasta su extremidad uterina.
E l extremo superimr, a l d i l a t
arse, forma e l f b r n i x en e l cual hace pretrlsién l a porcidn vag
-
i n a l del cervix uterine, e portio; las areas demarcadas por e l f '
-
6 r n i x y p w e l cgrvix fornaao%os fasndos de saco vaginales later
ales, anterior y posterior, siende eke último e l más profunds y
-
e l s i t i o donde se acumulans las cklulas exfoliadas y l a s secreci
-
enes de l a s glándulas utcrinas; e l fendo de sac@ vaginal pester
de acceso a l a cavidad abdominal.
l a mujer virgen l a muce
l a vagina presenta una serie de arru
sa de l a pared anteriar
i o r tiene también interksequirúrgico por constituir una vfa fác
il
En
de
gas transversales y un pliegue central longitudinal que l e dan
-
e l aspecto de "árbol de l a vida"; en l a multfpara, cuyo conduct
o vaginal está claramente distendide ne hay pliegues. La pared
de l a vagina esta fermada
por tres capas: l a mucosa que tapiza
e l canal vaginal es l a más interna; por debajo de l a mucosa se
encuentra l a túnica muscular formada por dos capas de músculo &
l i s o ; l a capa fibrosa es l a más externa de tedas y
-17-
4
por te j i d e conectius pehians.
UTERO
L
.
."
E l 6tere es un lbrgane hueco, de gruesas paredes musculares, s i t
-
uade en l a pelvis entre l a vejiga per delante y el recto por de
tras, sostenido per una serie de ligamentoa y por hojas d e l per
itoneo que se reflejun y se extienden hacia los lados del 6tero
y se conocen c-e
paramtrios dereche e ircquierdo. Tiene forma
de pera o de p i r k i d e invertida; en l a nulípara mide alrededu
de ocho o nueve centfmetros de longitud, seis centímetros en su
-
parte d s ancha y unos cuatre centfiactros de espesor. Se encuen
.,
,
tra inclinado hacia adelante y hacia abajo.
E l btero esta fumado de dos partes: e l cuerpe y e l cuello. En
. .
.<
l a mujer pre*ber
y en l a saclnop&usica, e l cuerlw, es muy pequeño
pero durante l a edad reproaluctiva suele ten-
pulpmen aumentado
a consecuencia de l a eatimúacibn ovQrica."
La pared del atere est& formada por delgadas capas de nñsculo
liso capaces de aumentar considerableumnte su peso y tanafie durante e l enbaraso. La Wrcibn superier del cuerpo se denomina
fondo o fundus; e l h g u l o que marca a une y otro lade e l erigen
aparente de l a s trompas recibe e l nombre de cuerne. E l c&rvix o
es una estructura tubular de aproximadamanta cuatro centímetros
de longitud y trea centímetros de diámtro; de su longitud total
- (3-
cerca de l a mitid se encuentri cerca de l a vaqini y constituye
1 i pertie, e perción va9inii del cérvix; e l resta e s t i ecuíto
..
per l a pared vaginii y se centinóa cen e l cuerpo del dtera.
US
-
reiacienes anitómicis del c d n i x sen con ii vegiga anteriuc
mente y c m wnbos urcteres hacia los lados.
Las paredes del 6tere fuman una cavidad que sigue l a f o r m del
cuerpo uterine; es cbnica, su base corresponde a l fend. y e l
vCrtice a l esteurn interne u a r i f i c i e cervical interne. La cavidad del cuerpo uterino se cbnece coii). cavidad endometrial y se
cemunica cen e l can81 endocervical hacia abaja, La abertura del
canal endecervical en l a v8jiAa se c m e e c « w osteuffl extern. u
-
u b f i c i r a e r v i c i l &%terne, y elbpinte de transicibn del canal e
ndocervical y l a cavidad endemetrial es e l osteum interne mencion ado.
La mucesi del c u e r p uterine recibe e l nombre de endometrio; e l
-
e g i t e l i e endocervicil recubre e l aanal correspondiente. El estr
0 ~ esta
1
ferrnado per tejido conective. U túnica muscular, for-
mada par fibras musculares l i s a s , entrecrumdas, recibe e l neut-
-
bre de Biiametrio. La tónica serese está formada por e l peritene
o que cubre tedo e l cuerpe uteririm;-
Las trampas de falopie miden entre eche y doce centímetros de
-
large par tres a cinc. centfmetras de d i h t r e , Sus extremos pr
oximaies se cemtinóm can ci cuerpo uterino, mientras que íos
- 9-
distales con su perci6n fimbriada se abren libremnte dentre de
l a cavidad abdeminal. A t d a su lengitud l a t r w p a presenta un
estreche canal gue asegura l a csmunicacibn directa entre l a v b
gina y l a cavidad abdominal (de impertancia para l a dliscainc
cibn de infecciones e de tumeres malignos).
..
.
Les avaries, une derecho y une izquierdo, sen estructuras w e i des que miden alrededer de cuatre H r des por des c e n t h t r e s .
Se lecalizan anatemicumnte en l a vecindad de l a perci6n fimbriada de las treapas, aunque ne directamente centigues a l a luz
..
tubaria. Les evarios a l iyual que las trorapas se encuentran
-
80s
tenides dentre de l a pelvis per medie de repliegues peritutb.ít
les.
EL UTERQ
L.S
Y LAS TRBDMPAS
DE FALfbBIO
ovarios, e l ótere y sus trempas, y l a vagina constituyen l e s
$r,rcganes internos de repr.ducci(in en l a mujer. Les wvaries se encuentran en l a pared pesterior de l a pelvis, sostenidos por e l
-
ligamente susgenserie d e l @vario (que lleva les Vasos), e l ligamen
to o v b i c o y una extensidn del ligamento ancho. Ea esta vista, los
ovaries han side fraccionados hasta una p s i c i 6 n hwizmntai para
..
..
esclarecer sus relaciones cnn l a s treinpas o tubas uterinas. Las
ttowpas o tub*s,uterinas sen extensimes laterales del titer.,
re-
-
vestidas de e p i t e l i o columnar c i l i a d o sestemdole pwr t e j i d e cenecti
vo y músculo liso. US
centraccienes rltmicas de este másculo ayu-
dan a l $vule en su v i a j e hacia l a cavidad uterina, y l a s c(5lulas
de recubrlipiento l o mantienen nutricionalmente. La trompa muestra
-
tres porcienes bien definidas: l a fimbria, que encierra l a s u p r f i
tie antcrier y superior de1 ovario,
**atrapa*@
e l 6vuie expulsade y
1s mueve rápidamente hacia su interior; l a ampolla o perci&n nias
ancha de l a trompa y e l istmo, cuya luz se estrecha cenforme pena-
-
t r a en l a pared uterina. E l óters es una estructura en ferma de pe
r a cuy0 cuello (cervix) entra en l a porcibn superior de l a vagina
y cuy@ cuerpe/fona¡o est$ flcrxienade ( a n t e f l e x i h ) e inclinado ha*.
- .
c i a adelante (anteversibn) sobre l a vejiga. La fiexiQn e inclinaci$n hacia atrás (retr,flexi(ln/retroversibm)
-\\
-
es frecuente, parti-
cularncnte en mujeres que ya han dade a luz. Esta última posici6n
del dtero ("inclinada"),
-
s i es importante, puede conducir a una va
riedad de molestias, desde darlor hasta infertilidad. La situacibn
tambidn predispone a un ligero deslizamiento del útero dentre de
r
l a vagina (prolapse), ya que este pone a i dtero más e menos en e l
e j e del cuelle y l a vagina. La pared del btero, como puede obser-
c
varse, es en su mayorfa d s c u l o l i s o (miornetriel recubierto cen
-
una capa glandular de greser variable (endmetrie) que es extrema
damente sensible a las hormonas estr6qens y progesterena. En el
lugar donde e l cervix penetra en l a vagina, se forma un canal e
fesa alrededor de d l ifendo de sac8 ar fornix vaginal). Esta &ea f
.
-
fibroelástica se expande considerablemente durante e l ceito. Justa
mente a l lade del cervix e cuerpo del útero, para e l ureter.
en su
camine hacia l a vejiga, my cerca de l a arteria uterina (una rela-
-
ciQn importante para e l cirujano gincc6logo). Debido a l a potencia
lmnte precaria posici6n del útere, e1 sastdn ligamentario de esta
* ,
..
.
estructura es crucial. E l ligamente ancho, una capa de peritoneo,
en fmna Be s I b M a sobre e l btere, sus trampas y les Bvarias, j u r
ga e l papel de sost& mas importante, en asociaci6n con otros.
UTERQ Y TRBHPAS DE FALLOPIO
.
.
,-1
...
c.
.,.
,'.
.
.
c
Utera:
a) fendo
Estructuras relacionadas:
i) ligamento anche (peritoneo)
b) c u e r w
3) ligamento redonde
c ) cueil.,
k) ligamente o v k i c e
canal cervical
L.
d) aavidad utarina (endometrio)
1) ligamente suspcnserie del
P-
e) miematrie
a) arteria uterina
a. vesical superior
CI
r'
-.
rC..
c-
Trampas de Falopi*:
O ) avarie
f ) istmo
p) vayina
g) ampolla
q) uretere
h) fimbria
I
c
-13-
p?) fend. de saco
vario
_L_1_L.
REPReDUCTM
AIARATTO
FEMENINO
b
a
.c
cBrganes reprducteres internos:
Estructuras jenitales externas:
a) ovario
e ) mente ae Venus
f ) labies mayores
b) trompa de Falepie
h) labios menores
i ) prepwio del c l f t e r i s
j ) f r a n i l i e del c l f t e r i s
k) herquilla
Vestíbule
1) uretra
ni) oi&adulas parauretrales
n) himen
e ) u i f i c i e vaginal
p) gllndulas vestibulares
9) cuarpa, perineal
g) C l f t c P r i S
c ) 6tero
d) vagina
- \4
r ) ane
E]. aparate reprducter femenina esta fermado gor partes inter-
nas y externas. E l &gane primarir de este aparate es e l evarie,
e l cual prduce l a s c6luias germinales femeninas (bvuies) y secreta l a s hermenas estrQSenes y pregesterana. US
estrkenos son
l a principal harmma femenina y es respemsablc, entre @tras cesas,
de l a s características sexuales secundarias (desarrelle ale turnas,
ensanchamiente de las caderas, crecimiente del v e i l e P b i c e , etc.
asf cemo del desarrelle de las glgndulas y cenductes d e l tract. rzp r d u c t e r en l a pubertad, EI evarie, a l igual que e l testlcuim, se
origina en l a pared pesterior del abdomen (a un lade de les r i i i t
nes) durante e l atesarrelle f e t a l tcmprane. Tanibi6n desciende a l e
large de esa p a r d cele e l testfcule, pure es interrumpido tempranamente en su camine, per un par de l&ganbsntos hacia e l ffarie y
e1 dtero y es detenido en l a pelvis,
úkere sirve c m l u g a r de
implantaci6n y nutxicibn de1 nueve embrih ; l a s tubas uterinas e
trompas de Falopie sen e l vehícule para l a cenduccibn del hueve
-
f e r t i i i a a d o e ne- hacia e l dtere. La vagina, una vaina fibrmtuscu-
lar, recibe e l semen preveniente d e l pene, l e transmite a l dtero y
actda came canal del porte, del dtere a l exterior, para e1 recida
-
nacide. A pesar de que l e s @varies y les testículos cempwten un e
rigen c d n , es1 c m l a s estructuras genitaies externas aascuii-
na8 y fedninas, e l dtere, sus trompas y los des tercies superleres
- \5 ~
-
.."
de l a vagina se originan de
un sistema de conductas m y diferentes
-. .
*.
,
-..
a ZWI del
VU&.
El monte da Venus y los labios mayores canparten una estructura cz
d n r a1 t e j l d e fibroso y adiposo. Los labies menawes
d S
PcqUQñOs
y pigmentades confuibuyen a formar e l capuchb del c l f t e r i s hacia
i-
r-
l a parte a n t e r i a y en l a parte p e s t e r i s se unen a l e s labiss me-
1..
yores feriaando una horquilla. La hendidura entre las labios menores
._I
es e l v e s t h ~ l oy recibe
.-.
na y los pqueflos conductss de cuatre glcndulas. Las e r i f i c i o s de
.--
enSf&c&osde l a uretra corta, l a vagi-
l a uretra y l a vagina se abren aquf: nernulmcnte se encuentran colapsados y cerrados. E l himen es una c a y de aucasa que cubre
col^
pleter e incompletamente e l csrificio vaginal en l a mujer ndbil. A
menudo se pueden ebservar restas de él en l a mujer sexualmente activa. E l cuerpo perineal, por debajo de l a piel, en e l l u g a r indicado, es una masa fibremuscular que sirve cem tendbn certral a un
ndmero de ~ s c u l o spurineales tealos los cuales ayudan a estabilizar
y dar sostbn a las estructuras petrineales y a l titer..
CITOLOGIA EXFOLIATIVA
La citelegfa es una ciencia, rama de l a bieiogía que -.mo e
su ncimbre l e indica se encarga del tratado de las células. La
c i t o l e g í a diagnbstica es e l &te de l a interpretaciba de las
cglulas del cuerpb humana, descamadas empentáneamente de las
superficiees epiteliales u obtenidas de los tejidas mediante
diversos procedimientos clínices.
HISTORIA
La descripcibn del contenido celular del calostro hecha per
Denn& en 1838 fue seguida de las de Pouchet en 1847, quién RH
-
ncienb l a aparicibn de células en l a secreci6n vaginal. En 18
43 Walshe encontré tejidos malignes en espute, y Tessenbach en
-
lavados gástrices hacia 1882. En 1853 Donaldson [email protected]$ chi1
ulas malignas en líquidos de derrames de cavidades, mientras
que b a l e , en 1860, las recenecid en e l esputo. S i n embargo,
l a apiicaci6n de l a citología en problemas diagnésticos se ha
basade en e l trabaje de George )apanicolacñr,
-
realizade en col
-
aberaci6n cam Stockard en 1917 que aiencien6 los cambios c í c l i
ces ebservados en l a s cblulas exfeliadas del epi-lie
, dands fundamente para e l
vaginal
-
use subsecuente del mé[email protected] para ce
necer l a f u n c i h hermonal. En 1928, Papanicolaou describi6 e l
papel d e l f r e t i s v a g i n a l dn e l diagn6stice del cáncer, y en19
43 €1 y Traut publicaron su trbajs clásica E l diagnbstice del
cáncer uterine por medio del f r o t i s vaginal.
FUNDWENTO
Come un mbtode diagnibtico l a citología se basa en e l heche
,.,.
,...
.
.
de que l a s cdlulas exfoliads o colectadas de una superficie o
colectadas de una superficie, reflejan muchas de l a s caracter'
í s t i c a s del tejido del cual previenen.
La exfoliacidn espentánea de las células es e l resultado d¿
-
continue crecimienta del epitelio, cuyas células más superfic
i a l e s son centinuamente descamadas y rempiazadas per células
m8s jovenes. Las células descarnadas meden acumularse en l a s
, cemo
diversas cavidades del cuerpa
suceden en l a vagina, de
dende pueden obtenerse para e l examen micrescopico.
Aplicacidn
En l a actualidad
, la
-
tconica citelbgica se aplica pkincipa
lmente en e l diagnhtico oportuno del cdncer, especialmente
en e l trctat genital femenino
.
S i n embargo e l rn6tsdo puede se
r usado para etres prspbsites,
Las ventajas de l a citologfa
exfoliativa para diagnbstics
son l a s siguientes :
-
En l a mayoría de les casos se puede obtener especímenes a
decuades con un rnfnime, de mblestias para e l paciente.
-
En algunes sitios, e l uso de l a citología puede hacer in:
ecesaria l a biobsia.
-
Las técnicas de bbfencih de las muestras usualmente re%
ieren de
POCC
equipa de bajo csste.
- La c i t b l o g h puede revelar anormalidades celulares en eta
pas tempranas del cáncer cuande a6n no sen aparentes a l a
exploración clínica, i o que significa un me#er pronhtico par a l a paciente.
- \a-
- La considerable sensibilidad del método para r e f l e j a r al;
eraciones en e l
hermana1 unida a l bajo costo y a l a f a
esjrddo
cilidad de hacer exhenes seriados hacen de l a CitQlOgf8 e l
método ideal para estudios a largo plazo.
Desafortunadamente
, l a citcslegfa diagnbstica tiene
también
limitacisnes y desventajas pues si se desconocen e ne se teun
an en cuenta, fácilmente se puede abusar del d t s d o con res-
-
tados negativas. E l primer pase es centar con una histeria c l ’
fnica adecuada que propercione dates erientadores relacionado
s con e l padecimiente y l e s antecedentes d e l paciente; el sig
uiente, es l a ebtencibn y f i j a c i b n adecuada de los especfmen-
-
es para estudio; ecacienalmente l a f a l l a en e l diagn6stico ac
urre per l a inaccesibilidad de un kgano para hacer una toma
..
directa, & bién, s i e l material ne se obtiene del s i t i o adecu
ado
Q
-
no se siguen l a s instrucciones para una fijaci6n cerrec
ta, E l manejo y process que se da a l es&cfraen en e l labmat?
rio también es importante ya que se puede a l t e r a r l a calidad
del material y disminuir l a s pesibilidades de emitir un diagn*
fiestice cerrecte. Otro problema es
l a necesidad de hacer un
cuidadese escrutinio [email protected] de l a totalidad de lamuesta
, pues ecasionalmenteun hallazge anormal permite hacer e l
-
dia
gnbstico inmediatamente pero en l a mayoriade los cases es neg
esario examinar tatalmente les especimenes para llegar a un
diagndstico definitivo; esta actividad requiere tiempo y frecuentemente resulta mmltona.
Per 6itimo existen f a l t a s inherentes a l método. Laopinibn #
-\3-
-
f i n a l del observador es una cembinacibn de l a informacih der
ivada de l a histmria c l h i c a del aspecto general del f r o t i s y
-
de l a apariencia individual de las c&lUlaS o de los grupcps ce
lulares; s i n embargo en e l diagn6stico del c h c e r nunca s - p:
ude estar seguro de que un informe negativo no se deba a una
lesi6n innaccesibae o a una toma de muestra inadecuada; t a l
-
informe negativo can ausencia de sospecha clínica, excluye ua
a investigacibn u l t e r i m que podría pejudicar a l paciente.
E l informe positivo tiene tambih sus limitaciones, pues e
1 tipo y l a diferenciacibn del tumor a menudo se estiman cit6
lbgicaniente, pero puedan ser confundidos con tumores que muez
tran varios grades de diferenciacien. E 1 s i t i o geegráfico
del turner piede ubicarse cuando l a toma de l a muestra es dire
cta, no a s í cuando e l material procede de cglulas exfoliadas.
-
Laprecencia o ausencia de invasi6n no se puede asegurar c i t
ologicamente, aunque e l aspecto y distribución de las células
pueden sugerirlas. Por t a l motivo, a un diagnbstico positivo
siempre debe seguir l a confirmación histoibgica antes de que
e l paciente se someta a una cirugia mayor o a una radioterapia
Li
c
...
."
.
,I
c
..1
.
L
i
-
c
L"
-.
Li
r
._
I
.I
-22-
y wjipa urinaria
t
t
carcinoma in *it" (fase o1
Adenoacantorna
-23-
I
p"
-
P
P
L
P
L
F
. ..
r-
. ..
I -
-24
-
L.
CARCIBOMA
EN EL CUELLO UTERINO.
EL carciroma d e l c u e l l o d e l ú t e r o e s quizá i a m8s impox
t a n t e de l a s afeccjiones
r
i
i
L
i
-
que e l ginecdlogo t i e n e que enfren
t a r s e . Es c i e r t o , no sdlo l a frecuanciii. con que s e encuentra2
-
l a s fonnas preinvasora,5e in.vasora de l a enfermedad, sino tam
b i 6 n porqke s e sabe mucho sobre l a h i s t o r i a n a t u r a l de é s t e
c a n c e r que puede s e r v i r como modelo para e i diagfidstico y e l
tratamineto tempranos de o t r o s cánceres.
Se sabe más sohre l a h i s t o r i a n a t u r a l d e l carcinoma epi
dermoide d e l c u e l l o de c u a l q u i e r o t r o c h c e r . EL c h c e r s i n tom4tico i n v a s o r puede d e s a r r o l l a r s e a p a r t i r del e p i t e l i o normal por un proceso l e n t o que i m p l i c a muchos años. Q u i z á s
transcurran io a ñ o s o más desde que aparece e i carcinoma
--
i n t r a e p i t e l i a l , hasta que invade e l t e j i d o c o n e c t i v o cercana,
Diversos a u t o r e s consideran que algunas a l t e r a c i o n e s microsa
cdpicas en e l e p i t e l i o , i n c l u s o a n t e s que b r o t e corn 'letamen-
t e e l carcinoma e p i t e l i r t l , son de n a t u r a l e z a precancerosa.Si
e s t o s sisas a t f p i c o s s e tornan como e l primer grado hacia l n
-
m a l i g n i d a d , t a l vez transcurran 1 5 años o m46 desde e l comien
zo de l a s primeras a l t e r a c i o n e s hasta l a muerte p o r c á n c e r
-
epidermoide d e l c u e l l o , s i n tratamiento.
FACTOREd ETIOLOGICOS.
En s e n t i d o b i o l ó g i c o s e desconoce, como en todo c á n c e r ,
l a c a u s a d e l carcinoma d e l c u e l l o . No o b s t a n t e , algunas c i r n
c u n s t a n c i a s e s t & t a n íntimamente r e l a c i o n a d a s con 6 1 que
-
pueden c o n s i d e r a r s e como factores e t i o i ó g i c o s .
Se han hecho muchos es:Euerzos para e s t u d i a r los diver-so8 f a c t o r e s económicos y s o c i o i 6 . ~ i c o sque pueden c o n t r i b u i r
-2s-
. .-
a ’ l a f r e c u e n c i a v a r i a b l e d e l c&ncer c e r v i c a l . Todos c o i n c i d a
en que e l cáncer c e r v i c a l es r e l a t i v a m e n t e r a r o en l a mujer
judfa; Wynder y colaboradores indican que su f r e c u e n c i a es aproximadanente l a octava p a r t e de l a que s e observa en grupos s i m i l a r e s de g e n i t a l e s . Rothman y colaboradoses, en una
-
r e v i s i ó n de e s t e tema, basada en e l riqulsimo material. d e l
h o s p i t a l Mount S i n a i , de Nueva York, encuentran que e l c&n-c e r d e l c u e l l o u t e r i n o es nuesre veces más f r e c u e n t e en mujer e s na judías que en l a s judfas. %
I i o que s e r e f i e r e al c a _.
r
cinoma endometrial no s e observan d i f e r e n c i a s de f r e c u e n c i a
en ambos grupos.
La b a j a f r e c u e n a i a del. cáncer c e r v i c a l u t e r i n o en mujer e s j u d í a s ha hecho sospechar que e l c o i t o con un varón s i n
rL
c
c i r c u n c i s i ó n puede actuaren a l m a forma como i n f l u e n c i a cau
s a l , q u i z á a base de poca h i g i e n e d e l pene, y se sospecha de
i s importancia que pudiera t e n e r e l esmegma, según señala
Fischer.
P-
Se ha observado recientemente e l d e s a r r o l l o
del
.
--
c ‘ n c e r c e r v i c a l en algunas cepas de ratones sometidos a e s t &
muiación durante 1 4 meses p o r l o menos con esmegma humano.
Wynder y colaboradores han publicado una amplia r e v i s i ó n so-
-
b r e l a importancia de d i v e r s o s f a c t o r e s extraños y han l l e g a
do a i a conclusión de que ]-a c a r c i n o g é n e s i s puede considerar
s e solamente como resulatdo de v a r i o s estimulos endógenos
y
exógenos. Hay que t e n e r siempre presentes f a c t o r e s como pro-
-
miscuidad sexual, C i r c u n c i s i ó n incompleta y f a l t a de adheren
L
c
c i a a l a d o c t r i n a mosaica, que impide e l contacto sexual. dur a h t e una semana después de l a mentsruacibn. S i n embargu Jon e s y colaboradores no han observado d i f e r e n c i a s n o t a b l e s en
e l c4ncer c e r v i c a l , sea cual sea e l estado de c i r c u n c i s i ó n
-
d e l compaiiero. ih un informe muy importante de l a universidad
-26-
de Kadrás, Rewell h a comprobado igdal f r e c u e n c i a de c á n c e r
-
c e r v i c a l en musulmanes ( v a r o n e s con c i r c u n c i s i ó n ) e hind6es
( s i n circluicisibn)
. Lilienfeld
y Graham estudiaron rigurosa-
-
mente l a supuesta r e l a c i ó n e n t r e c i r c u n c i s i d n y c á n c e r c e r v i
c a l demostrando que h a b í a un 34.4 $ de desacuerdo e n t r e l a
información de 2 1 3 varones a c e r c a de &u c i r c w c i s i ó n y l a
-
verdadera s i t u a c i ó n d e l prepucio según demostraba e l examen.
En resumen, solo caben suposiciones a c e r c a de l a supuesta r e
l a c i ó n e ti o16 gi c a.
Por o t r a p a r t e , durante l a &tima década una media doce
n a de e s t u d i o s importantes han examinado l a r e l a c i ó n e n t r e e i c o i t o o e i matrimonio y e i c á n c e r c e r v i c a l .
Todos los es-
-
t u d i o s concuerdan en que e l r i e s g o d e l c á n c e r c e r v i c a l sumen
t a con e l matrimonio temprano o e l primer c o i t o a edad teme-
prana
. Rotkin
ha estudiado e l problema nuevamente y s u s re7
s u i t a d o s parecen i n d i c a r que de los 1 5 a l o s 20 a ñ o s e s e l
período s u s c e p t i b l e en que l a primera r e l a c i ó n s e x u a l y
s i g u i e n t e s predisponen
-
las
u l t e r i o r m e n t e al cáncer. El tiempo
-
medio d e l periodo de l a t e n c i a e n t r e e l primer c o i t o y e l des
cubrimiento de c á n c e r f u e de unos 30 arios. Ponen de r e l i e v e
l o s f a c t o r e s sedalados p o r d i v e r s o s a u t o r e s , como pobreza, p o r ejemplo; c o i t o , casami!ento y embarazos tempranos; Rewell
tambdn s e n a l a que en l a I n d i a l o más f r e c u e n t e e s que l a s
-
mujeres s e casen e n t r e los 1 4 y 1 5 años, y que en e l l a s e l
-
c á n c e r s e descubre unos d i e z años a n t e s que en l a edad acostumbrada.
Durante anos s e h a t e n i d o l a impresión de que l a s muje-r e s con h i j o s t i e n e n mucha mayor t e n d e n c i a a s u f r i r c á n c e r
c e r v i c a l que l a s o l t e r a . Indudable:nerite, l o s datos más
impre
-
s i o n a n t e s a i r e s p e c t o fueron l o s proporcionados por Gagnon,
-. 21-
-
quien, en una r e v i s i ó n de l a s h i s t o r i a s y c e r t i f i c a d o s de
defunciiín de no menos de 13 O00 monjas canadienses, no logró
-
descubrir un s o l o caso de c h c e r c e r v i c a l . P o r o t r a parte,
un estudio más r e c i e n t e de T o m e sobre un n h e r o s i m i l a r
de
-
mujeres c é l i b e s r e v e l 6 s e i s casos de carcinoma d e l c u e l l o ,
aunque e s t e dato representa, f r e c u e n c i a extraordinariamente
baja. Considerando nuestra e x p e r i e n c i a con algunos hospita--
l e s c a t b i i c o s , y l a de v a r i o s autores de f i l i a c i o n e s r e i i g i o
sas s i m i l a r e s , no descubrimos uno s o l o di: t a l e s casos hasta
que finaimente o c u r r i ó uno.
Sin embargo, l a monja que l o s u f r í a provenfa de c i e r t a
casa que aceptaba muchachas jóvenes, y e l i n t e r r o g a t o r i o dir e c t o efectuado de e s t a mujer jóven, que colaboraba bien,reve16 que en su v i d a pasada habfa tenido contactos sexuales.
P o r l o tanto, es muy probable que e l tener h i j o s p e r s e
no e s acontecimiento causal importante, sino más b i e n l a ex-
-
p o s i c i ó n sexual. "1 e s t e sentido, e l carcinoma epidermoide
d e l c u e l l o puede considerarse una enfermedad venérea.
Estudios epidemioi6gicos han culpado d i v e r s a s o t r a s c i r
cunstancias que s e acompañan de carcinomas c e r v i c a l e s de poco o mucho p e l i g r o . La demostración e p i d e n i o i ó g i c a e s esen-c i a 1 y básicamente circunrbt;ancial. Por l o tanto, debe u t i l i ?
zarse e l sentido c o m b al i . n t e r p r e t a r l a i n f o m a c i b n . Asl l a
i n f o m a c i b n d e l cuadro s i g u i e n t e r e s u l t a muy i n t e r e s a n t e , pr o e s una cosa muy d i f e r a n t e a d m i t i r una o más de e s t a s ca-r a c t e r l s t i c a s como causa de l a enfernedad. Se t r a t a e s t e pun
t o en un t r a b a j o sobre r e l a c i ó n e n t r e tabaco y carcinoma de
c u e l l o . Sus conclusiones son ineteresantesr
I*
Los resultados
indican que habfa una a s o c i a c ~ ó ns i g n i f i c a t i v a e n t r e e l em-p l e o d e l tabaco en rapé, el. tabaco mascado, o ambos, y e l
--.
cánc(?r do los g e n i t a e s femeninos, y que los c i g a r r i l l o s n o -
-
guardaban r e i a c i d n con l a enfermedad estudiada’! Hecho i n t e r -
s a n t e , s e consideraba seriamente que e l rapé y e l mascar tsc
baco pudieran t e n e r a l g u n a r e l a c i ó n con l a causa.
¿Puede
c r e e r s e que e l tomar rapé y nascar tabaco son a r e n t e s cau
saLes?
-
Tampoco cabe admitir que e l a c u d i r a los s e r v i c i o s
r e l i g p o s o s PO s i mismo e v i t e , como s i dijeramos, l a enfermedad, segun s u b i e r a deducirse de l a información del cuadro.
Grupos de poco p e l i g r o
Mujeres musulmanas (kmet y colaboradores, Wynder y colaborad o r e s , 1954)
Mujeres amish
Mujeres j u d i a s (Haenme1 y H i l l h o u s e , 1959;kenaway,1948)
Mujeres a d v e n t i s t a s del s é p t i m o d í a (Lemon y colaboradores,
1954; Wynder y colaboradores, 1 9 5 9 )
Mujeres inmigrandee i r l a n d e s a s ( H a e n s ~ e l , 1961)
Mujeres inmigrantes italianas (Haenseel, 1961)
Mujeres p r o $ e s t a n t e s y católicas que acuden regularmente a
los s e r v i c i o s r e l i g i o s o s (Naguib y colaboradores, 1966)
Mujeres de s i t u a c i o n economica alta (Dorn y C u t i e r , 1959)
Mujeres de medio rural. (Levin y c o l a b o r a d o r e s , 1 9 6 0 )
Grupos de gran p e l i g r o
Mu j e r e a . de Puerto Rico (Haenseel y Hillhouse 1959)
Inmigrantes mexicanas (Haenszel)
Mu j e r e s negras
(Haenszel y Hillhouse )
Reclusas de una p r i s i ó n femenina ( P e r e i r a )
-29-
Prostitutas (Rojerl)
P a c i e n t e s de una c l í n i c a de enfremedades venéreas (Greene)
kíujeres de n i v e l económico b a j o (Dorn y C u t l e r )
Mujerees de medio urbano.
Los d a t o s epidemioldgicos
, como
s e ha dicho, implican
una e x p o s i c i b n s e x u a l temprana, especiaimente con compderos
-
d i v e r s o s , como causa e t i o l d g l c a importante. E s t o , naturalmen
t e ha hecho sospechar que pudiera i n t e r v e n i r un a g e n t e i n f e c
c i o s o . En
al&
-
momento s e ha sospechado semen, esmegma, tri
comonas, CUamydias y condilomas. Coppleson y colaboradores
han presentado pruebas que s u g i e r e n que e l agente e t i o l d g i c o
e s e l DNA d e l espermatozoide.
Al p a r e c e r , e l v i r u s d e l Herpes simple de t i p o 2 s e r e l a c i o n a claramente con e l c á n c e r c e r v i c a l , pero & no s e ha
comprobado s i e s una r e l a c i ó n de causa y e f e c t o más que l a
-
i n c i d e n t d de un agente i n f e c c i o s o comilui en mujeres sexual-mente a c t i v a s .
TIPOS ANATOMOPAMLO GICOS.
Hay dos t i p o & p r i n c i p a l e s de c h c e r c e r v i c a l , que nacen
respectivamente de l a s dos c l a s e s de e p i t e l i o
-
que recubren
e i c u e l l o , Se r e c o r d a r á que e l e p i t e l i o que t a p i z a l a superf i c i e e x t e r n a o v a g i n a l de l a porcidn vagiuiai d e l c u e l l o pey
-
t e n e c e a i a variedad pavimentosa e s t r a t i f i c a d a , que s e c o n t f
nua con e i e p i t e l i o pavimentoso e s t r a t i f i c a d o de l a vagina.
c e-r
De 61 nace e l carcinoma de cdlulas pavímentosas o epidermoi~:'
de. P o r o t r a p a r t e
, el
e p i t e l i o c i i i n d r i c o d e l conducto
v i c a l o r i g m a a l adenocarcinoma d e l c u e l l o . L a s l e s i o n e s de
c é l u l a s e s c a n o s a s corresponden a un 90
5
d e l t o t a l de
los
-
c á n c e r e s c e r v i c a l e s , aunque en l o s ú l t i m o s &os
los adenocax
cinomas c o n s t i t u y e n una proporción cada vez mayor. Ambas t i e
nen malignidad mayor que e l carcinoma d e l cuerpo u t e r i n o . Es
d i f i c i l d e c i d i r c u a l de l o s dos t i p o s de carcinoma c e r v i c a
-
e s más maligno; no o b s t a n t e , e l adenocarcinoma, p o r o r i s n a r
s e con f r e c u e n c i a i n s i d i o s m e n t e en e i i n e t r i o r d e l conducto
comporta q u i z á s un pron6sti.co mas desfavorable.
Se h a comprobado que a l g u n o s tumores t i e n e n componentev
c e l u l a r e s de ambos t i p o s , adenomatoso y palno, y s e han c l s ,
-
s i f i c a d o de carcinoma adenoescamoso o adenoepidermoide d e l
c u e i i o . T a l e s tumores pareoen t e n e r un pronóstico m4s s e r i o
que l o s tumores puramente de c é i u i a s p l a n a s o e l adenocarcinoma.
Los sarcomas d e l c u e l l o u t e r i n o s e han encontzsdo muy
-
raras v e c e s y s e han observado unas c u m t o s melanomas prima-
r i o s . También es r a r o e l sarcoma b o t r i o i d e o , que en ocasio-n e s puede a f e c t a r e l c u e l l o p e r o , como se ha comentado, en
la mayoria de los c a s o s e l d e a a r r o i i o primario o c u r r e en
-
la
vagina.
ANATOHOPATOLOGIA MACROSCOPICA.
Etapa Precifniaa.
No hay a l t e r a c i o n e s a n a t o m o p a t o l b ~ c a scaracterísticas -
-
que ayuden a i d e n t i f i c a r un carcinoma i n t r a e p i í z e l i a l d e l cue
110 p o r e x p i o r a c ~ 6 nmacroscbpica. De hecho, algunos c a s o s d e
c e r v i c i t i s a r ó n i c a c r e a n un aspecto m4s anormal que un c u e l l o
con carcinoma i n t r a e p i b e l i a l .
-3d-
._._I_
.
---
Primeros ea t a d i o s
---
.
En sus f a s e s m4s tempranas, e l c á g c e r c e r v i c a i s e pre*r-e
-
senta i a más de l a s veces, como una pequeña l e s i ó n que a s i e n
t a a n i v e l d e l o r i f i c i o externo o próximo a Qi, e s d e c i r , en
l a unión de l o s dos t i p o s de e p i t e l i o c e r v i c a l . Aparece como
una pequeña zona endurecida y granulosa, que a l a palpación
s e p e r c i b e como un l i g e r o s a l i e n t e en r e l a c i ó n con l a superf i c i e vecina.
Ai examen con. e l espécuio, l a s u p e r f i c i e en
--
c u e s t i ó n s e presenta granulosa y ligeramente prominente, san
grando a l menor contacto.
-
A veces 1 a : m u p e r f i c i e puede h a l l a r
s e c u b i e r t a en e s t a temprana f a s e p o r excreencias finamente
-
p a p i l a r e s . Las porciones vecinas d e l cue1l.o pueden s e r norma
l e s , pero es m4s frecuente que sean a s i e n t o de una a f e c c i ó n
c r ó n i c a infiamato r i a .
De hecho como en e l carc:inoma i n t r a e p i t e l i a l , s u e l e res u l t a r imposible d i s t i n g u i r p o r simple inspección e l cáncer
i n v a s o r i n i c i a l de lesionesi benignas como erosiones y ever-s i ones.
Estadio moderaelamente avanzado.
Desde su a s i e n t o primi.tivb, e l cáncer s e propaga hasta
que abarca l a mayor p a r t e o l a t o t a l i d a d de uno de l o s la-b i o s d e l c u e l l o , o porciones de ambos. A medida que avanza,
muestra una de sus dos c a r a c t e r í s t i c a s p r i n c i p a l e s . Puede
-
p r i v a r l a tendencia p a p i l a r , creciendo l a neo formación sobre
l a s u p e r f i c i e principalmente; de aquf que l a l e s i ó n a f e c t e
l a forma de c o l i f l o r , constituyendo l a variedad e v e r t i d a
o
extendién
-
e x o f i t i ca. Por o t r a p a r t e , q u i z 4 no se produzca formacidn
s u p e r f i c i a l , o é s t a tenga I.ugar en perueña medida,
dose l a l e s i ó n en l o s t e j i d o s c e r v i c a l e s y produciendo una
induración muy firme, a veoes p é t r e a , aunque practicamente
-32-
._
F-
siempre hay alguna u i c e r a c i ó n . Este t i p o se denomina invert i d o o endofftico.
La i n f i l t r a c i ó n puede a f e c t a r desde e l p r i n c i p i o e l f o n
-
-
do d e l saco v a g i n a l vecino, y l o s ligamentos anchos pueden
p a r t i c i p a r en e l l a .
Estadio Avanzado
sus a t i n a s f a s e s , e l d e s a r r o l l a d e l proceso cancer2
so provoca una destrucción cada v e z mayor d e l c u e l l o , al que
reemplaza entonces una cavidad excamada y ulcerada, de parédes ásperas y f r i a b l e s ; de aquf que de no hacerse un examen
extremadmente d e l i c a d o , se produzca una hemorragia franca.
-
Las paredes v a g i n a i e s prdximas a i c u e l l o son r e s i s t e n t e s y
s e h a l l a n induradas a consecuencia de l a i n f i l t r a c i ó n canceL
r
L
i-
rosas. EL ligamento ancho :muestra también extensa i n f i l t r a +
c i d n no sólo como resultado de i a e.:tensión d e l cáncer, s i n o
como consecuencia de i n f i l t r a c i ó n i n f i e m a t o r i a , secundaria a
l a l e s i ó n ulcerosa séptica del cuello.
i
-
rL..
P
-
En los casos en que l a p r o i i f e r a c i d n e s e x o f i t i c a , s e
-
forma una masa de aspecto de c o l i f l o r que puede ocupar c a s i
toda l a vagina. En e s t a variedad l a i n f i i t r a c i ó n macroscópica de l o s t e j i d o s vecinos pueden s e r asombrosamente escasa,.
-
atin en e l caso de qhe l a t;umoración c e r v i c a l sea de gran t a
maña.
si progreso u l t e r i o r de i a enfermedad s e c a r a c t e r i z a
porque va invadiendo y destruyendo cada v e z nuevas estructuw
L
ras, i n f i l t r a n d o en grado c r e c i e n t e l o s ligamentos ancfios,
-
con bloqueo de un u r é t e r o de l o s dos, y afectando con fre-cuencia v e j i g a o r e c t o . A menudo s e producen t r a y e c t o s f i s t u
-
l o s o s e n t r e cua1esquYk-a de e s t o s órganos y l a vagina.
- 33-
c--
,afermedad metastática.
c
A medida que avanza l a enfremedad, hay m e t á s t a s i s en l o s
g a n e i o s i i n f á t i c o s p é i v i c o s y finalmente en los g a n g i i o s a6r
c
I
t i c o s y distantes.
En l a enfremedad avanzada, también aparecen m e t 4 s t a s i s
-
en pulmones, hígado, t e j i d o óseo y c e r e b r o .
Con e l empleo c r e c i a n t e de l a q u i m i o t e r a p i a y técnicaspara r a d i a c i ó n nuevas, s e ban observado cada vez con mayor P
L
f r e c u e n c i a m e t á s t a s i s en s i t i o s "raros", por ejemplo p i e l , p r a r e n a i e s y meninges. L a causa más común de l a muerte en p a
c i e n t e s con c 4 n c e r cervical. e s l a o b s t r u c c i 6 n u r e t r a 1 por e l
n
c á n c e r p é i v i c o en c r e c i m i e n t o con l a c o n s i g u i e n t e i n s u f i c i e c i a r e n d . También e s común que haya s e p s i s , i n s u f i c i e n c i a
-
r e s p i r a t o r i a , hemorragia e i n s u f i c i e n c i . a h e p á t i c a .
Desde h a c e a ñ o s s e sabe q.ue s e puede d e s c u b r i r c é l u l a s
tumorales en e l t o r r e n t e v a s c u i a r , especialmente después de
c u a l q u i e r traumatismo que s u f r a e l tumor, % todo c a s o , e l descubrimiento dd c é l u l a s malignas en l a sangre no implica necesariamente que s e producirán m e t 4 s t a s i s c l í n i c a s . Por i o
t a n t o , t i e n e poco v a l o r pr"16stico
en p a c i e n t e s que r e c i b e n
t r a t a m i e n t o o s i n 61.
Tiene que a d m i t i r s e c i e r t o grado de r e s i s t e n c i a o s e n s i
-
b i l i d a d d e l huésped p a r a que l a s c é l u l a s c i r c u l a n t e s puedan
c r e c e r y formar un foco m e t a s t á t i c o ; de todas maneras, pare_...
c e que una b a c t e r i e m i a de c É l u l a s tumorales t r a n s i t o r i a no
e s rara en muchas enfermedades n e o p i á s i c a s .
e
-34-
-
ANATOMIA PATOLOGICA NICROSCOPICA DEL CARCINOMA EPIDERMOIDE.
Pato genia
Como antes dijimos, se sabe ms'
sobre l a e v o i u c i ó n d e l -
carcinoma epidermoide d e l c u e l l o que de l a de c u a l q u i e r o t r o
cáncer. A pesar de e l l o , aún es motivo de c o n t r o v e r s i a l a se
cuencia exacta de l a transformación neopiásioa en c u e l l o .
Uno de l o s h a l l a z g o s importantes en e l carcinoma micros
-
tópico i n i c i a l e s 4a constancia de su o r i g e n en l a unión esc m o c i i í n d r i ca.
J b l a s prepfiberes, l a porción v a g i n a l d , . l e x o c e r v i x es-
t a compuesta de estroma c e r v i c a l s i n glfmdulas, c u b i e r t a p o r
e p i t e l i o escamoso n a t i v o o maduro.
i%iendocervix, que anatómicamente corresponde a i condug
t o que se encuentra por a r r i b a d e l o r i S i c i o c e r v i c a l externo
-
y abajo d e l i n t e r n o , s e encuentra c u b i e r t o de e p i t e l i o c i l i r i
d r i c o , q.ue no s610 recubre l a s u p e r f i c i e d e l conducto sino también l a s g l h d u l a s endocervicales, que s e encuentran en
-
e l estroma.
Antes de l a menarquia, l a unión e n t r e e l e p i t e l i o esca-
moso d e l e x o c e r v i x y e l e p i t e l i o c i i i n d r i c o d e l endocervis s u e l e s e r p r e c i s ? . Sin embargo con e l c r e c m i e n t o d e l c u e l l o
durante ia menarquia, y en e s p e c i a l con l a eversidn f i s i o i ó -
-
g i c a d e l endocervix durante e l embarazo, e l e p i t e l i o c i l f n d r i
co y l a s gl.&ndulas d e l endocervix s a l e n hacia e l e x o c e r v i x
-
anatómico.
Este e p i t e l i o c i l f n d r l c o que s e encuentra ahora en e l
-
e x o c e r v i x e s t á expuesto al medio ambiente vaginal. Por in-f l u e n c i a de é s t e , debido probablemnte
en
-,
p a r t e al PH,
e l e p i t e l i o c i l f n d r i c o e x o c e r v i c a l e s substituido graduaimen
t e p o r e p i t e l i o escamoso.
-3s -
La zona anatómica d e l c u e l l o en l a que ocurre s e drnomina zcr.
na de transformacidn y e s a h í donde p o r primera v e z se obser
van l a s anormalidades e p i t e i i a l e s c i i f ndricas, consideradas
precursoras d e l carcinoma invasor.
A pesar de l o s años de intenso estudio p o r muchos i n v es
t i g a d o r e s , aún s e d i s c u t e l a forma exacta en que ocurre e s t e
-
proceso. Hoy en d i a , l a t e o r í a más aceptada es e l concepto
de l a metaplasia.
Todavía s u c i t a c o n t r o v e r s i a p r e c i s a r cu4l e s l a c é l u l a
exacta que p a r t i c i p a en e l proceso de metaplasia, pero eo--
ppleson y R e i d señalan que e s t a c é l u l a c i l f n d r i c a en s í , t a l
-
v e z aunada a l a s c é l u l a s d e l estroma subyacente que son tram
formadas directamente en ” c é l u l a s b a s a l e s neoescamosas“
.
Por
debajo de gruposde e s t a s c é l u l a s se forman una nueva membrara
b a s a l y de e s t a manera “ e l e p i t e l i o escamoso metap16sicor‘ ha
s u s t i t u i d o a l e p i t e l i o c i l f n d r i c o . Aún no se sabe qon seguridad s i l a s p r o p i a s c é l u l a s c i l í n d r i c a s s e d i f e r e n c i a n nuevamente en c é l u l a s e p i t e i i a l e s escanosas o s i son s u s t i t u i d a s
-
p o r una p r b l i f e r a c i ó n d s una cE1ul.a. i n d i f e r e n c i a d a d e l estral
ma endocervical denominada
s u b c i l f ndricas”
.
‘I
-
c é l u l a s de r e s e r v a o a é i u l a s
El proceso de l a metaplasia es especialmente a c t i v o des
-
pués de iü e v e r s i ó n d e l e p i t e l i o c i i f n d r i c o que ocurre d u r m
t e l a menarquh y e l primer embarazo. Es un proceso f i s i o i ó g i c o normal cuyo resultado puede s e r nuevo e p i t e l i o escamoso
b i e n d i f e r e n c i a d o , que cubre e l exocervix.
Los f a c t o r e s m a l c a r a c t e r i zados denominados en l a “ r e s
puesta d e l huéspedr1 pueden e v i t a r e l d e s a r r o l l o c l í n i c o de
una n e o g l a s i a c e r v i c a l , a pesar de l a combinaci6n de a l t o
r i e s g o que piantean l a s circunstancias.
-36.-
--
--
S i n embargo e s t a t e o r í a s e :ajusta muy b i e n a l a s observaciones epidemioiógicas de que e i c o i t o a edad temprana (durante
-
e i período posmenarquía de metaplasia a c t i v a ) y m ú l t i p l e s
-
compañeros sexuales ( e x p o s i c i ó n a un agente carcinógeno tram
m i s i b l e ) son factore.:. de a l t o r i e s g o de cáncer c e r v i c a l .
-
No obstante, cualquiera que sea l a c é l u l a o e l o r i g e n ,
pruebas f i r m e s sugieren que l a n e o p l a s i a d e l c u e l l o s e o r i gi
na de una c é l u l a a i s l a d a más que de d l t i p l e s s i t i o s . Park y
Jones al e s t u d i a r mujeres que fueron h e t e r o c i g o t o s para l a
forma A y B deshidrogenasa gLucosa-6-fosfat0,
.-
encontraron qve
e l t e j i d o canceroso t e n f a ~ ~ 3 1forma
0
A o B. Ya que l o s genes
de e s t a enzima s e encuentran en e l cromosoma-X,el resiultadopuede e x p l i c a r s e p o r e l d e s a r r o l l o de una masa tumoral a par
t i r de una c é l u l a simple que f u e a c t i v a d a p o r u:ia u o t r a
f or
ma de e s t a enzima.
rL
h v a s i d n temprana d e l estroma
P
L
Es d i f i c i l e l d i a g n ó s t i c o d e l o s grados tempranos de in-
-
v a s i ó n d e l estroma, ya que posiblemente l a s glándulas sean
L
c
L
s u s t i t u i d a s d e l todo p o r b r o t e s d e l tumor s i n que haya una i n v a s i ó n verdadera. S i i s e s t r u c t u r a de l a glándula s e con-
-
s e r v a b i e n y e s l i s a , no hqy invasión. S i e l contorno es gri
sáaeo y confuso, probablemite l a i n v a s i ó n por p a r t e d e l es-troma. Casi siempre Bas c é l u l a s escamosas invasoras parecen
más maduras con citoplasma e o s i n o f f i i c o aumentado cuando
comparan con l a e c é l u l a s i n d i f e r e n c i a d a s de t i p o b a s a l de
se
--
carcinoma i n s i t u . Los c o r t e s tangenciales, l a i n v a s i ó n de l a s glándulas, y una mala preparación d e l t e j i d o suelen d i f-i
c g i t a r en extremo e s t e diagnóstico.
L
L
c
-3%-
Carcinoma i n v a s o r
i g u a l que sucede en caso de que e l carcinoma a s i e n t e
-
en o t r o punto, e l d i a g n ó s t i c o de e s t a a f e c c i ó n a n i v e l d e l
c u e l l o s e basa en dos c a r a c Q e r f s t i c a s p r i n c i p a l e s ; 1)una exp o s i c i ó n o a r q u i t e c t E r a anormal, 2) anomalías en l a s c é l u l a s
que l o constituyen.
Eh tanto que en l a s u p e r f i c i e e p i t e i i a i normal l a s c é i u
destru
-
l a s e p i t e i i a l e s s e encuentran claramente separadas d e l e s t r o
ma p o r l a membrana basal, en e l cheer e s t a 6 l t i m a se
ye; de aqul que e l e p i t e l i o s e introduzca en e l estroma, al
i
i
p r i n c i p i o a manera de pequeños botones, más t a r d e en forma u
de l a r g a s colwnnas que crecen profundamente en qquel, seme--
-
jando l a forma en que l a s r a f c e s de un á r b o l s e introducen
en l a t i e r r a .
En toda8 l a s f a s e s excepto l a s ms'
-
tempranas
puede hacerse un d i a g n ó s t i c o f i m e con un pequeño aumento, fu
puesto que é s t e pone da m a n i f i e s t o i a desordenada y anormai
i n v a s i ó i i d e l estroma p o r el. e p i t e l i o . Es ese c a r á c t e r inva-s o r , junto con l a propagación de l a s c e i u i a s a t r a v é s de l o s
i i n f á t i c o s , e l causante de l a s cualidades t í p i c a s que t r a d i 3
cionalrnente s e asocian a l a malignidad, a saber:la i n f i l t r a ;
c i ó n l o c a l , l a s metástasis y l a s r e c i d i v a s
--
después de l a
e x t i r p a c i ó n incompleta.
-
En tanto que e l e p i t e l i o normal e s t a c o n s t i t u i d o por c é
i u i a s d i f e r e n c i a d a s de t i p o a d d t o , l a s d e l cáncer muestran
un grado v a r i a b l e de inmadurez. Hist916gicamente a s t o s e pon e de r e l i e v e p o r l a b s p a r i d a d d e l tamaño c e l u l a r y de
los
núcleos, hipercromatoais, m i t o s i s normales o p a t o i d g i c a s y
-
c a r r i o r r e x i s.
-
Aunque s e han propuedto a t r a v é s de l o s años v a r i a s c l a
s i f i c a c i ó n e s d e l carcinoma c e r v i c a i de c é l u l a s escamosas y
-
actualmente i s mayoria de l o s anatomopatóiogos d i v i d e n e l
C-
cinoma e p i d e m o i d e d e l c u e l l o en t r e s t i p o s r 1) célula gran*
-
d e , q u e r a t i n i z a ü a ; 2) c é l u l a grande, no q u e r a t i n i z a d a y 3)
c 6 l u l a pequeña. Los c á n c e r e s q u e r a t i n i z a d o s e s t á n compuestos
-
de c é l u l a s grandes con n o t a b l e pleomorfismo, r e l a t i v a m e n t e
p o c a s m i t o s i s , y fonnación de p e r l a s escamosas. Los tumores
no q u e r a t i n i z a n t e s de c 6 l u l . a s grandes se c a r a c t e r i z a n p o r c-
variedad -
i u i a s con pleomorfismo moderado y m i t o s i s f r e c u e n t e s , y gran
des n ú c l e o s que s u e l e n c o n t e n e r macronuciéoios. La
d e l carcinoma c e r v i c a l de c é l u l a s pequeñas e s t 4 compuesto de
una lámina monótona de c é l u l a s pequeñas
indiferenciadas s i n
p e r l a s de q u e r a t i n i n a . Las m i t o s i s son f r e c u e n t e s .
En un e s t u d i o l a s enfermas con tumores no q u e r a t i n i z a d w
de c é l u l a s grandes evolucionaron m e j o r que los tumores que-
-
t i n i z a n t e s y e l p e o r í n d i c e de s u p e r v i v e n c i a s e obervd en mu
j e r e s de c h c e r de c é l u l a s pequeñas.Swan y Roddick
pudieron
c o n f i r m a r e s t o s h a l l a z g o s en p a c i e n t e s tratadas con r a d i a c i ó n
pero no en i a s que s e hizo cirugía en i a s que no Encontraron
d i f e r e n c i a en l a s u p e r v i v e n c i a e n t r e los t r e s t i p o s histológ i c o s . Gunderson y colaboradores no encontraron d i f e r e n c i a s
i
P
L
de s u p e r v i v e n c i a en 1 7 5 p a c i e n t e s tratadas por r a d i a c i ó n y
o t r o s han señalado ambos t i p o s de r e s u l t a d o s . En consecuen--
-
c i a , cabe c o n c l u i r que atin e s i n c i e r t a l a importancia pronós
tics de e s t a c l a s i f i c a c i b n .
L a c l a s i f i c a c i ó n de B r o d e r s , que c o n s i s t e en d e s i g n a r
d i v e r s o s grados numérbcos según l a d i f e r e n c i a c i ó n c e l u l a r ,
también s e u s a con f r e c u e n c i a . Los tumores de p a d o I son
.-
-
l o s m e j o r d i f e r e n c i a d o s y los d e l grado IV los más i n d i f e r e n
ciadoa. Nuevamente, l a im&lortancia p r o n ó s t i c a de e s t a clasif i c a c i ó n e s motivo de c o n t r o v e r s i a . Desde e s t e punto de v i s t a p r o n ó s t i c o e s mucho más importante l a etapa de de l a en-
fermedad que d e l t i p o c e l u l a r o l a d i f e r e n c i a c i b n .
-39-
ADENOCARCD.TOMA Y CARCINOMA ADENOESCAMOSO D'EL CUELLO
EI adenocarcinoma c e r v i c a l es menos c o m h que l a varie-
-
dad de c é l u l a s escamosas. Segihestudios r e c i e n t e s , correspon
de aproximadamente ai io$ d e l t o t a l de l o s cánceres cervica-
-
l e s . Aunque p o r l o general s e i n i c i a dentro d e l conducto c e r
r
v i c a i , l a l e s i ó n primaria puede asentar en e l o r i f i c i o e x t eno o próxima a éi, formando en etapas avanzadas W a ve,Teta--
-
c i ó n de gran tamaño a n i v e l de l a s u p e r f i c i e v a g i n a l d e l cue
110. A p e s a r de l o b c h o , e s más t í p i c o que e l proceso afec-
-
t e en proporción c r e c i e n t e e l c u e l l o y los t e j i d o s v e c i n o s
s i n p r o d u c i r l e s i o n e s e x t e m a s en l a s u p e r f i c i e v a g i n a l .
-
~ i c r o s c d p i c a m e n t es e c'a a c I ; e r i z a p o r l a d i s p o s i c i ó n @an
g lh
d u l a r a b f p i c a tan d i s t i n t i v a d e l adenocarcinoma, en n o t a b l e
c o n t r a s t e con e l aspecto y d i s t r i b u c i ó n ordenada de l a s
duias c e r v i c a l e s n o w a i e s .
En algunos casos e l ppartaniento
de l o normal es mod rado; e.n o t r o s l a d i s p o s i c i ó n @andular
anoi-mai e s i n t r i n c a d a y a t f p i c a en extremo. Los mismos pa-dos de v a r i a c i d n s e a p l i c a n a:.,las c é l u l a s . En a l g u n o s adenocarcinomas, e l e p i t e l i o glandular puede e s t a r formado en su
--
mayor p a r t e p o r una s o l a capa c e l u l a r , y conservar dichos
elementos l a forina c i l í n d r i c a adulta d e l e p i t e l i o c e r v i c a l
-
normal. En o t r o s puede h a l l a r s e formado p o r m ú l t i p l e s capas,
y e s p o s i b l e que l l e g u e n a i n v a d i r tan amplianiente e l conduc
t o que en zonas a i s l a d a s parezca que e l carcinoma pertenece
a l a variedad epidermoide s ó l i d a .
Tomando como base e s t a s v a r i a c i o n e s en laL- mismal
famila
de tumores, s e han popularizado d i s t i n t o s sistemas de gradac i ó n h i s t o i b g i c a , siendo e l más myileado e l de Broders. La
-
v a r i e d a d menos i n d i f e r e n c i a d a c o n s t i t u y e e l grado I; a q u e l l a
en que el proceso e s .n4xima, e l grado I V , d e s i w n d o los
dos I1 y I11 l a s variedades intermedias.
-40-
gra
Quizilbash y F r i e d e l l y McKay han d e s c r i t o l a s alterac i o n e s n e o p i á s i c a s i n i c i a l e s dentro de l a s glándulas d e l endocervix, y s e r e f ' i r i e r o n a e l l a s como adenocarcinoma i n sL
-
t u en e l c u e l l o . Suelen ser' cambios f o c a l e s , y a d i f e r e n c i a
de sus c o n t r a l t a r t e s de c é l u l a s escamosas, aún no s e estudia
b i e n su h i s t o r i a natural.
dn ocasiones, l o s tumores que surgen d e l c u e l l o contendrán elementos mali6gos del, e p i t e l i o tanto escamoso como g
l
duiar. Cuando s e i d e n t i f i c a n claramente l o s elementos escamo
sos pueden denominarse carcinomas adenoescamosos. S i n embargo en muchos casos, e l componente epidermoide e s t a mal d i f e -
renciado s i n formación de j i e r l a s de queratinina, y Juiian y
colaboradores han recomendado que e s t s o tumor,es s e c l a s i f i 3 q
quen separadamente como carcinoma adenoepidermoide.
Las p a t
c i e n t e s con e s t o s tunores combinados t i e n e n al p a r e c e r peor
p r o n ó s t i c o que l a s enfermas con adenocarcinomas o con cáncer
de c é l u l a s escamosas únicamente. Cuando s e observa e s t e patr6n h i a t o l ó g i c o , e s importante t e n e r t e j i d o endometrial p o r
.,
d i l a t a c i ó n y raspado o b i o p s i a , para comprobac s i hay i n v a e i
s i ó n d e l cuerpo uterino.
S i e s t a afectando, debe eonsiderar-
s e e l tratamiento comb.inado que i n c l u y e radiacibn además
de
h i s t erectomfa.:.
En ocasiones, s e observan o t r o s t i p o s h i s t o i ó g i c o s de cáncer d e l c u e l l o . El. sarcoma, e l adenoide de c é l u l a s bas&l e s , y e l melanoma pueden s u r g i r en forma primaria en e l cue
110, y s610 en r a r a s ocasiones s e observa cáncer metast&tico
de o t r o s s i t i o s o linfomas.
CARACl'ERISTlCAS GLINICAS DTL CARCINOMA CERVICAL
La edad media de l a s p a c i e n t e s con carcinoma sintomátia
co d e l cuelLo es de aproximadaraente 45 años. En contraste,
-
~
l a s enfermas con carcinoma L n t r a e p i t e l i a l t i e n e n en promedio
32 años y no e s r a r o e s t e d i a g n ó s t i c o en adolescentee o j ó v ~
nes en l o s primeros &os
de l a segunda decada de vida.
-
EL
carcinoma i n t r a e p i t e l i a i es c a s i sie,npre asintomático, y
el
-
d i a g n b s t i c o s e e s t a b l e c e a l inoaento de un f r o t i s c e r v i c a l
r e a i i z a d o como estudio sisterntico
.
El. d o l o r no c o n s t i t u y e un sintoma d e l carcinoma c e r v i c a
hasta l a s a t i m a s Pases de l a enfreiaedad. % descoiioriiniento
-
de e s t e hecho c a p i t a l e s uno de l o s obst&culos m&s graves
que s e encuentran en l a campaña t e n d i e n t e d e l reconocimiento
prematuro d e l cáncer. En l a mayorfa de los casos, e l primer
-
síntoma e s l a hemorragia, l i g e r a p o r l o general. S i p o r su
edad l a p a c i e n t e s e encuentra aim en p e r í o d o reproductor, e s
t e sangrado adopta tfpicamente la variedad intermenstrual.
*
-
Pueden producirse después d e l c o i t o , e s f u e r z o s v i o l e n t o s o
de una defecation d i f í c i l . Es c a r a c t e r f s t i c a sobre todo l a
hemorragia " p o r contacto",
que sigue a l c o i t o o a un simple
examen p é l v i c o . Por desgracia, en muchos casos de hemorragia
no s e produce hasta que l a enfermedad e s t á arraigada y se ha
-
extendido a l o s l i n f á t i c o s ; de aquf que hasta una p a c i e n t e
i n t r r l i g e n t e y a l e r t a puede e s t a r condcnada a muerte antes c p
-
s e presenten l o s síntomas. Más a h s i e l tumor e s t á l o c a l i z a
do en e l endocervix, hay propensi6n a que l a hemorragia
spa%
r e z c a más tardiamente porque l a l e s i ó n e s t á en un s i t i o más
pro t e gi do.
A veces.puecle n o t a r s e un f l u j o anormal, p o r l o general
acuoso, aún antes de que aparezca l a hemorragia especialment e en caso de adenocarcinoma. Sin embargo, tarde o temprano
e l f l u j o aparece teñido de sangre. A medida que progresa l a
enfermedad, t a n t o l a hemorragia como e l f l u j o áe hacen más
-
p e r s i s t e n t e s y profusos, ali p r o p i o tiempo de l a ulceracibn,
cada v e z mayor, y l a i n f e c c i ó n secuhdaria van c o n f i r i e n d o
-42-
a
u
.
-
il
,,;.
producto de s e c r e s i d n un o l o r cada v e z más desagradable. ñLg
d::n p r e s e n t a r s e o t r o s sfntomas como i r r i t a b i l i d a d v e s i c a l , d bido a que e l proceso va tomando e l tabique v e s i c o v a g i n a i ,
con l a correspondiente sensación de m o l e s t i a r e c t a l , que
se
e x t i e n d e h a c i a l a p a r t e p o s t e r i o r . El d o l o r g r a v a t i v o , peno-
so c o n s t i t u y e p o r i o g e n e r a l un síntoma destacado, que puede
hacerse severo a medida que avanza l a afeccibn.
a
-
dolor per
s i s t e n t e en l a r e g i ó n lumbcsacra en e s p e c i a l cuando s e acompaiiade iinfedema de i a p i e r n a , es un s i g n o de muy mal pronbg
t i c o . La " t r i a d a t e r r i b l e " ,
d o l o r sacro, linfedema u n i l a t e r a l
y obstrucci6n '..citzrsteral u n i l a t e r a l , i n d i c a una enfermedad
m u y avanzada, p o r l o general incurable. Es p o s i b l e que s e
-
produzcan f i s t u l a s en v e j i g a o r e c t o , haciendo más t e r r i b l e
e l padecimiento. La m f i i t r a c i 6 n l a t e r a l c r e c i e n t e obstruye
l o s u r é t e r e s , siendo i a uremia i a causa ihbtima
de muerte
-
q u i z á en el gru;io mayor de casos.
M A R C O
I N G E N I E R I L
DIAGNOSTICO DEL CANCEñ CERVICAL
En &os
r e c i e n t e s s e :han revisado completanente n u e s t r a
métodos de v a l o r a c i ó n de c o n s u l t o f l o en i o que s e r e f i e r e . BL
d i a g n 6 s t i c o de c4ncer u t e r i n o , especialmente c a z v i c a l . De he
cho s e i n s i s t e en l a importancia de preguntar cuidadosamente
s i había hemorragias de contacto, c o n s i d e r a r e l aspecto ma---
-
c r o s c ó p i c o d e l c u e l l o , e l v a l o r de l a palpación y o t r o s méto
dos que aiiora hemos de c o n s i d e r a r simplemente secundarios.
L.
r.
F r o t i s de Papanicolaou
I..
El d e s a i r o l l o de un método c i t o i ó g i c o p r e c i s o para es%
C"
d i a r mujeres asintomáticas con c u e i i o de aspecto cornpietamel'
I
t e normal nos ha permltido en muchos casos e l d i a g n ó s t i c o de
P
-43
-
c h c e r temprano l a r g o tiempo antes que hubiera síntomas o
anomaifas patoh5gicas manifiestas.
-
poda mujer sexualmente a c
t i v a debe p r a c t i c a r s e un f r o t i s para cáncer, que no e s c a r o ,
y que como aclaramos sinceramente a nuestras enfermas, e s lw.
mejor i n v e r s i ó n que pueden kacer. No obstante hay que tomar
una h i s t o r i a completa sobre hemorragia intermentrual o p o r
-
contacto, p a l p a r cuidadosamente e l c u e l l o y e x p l o r a r l o con
-
e l esrPjo. "onde hay una i e e i b n v i s i b l e en e l c u e l l o , se debe
obtener una b i o p s i a además de un f r o t i s , tenga o no l a l e s i ó n
aspecto de c h c e r . De hecho, l a s c é l u l a s n e c r ó t i c a s e i n f l a k
matorias de un cáncer invasor u l c e r o s o pueden d i f i c u l t a r o
i m p o s i b i l i t a r e l d i a g n ó s t i c o c i t o l ó g i c o de una l e s i ó n muy
--
evidente. Papanicolaou y Tsaut i n t r o d u j e r o n i n i c i a l m e n t e
l a
t é c n i c a de l a c i t o l o g f a en l a a e d i c i n a c l í n i c a . Las muestras
de c é l u l a s e x f : > l i a d a s o raspadas de l a s u p e r f i c i e d e l c u e l l o
y de l a vtiggina s i r v e n como microbiopsias,
-
en l a s que e l c i t o
p a t ó i o g o e s t u d i a los múlti.pies procesos de l a salud y l a enfermedad. Aunque suelen p r o v e n i r de l a s u p e r f i c i e de órganos
como e l c u e l l o , e s t a s muestras r e l l e j a n con p r e c i s i ó n proce-
sos más profundos. Abarcan una Brea para e x h e n más amplia
F
de l o que suelen pennii&vlo l a s b i o p s i a s , no e x t i r p a n t e j i d o
v i a b l e y no producen procesos i n f i a m a t o r i o s o de reparación
o e s muy pequerio.S& han d e s c r i t o muchas t e c n i c a s para o b t e n a
muestras c i t o i ó g i c a s , e l denominado f r o t i s Papanicolaou. s-,i n
importar e l procedimiento u t i l i z a d o , e s necesario recordar
v a r i o s p r i n c i p i o s ; 1 ) l a t é c n i c a para obtener muestras debe
s e r óptima a f i n de l o g r a r l a s c e l d a s que proporcionan l a
información
an&s
-
p r e c i s a d e l transtorno en i n v e s t i g a c i ó n ; 2)
l a muestra debe f i j a r s e inmediata y adec admente para perm&
tir l a mejor i n t e r p r e t a c r b n , y 3 ) e s n e c e s a r i o informar a i
-44
-
-
-
c i t o p a t ó i o g o de c u a l q u i e r h a l l a z g o o h i s t o r i a c l í n i c a poco
comfmes, asf como s e n a l a r l e c u d q u i e r pregunta o preocupaciái
e s p e c í f i c a , % consecuencia,
s i e l o b j e t o e s hacer un estudiD
para d i a g n o s t i c a r cheer c e r v i c a l , hay que obtener una buena muestra d e l á r e a de l a unión escamocilíndrica d e l c u e l l o ,
pero s i s e desea v a l o r a r e l estado hormonal de una paciente,
es p r e f e r i b l e hacer un raspado de l a pared l a t e r a l de l a vagina. Para proporcionar la, i n t e r p r e t a c i ó n ms'
de l a muestra,
precisa y ú t i l
e s importan.te enumerar algunos datos de l a pet.
-
c i e n t e como laedad, ú l t i m a . meiiei6,ruación, t i p o de anticoncep4
t i v o ( s i emplea alguno) y diagnóst.!.cos y tratamientos p r e v i a
p o r ejemplo b i o p s i a o r a d i o t e r a p i a .
La t é c n i c a que hemos u t i l i z a d o para l a detección d e l .:cáncer c e r v i c a l i n c l u y e l a . de obtener muestras d e l fon.do común v a g i n a l , e l e x o c e r v i x y e l conducto endocervical.
hk cue
-
iio s e expone y se elimina. suavemente e l exceso de secresióri
Se u t i l i z a e l extremo redondeado de una espátula de p l á s t i c o
o de madera para obtener un poco de m a t e r i a l
del f6rnix
va-
g m a l p o s t e r i o r . Después s e usa e l o t r o extremo para raspar
e l c u e l l o en 360
. Por
úIt;imo, un a p l i c a d o r c u b i e r t o con al-
godón humedecido en s o l u c i ó n s a l i n a s e introduce en e l conducto endocervical y se hace g i r a r para obtener una muestra
-
d e l endocervix. Cuando s e han tomado l a s muestras, s e disemi
nan l o más rápidamente poe:ible en un p o r t a o b j e t o s y s e fijan
de insiediato en a l c o h o l e t S l i c o a l 95 $. Algunos autores recomienda
una p i p e t a pequeria p r o v i s t a de un bulbo de caucho
para obtener una muestra endocervical, pero Shingleton y colaboradores encuentran que son igualmente e f i c a c e s e l a p l i c a
dor c u b i e r t o en un extremo con algodón y l a a s p i r a c i d n endo-
-45-
c
c e r v i c a l . Cuando s e sabe que hay d i s p l a s i a c e r v i c a l u o t r a
anormalidad, puede s e r ú t i l . usar dos o t r e s p o r t a o b j e t o s y
-
para i a i n t e r p r e t a c i ó n d i a g n o s t i c a r e m i t i r separadamente cadi
muestra en un p o r t a o b j e t o , a f i n de proporcionar una disemi*
nación c e l u l a r más amplia. Los f i j a d o r e s comerciales en aer e o s o l o e t a s son bastantes s a t i s f a c t o r i o s , s i e l f r o t i s no
s e d e j a s e c a r a i a i r e hasta después de l a f i j a c i ó n . Una v e z
que se f i j a , puede secarse e l p o r t a o b j e t o s para e n v i a r l o a i
l a b o r a t o r i o de c i t o l o g f a .
Originaimerite s e propuso un sistema de gradación para
-
i n d i c a r e l diagnós .ico de i . a c i t o : o g í a c e r v i c a l de l a c l a s e
I a l a V,
se&
l a gravedad de l a anomalfa. S i no s e observg
ban c é l u l a s anormales, e l CLiagnnSstico correspondfa a l a c i a =
I, y s i habfa c é l u l a s que smgerfan cheer i n v a s o r franco, se
indicaba una l e c t u r a c l a s e V. Muchos c i t o p a t ó i o g o s informan
actualmente un d i a g n ó s t i c o en términos h i s t o l 6 g i c o s como "beni;;no",
" d i s p i a s i a moderada",
o "carcinoma
de c é l u l a s
esta.-
-
masas", e indiaan su inh$ryiretación de l o que hubiera mostra.
do una b i o p s i a con base an l a muestra c i t o l b g i r a .
La t é c n i c a d e l f r o t i s de Papanicolaou e s relativamente
barata, indolora, y p r e c i s a para e l d i a g n ó s t i c o de d i s p l a s i a
c e r v i c a l y cáncer. como t a l , e s i d e a l en e l d i a g n ó s t i c o de
-
grandes grupos para descubrir una n e o p l a s i a c e r v i c a l asintomática temprana. La i n f l u e n c i a de l o s grogramas de s e l e c c i ó n
en i a f r e c u e n c i a d e l cáncer c e r v i c a l y l a d i s p i a s i a ha s i d o estudiada. en muchos s i t i o s d e l mundo.
M u y pocos l a b o r a t o r i o s ; de c i t o p a t o l o g f a están conscien-
t e s de e s t a s n e g a t i v a s faisias, p o r l a simple razón de que
-
l a s pacien.tes con f r o t i s n e g a t i v o s normalmente no s e control a n después, de ninguna. maner;..
-
Sia embargp, e s p o s i b l e c a l -
c u l a r e l promedio de ne::a.tivas
f a l s a s de l o s dcztos obtenidos
en estudios r e p e t i d o s , y [email protected] de ellos poe ejemplo e l de
S i l b a r y Woodruff han desarrollado c i f r a s de n e g a t i v a s fal-sas de o t r a forma. De é s t o s y o t r o s e s t u d i o s s e ha obtenido
una c i f r a aproximadamente 204a de neAativas f a i s a s para l a c._
i
tologfa exfoliativa.
Esto no s i g n i f i c a que l a c i t o l o g f a e x f o l i a t i v a no sea
-
procedimiento d e v a l o r , pero q u i e r e d e c i r que l o s g i n e c o i b p
deber&
tomar en cuenta. e s t a p o s i b i l i d a d en e l empleo d e l m é.-
todo c i t o i b g i c o .
a
,
problema de l a s n e g a t i v a s f a l s a s es une.
de l a s p r i n c i p a l e s razones para r e a l i z a r un estudio anual,
pero s i l a p a c i e n t e t i e n e dos o a&
t i v a s Fn
ter
u11
mejor t r e s pruebas ne,F-
p e r í o d o de t r e s años, l a evoiucidn l e n t a d e l cán
-
d e l c u e l l o u t e r i n o indican que no e s necesario un f r o t i r
subsecuente anual. sino que puede r e a l i z a r s e cada t r e s , cuats,
arios.
Sin embargo, l a s ¿pacientes de " a l t o riesgo" p o r t e n e r
una o más c a r a c t e r l s t i c a s e p i d e m i o l b g ~ c a smencionadas o que-
han tenido f r o t i s anormales en años a n t e r i o r e s , dehen some-
-
t e r s e a e x h e n e s cuando menos una v e z a l allo. También hay
que recordar que un f r o t i s < c e r v i c a l de Papanicolaou no e s e l
único be:ieficio médico de una v i s i t a anual c t l ginecoiógo.
Desde un punto de v i s t a t e b r i c o , e l empleo frecuente de
un e x h e n c i t o i d g i c o con i n t e r v á i o s adecua.dos pudiera dismin u i r l a muerte p o r e s t a enfermedad, descubriéndola. más tern-prano y en etapa curable. La p o s i b i l i d a d de c o n t r o l a r e l
ter
-
CELL
de c u e l l o u t e r i n o p o r c e i e c c i b n c i t o i 6 g i c a . ahora e s de
-
plena catualidad. HL&O un tiempo, cuando aparecí6 e l método
[email protected],
en oue p a r e c f a como s i e l problema d e l cáncer
-
c e r v i c a l se pudiera resolver p r o p o r c i o m d o e s t u d i o s c i told*.
g i c o s anunciando su d i s p o n i b i l i d a d , y pensándose y esperarido
que todas l a s mujeres h i c i e r a n l o n e c e s a r i o para s e r examina
-
das q a c i a s a l o s medios que normalmente u t i l i z a b a n como ser
v i c i o médico r e g u l a r . P a r e c e comprobado que e r a ingenuo espe
r a r un c o n t r o l amplio y e f h a z d e l cáncer d e l c u e l l o en e s t a
forma.
Esto no s i g n i f i c a que una p o b l a c i ó n l i m i t a d a de mujeres
no pueda l o g r a r p r o t e c c i 6 n con e s t e método, y hay muchos infames publicsdos de grugot3 de p a c i e n t e s p w t i c u i a r e s que l o
confirmati. Con e l paso d e l tiempf.> y e l aunento d e l número de
f r o t i s r e p e t i d a s , ha disminuido netariente l a proporci6n de
-
descubrimientos a .iiediaa que los carcinomas en l a poblucidn
s e han i d o extirpando.
Aunque e s t e enfoque procede, cabe se-
5alar. que, en t é m i n o s de (control d'-e una comunidad y d e l t o
t a l de e n e r g f a gastada en proporcidn de l o s cánceres descub i e r t o s , s e t r a t a de una t'écnica poco e f i c a z . Irn estudio de
-
Kashgarian y colaboradores acerca de l a d i s p o n i b i l i d a d públ i
c a de l a c i t o l o g f a u t e r i n a en e l condado Shelby de Memphis,
-
Tennessee, en un p r o y a m a adecuado, demostró que, i h c l u s o en
t r e a q u e l l a s mujeres que s'e s i e n t e n impulsadas p o r l a propaganda a buscar un f r o t i s o r i g i n a l , después de un p e r i o d o de
c i n c o arios dki e s f u e r z o de .propaganda muy c o n s i d e r a b l e y prolongado s o l o dejd l a quintis. p a r t e de l a pi;blacibn dcsezndo
e f e c Luar ex&menes anuales r e p e t i d o s , i n c l u s o cuando t a l e s
-
examenes eran t o t a l n e n t e g r a t u i t o s .
f o r l o t a n t o e l problema primario en e l c o n t r o l publico
d e l cancer c e r v i c a l no e s de t e c n i c a n i d i s p o n i b i l i d a d de
-
mv
d i o s , sino mas b i e n de organizaci.6n y motivaci6n de dicha co
-munidad.
-48-
Prueba de S c h i l l e r
Se basa en que
e l e p i t , e l i o canceroso n o c o n t i e n e glucó-
geno y , por tanto no capta. e l yodo como e l e p i t e l i o normal
-
d e l c u e l l o , o de l a vagina, que son r i c o s en giucógeno. As1
l a a p l i c a c i g n de una s o l u c i ó n de yodo ( d e S c h i l l e r a l 0.3% de Lug01 a l
56) 2uede mostrar e p i t e l i o normal
de c o l o r caobm
mientras l a s zonas de displ.:isict y cancer quedan s i n t e f i i r ;y
ne tamerite l i m i ta.das.
p o r desm:racia, e l e p i t e l i o c i l l n d r i c o
>-
d i v e r s o s cuadrm
i n f l a m a t o r i o s benignos q u i z a tampoco se tinan,Jo que conduce
una "prueba p o s i t i v a de S c h i l l e F , y e l empleo más l i b e r a l
n
e s p e c ia
son dudo
-
de f r o t i s y b i o p s i a s ha l i m i t a d o hasta c i e r t o punto l a u t i l i
t i e n e Zran v a l o r
z a c i ó n de e s t e método. > i n embar@,
mente cuando LOS f r o t i s son p o s i t i v o s y l a s b i o p s i a s
sas; s i el. f r o t i s se vulve p o s i t i v o d.espues de h i s t e r e c t o m h
p o r un cáncer i n s i t u , sefiala l a necesiadd l ó g i c a de una b i o c
sia.
-
- ._-__
-.
--
-
"1.Hallazgos colposcópicos normal es.
A. E p i t e l i o esc&mosos o r i g i n a l .
Se obserara en c u e l l o y va
-
S n a un e p i t e l i o o r i ¿ - i n a l l i s o , rosado y s i n rasgos p r o
pios. No e x i s t e n v e s t i g i o s de e p i t e l i o c i l l n d r i c o quepueden i d e n t i f i c a r s e , a s i como e p i t e l i o que s e c r e t a moco, hendiduras abier.tas o a u i s t e s d e Naboth
B. E p i t e l i o c i l i n d r i c o . Es una capa de e p i t e l i o simplela&
t a l que produce moco, que se extiei4fle e n t r e e l endome4
t r i o p o r a r r i b a y e l e p i t e l i o escamoso oric&nal o e l
-
e p i t e l i o metaplásico p o r abajo. Lz s u p e r f i c i e cubierxa
con e p i t e l i o c i l i n d r i c o es i r r e g u l a r con p á p i l a s grander; de estroma y hentiiduras profundas.
l a colposco-
-
p i a despues de l a prueba de ácido a c é t i c o t i e n e estruc
tura t l p i c a semejante a uvas.
-43-
E l e p i t e l i o c i l f n d r i c o puede e s t a r presente en e l endo-
c e r v i x sobre una p o r c i ó n o toda l a vagina.
C. Zona de transfonnaci6ri. Es l a s u p e r f i c i e e n t r e e l e p i t e
-
l i o plano o r i g i n a l y el e p i t e l i o c i l l n d r i c o con d i f e r e n
t e s graüos de madurez que pueden i d e n t i f i c a r s e . Los com
ponentes una zona normal de transformación puede e s t a r
a i s l a d o s d e l e p i t e l i o c i l i n d r i c o que l e rodea p o r e p i t e
lie plano metaplástico, "aberturas giandulares"
y quis-
t e s de Naboth. En l a s zonas de transformacidn normales
no s e enc entran a l a colposcopia estructuras que hagan
pensar en neoplasia c e r v i c a l .
L
c
11. H a l l a z g o s colposcópicos anormales.
-
A. Zona A t i p i c a de transfomaci6n.Una zona dP transforma
c
c i ó n en l a cual e x i s t e n h a l l a z g o s colposcópicos que
L.
hacen pensar en n e o p l a s i a c e r v i c a l .
r
.
1. E p i t e l i o blanco- despues de l a prueba con ácido a6
c é t i c o s e observa con e l colposcopio una zona anor
-
c
mal f o c a l .
E l ep:ttelio blanco e s un f e n h e n o trail-
L
s i t o r i o que s e observa en l a s u p e r f i c i e de mayor
c.
d+nsi dad nuclear.
2. P u n t i l l e o
- anormalidad
coiposcópica f o c a l en l a
-
que s e observan c a p i l a r e s de manera punteada.
L
.
.
.._
...
.
3. Blosaico
-
anormal-idad colposc6pica f o c a l en l a que
e l t e j i d o semeja un inosoico.
Los campos del mosai-
co están separados s o r bordes enrojecidos.
4. Leucoplaquia
-
anormalidad colposcdpica f o c a l en
La que hiperqueratos
IS
-
y paraqueratosis parecen u-
na p l a c a blanquecina elevada. Esta p l a c a blanqueci
na s e i d e n t i f i c a antes de l a a p l i c a c i ó n dr. acid0
ace t i c 0
.
-
.
_I--
5. Vasos S a n p i n e o s anormales
-
-
anormalidad colpos-
c ó p i c a f o c a l , en l a que los vasos sangulneos no
son puntileados, como mosaicos o con pequenisi-mas
ramificaciones,
sino que son vasos i r r e g u i a
r e s con cursos sinuosos que parecen comas, tirabuzones o espagueti.
B. Sospecha de cáncer invasor. Es e l cáncer que en l a
exploraci6n c l l n i c a no puede observarse.
111. H a l l a z g o s colpóscópicos i n s a t i s f a c t o r i o s .
-
Casos donde l a unión p l a n o c i l i n d r i c a no puede observar
se.
TV. Otros H a l l a z g o s colposc6picos
A.
V a g i n o c e r v i c i t i s . Es un molde colposcópico d i f u s o de
hiperemia en l a oue l o s vasos sanguineos parecen
-
a r r e g l a d o s en forma. d i f u s a semejante a l p u i i t i l l e o
-
vascular.
B. Erosión Verdadera. Es una supe:.ficie s i n e p i t e l i o p r 2
vo c ada generalmente p o r traumat i smo
6. E p i t e l i o a t r 6 f i c o .
.
Es un e p i t e l i o plano con carencia
de estr6gerios en que e l molde vascular se i d e n t i f i c a
pronto p o r e l adelgazamento r e l a t i v o d e l e p i t e l i o
-
plano suprayacente.
D. Condiloma, papiloma. Son l e s i o n e s e x ó f i t i c a s que pue
den e s t a r dentro o f u e r a de l a zona de transformación
-51-
I
.
P-
-..
c
c_
rC..
rL.
-
_I
P
L
r
L
c
L.
r-
-
52-
Colposcopia
El colposcopio es un instrumento g r a c i a s a l cual puede
observarse e l cuellm en plena l u z con aumento de 10 a 40.
La t é c n i c a de exámen es rápida; r e q u i e r e practicarnente e l
-
mismo tiempo que l a i n s p e c c i ó n d e l c u e l l o a simple v i s t a .
Después de obtener muestras para cittblosfa, puede l i m p i a r s e
e l c u e l l o con un hisapo d? algodón, enfocarse e l eolposcopio
y e s t u d i a r cuidadosamente todo e l c u e l l o , primero con iluwinación normal, luego con f i l t r o verde para mejolrr l a v i s i b i l i d a d de l a s imágenes vasculares. &l c u e l l o s e asea adicio-nalrnente con una solución de acido a c é t i c o a l 3% que proporo
c i o n a una .nejor d i f e r e n c i a c i ó n d e l e p i t e l i o c i i l n d r i c o de
la
zona de transformaci6n. P o r último pueae t e ñ i r s e con solución
de Lug01 para delinear. &rea,s n e g a t i v a s a l giucógeno.
Los ha1 ~ a z g o scolposcó,icos
categorlas utilizando l a
pueden d i v i d i r s e en v a r i a s
erminoiogia adoptada p o r l a s o c i e h
dad Estadounidonse para Colposcopia y P a t o l o g l a C e r v i c a l .
-
Hay v a r i o s estudios que comparan l a p r e c i s i d n d i a g n ó s t i
ca de l a colposcopia con l a de l a c i t o l o g í a . Los resultados
-
presentan que dependen principalmeate de l a e x p e r i e n c i a de
l o s autores con cada una de l a s tecnicas. U t i l i z a n d o ambas mejora l a , p r e c i s i ó n diagnóstica, porque l a s v e n t a j a s d~ ambcs
metodos s e complementan.
E l p r i n c i p a l inconveniente de l a coiposcopia e s que só-
l o permite examinar l a p a r t e v i s i b l e d e l c u e l l o . Por l o tanb
s i no puede v i s u a l i z a r s e l a unión p l a n o c i l l n d r i c a , e l exilmen
colpocópico e s poco s a t i s f a c t o r i o y hay que c o n f i a r solo
l a citologia.
mente, de
Esta s i t u a c i ó n e x i s t e en e l 1 5
en
6, aproximada-
l a s mujeres premenopausicas.
P o r o t r a p a r t e , l a c i t o l o g l a t i e n e e l inconveniente de que
-53-
-
L..
no i n d i c a precisamente donde nacen l a s c e l u l a s svspechosae
c
d e l c u e l l o . P o r colposcopia s u e l e poderse l o c a l i z a r l a zona
c_
sospechosa, v a l o r a r sus dimensiones y su gravedad,y tomar
r-
b i o p s i a d i r e c t a para e s t a b l e c e r e l di;.gnóstico h i s t o l b g i c o .
-
La c o r r e l a c i 6 n e n t r e los datos c i t o l ó g i c o s , colposc6picos e
-
h i s t o l ó g i c o s permite un tratamiento más r a c i o n a l de l a e n f e r
ma que fiando solamente a l método.
Aunque l a colposcopia y a f u e creada p o r Hinse1.mm en
1 9 2 5 , y ha s i d o ampliaqente! u t i l i z a d a en Europa y en América
del Sur para diagn6stico de l e s i o n e s c e r v i c a l e s , ha tardado
en l o g r a r popularidad en Estados Unidos, probablemente por
e l d e s a r r o l l o más temprano de medios d i a g n ó s t i c o s para c i t p logia.
En los L i t i m o s años l a s t e c n i e a s c i t o 1 6 g i c a s s e han i d o
perfeccionando tanto que
L
c
82)
presentan cada vez mas problemas
-
a l d e s c u b r i r i ,pequeiias zonas sospechosas en e l c u e l l o por i n
v e s t i g a c i ó n c l i n i c a . Las b i o p s i a s a c i e g a s en casos de s i t o -
-
L
l o g l a sospechosas o p o s i t i v a , en l a s c u a l e s no s e descubre
c
una l e s i b n , puede t e n e r p o r consecuencia un diagnlbstico histt o l ó s c o n e g a t i v o f a l s o . P o r o t r a p a r t e , procedeer directa--
c
mente a l a c o n i z a c i ó n basándose en una s o l a muestra c i t o 1 6 g i
-
ca sospechosa originaráun nfunero muy elevado de conizaciones
innecesarias, que no dejan de t e n e r morbilidad.
LI
c
Es p r e f e r i a -
b l e u t i l i z a r e l colposcopio para buscar l a zona de biopsia.
Con e l descubrimiento de l e s i o n e s cada vez in&s semprh=!
nas durante l a s e l e c c i ó n s i s t e m á t i c a y e l estudio de l a s poblaciones,
e l colposcopio est$. adquiriendo un l u g a r cada v e z
más importante en e l tratamiento x's
c .i t 010?:la ano m a l .
-54-
moderno de p a c i e n t e s con
-- 6 5 -
Estas son a l v a s de las caracteristicas del Cblposccpio cuando onpaado Ccn otros
en e l nercado:
-
optica supeiior que p x m e
-
del anpiio canp visuai, i a profiindidad del misrrr, es excelente. ü s t d
no enantrará esto aun si io omrpara mn cualquier otro aziposccpio VendIda
a myor predo. üsteü se dara cuenta de su superioridad hadendo exdnaciionos
prácticas ya qm es muy dificil expresarla w n nomiros.
-
Nueva distancia focal de 2 8 h
-
Nueva canara Polaroid Sanar SLR om aditamento para acercar l a ha-.
-
Capacidad para fotografía an canaras Polaroid y Oe 35m.
a l otro imm3 menoc de M IIdnUto.
-
La
-
claridad. hasta e l borde
misrm de1
rxmp
visual.
Tiene un sistem Tri-Optiw, el cual elimina l a necesidad de dividir e l p d s
de haene l exanrinadDr y e l visor para fotografía.
ndemas
IiimrinacFbn superior del c a p o visual (3011 una luz de intensidad ajustable la cual
provee el 25 por ciento mas de luz que e l rrejor CblpscopiO decpies del nuestm
Canbiar de un slstem
camara foingrafica se p&
dejar posidonada en e l B l p s a p i o sin que esto
interfiera am e l operador o mientras se tcrrian nuestras.
El ajuste horizmtal es 2 pulgadas mas grande quo en cualquier otro b l p o s ~ u
l o cual permite hacer e l enfoque mas f a d m t e .
-
E l tubo de enseibnza es disponible am la mism seleccih de 0 4 - s
de ltmkque e l observador y e l examinador puedan ver exactammte l a ndsnia i m p .
-
E l film esta amvenienerrente integra& a l Cblposapio.
E l m t a j e se ofrece en cuatro m r a s diferentes: base &te
ligera; &ut7
o base mdante
para nesa wn bram de pimte; mntura para pared an brazo de pipara uso pesado wn brazo de piwte.
-
pdemac
-
Gran e
de estas hay muchas otras caracteristicas ~IE hacen gue nuestro BJposcq>io
sea superior y mudio mas practim a otros en su clase.
d
a
- Si usted aprpara cualquier otro Cblposwpio an la mima
capacidad, usted enwntrara
opinion, m o debe verlo y 10-
-rge Garamy
vice PXeSident
international Madceting
una gran e w n d a
a l omprar e l nuestro.
w n otros m3oelos.
En nuestra
LAPAROSCWIA
LO laparoscopia es una tecnica utilizada por varias especiaA
idades médicas c p » :
ginecolegfa, cirugia y gastrrnterología.
Desde su intreduccibn clínica en los inicies de este s i g l o
l a laparescopia ha llegado a jugar un papel impmtante, Ha sido utilizada en gastranterología para evaluar enfermedades de
hfgade, permitiendo l a visuaiizaci6n directa del hfgade y l a
biopsia directa de defectos focales.;
desde e l advenimiente de
l a tomograffa cemputarizada (CT) su papel ha disminuido
cirujanes se hen advecado
Los
en l a laparascapia en lugar de l a
laparetaniia en pacientes selectes
ción de dafio metastatid
.
, cmo en e l
caso de evalum
(hepático) y evaluación de abseses 1
de hfgade. Puede ayudar a evitar l a laparetomia innecesaria o
puede ayudar a determinar
k
t
neaissidad de operacib.
-
En l a g i
necmlogía es su area da mayiyor aplicación. Es ú t i l para e1 diag
ndstico da infertilidr¿, amunalfas cong&nitas, endemetriesis y
más recientemente
i i i
, para recuperar
hu v o
para fertiiizaci6n
(8.
E l equipo es variable dependiendo de l a especialidad en l a
que se va a utilizar. Come en otras areas,' l a introduccih de
fuentes luainasas de fibra-6ptica ha rnejuade l a iiumunaci6n
y eliminade e1 riesgo de quemaduras debido a una fuente de luz
-51-
caliente. Existen una gran variedad de modelos de laparescopies
, con alguna variacien en e:L angula de visi6n.
-
También se re*&<
ere una fuente de gas para insuflacidn de l a fiao&dad abdominal
y un menitor para estimar presibn entre e l abdomen y l a canti-
-
dad de gas infiltrado. Los acceshrios varian cen l a s indicacion
es; pruebas, forceps para bibpsias, equipo para cauterie, instrumentos ligantes para amarre de tubes
* instrumentos monitcrri-
sadores de presibn para medicibn de presiencs intravasculares,
sen algunes ejemplos de accesorios para laparascopia.
-
Elpracedimiente puede ser realizado bajo anestesia lecal cr ge
neral: l o s examenes sencillos de diagnbstico sen frecuentemente
hechos bajo anestesia lwal en e l sits de insercibn y s e d a c i h
intravenesa suplemental; los prwedimientos mas largos y l a may
@ r i a de de l o s precedimientes terapeutkes sen hechas baBe m e s
tesia general. E l s i t i o de insercibn es usualmente apenas aril,
ba deabajo del ombligo y requiere sela una muy pequefia incisión
tlcm). La cavidad abdeminal es distendida cen 2-4 L de gas para
prevenir daño a las viceras cuando e l instrumento es insertade
-
y para prevenir una completa visualizacibn del contenido abdemi
nal.(El
cia e l
gas mas frecuentemente usado es Ce2 y en mener frecuem
Ne2&% se u t i l i z a r a aire,
s o l a una pqqueña cantidad (ie
201111) intravascular bastaria para causar una embelea f a t a l ; el
- S8-
(b2
ne se u t i l i z a perque evitaría e l us,
-
de cauterio. Ne2 es u t i
lizade mas frecuentemente cen anestesia local pwque es nenes
rritante para e l peritonean
&
-
1. Despi&s de l a induccibn de pnaurn
aperitonrum, e l instrumento es insertado, y se l l w a a cabs una
inspeccibn. Se puede inspecc:ionar y probar, con biopsias temadas si se requiere; e l periteneum, l a superficie
serosal del
intestine, les organos pelvicos, e l hígado, l a vesícula b i l i a r
y frecuentemente e l apéndice, También se pueden llevar a cabo
maniabras terapdticas. Despu6s de del procedimiento, alge del
aj86
-
i n f i l t r a d o se escapa per e l area; 1. que queda es sistemáti
camente abs8rbj.de y excretado en las horas siguientes a e l procese.
-
Les resultados del precedimiento sen exelentes. Es d í f i c i l de
terminar de manera exacta
BU
razbn de exactitud en e l us@ diay
nbstice per8 se habla del 96% de exactitud, con 72% de p a c i e n b
-
es que evitaron l a laparatomía. E l use de l a laparoscopia segui
r& jugando un papel importante en obstetricia y ginecelegfa.
-F9-
.._
Colpomicroscopia
.
e
EI. coipomicroscopio proporciona una a m p i i f i c a c i 6 n mayor
que e l colposcopio.
L
El e p i t e l i o c e r v i c a l s e t i ñ e v i t a l m e n t e
con h e m a t o x i l i n a de Meyer o a z u l de t o l u i d i n a y pueda observ a r s e l a h i s t o i o g í a s u p e r f i c i a l d e l e p i t e l i o cervical.
Como
e s t e campo de v i s i 6 n e s l i m i t a d o , y i a profundidad de foco
no e s mucha, e l método r e s u l t a d o complicado y no t a n popular
que pueda emplearse si-stem4t icamente.
S i desea información a d i c i o n a l s o b r e e s t a t e c n i c a , e l
l e c t o r debe c o n s u l t a r l a e x c e l e n t e r e v i s i ó n de Wilbanks.
BlopsisL
L
L
!\ (
Nuestra p r e f e r e n c i a p e r s o n a l e s o b t e n e r una b i o p s i a di-
r i g i d a con colposcopio cuando e l f r o t i a de Papanicolaou e s
a n o r m a l o so hay una l e s i ó n c e r v i c a l p e r c e p t i b l e a simple
c
L_
c
L
P
vista o con e l colposcopio.
Debe r e c o r d a r s e que e l anatomo-
patóiogo ~ 6 1 0puede b a l o r a r e l m a t e r i a l que s e l e proporcion a y e s muy importante obbener p o r c i o n e s adecuadas de t e j i d o
aunque l a b i o p s i a en o c a s i o n e s r e s u l t e técnicamente d i f í c i l .
Para e s t e f i n , s e cilspone (&e v a r i o s instrumentos d i s t i n t o s
L_
para b i o p s i a s .
P
hemorragia s u e l e s e r m i n i m y f á c i l de c o n t e n e r con p r e s i ó n
L
r-
i.
Extirpan una porción pequeña de t e j i d o y l a
o l a a p l i c a c i ó n de n i t r a t o de p l a t a o de s o l u c i 6 n de Monsel.
Cuando no s e dispone de un e s p e c i a l i s t a entrenado en
c o l p o s c o p i a , s e adopta ampliamente l a t e c n i c a de l a b i o p s i a
mfLLtiple de t b l a d r o , con dos bocados de l a p i n z a a l a s 1 2 y
6 h o r a s d e l r e l o j -que son los s i t i o s mas f r e c u e n t e s de p r i n -
c i p i o de una n e o p l a s i a - o con c u a t r o mordidas de l a p i n z a a,
l a s 1 2 , l a s t r e s , l a s s e i s y a l a s nueve d e l r e l o j de l a c i r
c u n f e r e n c i a c e r v i c a l a n i v e l de
l a
unión p l a n o c i l i n d r i c a .
-
S i n embargo, no es un método muy s t i t i s f a c t o r i o para v a l o r a r e l
cuello.
S i l b a r y Woodruff señalaron que con l a b i o p s i a en los
cuatro cuadrantes se pasaron por a l t o 11 de 185 (6s) carcinomas
invasores.
Por tanto, a menas que e s t a s b i o p s i a s en sacabocado
tomadas más o menos al a z a r muestran un cáncer invasor, s u e l e
s e r neceshrio obtener una b i o p s i a c i r c u n f e r e n c i a 1 amplia deno
minada b i o p s i en cono a f i n de e s t a r seguros de no pasar por
al
-
t o un f o c o de cáncer invasor.
Lo dicho antes no s e a p l i c a a l o s casos m4s avanzados,
en
los cuales c u a l q u i e r parci6n d e l t e j i d o e x t r a i d o de c u a l q u i e r
-
p a r t e de l a l e s i ó n macroscópica demostrará cáncer, y generalmen
t e nada más que cáncer.
incluso en l o s casos más avanzados s e
recomienda l a b i o p s i a para d.eterminar e l t i p o anatomopatoi6gico
exacto de l a l e s i ó n , tambien para e l i m i n a r l o s r a r o s casos en
-
l o s cuales l e s i o n e s como l a t e b e r c u l o s i s c e r v i c a l pueden e s t i m
l a r e l cuadro macroscópico d e l cáncer.
Aunque tenemos e l mayor
r e s p e t o para los métodos c i t o l ó g i c o s , e s eaeo que e l tratamien-.
t o s e e s t a b l e z c a antes de obtener b i o p s i a s de confirmación.
Gonización d e l cuello
Cuando es imposible v i s u a l i z a r l a uni6n escamocilindrica
-
con e l colposcopio, l a i e s i 6 n s e extiende a l i n t e r i o r d e l conduc
t o y no es p o s i b l e v e r l o s l.imite8 superiores, o no concuerdan
e l f r o t i s , l a v a l o r a c i ó n col.poscópica, y l a b i o p s i a d i r e c t a , e s
t á indicado p r a c t i c a r una b i o p s i a en cono.
-
S i no s e dispone de
un colposcopio y no hay 1esi.ones macroscbpicas en e l c u e l l o n i
zonas s i n t i n c i 6 n en l a grueba de S c h i l l e r , también e s t á i n d i c a
do hacer una conizaci6n diagn6stica.
Esta t é c n i c a se ha
denomi
-
nado con f r e c u e n c i a “cono con c u c h i l l o f r i o “ para d i f e r e n c i a r l a
-
de una c o n i z a c i ó n c a l i e n t e , que se pracGica en t r a s t o r n o s i n f l a
matorios cor1 e l e c t r o c a u t e r i o , carboniza e l t e j i d o y r e s u l t a i n a
decuada para estudio c l t o p a t o i 6 g i c o .
-6
\
-
La conización d i a a o s t i c a
e s fit11 no s o l o para e s t a b l e c e r e l d i a g n o s t i c o de carcinoma i n
s i t u , sino e s además muy importante para d e s c a r t a r un cáncer
invasor, pues sieinpre hay que recordar que e l carcinoma i n S i t u
s u e l e e x i s t i r en l a p e r i f e r i a de un cáncer i n v a s o r verdadero.
Una b i o p s i a c e r v i c a l en cono i m p l i c a simplemente e x t i r p a r
una p i e z a de t e j i d o c e r v i c a l en foriua cónica que i n c l u y a toda
l a zona de anormalidad.
Desde e l punto de v s t a d e l l a b o r a t o
-
r i o , l a s dos p r i n c i G a i e s necesidades son l a s siguientes: i) ten e r l a seguridad de que e l t e j i d o s e e x t i r p a de manera que r e
-
v e l a l a mucosa endocervical coma e l e p i t e l i o plano d e l c u e l l o
v a g i n a l , y l a porción n e c e s a r i a subyacente d e l estroma y &&du
-
l a s ; 2) preparar un número de bloques y c o r t a r lo s u f i c i e n t e , a
n i v e l e s v a r i a b l e s , para t e n e r l a seguridad de oue no se d e j e
i n a d v e r t i d a una pequeña zona, de carcinoma i n t r a e p i t e l i a l .
La
tam -
primera de e s t a s precauciones parece evidente, ya aue a s i e l ana
tomopatóiogo puede v a l o r a r n.o s o l o l a l e s i ó n i n s i t u sino
b i é n l a p o s i b i l i d a d ,de i n v a s i o n temprana.
P o r l o que s e r e f i e r e
a l segundo punto, debemos i n s i s t i r qn que un informe anatomopat o i ó g í c o basado en un s o l o c.orte puede s e r causa de grave e r r o r
d i a g n ó s t i c o en etapas muy tempranas d e l cáncer.
Desde l a i n s t i t u c i ó n de l a colposcopia mas o menos como est u d i o de r u t i n a para l a v a l o r a c i o n de anormalidades c e r v i c a l e s ,
l a mayor p a r t e de l o s cen-trcis ha observado una disminución impor
-
t a n t e en e l númer; de biopsi.as diagnósticas en cono como s e ha
comentado, S t a f l y uattinglgr p r a c t i c a r o n sólo 80 conizaciones en
una s e r i e de 1410 p a c i e n t e s turnadas para v a l o r a c i ó n coiposcópic a por c i t o i o g i a c e r v i c a l ariormai.
Cincuenta y cuatro (4.3$),
en quienes l a v a l o r a c i ó n coi.posc6pica n o fue s a t i s f a c t o r i a , nec e s i t a r o n conización diagnóstica.
RI p a c i e n t e s embarazadas l a morbilidad por hemorrag-a (9.4%)
y parto prematuro (7.4)96)
complicó l a b i o p s i a en cono entre 180
p a c i e n t e s que dieron a conooer A v e r e t t e y c o l .
-62-
La v a l o r a c i ó n
....
coiposcópica de p a c i e n t e s embarazadas con f r o t i c de Papanicol a o u anormales ha elimrnado c a s i por com l e t o l a necesidad de
l a b i o p s i a en cono en e s t e grupo.
Lurain y Gallup, Talebian
y colaboradores y De P e t r i l l o y colaboradores han comunicado
-
resultados e x c e l e n t e s usando colposcopia para v a l o r a r a pacien
t e s embarazadas.
CLASIFICACION CLINICA DEL CANCER CERVICAL.
Es o b v i o que para un estudio camparativo de los informes,
e s t a d i s t i c o s y los r e s u l t a d o s t e r a p e u t i c o c d e l cancer c e r v i c a l
rL
s e n e c e s i t a un sistema uniforme de c l a s i f i c a c i ó n por etapas pa
-
r a ii,ue l o s grupos de p a c i e n t e s sean similares.
Ya que muchas
enfermas con cáncer cervical. s e t r a t a n con r a d i o t e r a p i a y nunL-
ca s e someten a laparotomia exploradora o b i o p s i o de g a n e i o s
r
linfátidos,
L
e l sistema de c i a s i f i c a c i ó n por etapas debe s e r
-
c i i n i c a mas que q u i r ú r a c o , para i n c l u i r a l a s p a c i e n t e s t r a
tadas con todas l a s modalidmies.
s i f i c a c i o n e s a t r a v e s de
1021
Se han propuesto v a r i a s ela-
apos, pero l a i n t e r n a c i o n a l adop-
-
tada p o r l a Federación i n t e r n a c i o n a l de Ginecologfa y Q b s t e t r i
c i a (FIGO) se usa amp1iameni;e hoy en día.
En e l sistema de
'<
c l a s i f i c a c i ó n por etapas uti.iizado en e l informe h u a i de l o s
Resultados d e l Tratamiento d e l Carcinoma d e l Utero
b a j o e l p a t r o c i n i o de l a Federación internacional.
JC
publicado
-
Las pacien
t e s s e c l a s i f i c a n antes de someterlas a n i n e tratamiento u t i
-
l i z a n d o s610 l a s t é c n i c a s d i . a g n 6 s ~ i c a sd i s p o n i b l e s comunmente,
como b i o p s i a , examen b a j o anestesia, c i s t o e o o p i a , proctoscopia
r a d i o g r a f f a s , pielograma intravenoso y enema de b a r i o .
Para
e s t a b l e c e r l a etapa de una n a c i e n t e dentro de l a c i a s i d i c a c i 6 n
i n t e r n a c i o n a l no s e permite l a laparotomia exploradora con b i o
p s i a de ganplios l i n f á t i c o s y l i n f a g i o g r a f í a , dos t é c n i c a s de
uso comun algunas partes.
E l l o no s i g n i f i c a , por supuesto,
que no deban hacerse e s t a s pruebas d i a g n o s t i c a s n i oue no s e
consideren l o s resultados a l planear e l tratamiento.
Sino,
-
simplemente, que no pueden considerarse en l a c l a s i f i c a c i ó n o f i
c i a l por etapas de las pacientes con cáncer c e r v i c a l .
Se ha reconocido ampliamente que hay un e r r o r i m p l i c i t 0 en
e s t e sistema c l i n i c o de c l a s i f i c a c i o n por etapas.
V e n Nagel ha,
encontrado que e l e x w e n c l i n i c o s610 da un 7596 de p r e c i s i o n
cuando s e compara con l a c l a s i f i c a c i ó n p o r etapas en exáaenes
b a j o anestesi.
En un grupo de 1 2 3 pacientes que s e sometieron
posteriormente a l a c l a s i f i c a c i ó n por etapar quirúrgicas,
el
examen c l i n i c o l a e s t a b l e c i o correctamente s610 en e l 54% de l a s
ocasiones, y e l examen b a j o anestesi en e l 74% de l a s pacientes
E l l o depende en g r a n p a r t e de l a s metástasis en g a n g l i o s i i n f & -
t ra
t i c o s p é l v i c o s que sólo pued.en d i a g n o s t i c a r s e mediante l a p a r o t o
mis, y y a que e e t o s h a l l a z g o s no i n f l u y e n actualmente en e l
tamiento, s e aonsidera que l a laparotomia exploradora no e s un
método n e c e s a r i o o s i s t e m á t i c o para la c l a s i f i c a c i ó n por etapas.
Meiga y Brunschwig han sugerido un sistema q u i d r g i c o para l a
c l a s i f i c a c i 6 n por etapas de iJacientes tratadas en forma prima
-
r i a con c i r u g x a r a d i c a l , pero no s e ha u t i l i z a d o extensamente.
-
Toda p a c i e n t e con c h c e r c e r v i c a l debe someterse a l a v a l o
raci6n d i a e ó s t i c a adecuada y s e r c l a s i f i c a d a fomalmewte antes
de i n i c i a r e l tratamiento.
-
Una vez e s t a b l e c i d a l a etapa, es PO
s i b l e que no se cambie a medida que sukgen nuevos acon8ecimient o s en e l curso d e l tratamiento o de l a v i g i l a n c i a U t e r i o r .
TRATAMIENTO DEL CARCIñOKA IHTRBEPITELIAL.
EL tratamiento d e l carcinoma i n s i t u d e l c u e l l o e s t 4 l e j o s
de haberse estandarizado y en l a d e c i s i ó n t e r a p é u t i c a i n f l u y e n
muchos f a c t o r e s .
Aunque l a r a d i o t e r a p i a r a r a v e z s e usa, e s un
tratamiento e f i c a z y puede considerarse para enfermas que q u i z a
no t o l e r e n l a c i r u g i a por su estado médico y su edad.
-64-
C l a s i f i c a c i ó n p o r etapas q u i r u r g i c a s d e l carcinomsc c e r v i c a l
se&
Melga
- Brunschwig.
Clase O
Carcinoma i n s i t u , también denominado carcinoma preinvasor
o i n t r a e p i t e l i a l , o microcarcinoma.
Clase R
El carcinoma s e h a i l a estrictamente l i m i t a d o a l c u e l l o .
Clase Ao;
DespuBs de una b i o p s i a p o s i t i v a para carcinoma i n f i l t r a n t e
no s e descubre tumor c e r v i c a l en l a p i e z a q u i d w g i c a .
Clase B
B1 carcinoma s e extiende partiendo d e l c u e l l o para abarcar
l a vagina, exceptuando su t e r c i o i n f e r i o r .
extiende por e l cuerpo.
na y cuerpo.
El carcinoma s e
Puede a f e c t a r p a r t e a l t a de vagi-
La extensión v a g i n a l , l a u t e r i n a , o ambas,
pueden s e r directamente o por metástasis.
Clase C
E l carcinoma ha a f e c t a d o t e j i d o s p a r a c e r v i c a l e s , paravagi-
n a l e s , C ambos, por extensi6n d i r e c t a por vasos l i n f á t i c o s
o en g a n g l i o s i n t r a t i s u l a r e s .
Hay metastasis v a g i n a l , exten-
si6n d i r e c t a , o ambas, en e l t e r c i o i n f e r i o r de vagma.
Clase D
Vias y g a n g i i o s i i n f á t i c o s afectados mas a l i a da l a s r e
giones p a r a c e r v i c a l 3 paravaginal.
-
Ello i n c l u y e todos los
vasos l i n f á t i c o s , g a n g l i o s , o ambos, en l a p e l v i s verdader a , excepto en l a forma correspondiente a l a c l a s e C.
t a s t a s i s en o v a r i o o trompa.
--h
s-
Me-
C l a s i f i c a c i o n por etapas d e l carcinoma c e r v i c a l adoptada p o r
l a Federación Internaciona2 de Gineaología y O b s t d r i c i s (FIGO)
Carcinoma preinvasor
Etapa O
Carcinoma i n s i t u , carcinoma i n t r a e p i t e l i a l .
Los casos en etapa O no deben i n c l u i r s e en ninguna e s t a d i s t i c a t e r a p é u t i c a de carcinoma invasor.
Carcinoma invasor
Etapa 1
Carcinoma estrictamente c i r c u n s c r i t o a l c u e l l o
(no s e debe tomar en cuenta ninguna extensión
a i cuerpo u t e r i n o )
Etapa l a
Carcinoma microinvasor ( i n v a s i ó n precoz d e l eat roma)
Etapa l b
Todos l o s demas casos de l a etapa 1. El cáncer
o c u l t o debe marcarse
Etapa 11
r
t
~
~
n
E l carcinoma se extiende mas a l l á d e l c u e l l o ,
-
pero no ha l l e g a d o a l a pared p é i v i c a . EL c a r c i
noma a f e c t a a l a bagina, pero no a su t e r c i o i n
-
feriar.
Etapa lla
No hay a f e c c i ó n e v i d e n t e d e l parametrio
Etapa llb
Evidente invasi6n d e l parametrio
Etapa 111
El carcinoma s e ha extendido a l a pared p é i v i c a
E l examen r e c t a l r e v e l a que e n t r e e l tumor y l a
pared p é l v i c a no hay nin&
chcer.
espacio l i b r e de
E l tumor abarca e l t e r c i o i n f e r i o r de
l a vagina.
Deben i n c l u i r s e todos los casos con
h i d r o n e f r o s i s o ririon carente de gunción, a menos que s e sepa que s e deben a o t r a causa.
Etapa l l l a
E l carcinoma no s e ha extendido a l a pared pei-
vica.
- 66-
Etapa lllb
Extensión a l a pared p é l v i c a , h i d r o f r e n o s i s o
riñón carente de función.
Etapa 1 V
EL carcinoma se ha extendido mas a l l á de l a
p é l v i s verdadera o ha afectado clinicamente l a
mucosa de l a v e j i g a o e l recto. Un edema v e s i cuioso en s i no permite i n c l u i r e l caso en l a
etapa i V .
Etapa 1Va
Extensi6n de l a n e o p l a s i a a los órganos adya
-
cates.
Etapa 1Vb
Extensión de l a n e o p l a s i a a órganos distantes.
& l a s p a c i e n t e s t i p i c a s que han completado su f a m i l i a y
-
no ob j e t a n l a h i s t e r e c t o m i a , l a i n t e r v e n c i ó n q u i n í r g i c a acepta
b l e para e l carcinoma i n s i t u debe s e r l a h i s t e r e c t o m i a t o t a l ,
puede p r a c t-i
con cuidado de e x t i r p a r todo e l c u e l l o y conservar en l a s muje
r e s jovenes uno o ambos o v a r i o s .
La i n t e H e n c i 6 n
c a r s e por l a s vias abdominal o v a g i n a l .
La v a l o r a c i 6 n preope-
r a t o r i a cuidadosa mediante l a oolposcopia y l a prueba de S c h i 4 :
l l e r delinean l a extensión de l a l e s i ó n y determinan l a variac i ó n en l a cantidad de l a cúpula v a g i n a l , s i acaso alguna, que
-
j
c
L
debe e x t i r p a r s e .
S i l a p a c i e n t e desea conservar su p o t e n c i a l de reproduc-
atón u o b j e t a l a h i s t e r e c t o m i a por c u a l q u i e r motivo, pueden
p r a c t i c a r s e conización, c a u t e r i z a c i 6 n o c r i o t e r a p i a , como a i
t e r n a t i v a s a l a histerectomia.
-
Los p r o s y contras de e s t o s
métodos s e comentan despúés y es importante que l a paciente
-
p a r t i c i p e an e s t a d e c i s i ó n , que s e l e debe e x p l i c a r l o mas c l a
ramente p o s i b l e .
Conieación
La c o n i z a c i ó n como tratamiento s e ha empleado mas que l o s
o t r o s dos métodos simples y fue usada primero en l u g a r de hist e r e c t o m i a para p r e s e r v a r l a f e r t i l i d a d .
Muchos estudios en
l a s decadas de los d n c u e n t a s y a l p r i n c i p i o de l o s sesentas,
antes de que aumentara e l empleo d e l colposcopio para d e l i n e a r
-
cuidadosamente los i f m i t e s de l a l e s i ó n , mostraron que l a c o n i
zación t e n i a una f a l l a de aproximadamente 20% como pudo compro
-
barse por los f r o t i s subsecuentes a l a conización, o por $umo-
--i
L
r
.-
rea r e s i d u a l e s en p i e z a s q u i r ú r g i c a s de histerectomia.
Estu
-
-
d i o s mas r e c i e n t e s , por ejemplo de E r i e g e r y McCormack, han de
mostrado una f r e c u e n c i a menor (9%) de enfermedad r e s i d u a i después de l a conizacibn, debido a e x p l o r a c i ó n p r e o p e r a t o r i a cuidadosadel c u e l : ~ oy a que no e x i s t e duda en cuanto a l a exten
-
*-
-68-
s i ó n de l a s l e s i o n e s .
Una de l a s v e n t a j a s de l a c o n i z a c i ó n e s que conserva l a
fertilidad.
S i n embargo, l a c o n i z a c i ó n p o r s i mismo p u d e te-
n e r e f e c t o adverso sobre l a función reproductiva.
MarVicar y
W i l l o c k s estudiaron c e r c a de 632 p a c i e n t e s que s u f r i e r o n c o d zacion con a n t e r i o r i d a d .
Eiicontraron una f r e c u e n c i a de f e r % i -
-
l i d a d de 64 $, pero aunque quedaron embarazadas, 2 4 $ presenta
ron aboryo o p a r t o greniaturo.
-
Otro estudio de Kullander y S j o
b e r g demostraron e x c e l e n t e s p o r c e n t a j e s de f e r t i l i d a d despues
de l a conizaci6n c e r c a de 83 $, pero de nuevo s e presentaron
aborto y p a r t o prematuro en 21 $ de l o s casos.
U t i l i z a n d o co-
mo t e s t i g o s l o s antecedentes de l o s embarazos de l a p a c i e n t e
antes de l a conización, B j e r r e y colaboradores señalaron un 17.
-
1 $ de aborto4 espontaneos despues de l a b i o p s i a en cono en com
paración con e l 11.6 $ antes de l a conización.
La f r e c u e n c i a
de premadurez f u e identAca en ambos grupos.
La p r i n c i p a l complicación de l a b i o p s i a en cono es l a hemorragia.
V i l l a s a n t a y Uurkan señalaron que en su s e r i e un lO$
de 200 p a c i e n t e s tuvo hemorragia posoperatoria que n e c e s i t 6
t r a n s m s i b n de sangre en e l 3.5
% de l o s casos.
hay i n f e c c i ó n p é l v i c a y e s t e n o s i s c e r v i c a .
En ocasiones,
Eh p a c i e n t e s embara
-
zadas hay e l p e l i g r o de i n d u c i r e l parto.
Cauterizaci6n y c f i o c i r u g i a .
El t r a t a n i e i i t o de a t i p l a c e r v i c a l y carcinoma i n t r a e p i t e i.-
-
l i a s l por c a u t e r i z a c i o n o congelacion s e ha d e s a r r o l l a d o mas r e
cientemente y t i e n a l a granventaja de que puede r e a l i z a r s e en
p a c i e n t e s externas.
Los r e s u l t a d o s a l a r g o p l a z o no e s p o s i b l e v a l o r a r l o s a h ,
pero s i pal timos de f r o n t i s p o s i t i v o s p e r s i s t e n t e s después d e l
-
t r a t a n i e n t o , Wilbanks y cola,boradores encoritraron que con caute
.- ...
r i z a c i ó n l a a t i p i a f u e curada en 84 $ de l a s pacientes.
-
Existen muchos defensores de l a c r i o c i r u g i a para t r a b a r neo
p l a s i a s i n t r a e p i t e l i a l e s de c u e l l o u t e r i n o .
-
Creasman y colabora,:!
dores p u b l i c a r o n un a r t i c u l o sobre l a p e r s i s t e n c i a d e l tumor en
p i e z a s q u i r ú r g i c a s histerectomizadas de 75 p a c i e n t e s con b i o p s i a
que mostró caccinoma i n s i t u o d i s p i a s i m severa.
Con l a congela-
c i ó n doble, 82 $ de l o s e pecimenes no mostraron tomor r e s i d u a l ,
pero s i s e empleo una s o l a congelaci6n hubo p e r s i s t e n c i a d e l tu-
mor de 48 $ en l a s p i e z a s qui.nirgicas extraidas, Kaufman e Irwin
encontraron un 5.8 $ de f r a c a s o s en 238 pacientes tratadas con
c r i o t e r a p i a después de uALraspado endocervical n e g a t i v o , pero un
20
$ de f r a c a s o s en 48 mujeres en quienea e l respaldo d e l endo
c e r v i x mostró c é l u l a s anormales.
c
i
i
c
P
Como i o s e n d a Townsend, e l
ta
malo de l a l e s i ó n tambien s e r e l a c i o n a directamenTe con e l i n d i c e de e x i t o s de l a c r i o t e r a p i a .
-
Sevin y colaboradores han p u b l i
cado ocho casos de carcinoma i n v a s o r d e l c u e l l o despues de c r i o c i r u g i a d e l misino.
Probablemente dependieron de una v a l o r a c i ó n
pretratamienGe inadecuado y tambien f r a c a s o d e l método, pero es
importante no o l v i d a r l a
os:tbJidad de p e r s i s t e n c i a o recurren-
L_
c i a de una n e o p l a s i epite1ia:L despues de e $ t e o c u a l o i d e r o t r o
P
tratamiento, incluyenco l a h i s t e r e c t o m i o por carcinoma i n s i t u .
Ha hahido muchas c o n t r o v e r s i a s , algunas de e l l a s acaloradas
i
-
i
sobre los me*itos r e l a t i v o s ide l a c a u t e r i z a c i o n y l a congelacion
-
Sin embargo, en base a l o s datos publicados en un i n t e r e s a n t e es
t u d i o comparati o de M i l l e r y E i s t e i n concluimos "no s e comprobó
a i f e r e n c i a n i en e l número de curaciones sintomáticas n i en e l de
curaciones o b j e t i v a s "
L
.
BI res..imen, l a s pruebas r e c i e n t e s parecen mostrar que momo
tratamiento d e f i n i t i v o y con menos problemas cie c o n t r o l p o s t e r i o r
...
r
L
c
e l método de e l e c c i ó n s i g u e ,siendo l a n i s t e r e c t o m i a ya sea abdominal o vaginal.
Sin embargo, para l e s i o n e s de bordes definirlos
o cualido es de g r i m o r d i a l importancia p r e s e r v a r l a f e r t i l i d a d ,
l a conizaci6n, c a u t e r i z a c i ó n o congelación pueden emplearse como
métoaos de substitución.
S i n importar e l metodo seleccionado,
t a n t o l a p a c i e n t e eomo e l medico deben e s t a r de acuecdo sobre e l
periodo prolongado de v i g i l a n c i a .
La transformacibn n e o p l á s i c a
que o c u r r i ó una vez en e l c u e l l o puede i r e s e n t a r r e de nuevo, so-
-
b r e todo s1 hay una zona de transformación a c t i v a , y l a unica ga
r a n t i a contra e l d e s a r r o l l o de un cancer i n v a s o r es l a vigilanc i a continua.
R E S U M E N
La c i t o l o g i a e x f o l i a t i v a representa e l método más p r á c t i c o
Aunq.ue e l i n d i
para l a detección temprana d e l cáncer c e r v i c a l .
- 20
c e de f a l s a n e g a t i v i d a d es extraordinariamente a l t o (10
q6)
en una s o l a muestra, é s t e puede s e r considerablemente disminuído por medio de r e p e t i c i ó n de l o s f r o t i s c i t o l ó g i c o s .
Eh presencia de c i t o l o g i a anormal se debe l l e v a r a e s t o
evaluación u l t e r i o r d e l c é r v i x .
l a
La prueba de S c h i l l e r der4 re-
sultados c o n f i a b l e s s o l o en un b a j o porcentaje de pacientes en
riesgo.
Con l a t é c n i c a de l a b i o p s i a c e r v i c a l c i e g a de cuatro
cuadrantes e l e r r o r d i a g n ó s t i c o estimado e s p o r encima d e l 20 $.
Sin embargo, cuando s e u t i l i z a l a prueba de S c h i l l e r para d i r i g i r l a s b i o p s i a s c e r v i c a l e s y cuando l o s especfmenes de b i o p s i a s
m f i l t i p l e s s e examinan con numerosos c o r t e s histo16gicos,
puede obtener mayor c e r t e z a .
La conización d e l
t a un método c o n f i a b l e de evaluación c e r v i c a l
C6rViX
, pero
se
-
represen
su c o s t o
-
e s elevado y e l í n d i c e de complicaciones es importante, en espe
c i a en l a s gestantes.
La coiposcopia es un método p r á c t i c o ,
barato, c o n f i a b l e , de b a j a niorbilidad, e x c e l e n t a para evaluar e l
c é r v i x de mujeres cuya c i t o l o g i a e s anormal.
De acuerdo a l o s
h @ l l a a g o s colposcópicos l a s pacientes se d i v i d e n en 3 grandes
grupos$
1) Normales.-
Sólo s e r e c o m i a i i a c o n t r o l cito16gico:
sólo en l o s
casos en qe l a c i t o l o g i a s e presente repetidamente sopechosa o
s e transforme en p o s i t i v a ,
e s t a indicada &a conización.
2 ) Coiposcopia " i n s a t i s f a c t o r i a " . -
La c o n i z a c i h e s t á indicada
para d e s c a r t a r n e o p l a s i a s i l a c i t o l o g i a e s p o s i t i v a o peresisten tement e sospechosa.
3) Anormales.-
Siempre e s t á indicada l a b i o p s i a b a j o c o n t r o l
-
colposcópico; s i l a l e s i ó n s e extiende dentro d e l c a n a andocer
v i c a l e s t á indicada l a conización; S i e l d i a g n ó s t i c o de l a b i o g
s i a d i r i g i d a e s de d i s p l a s i a l e v e o moderada, s e puede a p l i c a r
un tratamiento conservador; cuando s e t r a t a de d i s p l a s i a s e v e r a
o carcinoma i n s i t u s e puede a p l i c a r e l tratamiento d e f i n i t i v o
( conización t e r a p k t i c a , h i s t e r e c t o m i a ) s i n necesidad de coni-
eva -
zación diagnostica;. cuando e l d i a g p ó s t i c o e s de carcinoma mi
croinvasor, l a conizaci6n diagnóstica e s t á indicada para
l u a r exacta.mente l a extensián y profundidad de l a invaeión.
ih e l caso de cáncer invasor, e l tratamienfio dependerá d e l es-
t a d i o c l i n i c 0 de l a enfermedad.
Uno de l o s usos más v a l i o s o s de l a colposcopia e s l a cert e z a d i a g n ó s t i c a y m w j o de l a s pacientes embarazadas con c i t o l o g l a anormal.
a -
Las p a c i e n t e s embarazadas son e x c e l e n t e s can
d i d a t o s para l a evaluación coiposcópica de un Papanicolaou
normal, ya que l a e v e r s i ó n f i s i o l ó g i c a que s e presenta en e s t e
estado permite muy b i e n l a visuaLizaci6n de l a unión escamocolumnar completa.
ih esta€! pacientes, cuanto l a b i o p s i a d i r i
-
g i d a por colposcopia encuentra d i s p l a s i a severa o carcinoma i n
s i t u , e l tratamiento d e f i n i t i v o puede posponerse hasta después
d e l pai%o, e v i t á n d o l e s los r i e s g o s de l a conización en estado
de gestación.
-13-
B I'B L I O G R A F I A
-
-
STAFG, A. ERIERICH ,J.R.E. G. & MATTYNGY, RF.
Detection of cervical Neoplasia
Reducfmq the risk of
error; Clinical Obstetrics & Ginecoloqy. pp 16
22; 238-260.
June 1973.
-
-
DR. GRETCHE HINDERSMAN
PRACTICA Privada de Obstetricia y Ginecoloqía, Pruebas sistematicas de Papanicolaou. Mundo Médico pp. 81 - 85. Junio 1986.
DR HOWARD W. JONES, DRA. GEORGEONNE SEEGAR.
.Tratado de Ginecoioqfd
pp 299
332.
.
-
- MEHRAN
M.
The impact of diagnostic ultrasound on the prediction intrauterine
growth retardation in devolopinq countries.
Int. J. Ginecoloqy
Obstetrics Vol. 26 1988 pp 315 - 378.
-
-
FRANK H. NETER
Coleccidn Ciba de ilustraciones Médicas.
Tomo I1 Sistema Reproductor.
-
WEBSTER JHON G.
Encyclopedia of Medical Devices.
-
DR. PINO JUAREZ VERGARA, DRA, E MEZA BAND.
Cáncer Cérvico Uterino, Histología Normal y Citología
Exfoliativa. pp 13-15 y 19-22.
-
WYNN
-'
W I T / LAWRENCE M. ELSON
Anatomfa Cromodindmica pp 96-98. Ferndndez Editores.
CANCER. A Journal of the American Cancer Society
'Vol 64 pp 2069- 2075, 2104
2109.
Vol 61 pp 679-687.
-
r
Human Papillomaviruses and the Pathogenesis
of Cervical Neoplasia
-
A Study by In Situ Hybridization
c-
t
t
t
t
C
JEAN W. GUPTA. PHD:.t
KAORU SAITO. MD,S AKEMl SAITO. MD,S
YA0 S. FU, MD.S AND KEERTl V. SHAH, MD. DRPH'
1
in a previous topographic shdy ofcenial conization specimens, cvndylommus changes were commonly
present III direct contact with inhaepitheüd neophsh dwere 8iwaya b a t e d dist8J (ectocervial side)
to the neoptuh.Viral DNA was detected by irsita hykidintba using "S-hbekd nick mmhted DNA
probn in 50 of '10 asea (71%) which bid adequate IesiOW: HPV-16 in 30, HPV-18 in ten, HPV-31 III
six, dmultipk types in fwr cases. HPV4/I1
detected wly once, in a multipk infeaiOii. As a ni*,
s positive cervix contained a single v h s type, dthe same v i m type m a found in mudyiwiitcui d
neophetic ueas. Thc results tsthat the neoplutic process is initiated in the Mof condyloma
toward the endocervix d,
once established, extends proximally toward the cervical cand. Capsid aotigen
was detected b 19 aseq hdicntiw that a proportion of thee-h
lesions ispotentiilly infecüom.
Cancer 64:2104-2110, 1989.
t
(
S
a
c
I
S
I
3
H
UMAN PAPILLOMAVIRUSES (HPV) of the genital tract
have been recovered from tissues showing a wide
spearum of pathologic lesions as well as from histoiogically normal tissues.' HPV-6 and HPV-I I have been associated predominantly with condylomas and HPV-I6
and HPV-18 with high-grade intraepithelial neopiasias
and invasive cancers.' HPV-31 is associated with both
mild and severe lesions.' Cervical intraepithelial neoplasia
(CIN) frequently coexists with condylomatous changes in
affected cenices.'" We have examined ceMcai conization
specimens removed for the diagnosis and treatment of
high-grade cervicallesions in an attempt to define the role
of human papüIomavinises in the pathogenesis of ceMcai
cancer. in an earlier report, the topographic relationship
of condylomatous changes to CIN was studied by an examination of 101conization specimens! More than 90%
of C I N and condyloma lesions were found within, or
trans$rrssing,the transformation zone. Seventy-five per-
cent ofthe tissues contained both CIN and condylomatous
lesions. Condylomatous changes were invariably distal
(toward the ectocervical side) to CIN and in a majority
of cases were in direct contact with CIN.
In this report, we describe the virologic correlation of
the topographic findings. Paraffin sections of the conization specimens were examined for the prevalence and distribution of HPV D N A by in silu hybridization, a technique which is well suited to correlate the morphologic
features of the lesions with the presence and type of HPV
DNA. The sections were also examined for HPV capsid
antigen. W e found that despite the high frequency ofcondylomatous changes assoCiated with the high-grade lesions
in cervical conization specimens, HPV-6 and HPV-I I
were virtually absent in these tissues. HPV-16, HPV-18,
and HPV-3 I were identified from a majority ofthe lesions.
As a Nk,the areas with different grades of seventy in a
conization specimen displayed the same HPV type. Also,
about a fourth of the tissues contained HPV capsid antigen, an indication that some of these lesions were potentially infectious.
I
t
2
I
S
t
I
\
I
1
\
t
t
c
c
-
Materials and Metbods
Coniralion Specimens
In the topographic study described previously> 101
conization specimens removed between 1981 and 1983
for ceMcaI dysplasia and carcinoma in situ at the Univer& of California, Los Angeles (UCLA)Medical Center.
2 104
-
HPV AND CERVI<:AL NEOPLASIA
No. IO
TABLEI. Rcqwney ofDe<ation of Human Rpülomavinu DNA
d Canid Antinen in Gmid Conization T k u s
No.mcd
No. &tive
1%)
~
HPV DNA
HPV ppvd pnIig.cn
HPV DNA or in&m
HPV DNA and iaIig.cn
Only HPV DNA
Only HW pnIig.cn
70
67
70.
70.
70
67
u)(71%)
19 (28%)
52 (74%)
17 (24%)
33 (47%)
2 (3%)
-
H P V humin papillmavinis.
Sixtylcvcn tisuer tmtcd for capsid anIig.cn
were analyzed. The ages of the women ranged from 2 0
to 89 years (mean age, 36). These specimens were recut
(233 blocks, one-three blocks per cone) to obtain paraffin
sections for in situ hybridization, immunoperoxidase test
and hematoxylin and eosin (HdrE) staining. Pathologic
diagnosis was made by criteria desCnbed previously6 and
is based on the most severe lesion found in the examined
sections.
in Situ Hybridization
Serial sections on sepanite slides were hybridized with
35Slabeled, nick translated whole genomic HPV DNA
probes. pBR vector DNA probe served as a negative control. Viral DNA probes of HPV-á+I I (mixed), HPV-I6
and HPV- I 8 (all.prepared from clones kindly supplied by
Dr. H. zur Hausen) and HPV-31 (from a clone kindly
supplied by Dr. A. Lorinn) were employed. Sections of
tissues known to be positive for HPV-6, I 1 and 16 were
included in each test as positive controls. Hybridization
was performed as described previously: with the following
modifications. Paraffin sections were cut onto organosilane-treated siides'; this treatment almost completelyprevented loss o f sectionsduring the procedure. Sectionswere
treated with a reduced concentration (IS &mi) of Proteinase K solution at 37°C for 15 minutes. The reduction
of Roteinax K concentration improved the preservation
of morphologic features while still allowing adequate
2105
Gupta el a/.
probe penetration of the tissue. Slida were hybnáized
overnight insiead of for 2 days. The final hi&-tcmperaturc
washes at 55OC were performed with 0.25 X SSC instead
of 0.5 x SSC. The increased stringency reduced nonspecific background hybridization as well as cross-hybridization between related HPV and permitted distinction between HPV-16 and HPV-31. Hybridized sections were
scored for the presence and type of HPV DNA, intensity
and location (neoplastic or condylomatous ana) of signal,
hybridization pattern (diffuse, focal or mixed), and for
basal cell hybridization. No attempt was made to distinguish between HPV-6 and HPV-I 1.
Test for Capsid Antigen
The immunoperoxidase test for capsid antigen was
performed using anti-BPV polyclonal antibody (Dako
Corp., Santa Barbara, CA) and the anti-rabbit igG ABC
kit (Vector Laboratories, Burlingame, CA) as described?
This polyclonal antibody prepared by immunization with
disrupted viral capsids recognizes capsid antigens o f
all HPV.
Results
Sections from one to three blocks of each conization
specimen were examined for these studies. Tissues available forthis study from 31 ofthe 101conization specimens
contained only normal epithelium, or no epithelium, or
mere traces of atypical epithelium. They were uniformly
negative for HPV DNA and capsid antigen and are excluded from further analysis. Ofthe 7 0 tissues which displayed adequate pathologic lesions and were examined
for HPV DNA, 67 were also tested for HPV capsid antigen. Human papillomavirus DNA and HPV capsid antigen were detected, respectively, in 71% and 28% o f the
tissues (Table I). Evidence o f HPV infection by the pmence of either DNA or antigen was demonstrable in 74%
o f the tissues. Only HPV DNA and only HPV antigen
were detected, respectively, in 47% and 3% o f the tissues.
TABLE2. DirVibution of Human PapillomavinisTyps by Moat S e v m Lepion REnt
No. of
DWOSiS'
FBSJ
Condyloma only
CIN I
CIN II
CIN 111
Tntal
_ _ _ _ ~ ~
~~~
~~
~
5
2
14
No.of DN k
positive ( a )
3
I
lO(71)
49
36 (73)
7n
5nr7ir
HPV type
6/11
16
O
I
O
O
I
O
2
26
O
M
18
31
Multiplet
I
O
O
O
5
4
5
I
O
2
I
10
6
4
I
~~
H W . human ppiuomawus; CM.emspl intncpthclipl neoplas¡r
* The most severe laion in the hybriduÍd d o n s .
t The multiple infdom wre: HW-18 and HW-31 in a condyloma:
HPV-611 I uid H P V J I , and HPV-16 and HPV-18 in two cssco ofCIN
I t and H W - I 6 and HW-18 in one CIN Ill.
-
2106
CANCER
November I S 1989
vol. 64
No.
Distribution ofHuman Papillomavirus Types
by Lesion Pathologic Features
Almost all conization specimens contained lesions o f
different grades of severity. The pathologic diagnosis ofa
tissue was based on the most severe lesion Seen in the
hybridized sections. The most frequent diagnosis was CIN
111,which accounted for 49 of the 7 0 (70%) cases (Table
2). Human papillomavirus DNA was found at about the
same frequency in tissues with a CIN 1
1
1diagnosis as in
lower grade lesions. A single HPV type was dein
46 of the 50 HPV-positive tissues; HPV-I6 in 30, HPVI 8 in ten, and HPV-31 in six tissues. The HPV-6/11 was
not detected in any of the cases which were infected with
a single HPV type. Four tissues showed double infections
(Table 2). In the study, HPV-6/1 I was identi6ed only
once, present along with HPV-3 I in a CIN I1 lesion. Both
HPV-I6 and HPV-31 appeared to be more frequent in
CIN I11 lesions, relative to HPV-18.
Both condylomatous cells and CIN coexisted in most
o f the tissues. In most instances, there was either a gradual
transition from condylomatous cells to CIN or condylomatous cells were contiguous to CIN. As a rule, in a
DNA-positive ceMx a single viral type was locahed in
both pathologic types. In 13 DNA-positive cases, there
were separate discrete lesions in the tissues which were
more than two histologic grades apart. The results of HPV
typing and capsid antigen tesis for each of these lesions
are shown in Table 3. In five cases, the same HPV type
was present in both lesions and in two additional cases at
least one HPV type was found at both sites. In four cases,
only one site was DNA-positive. In Case 79 and probably
TABLE3. Virologic Findingr in Cavs Showing Discrele Area Two
or More Hi?.IoI&e O d e s Ame
Condyloma or
atypical
condvlomala
CIN I1 or CIN 111
No.
I
2
3
4
5
6
7
8
9
IO
II
12
13
Case
4
9
18
20
26
33
39
66
61
70
78
79
91
DNA
HPV-I6
.
.. . .-
NCS
HPV-18
HPV-I6
HPV- I8
HW-I6
HPV-16t18
Ns
HPV-I6
HPV-I6
H P V W I t31
HPV-I6
HPV-16
AP
-
t
-
DNA
FIGS. IA AND 18.A condyloma in a conization opcimen. (A) H &
E, X312. (B)focal hybridization with HPV-16, X312. This specimen
also displayed HPV capsid antigen.
(H
an
Case 67, different HPV were present at the two sites. Thus,
the data suggested that in a majority o f the cases, histologically diverse and spatially discrete lesions in a tissue
were caused by the same HPV type. In these 13 cases,
viral capsid antigen was detected in one o f CIN and eight
of the condylomatous lesions.
cr
flc
Characteristics of Hybridization
h:
Ap
NHFV-3 I
HPV- I8
HPV-I6
HPV- I8
NS
HPV-18
HPV-18
NCS'
HPV-I6
HPVJI
HPV-18
HPV-I6
H P V human p p ü b m v i n u ; Al: antigen.
* A n HPV typ n a repremu in the pobamidc<sted in uiu
twe by nonrviiyat hybridization (unpubürhrd diu).
Hybridization was seen only in the altected epithelium
and never in the normal epithelium adjacent to the lesion
or in the stroma. Intensity of hybridization signal ranged
from banly deteciable over background to extremely
strong. It is probable that very faint signals were masked
by the background, which varied from test to test. Lesions
associated with HPV-3 I and with mixed infections showed
relatively stronger signals.
Two distinct patterns of hybridization, focal and diffuse,
were evident. The focal patternwas d t e d with signals
of moderate to high intensity, and with antigen-positive.
lowgradc lesions (Figs. IA and 18, and 2). It occurred
nt
al
o1
hi
N
di
ai
ir
al
lT
5
L
ti
ii
I<
fi
No. 10
HPV AND CERVICAL NEOPLASIA
2107
Gupra 91 01.
TABLE4. com*iion of Hybridization Panmi and Human
PaoillomavirusT w e
TOUI
Hybridization
DNA
positivc
HPV type
cases
6/11
16
18
31
Multiple
Mired
16
17
O
O
o
a
9
I
4
1
Total
50
O
30
3
3
0
6
4
O
5
Dsttem
Focal
m
u
s
e
17
3
5
IO
I
HPV human papillomavirus.
and 11% of DNA-negative tissues (Table 6). Productive
infection was found in a significant proportion of tissues
infected with HPV-16. HPV-18 and HPV-31.
Discussion
FIG. 2. Faal hybridization in a CIN I lesion with HPV-31 probe
(Hematoxylin stain, x400). This specimen alza diwlayed HPV capsid
antigen.
near the epithelial surface, with the signal strength increasing toward the surface. This pattern probably reflected productive viral infection. The diffuse pattern usually displayed faint signals distributed uniformly throughout the depth of the epithelium and was associated with
high-grade antigen-negative lesions (Figs. 3A and 38).
Many tissues displayed a focal pattern in one area and a
diffuse panern in another (Fig. 4). The diffuse pattern of
hybridization was relatively more frequent with HPV-I6
and HPV-18 infections than with HPV-31 and multiple
infections (Table 4). Basal cell hybridization was seen in
about one third ofthe DNA-positive cases and was found
relatively more frequently in HPV-I8 infections (Table
5). It was usually associated with CIN 11or CIN Ill lesions.
Dislribution of Capsid Aniigen
The presence of capsid antigen is indicative of producs
The antigen was detected most often
tive v i ~ infection.
in condylomatous areas, infrequently in CIN I or CIN II
lesions and rarely in CIN 111 lesions. The antigen was
found in 28%of all tissues, 35% of DNA-positive tissues
Overall, this study investigated the role of HPV infection in the pathogenesis of cervical neoplasia. In the topographic analysis reported earlier? 68% of C I N lesions
were found to be in direct contact with condyloma or
atypical condyloma. Condylomatous areas were always
located distal to the C I N areas, and koilocytosis was pres
ent in 90% and 78%. respectively, of CIN K I N I1 and
CIN 111.This observation confirmed the results ofprevious
investigators that in a cervical lesion the les severe pathologic type is located distal to the more severe pathologic
type.'Gl2
A major aim of the current study was to examine if
condyloma and CIN were different manifestations of the
same underlying process. The in siiu hybridization test
was an excellent tool for this purpose because it allowed
localization of genome to specific cells and therefore permitted a correlation between presence of vinis and cellular
pathology.'3-" The test for capsid antigen was useful in
two ways. First, it determined to what extent these highgrade lesions were still producing full viral particles. Second, the number of antigen-positive DNA-negative tissues
gave some indication of the extent to which infections
TABLE5. Correlation %tween Frequency nf Basal Cell
Hybridization and Human PapillomavirusTypcs
Total DNA
basalcell
hybridization
Positive
caws
Present
Absent
33
11
O
O
Total
50
O
HPV type
6/11
HPV human papillomavirus
2
7
30
31
Multiple
6
4
3
3
4
10
6
5
18
16
3
I
..
.,
2108
CANCER
November I5 1989
--7
Vol. 64
Ni
Al
DI
LT
U
CI
CI
th
st
K
tt
el
si
t>
tr
hos. 3A ANO 3B. A CIN 111!esion h y b r i d ¡ with HPV-18 probe. (A) Low power view (hematoxylin stain, XIS6). (B)Higher magnification,
displaying Faint, d i b hybridization(hematoxylin stain, X W ) .
O
a:
tl
with HPV types not represented in the probe panel were
involved in cervical neoplasia.
Sixty-three of 70 tissues which had adequate pathologic
features contained high-grade lesions (CIN I1 01 CIN Ill)
-
Flo. 4. WII
d8 CIN II-CIN 111laion hy'Iac
is íocilllcu tkc
sad
mamxyün 6%
~250).
to M HPV-IO
below (he-
and most of them also contained condylomatous cells.
Virologic evidence for the presence of HPV was obtained
in 53 of the 70 (74%) specimens. The two specimenswhich
contained capsid antigen but no HPV DNA very likely
reDresented infection with HPV tvoes not included in the
probe panel. They constituted aGUt a tenth ofaü antigenpositive tissues The 20 specimens (29%) with adequate
pathologic types in which HPV DNA was not detected
may represent infedon with HPV types not included in
the probe panel or low copy numbers of HPV types included in the probe panel, The 70% rate of detection of
HPV DNA in high-grade cervical lesions is in the range
of detection rates in other studies in which tissues are
examined with a similar panel of proba either by in sifu
hybridization or by Southern hybridization.'"2
Despite the presence of condylomatous a l l s in most
of the DNA-positive specimens, HPV-6/1I was virtually
absent from these tissues. Forty-six (92%) of the DNApositive h u e s contained a single viraltype. Among thesc,
HPV-16, HPV-18, HPV-31. and HPV-ó/II accounted
for, respectively. 30, ten, six, and zero specimens. The
finding of the predominance of HPV-I6 and HPV-18 in
tt
le
H
V
si
d
b
H
fc
11
C
e
P
g
O
r
I
L
H P V AND CERVICAL
NEOPLASIA
No. 10
TABU 6. Frequency ofClpiid Anlieen D M t i o n
in Conintion Specimens
Spcimas
No. t a c d
All Iissuer
DNA @¶¡Ve
t¡SSUa
HPV-611 I positive
HPV-I6 @tive
HPV-I8 positive
HPV-31 positive
Muhipk HPVs
DNA [email protected]¡W IissUts
CondylomsClN I
CIN I1
CIN 111
67
48
O
28
9
6
4
19
5
14
48
-
-
No. with
intipcn (%)
19 (28%)
17 (33%)
O
6
4
2
4
2 ( I 1%)
3(low
9 (64%)
5 (10%
HPV: humn papillama virus; CIN: cervical intraepitheli neoplasia.
these high-grade lesions is similar to that described in other
studies.’”’2
The combined evidence from the topographic and virologic studies of conization specimens clearly indicated
that, in most instances, the entire spectrum of pathologic
effects seen in a conization specimen is the result of a
single underlying process, vi,?., infection with one HPV
type. Nearly all condylomas and CIN were within, or
transgressed, the transformation zone. Condylomas were
often contiguous to CIN, or showed transition into C I N
and were always distal to CIN. A s a rule, a single HPV
type was found in these lesions and it was present in both
the condylomatous and the CIN areas. Even when the
lesions in the examined blocks of a conization specimen
were spatially discrete as well as histologically diverse, the
virologic evidence suggested that in a majority of cases, a
single HPV type was responsible for both lesions. In condylomas, and in koilocytotic cells within CIN, ihe hybridization pattern was focal whereas a diffuse pattern
was seen in high-grade CIN. These findings suggest the
following sequence of events: H P V infection of the epithelium of the transformation zone expressed as koilocytosis; neoplastic changes in the proximal part (toward
endocervical side) of the lesion; and extension of the neoplastic changes toward the cervical canal. The different
grades of severity in a conization specimen refleci the inh e n c e of local factors in the differentiation of C I N and
the evolution of the lesion over time. Cervical biopsies,
often taken from the most affected area, seldom display
the heterogeneity of the histologic changes in a conization
specimen.
Intensity of hybridization signal gives an indication of
relative copy numbers per cell. Of all viral types, HPV18 infections were usually faint or very faint, whether in
low-grade or high-grade lesions. Thus, the number of viral
Gupia el al.
2109
genomes per aU appeared to be lower in HPV-18 infections than in the other H P V types. The hybridization pattern appeared to be linked to the state of maturation of
the infected epithelium. An immature epithelium d i c
played an even, faint, or moderate level of hybridization
over every cell. in a maturing epithelium, such as Condyloma, or low-grade CIN, the hybridization pattern was
focal. Signal intensity varied widely from cell to cell, increasing toward the surface of the epithelium, with occasional “hot spots” of strong labeling. The focal pattern
resembled that of staining for capsid antigen, but involved
a greater proportion of cells, and was seen in antigenpositive as well as in antigen-negative condylomas. These
hybridization patterns are identical to those recently described by Schneider el al.” Basal cell hybridization was
seen in about one third of the HPV-positive cases and
was usually associated with high-grade CIN.
The antigen detection rates óf 28% in all tissues and
35% in DNA-positive tissues indicate that productive infection is not uncommon in high-grade lesions associated
with HPV-16, 18 and 31. Condylomatous cells were almost always present in association with high-grade lesions
and the capsid antigen, when present, was found in these
condylomatous areas. Condylomatous areas in high-grade
lesions are located toward the ectocervix where they are
likely to come in contact with the sexual partner. These
findings suggest that an individual may continue io remain
infectious despite the development of high-grade lesions.
REFERENCES
I . Macnab JCM, Waikinshaw SA, Cordiner JW el al. Human pap
illomavims in clinically and histologically normal tipsue of patienu
with pcnital cancer. N Engl JMed 1986: 3151052-1058.
2. zur Hausm H, Schneider A. The role ofpapillomavirus in human
anogcnital canar. In: %&man NP, Hmley P, eds. The Papovaviridae,
vol. 2. New Yo*. Plenum, 1987,245-263.
3. Lorinn AT, Lancaster WD, Temple GF. Cloning and character¡ration ofthe DNA ofa llcw papillomawuS from a woman with dysplasia
ofthe uterine a m x . J Vim11986: 58~225-229.
4. Kurman RJ, Jason AB, LancasterWD. Papillomavirusinfection
ofthe aMx: 11. Relationship io intraepithelial neoplasia bavd on the
p m n a ofspccihc viral structural proteins. Am J Surg Pniholl983,7
39-52.
5. Crum CP, Mitao M,Levine R U el ai. Cervical papillomaviruses
segregate within morphologically distinct PRcancemus lesions. J Viro¡
1985: 54675681.
6. Sito K, Saito A, Fu YS et ai. Topographic study ofarvical condyloma and intraepithclial neoplasia. Cancer 1987: 592064-2070.
7. Gupta JW, Gendelman HE, Naghashfar Z elal. Spaific identification of human pspillomavirus type in cmical smears and m
l
ñ
n
sations by in silu hybridization with radioactive probes: A Preliminary
communication. In1 J Gpmoi Pathol1985; 421 1-218.
8. Tourtellonc WW,Verity AN, Schmid P, Manine2 S,Shapshak P.
Covalent binding of formalin fixed wraffin~mbeddedbnin tissue scctioni to glass slider suitable for in situ hybridization.J Virol Mefhods
1987; 1987-99.
9. Gupta JW, Gupta PK, RoscnsheinN, Shah KV. Dnstion ofhuman papillomavirusin -cal
smears:A wmparison of in sifu hybrid-
I ) .
.
..
I
..
. .
r.
.L
C.
L
.
.
I
.c
._
w..
21 10
.
....
c
.
L.
r
.
....
..
C.
*"
L_
I,-.
r-
L
c.-
....
._..
.__
c
....
*.
CANCERNovember 15 1989
ization, immunocytochemistryandcytopathology.Acta Cyid 1987:31:
387-396.
IO. Reagan JW, Patten F Jr. DysplaJia: A basic rraction to injury in
the uterine cervix. Ann NYAcadSci 1962: 96662682.
I I. Johnson LD. The histopathologic approach 10 early m i c a l ne&
plasia. Obsiel C
~ SUNd1969: 24135-761.
12. Wielenga G, Old JW, von Haam E. Squamous carcinoma in situ
of the uterine cervix: 11. Topography and clinical comlations. Cancer
196% 18:1612-11621.
13. Stoler MH.Broker TR. In riru hybridizationdetection of human
papillomavim DNAs and messenger RNAs in genital condylomas and
a cervical cancer. Hum Pachol 1986; 17:1250-1258.
14. Crum CP, Nagai N, Levine RU, Silverstein S. In situ hybridization
analysis of HPV-I6 DNA sequences in early cervical neoplasia. Am .I
Paihd 1986; 123174-182.
15. k k m a n e AM, Myemon D. Daling J, Kiviat NM. Fenoglio CM,
McDnugall JK. Detation and localization of human papillomaviru
DNA in human genital condylomas by in silu hybridization With biotinylated probes. JMed Virol 1985; 16265-273.
16. Fuchs PG.GirardiF, pñstcr H.Human papiilomavinu DNA in
normal, mnaplartic, pnneoplastic and neoplastic epithelia ofthe c e M n
uteri. Inr JCacer 1988: 41:4,l45.
17. Reid R, G m n b M,JenJon AB el d.Sexually transmittedpap-
I
r
-
Vol. 64
illomavid infections: I. The anatomicdiatnbutionand patholoQc @
of neoplanic lesions nmisted with diaerent viral typcs. Am J Obsta
Gplecol 1987: 156212-222.
18. Millan DWM, Davis JA, TorM TE,Campo MS. DNA sequenas
of human papillomaviw t p I I, 16, and 18 in IeJiono of the uterine
cervix in the wesf of Scotland. Br Med J 1986: 29393-96.
19. Wilnyruki SP, &rssn S. Walker J, Liao SY, Pcailman LF. Human papillomaviwcs and cervical canm Anal* of hinopathologic
fcatum ayoOated with d i b n t viral types Hum Paibol 1988: 19697704.
20. Lonna AT, Temple GF, Kumian RI,Jason AB, LancaNrWD.
Oncogenic &tion
of e ü c human papillomavim t p with cervical neopluia. J Nail Cancer Ins1 1987; 79671477.
21. Gupta I, Pilotti S, Rilkc F, Shah K.A d t i o n of human pap.
illomavimtypc 16 with neopla5tic laionsoíthc vulva and other genital
rites by in silu hybridmtion. Am JPalhol1987: 127:206-215.
22. Caussy D, Orr W, Daya AD.Roth P, Reeves W, Itawls W. Evaluation of methods for detation human papilbmavinis deoxyribonucleotide sequences in clinical specimens.JClin Microbid 1988:26236243.
23. Schneider A, OltersdorfT, Schneider V, GisJmann L. Distribution
pattern of human papillomannu 16 genome in cervical neoplasia by
molecular in silu hybridization oftiJsue d o n . Ini J Cancer 1987; 3 9
717-721.
Adenocarcinoma del Cérvix Asociado al
Uso de Anticonceptivos Orales.
Presentación de i caso.
Dr. Alfonso Torres-Lobatón'
Dra. M c d a Hernández Gonzücz"
Dra.Guadaiupe Rojo-Herren' * *
Dr. Avissai Alcántara-Vázquez"
Dr. G c d o Mota-Cieafucgos' * * *
I
I Resumen
~
~
~~
Se presenta el caso de una paciente de 30 años de edad con diagnóstico de adenocarcinoma invasor del cénrix en
estadio Ib, el cual fue manejado con histerectomía total más Iinfadenectomía pélvica. Se menciona que la pacienre tenía
el antecedente de emplear de los 17 a los 20 años contraceptivos orales y que conraba con los siguientes factcfesde riesgo
para el desarrdlo de ccincer cérvico-uterino:proceder de medio socioscondmiw bajo. inicio a los 15 años de la acfividad
sexual. embarazo antes de los 20 arios. multiparidad, y no contar con estudio previo de papanicolaou.
Se llevan a cabo consideraciones acerca del efecto inductor o promotor de los contraceptivos orales, en la etiología
del adenocarcinoma cervical que se estd observando en mujeres @venes, los cuales podrían actuar en forma aislada o
en combinacidn con otros factores. Se trara de documentar esta relacidn. con observaciones que presentan a los contra
ceprivos orales asociados a la hiperplasia de las células de reserva del endocérvix, misma que se ha visto evolucionar a
displasia glandular, a adenocarcinoma in situ y a adenocarcinoma invasor.
Se senala que la paciente en estudio mostrd un tumor de menos de 4 cms. de diámetro y ausencia de meidstasis ganglionares, factores considerados como de buen prondstico; cabe esperar en ella su curacidn.
r Summarv
1
A 30 years da female with a history of oral contraceptwe use presented an invasive aaenmarcinoa of me C ~ N I Xsrage
,
Ib. tne treatment was total nysierecromy win pslvic node resecrion She had oral contraceptives from 17 to 2ü years of
age sne was also noted to have cenam risk lacrors for cervical cancer such as she came from a low socio~wnomicenvi
ronmer. sraned sexual activity at 15 years sne was also mulrparous and firsr became pregnanr before 20 years of age
She had no previous cpofogy sruay
There is some evidence thaf oral conrracepfives may play some role in the etiotogy of cervical adenocarcinoma in young
women,eitner aione or with omer facrors The evtdence for mts IS rhat me reserve cell hyperplasia of me endocerv*. asK)cIatea wnrh ora! conrraceptrve use nas been observed to develop into glandular dysplasia adenocarcinoma in situ as well as
invasive carcmoma
The prognosis for rne parnent is good because of mall size 01 the tumor and the abscense of lymph node metastases
I
rPalabns ciave
Adenocarcinoma, cérvix. anticoncepiivos
El adenocarcinoma del cérvix ocupa el segundo
lugar entre los cánceres de esta estructura anat6mimisma
ca con una frecuencia global de un 60/0,'.~
rt
*
**
***
..*
Jefe del Servicio de Gineoologia. Unidad de Oncologia.
Servicio de Patologia.
Servicio de Gineco-obstetricia.
MBdico Residente Unidad de Oncdogia.
Hospital General de MBxico. S.S.
Dirección: Or. Balmis 148 MBxico. D..C.P. 06820
que según estudios epidemiol6gicos recientes, tiende a incrementarse en algunas regiones de los Estados Unidos de Norieamérica, donde su frecuencia actual, fundamentalmente a expensas de mujeres de
menos de 35 años de edad, ha alcanzado cifras hasta de un 18%.3,4
Si bien es cierto algunos de los factores involucrados en su posible etiología concuerdan con los del
carcinoma epidermoide del &rvix, como el iniciotemprano de la actividad genital. la promiscuidad sexual
43
y las infecciones cervicales por algunos tipos de virus,5 y otros factores con un cierto predominio en
clases sociales altas, en pacientes nulíparac, en obesas y diaMticas. recuerdan a los factores de riesgo
asociados al cáncer endometrial en cuya etiología se
ha demostrado una clara influencia horm0nal.2~~~~
A partir de 1986, se han reportado casos clínicos
en forma aislada o en pequeñas series de pacientes
jóvenes con adenocarcinoma del cérvix, quienes habían tomado contraceptivos antes de los 20 años de
edad por 1 o más años, lo cual está relacionado el
incremento anotado en la frecuencia del adenocarcinoma cervical, con el empleo 10 años antes de contraceptivos orales.4J
En esta publicaci6n se muestra un caso de adenocarcinoma del cérvix en una paciente de 30 aiios
de edad, quien tenia el antecedente de empleo de
anticonceptivos orales y quien fue atendida en la Unidad de Ginecologia del Servicio de Oncologia del
Hospital General de México, S.S.Se analizan los aspectos etiopatogénicos. anatomoclinicos y terapéuticos correspondientes y se llevan a cabo consideraciones al respecto.
Con un diagn6stico de adenocarcinoma de cérvix
en estadio clínico Ib, radiografía del tórax sin evidencia de meastasis y ex8menes preoperatorios dentro
de la normalidad, se sometió a histerectomiacon linfadenectomía peivica conservando ambos anexos. La
evoluci6n postoperatorio fue satisfactoria y la pieza
quirurgica 8-89-4000report6 adenocarcinoma del
cérvix moderadamentediferenciado, sin tumor eq borde vaginal ni metástasis ganglionares (Figura 1 y 2).
La paciente asiste a sus controles sin evidencia de
enfermedad.
FlüURA 1
Preaente~lóndel caso
L.M.R. de 30 años con exp. 89/1112. residente en
el Edo. de MBxico y procedente de medio socimonómico bap. Antecedentesde menarca a los 13 años,
ritmo de 30x3, eumenorreica hasta hace 2 años en
que inició con hipermenorreay dismenorrea. Vida c8xual activa a los 15 años con un compai7ero sexual.
GestasVIII, parasVI, aborto I, cesáreas I.Primer parto a los 17 años, &ea
por presentacan peivica con
salpingoclasia a los 26 años. Utiliz6 de los 17 a los
20 años contraceptivosorales. Nunca se le había efectuado papanicolaou.
Un mes antes de su ingreso a la Unidad, present6
metrorragia abundante en una ocasi6n. lo que motivó consulta en medio hospitalario de su localidad. de
donde 89 canalizada al Servicio con diagnóstico de
adenocarcinoma del cérvix.
La expioraci6n física mostr6 al paciente en buen
estado nutricional, con una cicatriz media infraumbilical; la exploración ginecd6gica arrojó como datos
positivos la presencia de una lesi6n neoplásica exofitica de 3.5 x 3.5 cm limitada al cuello uterino.
4.4
I
Comontirio
En 1986, R.K. Peters y cds7 destacaron el incremento observado en la frecuencia del adenocarcinoma del Cervix en mujeresjóvenes del Condado de los
Angeles California. Este fue atribuido al uso de antiwnceptivos orales, al demostrar en este tipo de pacientes la asociaciónde adenocarcinoma in situ e invasor del cérvix. M.P. Hopkins y cok del Centro MBdico de la Universidad de Michigan? informaron en
1987 una frecuencia para esta lesión del 16.8%, haciendo notar que el 45% de las pacientes que contaban con 45 años de edad o menos tenian el antecedente de emplear anticonceptivosorales. Otras autores? han llegado a idbnticas conclusiones.
En cuanto al mecanismo a través del cual los contraceptivos orales influyen en el desarrollo del cáncer
endocervical, se han reportado evidencias de diversa indole que muestran una clara influencia hormonal ejercida por estos agentes en las células columnares del endocérvix y en las células subcdumnares
o de reserva de esta estructura. Algunos autores’ sugieren que en la adolescencia, el epitelio del cérvix
en su zona de transformación de epiletio columnar
a epitelio escamoso. es especialmente susceptible al
efecto de los cancerigenos ambientales, por lo que
el efecto inductor o promotor generado por estos
agentes bien sea en forma aislada o en combinación
con otras factores, podría llevarse a cabo precisamente en esta etapa de la vida de la mujer.
Por otra parte, desde el punto de vista experimental se ha demostrado que los progestagenos administrados a dosis altas, desencadenan en ciertos mamíferosel desarrollo de adenocarcinomas cervicales
similares a los de los humanos.3
El empleo de contraceptivos orales se ha visto asociado a hiperplasia microglandulardel endcdrvix, patologia que también se ha llegado a observar durante el embarazo y que no es considerada como pre
maligna. Sin embargo, existen reportes que señalan
su asociaci6n con la hiperplasia de las cBlulas de reserva del cérvix la que puede evolucionar a displasia glandular, a adenocarcinoma in situ y adenocarcinoma invasor.
La paciente presentada en este estudio tenia los
siguientes factores de riesgo para cáncer cervicouterino: el proceder de medio socioecon6mico bajo,
inicio a temprana edad de la actividad sexual, multiparidad y el no contar con un estudio previo de papanicdaou. Todos los factores. sobre todo los men.
cionados en segundo y tercer lugar, son considerados en la actualidad de alto riesgo para el desarrollo
de esta enfe~nedad.~
El hecho de que la neopiasia se haya diagnosticado a la edad de 30 años y no a los 45, como ocurre
para esta etapa clinica en las pacientes que acuden
al Servicio con cancer c8rvicuuterino invasor limitado al cérvix,’ sugiere la participación de otro u otros
elementos en su posible etiologia. por lo que el antecedente de haber utilizado durante’más de 3 años
anticonceptivos orales, permite establecer una relación causa-efecto, como se ha mencionado previamente.
Para esta paciente, resulta atractivo el plantear la
muencia de los siguientes eventos en la historia natural de su enfermedad: 1, inicio temprano de la actividad genital; 2, embarazo en la segunda década de
la vida; 3, empleo de contraceptivos orales durante
3 años por lo menos; 4, multiparidad; 5. periodo de
lactacia de 10 años y 6, presencia de adenocarcinoma cervical invasor.9
En cuanto al tratamiento instituido a esta enferma,
la histerectomia con linfadenectomia p4ivica constituye la terapéutica de elección para pacientes de 40
años o menos, en las que resulta fundamental conservar el funcionamiento ovárico, ya que el otro prccedimiento. considerado también de elección y que
es la radioterapia, termina prematuramente con la funci6n ovárica. lo que trae como consecuencia un climaterio a edad muy temprana.
La pieza quirijrgica reportó un adenocarcinoma del
cérvix de 3.5 cms moderadamente diferenciado, sin
tumor en bordes de sección y sin methtasis ganglionares, lo cual permite inferir una evoluci6n a largo plazo sin actividad tumoral.
Si bien es cierto que al adenocarcinoma cervical
se le ha atribuido, en términos generales, un pronóstico menos favorable que al carcinoma epidermoiel caso en particular presenta como factores
de buen pronóstico, el contar con una lesión menor
de 4 cms y el carecer de ganglios metasthsicos. M.P.
Hopkins y cdc4 obtuvieron curaciones del orden de
88% en 34 pacientes, con reporte de ganglios negativos para metástasis; por su parte J.S. Befek y colsl0
45
Admanminomi del ctrrk. T-Lobztón
-
A,, Hcmindcz-Gonzílcz M.. Rojo-Hwen G.. Alrinnrr Vízqucr A,, Mora-Cicnhicgm G .
alcanzaron, en lesiones de 4 cms,curaciones del orden de 69%.
I.Torres LA.Cruz TF. Plata NP et ai Cancer &rviwuterino
1983.1984. Experiencia de la Unidad de Oncología. Hospitd
General de MBxico, S.S. Cancercdcgia 1988; 3 4 617-622.
2. Brinton LA. Karen T. Tashima H el a i Epidemiology o1 Cervical Camer by wll type.Cancer Res. 1987; 47: 1706-171 1.
3. Valente PT, Hanlani P. Endocervical Neoplasia in Longterm
Users d oral Contraceptives:Clinical and Patholwic Observa.
, lions. Obstet and Gynscol. 1986; 67: 695.704.
4. Hopkins PM.Patrick S.Robert9 AJ: Prognostic features and
treatment of endocervicalAdenmrcincma of the cervix. G y m
cd Oncd. 1987; 27: 6475.
5. Menu- J. YaronSchfflwO. LiventonS, elal. HerpesvirusType
2 in Adenocarcinoma of the uterine cervix: A possible association. Canm. 1981: 48: 1497-1499.
6 Milson I. Gostafr berg L Primary adenocarcinomao1[ne Liter.
ne C B N I ~Cancer 1983. 52 942.947
7 Peters KR. Chao A. Macr MT. Bemstein L h noerson EB In.
creased frequency of adenocarcinoma of the uterine carvix id
young w a n e n in Los AngdesCounty. JNCl19BB; 76: 425428.
8. Chumas CJ. Beth N. Mann JW. et al. Microglandular Hyperplasia of the uterine ceMx. ObsfetG y d . 1985; 66 406409.
9. Torres LA. Roman BE. MaRinez MR. Cruz TF. CAncer cervicouterim. Evidenciasa favor de una etiología multifactorial.GI.
nec Obstet Méx. 1987; 55: 214.
10. Berek SJ,Castaldo WT. Hacker FN, et a i Adenocarcinoma of
the uterine cervix. Cancer 1981: 48: 2734.2741
i
46
Second Primary Cancer After Treatment
for Cervical Cancer
Late Effects After Radiotherapy
HANS
H. STORM. MD
U s 4 data horn t
k pprhlioii-bawdW s b Cancer Re&try. i
k relative risk (RR)of second priaury
Museued
24,970 w0-n wilh invasive cervlal m c e r (1943-1982)d19,4470women
rrHh eird.oii ir situ of l
k cervix. Tk aulyris was s h t i f k d
to treatment with (+) and
*ilLout (-) n d h l b a . For dl
p r h u k s c o m b i d , I RR + = 1.1 (95% mfdcrc htcnil (a)
= 1.06-1.18)diRR- = 13(95%CI = 1 . 1 3 - 1 . ~ ) ~ o b a c r v e d . f i r r i n v w i v c c e ~ l e i a e m u d i
RR+ = 3 5 (95% Ci = 1.4-7.2) dRR- = 1.1 (95% Ci = 0.7-1.6) fdl0wi.o in situ cancer. T k amll
oven11 excesa of second primuy
is m u n t e d for by 80 incruse of some -rs
such as Iwg.
W d e r , di
eo~~yrrenl
dnreue In others such as h i . Allhough mot stalisticslly dinefist from
.oiiimdbld tk RR iarrased with lime since treatment among irndbied invisive emhl 1).lints in o r p i s cbse to and i t in~ermediite
distance from lbe m i x , remchiag a maximum after 30 or
more y a n of follow-up (RR = 1.9; 95% Ci = 1.4-2.5). Altogether. for tbcse eles i n c x m s of 64 ases
per 10,oOO women prr y a r wen imikitabie to ndhtion amhug survivors of 30- y a n . The bilkst
risks immg lomg-term s d v m were observed for the foliowing: otber gaita1 organs (RR = 58; 95%
CI = I a-13.0)H e r (RR = 55;95% CI = 2a-95),~
V ~ ~ S(RR
SCW= 3.3 95% a = 0.4-12.0~
stonicb (RR = 2J; 95% CI = 1.1-4.7) d rectum (RR = 2.4; 95%Ci = 1.146).A s i p l R a i t deAcit
of risk for breast cancer (RR = 0.7.956 Ci = 0.6-0s) Mokcned for 10- y a n , my k imibutsbie
to lbe d C a of o v U t a ibbtim by ndiotknpy. I1 is spenilted that the mme decl iIS0 my expLin
lht observed dcfleita of k
i
i
otumors (RR = 0.6; 95% Ci = 0.4-1 .O) dskim m~lioonis
(RR = 0.6; 9%
Cl = 03-10). It b cmclded Ibat cancers itidbutable lo rdbtiai. i p u t from m t e Iioiiympbocytic
ieukrmts,l e d to ippeu lale (IO or more y e i n after ndiotknpy), dthat <berisk remiins denled
for mom than 30 y a n .
Cancer 61xÍ79488,1988.
-r
A
SUBSTANTIAL PART of our knowledge on the carci-
nogenicityto humans of ionizing radiation derives
from populations irradiated for medical purposes.' The
results of these studies are corroborated by those of
atomic bomb s~rvivors?~
although there are some discrepancies.Whereas high to moderatelevels of exposure
From the Danish Cancer Registry, Institute oiCanar Epidemiol-
ogy, Danish cuK?cr Society. Copnha&%n.
SuppoRed by a lcsepIch grant h m the Danish Canar Society,
which also suiupponr the Danish Cancm Rcgistiy.
The authon thank Mis. Hew Ni&n for thorough pmfmading,
prepamtion oftabla and typing thc tat;MIS.
Aase Lami for drawing
the fisurrs; Mr. Ni& Chri>tnisni for wnducting the computer anaiyand Un. Elizabeth Hsssltine for her skilled editing. The authors
also thank DI.
Ok M d a Jcnsen and Dr. John D.Lbice Ir. Radiation
e,
Epidemiology Branch, NCI, for valuable criticism during p-tion
of the manuscript.
Addreda for reprints: Huis H. Síorm, UD,Danish Cancer Registry,
Institute of Cnocer Epidcmidogy under thc auwh Cancer Society.
L.ndil;ronasr.de 66,4th Boor DK-2100,Copenhasen, Denmark.
Ampted forpublimtion Aue.ust 31,1981
to ionizing radiation increase the risk of cancer,' controversy still exists about the carcinogenicity of exposures to low-level ionizing radiation.'
Leukemia is the most prominent radiation-associated
cane+' although its induction appears to be a complex
function of dose, involving inactivation of bone marrow
cells at high
Radiation-induced solid tumors are
of greater numerical significancethan leukemia' and, in
contrast to leukemia, appear decades afier radiation exposure, with a temporal pattern of incidence (ie.,age at
diagnosis) that seems to follow the natural incidence.'
The excm risk does not appear to decline with time, as
it does for leukemias, bur few irradiated populations
have been foliowed beyond 30 years.
The majority of cervical cancer patients in Denmark
are treated with radiotherapy,''." and most as a benefit
of treatment survive for a long time, as in other countries.'* An international coUaborative study by Hutchisoni3 on irradiated cervical patients, withan extended
679
CANCERFebruary I S 1988
680
Vol. 61
Numkr of Women by Age at Dtagnmis. Total 'lumber of Womcn (1943-1977). and Women-Yenrs a1 Risk for Second h m a r y
Cancer Development í 1943-1982) A h irradiated or Not Irradiated Cervical Cancer or Canar In Situ of the Cenia in Denmark
TABLEI
Invasive
Radiotherapy
No. of women
20160
Women-yr at risk since diagnosis
<I yr
18292.0
1-9
97294.3
10-19
58895.8
20-29
23969.4
30+
4646.6
All
203097.9
No. of women by agc at diagnosis
<35 yr
1953
35-44
5151
45-54
5849
55-64
4170
2232
65+
Nc
fa<
mt
In siru
No radiotherapy
Total
Radiotherapy
No radiotherapy
Told
4810
24970
567
18903
19470
18827.6
139930.1
3 1548. I
669.2
7.1
190982.0
19382.2
I44 136.6
33599.2
855.8
12.7
197986.2
h>
ar
b!
hi
4246.4
29927.6
18303.8
5425.7
485.4
58388.7
22538.4
127221.9
77199.6
29395.1
5132.0
26 1486.6
554.6
4206.5
2051.1
186.6
5.6
7004.2
723
IS69
1281
672
2676
6720
7130
4842
66
183
I64
75
61
744
7574
W
di
h
W
1
.
fi
664
188
739
249
ti
b
t
F
follow-up,l' showed no increased risk of leukemia, in
contrast to a larger and more recent study." Excess risks
also have been observed for solid tumors in organs close
to and at intermediate distance from the cervix.ls However, information on long-term survivors (20 or more
years) has so far been limited in large part to resultsfrom
the Connecticut Tumor Registry.16
The current study reports the results of an extension
of follow-up through 1982 of cervical cancer patients
diagnosed in Denmark between 1943 and 1977; the
previous follow-up was through 1976.'" Emphasis is
placed on the risk of second primary solid cancers
among long-term survivors, since the present report is
based on more woman-years accrued among 30+ years
survivors than any previous publication o f cervical
cancer patients.
Materiala Ud Methods
Danish Cancer Registry
The study is b a d on incidence data collected routinely by the national Danish Cancer Registry between
1943 and 1982. The Registry receives notification from
hospital departments, pathology institutes, and practicing physicians. Reports on carcinoma in siru and severe
dysplasia of the CCMX have been requested since 1968.''
Unreported cases are identified by means of annual
linkages with the ~ t i o n a death
l
register." The Cancer
Reeistry is tumor-baJcd, with each tumor linked to the
unique personal identification number provided to all
Danish inhabitants. AU tumors Mcoded and clpsyfied
in accordance with a modified version of the seventh
revision of the Intmnationai Classification
which allow a distinction baween sarcomas and carcinomas
5
Multiple primary cancers in the same person are defined as tumors that arise in different organs, and these
are notified as new primary cancers or as separate
tumors with different morphologic characteristics." The
registry is known to be complete regarding invasive
tumors,2°*'1 although there is some underreporting of
multiple primaries, estimated to be approximately 16%
depending on site of first cancer?'
Study Population
All women with either invasive carcinoma (24,970) or
carcinoma in situ of the cervix uteri (19.470) diagnosed
between 1943 and 1977 were included if the woman
survived at least I month from the day of diagnosis
(Table 1). Altogether, 460,000 person-years were accrued. A total o f 2313 second primary cancers (expected, 2045) were identified 1 or more years &r the
initial cancer diagnosis by means of record linkage
within the registry and were distributed according to
site, notified treatment, and time since cervical cana%.
Nonmelanoma skin cancers were not considered in this
study. Most cervicalcancers (98%) but fewer of the second primaries (88%) were histologically verified.
The study population was subdivided into four CDhorts: invasive or in situ c e r v i d cancer and, within each
of the two, with or without radiation, the latter including
cases without information on radiotherapy. classiñcation of invasive and in situ cervical cancers by the
Cancer Registry was shown in an d i e r study to be
~0rrect.l~
A comparison of regisminformation and hospital record information showed that 98% of 2156 cerv i d cancer patients notified and coded as irradiated
had in fact undergone radiotherapy, whmps 24% of 423
cases classified by the Registry as not irradiated had in
i
(
I
- *
I
.
No. 4
ii
B
f
a
*-
c
.._
-
.._
-
>
j
Y
,
.<I
j
._
F'
,
u
'
SECOND PRIMARY
CANCER
AFTERTREATMENT
fact received radiotherapy as part of the primary treatment (Table 2).
The Same data were used to estimate the frequency of
hystcnctomies and oophorectomies among irradiated
and noninrrdiated cervical cancer patients as classified
by the Cancer Registry. Among irradiated patients 9 1%
had an intact uterus and 89% intact ovaries, compared
with 4% and 21%. mpectively, among those nonirradiated, among the nonirradiated women, a further 18%
had one ovary left. Altogether, 39% of the nonirradiated
women thus probably had intact ovarian function.
A random Sample of 4 1 patients classified by the Registry on the basis of a histologic diagnosis as irradiated
for in siru cervical cancer were selected for recoding of
the notified information. One patient (2%) was found to
have had invasive cervical cancer. The majority, 3 I patients (76%), were irradiated for an in s i f u lesion,
whereas nine (22%) were not. However, 19 (48%) of in
situ patients were classified according to the International Federation of Gynecology and Obstetrics
(FlGO)I2 as Stage I (a or b) or even Stage 11by the
reporting clinician, who thus disagreed with the pathologic classification.
Follow-Up and Analysis
Follow-up information on vital status is available
from the Cancer Registry records." Woman-years at
risk were calculated from the date of diagnosis of the
cervical cancer until the date of diagnosis of the second
primary cancer, death, or December 31, 1982, whichever occurred first. Expected numbers of second primary cancers were obtained by multiplying age, sex and
calendar time-specific incidence rates by appropriate
women-years, using a modified version of the program
developed by Monson?' Relative risks (RR) were measured as the ratio of observed to expeaed numbers of
incident cancers. Tests of significance and confidence
interval (CI)of the RR were calculated assuming a Poisson distribution?' Tests for trend over time from 10 to
30+ years were canied out utilizing programs for a programmable calc~lator,~'using midpoints of follow-up
intervals and 35 years for the 30+ year category. The
absolute risk attributable to radiation was calculated as
the difference between the observed and expected numbers divided by the woman-years at risk.
Results
Table 3 shows the overall pattern of risk one and more
years after cervical cancer or carcinoma in situ according to initial treatment with or without radiation.
Among irradiated (+) and nonirradiated (-) women
with invasive cervical cancer, significantly increased
68 I
Storm
TABLE2. Vdidity of Clusifiation (%) According to Radiotherapy
in thc Din¡& cinar Rsutry
Ciruification according io
hosmul rem&
CLirUficationin
Ridiothcrnpy
cancer @svy record
(5)
Irradiated
No1 irradiated
97.1
23.9
No
ndiothcrapy
Total no. of
2.3
76.1
2156 (100)
423 (100)
ptimía (5)
risks are observed for cancer of the lung (RR+ = 2.6;
RR- = 2.4). other genital organs (vulva and vagina predominantly; RR+ = 1.9; RR- = 3.2), bladder (RR+
= 2.9; RR- = 2.5) and connective tissue (RR+ = 1.5;
RR- = 2.3). Kidney cancer rates are increased among
nonirradiated patients (RR- = 2.6). Imspective of
treatment, nonsignificantly (NS) increased risks are observed for second primary cancer of the esophagus,
small intestine, rectum, and bone. Significantly decreased risks are seen for cancer of the uterine corpus
(RR+ = 0.6; RR- = 0.1) and ovary (RR+ = 0.5; RR= 0.5). The risk of breast cancer is significantly decreased among irradiated cervical cancer patients (RR+
= 0.7); similarly significantly decreased risks are Seen for
melanoma (RR+ = 0.6) and brain tumors (RR+ = 0.6)
after irradiated invasive cancer.
Among women with in situ cervical cancer, increased
risks are seen for cancer of the lung (RR+ = 2.2 [NS];
RR- = 2.1), other genital organs (RR+ = 14.3; RR= 7.1) and bladder (RR+ = 4.7; RR- = 1.3 [NS]).
The
only overall excess of cancers of the lymphatic and haematopoietic tissus (due to acute leukemias) is observed
after irradiated in siru cervical cancer (RR+ = 3.5). Significant deficits of breast (RR- = 0.8). corpus uterine
(RR- = 0.2), and ovarian cancer (RR- = 0.7) are seen
after nonirradiated carcinoma in situ.
Table 4 shows the trend in RR for cancers at selected
sites among long-term survivors irradiated (+) and not
irradiated (-) for invasive cervical cancer. The RR for
stomach cancer increases significantly with time since
irradiation of the cervical cancer (x2trend = 9.1; P
= 0.001). The RR for colon cancer is 10% to 40%higher
among nonirradiated than among irradiated women
throughout the follow-up period, although none of the
RR is significantly different from unity. The RR for
rectal cancer is similar in irradiated and nonirradiated
patients, with a significant increase in risk with time for
irradiated women only (xztrend = 16.9; P = 0.0002).
A n increasing RR (x2trend = 11.2;P = 0.004) for pancreatic cancer is Seen among irradiated cervical cancer
patients, however not present if including follow-up
years 1to 9. The overall increase in risk fG this cancer
.
~
I-
-
.
.. .-. .
1
CANCERFebruary 15 1988
682
T A B U 3.
Vol. 61
ObSe~edand O ~ W NtoCExpected
~
Ratio of Second Primary Cpnocr 1944-1982 at Lmst I Year ARer CeMcal Cancer and
Carcinoma In Situ of the Cervix 1943-1977. According io Notified Treatment With and Without Radiation
Carcinoma in situ cervix utcnt
Invasive cervical canwr
Second primary cancer
(ICD.7 code no.)
No rddiotherapy*
-
Radiotherapy
ObSeNed
Buccal cavity and pharynx
(140-146)
E-phapua ( I 50)
Stomach IIs I )
Small inicstine (152)
Colon (133)
Rmum(i54)
Liver and pllbladdcr( I 55)
Pancreas (I57)
Lun6 ( I 61)
Brcpst ( I To)
Uterine wpus (172)*
ovary ( 175)
Other genital (176)
Kidney I180)
Bladder [ I 8 I)
Melanomi (IW)
Brain(IO3)
Thymid t 194)
bne(I%a)
Connective tisue (197B
Lymphatic and
hematopoietictisues
(200-204)
All sites ( 140-204 except
191othcrskinand 171
cervi.i\
18
I5
75
6
131
85
30
48
I70
207
54
47
29
35
104
I3
20
8
5
26
59
1338
OIE
1.1
I .6
I .o
I .7
I .o
1.2
0.9
1.1
2.66
0.76
0.66
0.56
I .9(
I .o
2.96
0.61
0.61
observed
5
3
16
I
29
17
5
I2
42
69
3
10
IO
21
Radiotherapy
observed
OIE
1.3
I .8
1.3
O
O
1.1
4
-
1.3
I .2
0.7
I .4
2.4#
0.9
0.16
0.57
3.21
O
O
I
2
4
5
10
2
3
6
I
5
O
10
IS!
I
9
2.3
2.31
O
0.9
In
349
0.9
2.1
1.16
R
I
Inilihks radiotherapy unknown.
t Includes severe dyrplaua of the cervix.
t Includes uterus unspceificd.
disappears 20 or more years after a nonirradiated primary. The risk for lung cancer increases consistently
(RR = 1 . 3 4 3 ) imspective of primary treatment.
in mntmt, breast cancer risk remains consistently
below expectation (RR = 0.5-0.9). but significantly so
only if the cervical cancer was irradiated. After a nonirradiated cancer of the cervix, deficits of second primary
Corpus uterine cancer (RR 0.3) and of ovarian cancer
(RR 0.6)are observed. The overall deficits of corpus
uterine and ovarian cancers disappear 20 or more years
after an irradiated primary, and a significantly increasing tRnd in risk by time is seen for both sites (X' trend
corpus 9.9: P = 0.001; x' trend ovary = 7.2: P
= 0 . 0 0 . The risk of other genital cancers (predominan@ cancera of the vulva and the vagina) is significan* incnased, especially among long-term survivors
who irnived radiotherapy, and the RR increases significan@ with time (x' trend = 10.9; P = 0.ooOS). The
hcrraned RR of bladder cancer after either an irradiated
or a amirradiated primary W significantlydifferent from
unity only if the cervix received radiation. The risk
No radiothera~
O/E
2.66
2.56
1.1
I .2
20
O/E ~*dcrvedio expected ratio.
-
No. 4
1.1
0.5
2.1
3.3
2.2
I .o
0.7
1.1
14.36
I .o
4.71
Obamed
10
I
13
O
30
OIE
1.5
0.5
I .o
-
Red
0.81
0.21
BRí
0.7
1.3
I .o
ov;
Bla
28
37
9
I6
2.16
0.7W
7.16
O
5
1.2
7
3.5n
29
1.1
1.36
58
1.611
Col0
I55
8
77
0.2
I .4
0.6
I
Stom
16
I4
9
23
13
I
I
0.5
-
0.8
0.7
I .o
I .4
4.2
-
O
secar
(1'
568
0.6
1.1
5 Includes own siic-spific sarcomas
IIP < 0.0s.
Pam
iun
Cor
Ott
M<
BE
C<
di
TP<O.OI.
6 P < 0.001.
among these patients also increases significantly with
time since irradiation (x2trend = 26.0;P < O.oooO1).
When one combines all second primary cancers close
to (bladder, rectum, ovaries, vulva-vagina, pelvic bones,
connective tissue of the pelvis) and at intermediate distance from (kidney, ureter, colon, small intestine, stomach, pancreas, liver, peritoneum, retrope&oneum) the
cervix of irradiated women (Fig. 1), the RR increases
substantially and significantly with time (x'trend
= 225.6; P < O.oOo1). The excess number of tumors at
these sites amounts to 64/10000women-years and
17/10,OOOwomen-years among irradiated and noninadiated women, respectively, 30 or more years after initial treatment. The relative risks over time are not statistically different from those of the nonirradiiated patients (x2 = 9.6; 8 df, P = 0.30). The RR increases
significantly with time for nonirradiated patients also
(x'trend 25.1; P < 0.0001). However, the RR- start
from a higher level at the outset (RR = I .O) for nonirradiated than for irradiated and ends at a lower level (30+
years: RR- = 1.3; RR+ = 1.9) (Fig. I).
a
E
C
I
<
SECOND PRIMARY CANCERA m n TREATMENT
NO. 4
*
683
Sform
TABLE4. o b s e d and Eapsied Second Rimary C a n a n of Sc*aed Sita. Lon&Tcmi SuMvh8 Irndkld
and Not Irradiated* Cervical Qnar htients
Yr U n a aMcal cancer
10-19
Second primary cancer
(IcD7 d e no.)
Stomach (IS I )
Colon (153)
Rmum (154)
P n n m ( I S7)
Lung (162)
BrraJt(170)
Corpus uteri (172)
Ovary(175)
Other genital (176)
Bladder ( I8I )
Melanoma ( 190)
Brain (193)
Connmive tirsue (197)
1
N
I
N
I
N
I
N
I
N
O
Et
OlEt
O
E
OIE
O
28
9
42
14
29
7
25.2
1.1
2.0
0.9
I .4
I .2
14
O
14.0
2. I
29
28.9
I .O
0.0
I .o
9
I
6
13
6
43
16
4.5
44.5
10.3
24.8
S.8
15.2
3.6
24. I
7.5
2.15
0.75
0.9
0.7
0.2$
0.411
0.6
78
27
104.2
31.1
23
13
2
32.0
N
I
N
I
N
5
IO
6
34
8.6
5.1
I .2
N
5
3
I
1
N
I
N
7
5
6
8
4
1.2
0.9
I .7
1.81
I
N
I
N
I
M+
20-29
9.3
30.2
12.8
3.3
8.0
2.8
11.1
~~
3.3
5.5
I .5
6
25
4
9
2
20
5.3
14.5
2.7
10.3
I .9
14.9
3.4
6
28
7
S2.4
I
14
15.9
3.7
14.7
3.I
3.0
16
2.0
2
8
4.99
I
2.711
2.I
0.6
1.1
32
11.2
2
0.6
8.3
1.6
2
4.0
5.3
I .8
I
4
O
I .4
7
2.1
2
0.5
I .o
I .2
2.1
0.5
1.1
I
1.7t
9
I
6
O
7
1.5
0.9
1.1
I .3
I .8
0.51
0.6
2
8
I
I .o
7
0.3
O
3
O
1
.o
0.6
2.6t
1.7
3.911
1.3
5
O
I2
I
0.5
Io
.
0.8
O
O
0.0
2.6t
3.1
O
2
O
O
E
OIE
3.6
0.4
8.3
0.8
3.7
2.5.
2.8
0.7
1.2
0.4
2.7
2.9
0.3
3.7
0.4
11.9
I .2
2.0
0.0
I .9
5.3
0.7
0.8
2. I
0.0
3.3
0.3
3.1
0.3
0.9
o.I
2.2
0.2
2.4#
Io
.
0.0
5.88
0.0
5.511
0.9
o.I
4.6
0.0
0.0
1.1
0.0
o.I
0.6
o.I
0.0
3.3
0.0
0 O b X N e d , E CXpCCtCd; O/E ObXrVCd IO CX-d
d o ; 1: imdialed: N no1 irradiated.
Includes irradiated unknown.
t [email protected] based on two decimal point calculations.
t P < 0.05
5 P < 0.01.
Soft tissue sarcomas occur in excess in both irradiated
and nonirradiated women (Table 3), and the RR increases with time since diagnosis among those who received radiation (x2trend = 3.5; P = 0.03) (Table 4).
Almost 90% of the sarcomas are located at sites close to
or at intermediate distance from the cervix (Table 5),
which is not significantly different (x2= 2.66) from the
distribution among women in the Danish Cancer Registry material as a whole (70%).
Table 6 presents the data on women treated for carcinoma in siiu and followed for more than 10 years. The
findings are similar to those in patients with invasive
cervical cancer: elevated RR are seen for lung cancer
(RR+ = 1.E [NS]; RR = 2.3). other genital cancer (RR+
= 11.8; RR- = 4.9) and bladder cancer (RR+ = 4.7;
RR- = 1.4 [NS]).The RR for second primary breast
cancer among nonirradiated carcinoma in situ patients
is as expected (RR = 1.O), whereas a decreased risk (RR
= 0.5 [NS]) is observed after irradiation.
When one combines all irradiated patients, with both
invasive and in situ cervical cancer, 15 acute nonlym-
phocyiic leukemias (ANLL) are observed, with 7.3 expected (RR = 2.1; 95% C1 = 1.2-3.4); five ANLL occurred among nonirradiated patients during the first 19
years of follow-up, with 6.7 expected (RR = 0.81; 95%
CI = 0.2-1.7). The leukemia risk by time among irradiated women is shown in Figure 2. A significantly increased RR of 3.5 (95% CI = 1.3-7.6) for ANLL is
observed 1to 4 years after primary treatment; the RR
then decreases and no case is observed beyond 15 to 19
years. The RR for chronic lymphocytic leukemia (CLL)
remains constant at the expected level for the entire
follow-up period (20+ years; RR = 0.9).
IP<O.oOl
Discussion
Studies of multiple primary cancers may point to
common exogenous and endogenous risk factors for the
first and the second primary cancer, to genetic susceptibility or to an etiologic role of treatment for the first
~ancer.~~J~-*'
Cervical
'
cancer patients are probably
those mOSt extensively studied over the years with respect to second cancer de~eloprnent."~~~-'~*-~'
A Par-
684
CANCERFebruary IS 1988
.
Vol. 61
No.
TABLE5. Morphologic Findin@ of Second Primary Sarcoma 10 or More Yeam Aftcr Irradiated Invasive Cancer o f t k Cervix, According to
Distance From the Cervix and Treatment With or Without Radiation, Compared With Incident Soft Tisnue
Sarcomas Amom Women in Denmark 1978-1982
Irradiated
Denmark 1978-1982
Site relative to cervix’
Site relative to ceMx
Close
Morphologic features
Intermediate
1
Sarcoma NOS
Leiomyosarcoma
Carcinosarcoma
Fibmsareoma, fibrous histiocytoma
(malignant)
Myxolanama. lipovumma
An&sarmma
Other specified sarcomas
All sarcomas
No.
Percent
I
2
5
I
-
2
-
-
-
I
I
I
-
II
65
4
24
‘ C l o x female genital organs, unnaty bladder, m t u m , pentoncum-mmpntomum and connective tisue of plvis, intcrmediaie.
thi
Clm
Intermediate
I
-
64
103
32
13
33
3
32
I2
4
64
47
13
49
5
I
7
3
29
I2
8
91
372
65
IO
188
2
I2
60
I
do
an
Distant
I
lev
Ca
th
Distant
or
ct
ce
O‘
di
9.
T
ic
e
a
30
kidney, u e r , descending colon, small intestine, stomach, paand liver, distant: all other sites.
r’
a
E
ticular feature is the risk related to therapy among
women followed for more than 20 to 30 years, since only
limited information from uniform registration schemes
has previously been available for such long-term SUMV O T S .This
’ ~ study examines irradiation as a dichotoNO. OF CANCER CASES
(+)
(-)
216
52
r2a
u
l(H
U
23
4
m4
s
2
I
1-e
I
lo-I*
,
?a-=
I
!
L
ao+
TIME SDWX flRn TRE*lUENi (YEARS)
FlO. I. AtVibuPMe I¡&
(AR) pr 10,OOO woman-yM (WY)
and
rclativc nak (RR)with 93%mnhdena intervals for developing a sec.
onday prinuty unta 8t any Yle close to or at immdule diamnce
from the d a by time Once hirt tmtment among women irradiated
(+) or not imdiaud (4 tor d c d einar.
t
mous variable as reported to the Registry; no account
could be taken of other known risk factors for cancer,
operating among cervical cancer patients.”
Overall, no major difference between irradiated (RR
= 1.I ) and nonirradiated (RR = 1.3) invasive cervical
cancer patients is observed although a somewhat steeper
increase in risk over time is observed for irradiated patients. However, one fourth of the “nonirradiated”
group probably received radiotherapy, limiting the validity of using this group for comparison. Furthermore,
due to the smaller number of long-term survivors
among nonirradiated, confidence limits of the RR are
wider than among irradiated, thus rendering comparisons difficult. A valid comparison group could be nonirradiated carcinoma in situ patients, who also show no
difference in RR ( I . 1) from the irradiated invasive
group; however in siru patients in Denmark have not
been followed for a sufficiently long period for comparison with long-term survivors of invasivecervicalcancer.
Time trends in RR combined with risk differences between irradiated and nonirradiiated groups for specific
sites were thereforechosen to evaluate the possible influence of irradiation on the observed risks.
The extended follow-up provides further information
on the long-term hazards of radiation, inasmuch as pnviously demonstrated excess and decreased risks persist.” The risk attributable to radiation for cancer in
organs close to or at intermediatedistance from the cervix increases with time to 64/1oooO women yeam, as
daes the RR (to 1.9; 95% CI = 1.4-2.5) (Fs I). The
attributablerisk only changes slightly if the uterus is not
included in the calculation. Intmstingly, the attributable risk increases from a lower level at the OUM to
z
1
4
I
!
I
...
--’
SECOND PRIMARY CANCER
AFTER TREATMENT
No. 4
rs
.e
1I-
D
e
t
levels comparable to those published on data from the
Connecticut Tumor Registry.”
Other genital organs, uterus, and ovanes receive large
doses of radiation3’during treatment of cervical cancer,
and it is thus not surprising that the risk for cancer at
these organs increases with time since exposure and that
there is a significant excess of cancer of “other genital
organs.” Adjusting for uteri and ovary at risk does not
change the RR appreciably after an irradiated invasive
cervical cancer (RR uterus = 0.7; 95% CI = 0.5-0.9;RR
ovary = 0.6; 95% CI = 0.4-0.7), whereas the deficit
disappears in the nonirradiated group (RR uterus = 3.3;
95% CI = 0.7-9.7; RR ovary = 2.2; 95% C1 = 1.0-4.0).
These findings are at variance with those from Connecticut,mwhere cancers of the ovaries and uterus are above
expectation after radiotherapy; however, the time trends
are similar to those in the current study. The discrepant
results for the RR may be due to a more conservative
attitude towards accepting new primary cancers in
organs close to or adjacent to the cervix in Denmark
than in Connecticut.18*uThe same adjustment for uten
and ovary at risk for nonirradiated in situ patients (61%
uteri and 88% ovanes intact)I0 changed the R R uterus to
0.3 (95% CI = O. 1-0.6) and the RR ovary to 0.8 (95% C1
= 0.5-i.2), which agree well with previous findings.”
The urinary bladder receives radiation doses comparable to those of female genital organs” when a woman
is treated with radiotherapy for cervical cancer. A significantly increasing trend in nsk by time and a significantly increased RR for bladder cancer among longterm survivors are seen only among irradiated invasive
cervical cancer patients. This indicates that radiation
-.“
I,
TABLE6 Obse~ed
and fipmed Second Pnmary C a n a
Among Sei&
Pstlenu With Carcinoma In Siiu ofthc Cervix
Uten Foilowcd for 10 or More Yean According to
Trcatmcnt With or Without Radiation
?
-r.
Seeond primary w a r
(ICD7 code no.)
. . ” ,
r_.
L,..
,...
,
..
F
.
L.
O
E
Stomach(151)
Colon (153)
Rectum (154)
Panmas (157)
Lung (162)
Brega ( 170).
Corpus uim (172)
Ovary (175)
Other gmital(l76)
O
3
I
2
2
2
2
0.6
1.5
0.8
0.5
1.1
4.1
1.3
2
Melanoma (190)
Brain (193)
Connectiwtissue(197)
O
O
O
0.2
0.5
0.4
Bladdrr(181)
r
,
*P< 0.05.
f P < 0.01.
# P < 0.001.
Not irradiated
Irradiated
o
2
1.1
0.5
0.2
O
0.0
2.0
1.3
3.9
1.8
0.5
1.6
0.0
11.8.
4.7
0.0
0.0
0.0
F
0
E
OF
6
4.2
12.4
6.7
4.2
12.4
50.7
13.9
11.8
1.6
4.3
5.5
1.4
0.8
0.6
1.7
2.3t
1.0
O.2t
0.8
4.9#
1.4
0.7
0.6
0.0
10
4
7
29
50
3
9
8
6
4
3
O
5.4
2.1
storm
685
OBSERVED NO OF CMES
W
cu
RRC
6
2
4
2
2
1
3
2
I7.6
5
4
3
2
*NU
cu
1
C
1
I
I
I
1-4
5-8
10-14
15-le
TIME SINCE FIRST RADIATION TREATMENT (YEARS)
FIG. 2. Relative risk (RR) and 95% confidmcc interval for acute
nonlymphocyticleukemia (ANLL) and chronic lymphobktic leukemia (CLL) among women trratcd for cervical canar, by time since
fim radiation treatment
plays a part in the increased risks, and the findings corroborate well with those of other^.^'^'^." However, exc e s so
~ f~bladder cancer are observed after both invasive
and in situ cancers, irrespective of treatment. Cigarette
smoking, which is an identified risk factor for both cervical cance332and bladder cancer,)’tM may play an additional role.
The rectum also receives high doses,)’ and radiation
may induce some of the rectal cancer excess observed, as
seen in studies o f women treated for benign gynecological disease’.’ and cervical
No excess risk of
rectal cancer was observed among atomic bomb survivors’ or spondylitis patients,” in whom radiation doses
would be lower. In the current study also, the findings
for nonirradiated women are similar to those for irradiated patients although not attaining statistical significance.
During radiotherapy for ceMcal cancer, different sections of the colon receive widely varying radiation doses
from high levels (7000 rad) to low levels (4QO rad). In
this study the colon was not subdivided into anatomical
686
CANCERFebruary 15 I988
sections; in contrast to the rectum, no increased risk was
observed (RR = 1.0). Increased mortality from colon
cancer has been observed both among spondylitis patients)' and among people treated for metropathia hemmorrhagica? in which radiation doses are lower than
after treatment for invasive cervical cancer.
It has been ~peculated'~
that the low colon cancer risk
can be explained by the fact that low social class is a risk
factor for cancer of the c ~ M x and
' ~ high social class
colon cancer.38The low risk observed among nonirradiated in situ patients (RR = 0.8)supports this explanation, but not the consistently highest risk (10%-40%,
Table 4) among nonirradiated invasive cervical cancer
patients through all periods. However, the increased risk
seen among nonirradiated as compared to irradiated patients with invasive cervical cancer, is probably due either to chance or, although speculative,to an interaction
o f endogenous estrogen production with the colon
cancer risk among those with functioning ovarie~.'~
An increasing trend in risk by follow-up is observed
for both pancreatic and stomach cancer after an irradiated primary, reaching a maximum after more than
30 years of follow-up. This is in agreement with observations on irradiated spondylitis patients3' although the
radiation doses are lower (99-300 rad):) and in other
studies on irradiated cervical cancer patient^?^,^^ Stomach cancer and pancreatic cancer also were increased in
the Connecticut
however, no observation was
made after 20 to 29 years of follow-up. Since overall
elevations of risk also are observed among nonirradiated
patients, cigarette smoking, which has been related to
both stomach" and pancreatic cancer,'6 may be responsible in part for the increased risks, even among irradiated patients.
The lung cancer excesses observed among both irradiated and nonirradiated patients (Table 3 and 4) and to
some extent the excesses o f esophageal (Table 3) and
stomach cancer, also may be explained by risk factors in
common with cancer of the cervix, i.e.. smoking and
social class.'2 The excess of lung cancer agrees well with
the hypothesis of ~molon&''~~~
as Seen in other studies o f
cervical cancer patients."." However, the pattern o f risk
over time is at variance with the observationsof h i c e et
where the RR declined to normal levels after 20
years of observation. In our study, the RR remains consistently elevated throughout the follow-up period, indicating an effect on all patients and not an effect related
to misclassification of lung metastasis after the primary
cancer. Kleinerman et al." attributed a five times higher
RR of lung cancer among irradiated than nonirradiated
cervical cancer patients to differences in social class and
Stage of disease being associated. which should in turn
be related to radiotherapy; although plausible, this is not
Corroborated by the results from Denmark.
Vol. 61
One of the most striking findings is the significant
deficit of breast cancer after an irradiated invasive ceMcal cancer (RR = 0.7), a deficit that is apparent
throughout the follow-up. No deficit is observed IO or
more years after diagnosis of nonirradiated carcinoma
in sifu (RR = l.O), whereas a slight hut nonsignificant
deficit is observed after a nonirradiated invasive cervical
cancer. Ovarian function may be assumed to be intact
for all (insitu) or a large proportion (invasive) of the two
nonirradiated groups. Surgically and radiation-induced
menopause have been shown to reduce breast cancer
risk,''.'' and radiation seems to be etiicient in this respect even if women are irradiated at ages older than 50
years? The younger a woman is at artificial menopause,
the lower the breast cancer risk".? although age has not
been considered in this anaiysis, our previous follow-up
showed the lowest risk for breast cancer after irradiation
at young ages." Endogenous estrogens have been related
to increased breast cancer risk." whereas exogenous n
trogens given as replacement therapy increase the risk
only slightly, if at
A s in other s t ~ d i e s ' ~ . ~ ~our
-"
findings point to a protective effect of cervical cancer
irradiation on breast cancer, probably due to destruction
of the ovaries. In addition, cervical cancer patients have
risk profiles, e&, early age at first birth, low social class,
that would result in lower breast cancer rates compared
to the general population."
Other cancer deficits among irradiated women are
those of skin melanoma (RR = 0.6)and brain tumors
(RR = 0.6). Exogenous estrogens and reproductive factors may influence the development of melanoma," and
a protective effect of bilateral oophorectomy has been
described!6 Breast cancer was found in excess after malignant melanoma in one study" but not in another.q
Meningiomas and breast cancer may share hormonal
risk
since meningiomas constitute a substantial proportion of brain tumors, ovarian ablation may
influence the deficit.
The excess of connective tissue sarcomas is present
among both irradiated and nonirradiated invasive cervical cancer patients followed for more than 20 years
(Table 4); however, a significantly increased risk is Seen
only among irradiated women. Several reports of radiation-induced soft tissue sarcomas have been publi~hed,'.~.~'
all showing a clear relation between the site
of the sarcoma and the therapeutic fields. However, the
relative excess of 19% (Table 5) observed for sarcoma
development close to or at intermediate distance from
the radiotherapy fields is not statistically significantly
different from the expected distribution in the female
population. Wagone? showed a ratio 3: I for leiomyoSarcomas and carcinosarcomas among women irradiated for gynecologic disorders. In the cumnt study,
the overall distribution by morphologic features of soft
N<
ti!
Wi
is
r
i
fc
b
e:
e:
t<
d
Ji
C
tl
I!
I
a
t
t
I
1
(
1
I
!
No. 4
SECONDPRIMARY
CANCERA ~ E TREATMENT
R
Srorm
tissue sarcomas was no different from that among
women in Denmark.”
Although a significantly i n c d RR (3.5) for ANLL
is observed only during the first 5 years of follow-up, the
risk continues to be slightly elevated (nonsignificantly)
for 19 years. N o excess is seen of CLL,which has never
been assoCiated with radiation.’ Surprisingly, in most
leukemias were not observed in
early
exccss after radiotherapy for cervical cancer, in contrast
to the result of studies of patients who m i v e smaller
doses of radiation to the bone marrow?d In other largescale studies of cervical cancer patient^,'^.",'^ an increased risk was observed, although it was much smaller
than predicted on the basis of radiation dose to the
marrow.
It has been argued that very high doses of radiation
inactivate or kill cells in the bone marrow. The results of
a recent case-control study among cervical cancer patients in Denmark’ are in line with this hypothesis, as is
the analysis of medically irradiated populations.” The
highest RR (12.9) for A N L L is observed among in sifu
patients. An explanation to the high risk among irradiated in siru patients compared to invasive, could be
that the bone marrow receives lower doses, if in siru
cancers are treated with hrachytherapy alone as is often
seen for invasive Stage I cancers, or with reducedexternal-beam therapy. If so, less inactivation or cell killing in
the bone marrow is likely, and a higher leukaemogenic
potential of the radiation a possibility. The analysis on
leukemic risk would have benefitted if numbers of leukemia and quality of notifications allowed differentiation between carcinoma in sifu, Stage l cervical cancer
and other m e s .
Generally, in studies of cervical cancer patients, including population-based studies, the overall risk of second primary cancer is very close to that of women withoutcancer(RR= 1.1,’5RR=1.4,)0presentRR= 1.1).
Several caveats must be kept in mind when interpreting
the current data and comparing them with others. These
include differences in the definition and classification of
second primary cancers; lack of correction for organs at
risk:6 underreporting,2*and decreasing autopsy rates52
also may mask true risks. In addition, a substantial study
size and long follow-up are needed in order to demonstrate the relatively small increased risks of second primary cancer related to radiotherapy. A further problem
is associated with validity of exposure information.
a
In conclusion then, by preventing death from cancer
of the cervix, radiation for ceMcal cancer over a 35-year
period in Denmark has influenced the risk of second
cancer both directly, by increasing the risks of bladder,
other genital, rectal, and connective tissue cancers,
ANLL, and perhaps stomach and pancreatic cancers,
and indirectly, possibly through some hormonal mecha-
687
nism, by decreasingthe risks of cancers of the breast and
perhaps brain, skin melanoma, and colon cancer. The
role of radiation in these excmes and deficits will be
addressed in d&l in case control studies both nationally and internationally,” in which aciual organ doses
will be taken into account” and controlling for the influence of confounding factors is possible. However,
follow-up for life of irradiated populations is warranted
to evaluate fully the long-term carcinogenic risk of radiation.
REFERENCES
I . Boia JD.
C i n a r following medical irradiation. Cancer 1981;
47:1081-1090.
2. Schull WJ. Atomic bomb survivors: Patterns of canar risk. In:
B o k I D Ir, Fraummi IF Ir, eds. Radiation C u c i n ~ n m sEpidemi:
ology and Biological Significance. New York Raven Press, 1984;
21-36.
3. Darby SC. Nakashima E, Kat0 H. A patallel analysis of cancer
monality among atomic bomb suMvors and patients with ankylosing
spondylitisgkn x-ray therapy. JNafl Cancer Inst 198% 75:I-Zl.
4. Unitcd Nations Scientific Committee on the EK& of Atomic
Radiation. Ionizing radiation: Sources and biological dfau.In: Official Records ofthe General Auembly, ThinySeventh W o n , suppl.
no. 45. New York United Nations, 1982.
S. National Academy of Sciences Advisory Committee on the Biclogical Eñens of Ionizing Radiations. The &cas on populations of
exposure to low levels of ionizingradiation. Washington, DC National
Academy Rers 1980.
6. lchimm M,Ishimam T,Bel* JL. lncidena of leukemia in
atomic bomb survivors belonging to a fixed w h a in Hiroshima and
Nagasaki, 1950-1971: Radiation doze, ycan &r wpouire, age at
exposure, and typ of leukemia. Jpn Radiar Res 1978; 19262-282.
7. Smith PG,Doll R. Late dfau of x-irradiation in patients treated
for mctropathia hacmorrhsgica. Br JRadiol 1976; 49224-232.
8. Wagoner JK. Leukemia and other miliSiipncinfollowing radiation therapy for gynecological disorders. In: Boia I D Ir, Fraumcni JF
Ir, eds. Radiation carcinognicsis:Epidemiology and biological sinificana. New York Raven Rcsr, 1984; 153-159.
9. Storm HH. Boia JD Ir. Leukaemia after &cal
cancer irradiation in Denmark. inr J €pidemio/ 1985; 14363-368.
IO. Storm HH, Jcnscn OM. Secondary primary canars among
40518 womcntreatcdforcanaror~inomainsifuofllicanixuten
inDenmark 1943-1916.111: DayNE,BoiaJDJr,edr.SemndCanar
in Relation to Radiation Tnatment for Cewical Canar (IARC Sci
publ. no. 52). Lyon: International Agency for Rcscnrch on Cancer,
1983: 59-69.
I I . Storm HH, Ewcnz M. Scmnd war following cancer of the
female genital system in Denmark 1943-1980. Nafl Cancer Insf
Monogr 198% 68331-340.
12. Pettcru>n F. Bj8rkbolm E. Staging and rcporüng of arvical
carcinoma. Sernin Oncol1982; 9287-298.
13. Hutchison OB. Leukemia in patients with cancer ofthe arvix
uUn tnatcd with radiation: A report wvcring the 6rslS yean of an
intemationai audy. J Natl Cancer Inst 1968; 40.951-982.
14. Boia ID, Hutchison OB. Leukemia in women followingradictherapy for aMcal cancer Ten-year follow-up of an international
study. JNaflCancerInrt 1980; 65:115-129.
IS. Bois I D Ir, Day NE, Andervn A el a/.Semnd canars following radiation treatment for aMcal cancer An intrrnationaiwhboration among canm registries. J Natl Cancer imf 198% 14995-975.
16. Curtis RE, Hoover RN, Klcinrrman RA, lbwcy EB. Second
cancer following cancer ofthe female genital aynem in Connecticut.
Nat/CancerInstMonogr 198% 60113-137.
17. ClemmI. Statistical studies in the I*iology of d ¡ a n t
neoplasms: 111. Denmark 1958-1962. Acta Pafhol Microbid Scand
[Supp/]1969.
688
-
CANCERFebruary 15 1988
18. Jcrika OM, Storm HH, Jcnm H. Cancer ngistration and the
study of multiple primary canccn in Denmark 1943-1980. NalI
Cancerlmt Monogr 1985; 60245-251.
19. Danish Cancer ReButry. Cancer incidmce in Denmark I981
and 1982. Dan& Cancer Sonay, Copenhagen, 1985.
20. Storm HH. Complnenes of cancer regklration in Denmark
1943-1966 and cliicacy of record linkage praedum In1 J Epidemiol
(in pma).
21. 0stcrlind A, Jcnsen OM. Evaluation ofcaocer in casts in 1977
Preliminary evaluation of regklration of cancer casts by the Cancer
Registry and National Patient Registry. Ugeskr Laeger 1986;
1472483-2488(in Danish).
22. Storm HH. L y w E, Wründ A, Jmscn OM. Multiple primary
cancers in Denmark 1943-80: lnüucnce of poaslbk undcrreponing
and sugpatcd risk facton. Yak J B i d M e d 1986; 58547459.
23. Monson RR. Analysis of d a t i v e survival and proportional
mortality. Comput Biomed Res 1974; 7325-332.
24. Rothman KI, Boice JD Ir. Epidemiologic analysis with a m
grammable calculator. -on:
Epidemiology Rcsourq 1982:30-31.
25. Schoncnfeld D. Multiple primary canan. In: Schottenfeld D,
Fraumeni IF Jr. Cdr Cancer Epidemiology and Prevention. Philadelphia: Saundcn, 1982; 1025-1035.
26. Boia JD Jr, Storm HH. Cunis RE el al. Introduction to the
study of multipk primary canccn NalI Cancer INI Monogr 1985;
683-9.
27. Storm HH, Jenscn OM, Ewe* M el ai. Summary: Multiple
primary canan in Denmark, 1943-1980. Nall Cancer Ins1 Monogr
1985:68411-430.
28. Sarensen B. Late mults of radium therapy in cervical carcinoma. Ac~aRadld 1958: (Supil) 16995-100.
29. Dickson RS.Late mulu of d u m treatment of carcinoma of
the cervix. Clin Radio/1972; 23528-535.
30. Kkinemun RA, Cuflis RE, Boice JD Ir, Flannery JT, Frau-
mcni JF Ir. Second c a n m foliowing radiotherapy for m i c a l cancer.
J NaII Cancer Ins1 1982; 691027-1033.
31. Clarke EA, Krciger N, Spcngler RF. Ssond primary cancer
following treatment for cervical cancer. Can Med Assoc J 1984:
131553-556.
32. Keky JL, Hildreth NG. B W and gynecological cancer epidemiology. BaaRaton,FL:CRCPrcss, 1983: 117-141.
33. Stovdl M. Organ dosca from radiotherapy ofcancer ofthe uterine cervix. in: Day NE, bin JD Ir. eds. Second Cancer in relation to
radiation treatment for cervical cancer (IARC Sci publ. no. 52). Lyon:
international Agcncy for Raeardi on Cancer, 1983: 131-135.
34. Lynge E. Storm HH, Jenxn OM. Evaluation of trends in soli
tissue sarcoma according to dipsnaatic criteria and consumption of
phenoxy herbicida. Cancer 1987: M):1896-1901.
35. Lockwood K. On the etiology of bladder tumours in Kebenhavn-Frcdenkskg! An inquiry of 369 psticntsand 369 controls. Acla
Patho/ Microbid S c a d [Supp/)1961: 51:1-166.
Vol. 61
36. Doll R, Pelo R. Mortality in relation 10 smolrino: Twenty yean
obseivationonmakBrituhdMonBrM~J1976;21525-1536.
37. Smith Po,Doll R. Mortality among paticnu with ankyloYng
spondylitis aRcr a Einglymtment coune with x-ray* Br Med J 1982;
284449460,
38. Schottcnfcld D. Winawcr SJ. L ~ Q Cintestine. In: Schoncnfeid
D, Fraumeni IF Jr, cds. Cancer epidemiology and prevention. PhiL,
delphk Saunden, 1982; 703-727.
39. McMichacl AJ, P O W JD. Reproduction. end-nous
and exogenous sex hormones, and colon canA review and hypothesis 3
Not/ Cancer Ins1 1980: 651201-1207.
40. Hacnslcl W.Correa P. Developments in the epidemiology oí
stomach cancer in the past decade. Cancer Res 197% 353452-3459.
41. Winkelstein W Jr. Smoking and cancer of the uterine cervix
Hypoth& Am 3 Epidemid 1977; Ioó:257-259.
42. Winkelstein W Ir, Schillitoe U.Brand R. Johwn KK. Funher
comments on cancer of the cervix, smoking and berprvinu infection.
Am JEpidemid1984; 1191-8.
43. Trichopoulos 0, MacMahon B. Cole P. Menopause and b m
cancer risk. J Nall Cancer Ins1 1972; 48:605-6 13.
44. Feinleih M. Breas cancer and artificial menopsuse: A cohort
studv. 3 NaI1 Cancer Ins< 1968: 41:315-339.
45. HOUYEA. weis NS.LHIM. CurPneous melanoma in relation
to exogenous hormones and reproductive fanon 3 Nad Cancer I ~ I
1983: 70827-831.
46. Gallagher RP, Elwood JM, Hill Gü, Coldman Al. Threlfdl WJ.
Spinelli 11. Repmductive facton, oralcontraceptives and risk of maügnant melanoma: Western Canada melanoma study. B r J Cancer 1985:
52901-907.
47. Tucker MA, Boia JD Jr. HoiTman DA. Second cancer following cumncous melanomaand cancen of the brain. thyroid. connective
tissue. bone and eye in Connecticut. 1935-1982. .Vai/ Cancer INI
Monogr 198% 68: 161- I89
48. &ierlind A. Olxn JH. Lyngc E. Ewe* M. Semnd cancer rollowng cutaneous melanoma and canccn of &e brain. thyroid, COD
ncctiveiissue. boneandeye in Denmark. 1943-1980. S a t l C a n c r r l ~ i
Monogr 1985: 68361-388.
49. Schoenbeg BS. Christine BW. Whisnant JP. Nervous system
ncoplarmr and primary malignancies of other rites Thc unique ars00ation betuun meninpicomas and brrast cancer. \;woloXy 1975:
25:705-7 12.
50. Pinkston JA. Sckinc I. Postirradiation LZrmmi (malignant fibrous histiocytoma) folloWng ccrvia cancer. Cancer 1982 49434438.
5 I. Kim JH, Chu FC. Woodward HQ et al. Radiation induced SOR
tissue and tone wcoma. Radidom 1978: 129MI-508.
52. Storm HH, Andenen J. PerrrntsOe of auropaia in cancer po
tienuin Denmarkin 1971-1980.UgrskrLarger 1986.1~.1I10-1I14
(in h n i s h ) .
The Use of Magnetic Resonance Imaging and Spectroscopy
in the Assessment 07 Patients With Head and Neck
and Other Superficial Human Malignancies
W. GlLLlES MCKENNA, MD. PHD,. ROBERT E. LENKINSKI. PHD,t ROBERT A. HENDRIX. MD,S
KATHERINE E. VOGELE. BA.7 AND PETER BLOCH, PHD'
The proper demarcationof d i d tissue is important for radiation thenpy phnlng and trritwiit. The
volume to he irrndmted is vsuplly identified on radiographs or on x-ray computed tomography (cr)
soetions. Magnetic resonance (MR)-derived images of the proton Ti. relwtion times in prmul *el
sharper differentiation between nMm;il and
elements, typically 0.5 nun* or less, provide [email protected]
diseased tissue. The T2 vdws in tissuc depend on tbe tissue composition,histologic condition, and physiologic environment within the tumor. Furthermore, for many tumors the histogram of R v d a s h.s a
dear biphasic distribution suggesting thpt R m p s may he Usctol for the identifiation of necrMic or
hypoxic regions within tumors. The distribution of T2 values witbin the tumor bed shorn the geneml
pattern that the T2 vdues are elevated with a range greater than that seen in normal muscle. E h r t e d
T2 values are not by themselves dingamtic of mliginacy; however, they demonskate the heterogeneity
of ihe microenvironment present within a tumor. The spatid distribution of R vplnes ishe¡¡ explored
as a method for computer wsishnce in the delineation of the target volnme for treatment pbo¡n& In
addition, M R P-31 spectroscopic eximinntiws were performed on 30 patients with squimws cell carcinomasof the head and neck. Although hampered by m d e contaminationin Some P-31 spectra o b l 8 k d
with surface coil profile localintion techniques, significanttrends cm still be appredated ia our datu.
These trends include the following (1) the P-31 specbufrom millgnint tissue have well-resolved spectral
lines in the upfieid region that corres&
to Pi, pbsphomomester (PME), and phosphodiester (PDE)
not usuilly seen in normal muscle; (2) the PDE/B-ATP and PME/B-ATP ratios are greater tbia unity
In nU a s ; and (3) most of the tumors have bigher P M E peaks than PDE peaks. The P-31 speebi from
patients treated with ionizing radiation changed during and after tbenipy. Some of the c h g c s emld be
associated with alteration of the tumor metabolic activity or synthesis and breakdown of lipoproteins.
These studies suggest that mngnetic resotmm imaging (MRI) and magnetic resonmce speftmseopy
(MRS) studies my be useful for both radiotberapy treatment phoning and the noninvasive momking
of patients both hefore and during treatment.
Cancer 61:2069-2075, 1989.
T
HE EFFECnVE TREATMENT Of tUmOrS by radiaUOn
therapy depends on accurate delineation of the mah a n t tissue. Failure to treat the entire volume infiltrated
by a tumor will result in local recurrence. There also is
evidence to suggest that local failure may result from the
p m n c e of heterogeneous, hypoxic regions within a tumor
v-kme. Additionally, one of the assumptions underlying
fractionated radiotherapy is that it allows the reoxygen-
-
From the Dcparimcnis of *Radiation Oncology, tRadiolugy, and
$Otorhinolqngobgy,Hospital ofthe Univenity ofPennsylvania, Phil.dclohic
l v a-~~
nh.
- -h- n w
--,&\ad,
for npnnts: W. Gillin McKenna MD,PhD, Dmrtmerd Of
Radiation Oncolw, Hospital ofthe Univmity ofp~nrylvania,MMJ
spnioc Sheet, Philadelphia, PA 19104.
Accepted for publication May 26, 1989.
ation of hypoxic tumor tissue to take p W but the kinetics
and scope of such reoxygenation have not been studied
extensively in human tumors. Magnetic resonance imaging (MRi) and magnetic resonance spectroscopy (MRS)
offer the oncologist new methods with which to noninvasively study the extent and metabolism of tumors and
to follow the changesthat occur in tumors as they undergo
treatment.
A prospective study was initiated ofpatients with head
and neck and other superhcial malignancies before and
Our goal in these studies was to
whether MRi and MRS can be used to dan and follow
therapy. We used MRI to study whether iKtter definition
of tumor volumes can be
we can
develop automated computerized methods for doing so
I..
2069
V
,
-.
I
,.<.
:
pto. 10
.....
NONINVASIVE
MONITORING
OF RADIOTHERAPY PATIENTS
*
McKenna et ai.
207 I
.,
Results
Analysis of T2 Distribution in Patients
Figure I shows both the quality of the T2derived image
for a histologically confirmed liposarcoma in the thigh of
natient and the tissue heterogeneity within the tumor
&. The T2 distribution throughout the entire tumor and
n o i d surrounding tissue shown in Figure 2 was obtained
by evaluating the T2 values in 17 contiguous axial d o n s
7.5 mm apart. The means of the derived T2 values for
normal muscle and the liposarcoma were 35 and I 10ms,
respectively, with all of the T 2 values of normal muscle
king less than 45 ms. The maximal values of the T2
values within the tumor volume were found at the center
ofthe tumor volume. The range ofvalues ofthe T2 values
for the tumor volume shows a broad peak extending from
311?o 180 ms.
In Figures 3A and 3B, the T2 distribution within the
tumor and normal volume, respectively,are shown for a
patient with a T2NO squamous cell carcinoma of the floor
of the mouth. Once again, the T2 distribution shows a
mean value higher than that of adjacent normal tissue
and the range of values also is large (35 to 95 ms). In
Figures 3C and 3D, this is compared with the distribution
seen in a patient with a benign branchial cleft cyst that
was found to have an extensive lymphoid infiltrate and a
I rge cystic component at surgery.
Figure 4 shows the regional T2 distribution in multiple
axial slices through a large soft tissue sarcoma. In this
case, we show the change in the distribution of the T2
values in serial cuts through the tumor volume at 7.5mm intervals, not the integral value throughout the tumor
volume. Position I is defined as the most inferior part of
the tumor volume and position 16 as the most superior
an,
:din
n)-I
ninLI1
FIG. 4. The regionai distribution of TZ in a SORtissue sxoma. The
arc shown in 16 axial sliar xpratcd by 7.5 mm. Slice
I m m p o n d s to the moat inferior imaged d o n and slice 16 the most
superior. The nominnl cmvsl ponions of slica 6 thmugh IO show the
TZ -ms
bgm n u m h of regionally clevited T2.
part. This illustrates the changes that are seen in comparing one region of the tumor to another. Note that the
largest T2 values are seen close to the center of the tumor
volume (positions 6 through IO), although the values at
the edges of the tumor mass remain elevated above normal.Furthermore, there appearsto te several components
of the T 2 distribution, including one which is elevated to
higher levels and Seen most clearly in the central part of
the tumor volume. Of course, it is possible that some of
these components represent edema in surrounding normal
tissues rather than the presence of malignant tissue per
se. The analysis of this possibility will await better correlation of T2 maps with resected specimens after surgery.
+f
FIG. 3. Ibe TZ distribution within squamous ccll carcinoma of the
Boor of the mouth (A). The mean value and range of values ir m
t
e
r
Analysis of "P Spectra
thanthatseeninadjnanf ~ r m a l t i s ~ ~ ( B ) . A J i m i l s r p a t t ~ n ~ ~ f o u n d
in a patient with a nonmalignant branchial cleR cyst with lymphoid
Figures 5A and SB show "P spectra for two patients
tissue (C); however, mmpMsd with n
d adjaant tisue (D),
the T2
who
had T3NO squamous cell carcinoma of the lower lip.
distribution within the b e n i IsUon -s
a normal pius a sndc
The tumors masses in both of the patients were large
abnormal twional Ti.
~
c_
2070
CANCERNovember 15 1989
Vol. 64
-
No
as an aid to three-dimensional treatment planning. In addition, MRS was explored to see if we can identi@, among
patients of similar presentation, those patients with significant hypoxic subpopulations within their tumors. W e
wanted to assess whether this will assign them to groups
of varying risk of local or distant failure. Gatenby m al.'
used an oxygen electrode to directly measure oxygen tension within lymph node metastases in patients with head
and neck cancer. In this study, those patients who were
shown to have hypoxic tumors were at increased risk for
local failure. This report presents some preliminary data
gained from patients entered early into the studies currently underway at our institution.
An
for
ar
bec
o
40
80
120
160
no
by
1.5
no
res
be¡
Val
oft
for
200
Materials and Methods
MRI and spectra were performed on patients using a
1.5 Tesla GE-Signa whole body scanner (General Electric,
Milwaukee, WI). Typical pulse sequences were used to
obtain T I and T2 weighted images. In addition, the relative spin-spin relaxation times, T2, were determined in
small voxel elements (.7 mm X .7 mm X 6 mm thick)
throughout the imaged volume. Typically, 15to 20 axial
sections encompassing the lesions were imaged. A least
square fit of the amplitudes of four multiple spin-echo
signals was used to compute a pixel map of a T2 distribution? The pulse sequence consisted of a 90' pulse followed hy four 180" refocusing pulses with spinecho times,
TEs, corresponding to 20,40,60, and 80 msec. The pulse
repetition rate, TR, was long (2500 ms).The images were
transferred by magnetic tape to a V A X 780 minicomputer
(Digital Equipment Corp., Maynard, MA) where computation of the regional T2 image (256 X 256 pixels per
imaging plane) was performed in less than 1minute. The
FIG.I. Axial image through the thigh ofa patient suggesting r*onal
tissue hctnogcneity within the Liposarcoma
30
1
tur
pat
of
T2 (mSec)
Flc. 2. A histogram ofthe T2 disvibution throughout a lipoaamma
tumor (top) and normal adjacent tissue (bottom). The numbcr ofpixels
in 4ms T2 intervals is shown. The T2 values wcre evaluated from T2
m a p in 17 contiguous axial d o n s separated by 7.5 mm. Each axial
i2 image consisted of 256 X 256 pixek
uniformity in the derived T2 values throughout a large
CuS04 phantom (approximately 20 mi) was within 796,
and the long-term reproducibility during a 2-week period
was 3%. The T2-derived image was viewed on a Gould
8800 graphic work station (Gould, Sunnyvale, CA)and
the contour of the boundary of the lesion was outlined in
each section. Histograms were obtained of the relative T2
distribution within the lesion and adjacent normal tissue.
Using the results of the imaging studies, a 3-inch in
diameter, double-tuned surface coil was placed over the
lesion in such a location as to minimize the influence of
muscle tissue on the "P spectra. The coil was shimmeti
on the water signal. Approximately 100 to 200 90" radio
frequency (RF) pulses with a 4-second repetition rate were
applied to gather the "P spectra. Ail spectra were obtained
over a spectrum width of 2 kHz that was digitized into
1024 points. Spectral processing included exponential line
broading of IO H z with baseline correction and phasing
applied.
me
an<
Fig
see
war
lar$
I
axi:
cas
vali
mn
vol1
the
1,
rn
I.
I
II
,I
6
-
. ."
r .
Fn
nWr
than
in a
lissu<
distn
abno
I
.
CANCERNovc?mber 15 1989
2072
A
vol. 64
Discussion
U*
The Use of TZ Mapping to Delineate Tumor Volume
The distribution of proton nuclear magnetic resonance
(NMR) relaxation times in living tissue is the result of
contributions from tissues of different composition and
histologic condition (i.e., muscle versus fat), and from
O
B
-20
PPM
':
IC."
different physiologic processes occurring within the tissue
( i e . ,the extent ofhlood flow). It may be possible to gather
additional information on the physiologic environment
of a tumor in situ from measurements of regional proton
relaxation times. For example, the studies of Thulbom et
al'demonstrated a variation in T2, but not T1, as a function ofhoth hematocrit and oxygenation levels. This finding is attributed to the relatively large differences in the
magnetic susceptibility of oxygenated and deoxygenated
hemoglobin. This finding suggests that regional differences
in T2 values within a tumor may correlate with local vascularity. Interstitial oxygen in the tissues may act as a
weak paramagnetic ion to reduce the relaxation times of
protons assoCiated with water molecules?.' In addition,
recent measurementS in an animal model of proton MR
Pci
N
re
a!
ir
ki
ir
ai
ai
th
tu
ra
ar
sh
Ck
m
tu
tu
ar
gr
fir
tu
PC
an
of
f0i
th:
11
o
-20
m!
mi
see
PPM
FIG.5. "P s
q befon and after a single fmdon of 180 Gy for two
patients (A and B)with T3NO squamous cell caarcinoma of the lower tip.
(greater than 10 cm in diameter) and superficial. Therefore, "P spectra were obtained by placing a surface coil
directly on the tumor mass. Note that in Patient 1a clear
peak for phosphocreatine was present before any treatment was given and this did not change appreciably in
size after the fust fraction of I80 &y. In contrast, Patient
2 showed only relatively low levels of phosphocreatine
before treatmen$ however, after the fust treatment fraction, there appeared to he a clear rise in the levels of phosphocreatine present in the tumor mass.
Figure 6 shows spectra obtained from a patient with a
bulky, medically inoperable T3NO squamous cell carcinoma of the pyriform sinus. In this case, the tumor mass
was palpable in the neck. The surface coil was placed on
the palpable mass and its position verified relative to the
MR image that was obtained at the same session. Note
that the region corresponding to PME, PDE, and Pi was
large and relatively poorly resolved in the initial spectrum.
This region became smaller and better resolved in the
Scan obtained at 4800 &y, but was again more prominent
at the end of treatment (7400&y).
Slil
bri
va:
thc
in
Ue:
of
mi
PIS
erc
ofAn
I\
be
res
P i l o r t o Treatment
10
-10
-20
Chamical S h i f t (ppd
O
Flo. 6. Ckws in the "P spcira of a squamous cell carcinoma of
sinus before irratment and a k r 4800 and 7400 &y.
the pyiform
exa
anc
the
can
am
nor
in I
not
val,
c
C.
1. 64
,-..
-_
nce
oc-
md
omLsue
her-'
ent,
ton
1errnc-_
idthe,
ted
LesasE ar
of,
)n,
IR
P-
c
c
c
I
rb
.
c
..rL
n
L
c
If
c
No. 10
NONINVASIVE
MONITORING
OF RADIOTHERAPY PATIENTS
relaxation times in ischemic brain indicate that edema
associated with ischemia may be identified by an increase
in the regional T2 value! Many other factors also are
known to influence relaxation times within tumors. These
include cellular architecture and nuclear fraction? trace
amounts of paramagnetic metal ions: chemical radicals
and species such as hydrogen peroxide' and glycogen?
the cell cycle,l0 local inflammatory processes within the
tumor bed, and regional differences in cellular growth
rate." Kroeker et a/.!*looked at the distribution of T I
and TZ values in murine experimental tumors. They
showed that increases in the values ofboth variables were
characteristic of tumor growth rate, with T2 being the
more sensitive indicator.
W e have now demonstrated for a variety of human
tumors what had been previously demonstrated for animal
tumors, that the mean value of T2 for the tumor volume
and the range of values of T2 within a tumor volume are
greater than those seen in adjacent normal tissue. in the
first example in Figure I, the T2 distribution within the
tumor bed for a histologically confirmed high-grade liposarcoma of the thigh can be seen. This has been examined in multiple axial slices from the nominal center
ofthe tumor. When the integral distribution of T2 values
for the entire tumor volume is examined, it is apparent
that the mean of T2 values for the tumor is approximately
I10 ms whereas the value for normal muscle is only 35
ms. it also can be seen that all ofthe T2 values in normal
muscle lie below 45 ms. However, the range of values
seen in the tumor is much greater than this, with few
slices showing T2 values within the normal range. The
broad range o f T 2 values within the tumor impliesgreater
variation in microenvironments within the tumor. Furthermore, there appears to be a nonrandom distribution
in this regard. When the distribution of the mean T2 values is plotted along the length ofthe thigh, the mean value
of the T2 values is maximal at the center of the tumor
mass and decreases toward the periphery. This would imply for this rapidly growing tumor that the greatest heterogeneity in microenvironments is present at the center
of the tumor mass. However, other patterns are possible.
An increase in the value of the T2 values can sometimes
be seen at the periphery of a tumor mass, possibly r e p
resenting the presence of edema.
The same patterns can be seen in the patients we have
examined so far with squamous cell carcinoma of the head
and neck. Figure 3 shows the distribution of T2 values in
the tumor bed of a patient with a T2NO squamous cell
carcinoma of the floor of mouth. As in the previous example, the mean value of the T2 values is elevated above
normal and the range of values is greater than that seen
in normal muscle. However, in this case it also should be
noted that the separation between normal and malignant
values is not as clear cut as it was in the previous example.
*
McKenna et ai.
2073
W e are currently exploring the following possible explanations for this: ( I ) it may be a function of tumor histologic condition (the liposarcomas, which have a higher
fat content, might be expected to have higher T2 values);
(2) it may relate to tumor grade (higher grade lesions with
greater degrees of anaplasia may contain more heterogeneous regionswithin them); (3) it may relate to tumor
growth rate (rapidly growing tumon may have a greater
tendency to outstrip the blood supply, producing regions
of hypoxia and necrosis within them); or (4) it may be a
combination of all of these factors and others. W e hope
that some of these factors will become clear as we have
more opportunities to compare T2 maps with the histologic condition of grossly resected specimens, work that
we are only now beginning to do.
It should be noted, however, that an elevation of the
average values of T2 and an increase in the range are not
diagnostic of malignancy. In Figure 3 we show the T2
distribution of a patient who had a large parotid mass.
Note that here also we see an elevation in the average T2
value and a skewing of the distnbution to higher values,
although perhaps not to values as great as those seen in
the malignant lesions presented above. In this case the
resected specimen proved to be a benign branchial cleft
cyst, albeit with large regions ofliquidfaction necrosis and
an extensive lymphoid infiltrate.
Nevertheless, the T2 distribution may provide important information for delineating the extent of abnormal
tissue. For example, Figure 4 shows the regional T2 distribution in multiple axial slices through a large soft tissue
sarcoma. The large differences in T2 relaxation times between diseased and normal tissues may pmnit a computer
algorithm to be used for demarcating the boundary of the
abnormal tissue in each axial, coronal, or sagittal plane
through a tumor volume. The delineationoftarget volume
in each section is a prerequisite for treatment planning,
but currently this requires large amounts of physician
time. Computer assistance in delineating these boundaries
could significantly expedite the time required to identify
the target volume on multiple sections, thus removing
one of the major time constraints on three-dimensional
treatment planning.
Phosphaie Spectra in Tumor Tissue
spectroscopy can be used to gain noninvasive information on the status of metabolism within a tumor
volume."-" These studies tend to fail into the following
two distinct categories: (1) studies of untreated tumors
and (2) serial studies of tumors after therapy. Tumors in
experimental animal models may be characterized by elevated signal intensities from inorganic phosphate, phosphomonoesters, and phosphodiesters. The phosphomonoester and phosphodiester peaks have been associated
2074
CANCER November I5 1989
with the extent of cell membrane biosynthesis and degradation." Phosphocreatine (PCr) is a high energy storage
product usedby some tissues to maintain ATP levels when
demand exceeds the capacity for generation from oxidative phosphorylation. Therefore, the presence and
height of this peak may give some indication of the quality
of oxygenation within a tumor volume. Okunieff et
showed that the PCr/Pi levels in C3H murine fibrosarcoma and spontaneous mammary caminomas transplanted in mice were modulated by the inspired oxygen
concentration. In addition, they demonstrated lower PCr/
Pi ratios in larger tumors that are likely to have a greater
hypoxic component. This strongly suggests that changes
in the PCr levels during therapy may possibly be used to
follow tumor reoxygenation. The levels ofthe high energy
phosphates of nucleoside triphosphates tend to be more
constant and maintained in tissues until death. N g and
coworkersm have used "P spectroscopy to study the
growth of animal tumozx and their response to therapeutic
agents, chemotherapy, radiation therapy, and hyperthermia They concluded that tumors became less metabolically active as they grew larger and that tumor tissues
were more hypoxic than normal tissues. They also claim
to have seen rapid spectral changes in tumors in response
to therapy, and at least in their experimental system these
changes appeared to be specific for the type of therapeutic
intervention.
With the development of relatively high-field, wholebody scanners, many studies have been conducted on the
application of )'P spectroscopy to the study of intact human tumors.'3M23 These studies indicate that the MR
spectra of the tumors show marked changes in response
to therapy. For example, the studies conducted on neuroblastomas by Maris and Chance" indicated that the
level of phosphomonoester peak was a sensitive indicator
of tumor regrrssion. Most of these studies used surface
coils as the only method of localization, so it is likely that
the observed spectra were reflections only of the hulk or
average metabolic properties of the tumors. This may be
disadvantageous when studying relatively heterogeneous
lesions.
In Figure 5 we compare the spectra, before and immediately after starting treatment, of two patients had
large T3NOMO squamous cell carcinomas of the lower lip
(Fig. 5A). in one patient the peak for PCr was large before
any treatment was given, whereas in the other patient the
peak for PCR was small relative to the other peaks in the
spectrum. One interpretation of this result is that, in these
two patients with an essentially identical clinical presentation, one had a tumor that had a greater hypoxic fraction, as manifested by the much lower levels of PCr present
in the tumor tissue. Consistent with this hypothesis is the
fact that the calculated pH for the tumor tissue of this
patient was 0.3 to 0.4 units lower than in the other, which
Vol. 64
would be consistent with an accumulation of acidic products of metabolism in the tumor tissue. However, it is of
interest in this patient that after a single 180 cGy fraction
there appeared to be a rapid rise in the PCr peak. If our
original interpretation of this spectrum was correct this
would imply rapid reoxygenation of hypoxic tumor tissue.
The situation becomes more complex as we follow patients further into treatment. in Figure 6 the spectra are
shown of a patient with a medically inoperable T3NO
squamous cell carcinoma of the pyriform sinus. These
spectra were obtained before radiation treatments were
initiated, after 4800 cGy when an approximately 50%
regresion of the tumor was seen and after 7400 cGy when
the tumor had shrunk to approximately 15% to 20% of
its original volume. in this case the original spectrum
shows a relatively high peak for PCr, implying that the
tumor was relatively well oxygenated, however, a large
and poorly resolved shoulder is seen in the region where
the peaks for phosphomonoesters, phosphodiesters, and
inorganic phosphate are located. This implies the presence
of a significant necrotic and heterogeneous component.
At 4800 &y, this component is both reduced and better
resolved. At 7400 &y, this shoulder region is again larger
and poorly resolved. There are many possible explanations
for this result, ie., damage to normal tissue and its vasculature seen at the end of therapy, selection and growth
of radioresistant clones during treatment despite overall
continued shrinkage of the tumor mass, or simply a sampling error of this relatively insensitive and poorly localized "P spectra. This latter problem can be partially overcame by the use of onedimensional phase encoded spectroscopy,2' a technique that allows spectroscopic
information to be acquired from I-cm slices through the
tumor volume, which can in turn be tied to the tumor
image. We will be pursuing this route with our patients
in the future. it also should be noted that there is an intrinsic limitation to the information that can be derived
from "P spectroscopy because of the relatively large voiumes (e.& 30 mi) from which the spectra must be collected. This places an intrinsic limitation on the information that can be gathered from heterogeneous tumors
since it averages over too large of a volume. Additionally,
as the tumor shrinks it will tend to volume average information from adjacent normal tissue (often muscle with
a high PCr concentration). It is possible that some of these
problems may be overcome by using proton spectroscopy.
This may offer the possibilities of not only sampling from
a smaller volume, but also of obtaining information on
other useful metabolites for following the response of tumors to therapy such as la~tate.2~
Summary
These initial studies demonstrate areas where MRI and
MRS may be useful for the management of patients with
No. 10
NONINVASIVE
MONITORING
OF RADIOTHERAPY PATIENTS
head and neck and other malignancies. The displacement
of T2 values in tumors to higher values than are seen in
normal tissue may be useful both as a measure of the
heterogeneity of microenvironments within a tumor volume and for the development of computer algorithms to
outline tumor volumes for three-dimensional treatment
planning in a way that is practical and economical of
physician time. In addition, it is now possible to obtain
from MRS serial information about tumor metabolism
under therapy in a way that is benign and noninvasive,
offering us (for the first time) the ability to assess the true
role of tumor metabolism in determining the outcome of
therapy in human cancers. From this information may
come new insights into treatment strategies for these difficult patient management problems.
*
McKenna et d.
2075
patterns of intradlular water as a function of H e L ail cycle. Science
1976; 192:904-907.
I
I
.
Hollis DP, Saryan LA,EBskston PA, Moms HP. Nuclear ma~netic
-nam studies of cancer wrrlationship among spin-ipnia &tion
tima, grourth rate, and water mntent of Monis hepatomis. JNal Cancer
Inri 197% 541469-1472.
12. Krwker RM, stewvt O.,Bronskill MJ,Hcnkelman RM. Continuous distributions of N M R nla?ationtima applied to tumors bcfore
and after therapy with x-rays and cyclophosphamide. Magn Reson Med
1988: 624-36.
13. Maris JM, Chance B. Magnetic resonance spscimwpy of nMolarms. In: Krcnel HY, ed. Magnetic Resonance Annual. New York
&ven, 1986; 213-235.
14. Glickson JD. Evanochko WT. Sakai Ti.NP TC. In vivo nmr
~ p r r U m o p ,oftumon. in: ~ u p t aR1, cd. NMR spcMaropy ofcciis
and Organism%Boca Raton: CRC Reps. 1987: 99-134.
15. Gnffiths JR. Suverns AN, llles RA. Gordon RE, Shaw D. "P
N M R investigation of d i d tumors in thc living rat. Biok-I Rep 1982:
2719-725.
16. Schiffer LM. Braunrhweipr PG,Glickeron JD. Evanochko WT.
Ng TC. Reliminuy observations on thc codation ofpdifnative phrnomena with in vivo "P N M R spmmscop) añcr tumor chcmotherap)
REFERENCES
Ann Ni' AcadSci 1986: 158270-276.
17. Sostman HD. A m i y IM. Firher JJ. N M R in cancer. high
I
.
Gatenby RA, Kesdcr HB, R m b l u m JS B ai. Oxygen disvibution
molution spxmxopy in tumon M a n Reson /mag 1984; 2265-278.
in squamous all carcinoma melaslas%: Relationship to outcome of ra18. Cohcn JS. Phospholipid and cnerg) melabolkm of cancer cells
diation therapy. Ini JRadial Oncol Bid Phys 1988 14:831-838.
monitored by "P magncuc mnanrc rpcuoocopy: Poasiblc clinical Yo.
2. Brcger RK, Webrü FW.Charles C, MacFall JH, Haughton VM.
nihcana. Muw Clin Pmc 1988,631199-1207.
Reproducibility of relaxation and spin density paramaen in phantoms
19. Okunicff P.Ramuy J. Tokuhiro T.Hiuig B n al. Estimation of
and the human brain measured by M R imaging at 1.5 T. Magn Reson
tumor or)gcnaiioa and metabolic rate using "P MRS: Comlaiion of
Med 1986; 3:649-662.
longitudinal relaxation mth tumor grovth rate and DNA synthesis. Ini
3. ihulbom KR. Waterton JC, Matthewx PM. Radda G. Ox$CMtion
J Radia Onwl &o1 Phys 1988 14118s-I 195.
dependence of the transverse relaxation time on water pmtons in whole
20. Ng TC. Major AW. Mcancy TF,Thomas FJ. "P MRS nudy of
Mwd at high fields, Biochem Eiophys Acta 1982: 714265-270.
h u m a n t u m o n i n ~ n r t o r a d i a u o thcmpyunnga
n
1.5TMRlsyjtcm.
4. Hausicr R, Nasck F. I(crmagnstische relaxation und korciation
SMRM Absir 1986, 516.
ion systcm wasynaumtoff. Zeii FNaiwf1965; 20:1668-1675.
21. Semmln W, Gadcmann G. von Kaick. Zabel HJ, Lorenz W.I n
5. Krynicki K. Proton spin-iattia relaxation in pure water bstwecn
vivo "P specmxopy of human tumors and their m n s c to therapy
O'and 100" C. Phyiivs 1966; 32167-173.
usinpa 1.5 Tnla wholc body scanncr. S.WRM Abur 1986; 39.
6. Horikawa Y, Nawc S, Tanaka C, Hirakawa H, NSkawa H. Proton
22. Niedcckn AC. Muller S. AYCW P ei ai Earnmiry bone Iumors:
N M R relaxation in ischemic brain edema. Siroke 1986; 171149-1 152.
Evaluation by "P M R spnrce~opy.
Radiology 198% 157167-174.
7. Shah SS. Rande SS, Phadke RS, W u r i SR. Significance ofwafer
23. Balcnau D. Arnold DA. Segebartb C. Luytcn PR. den Hollander
proton spin-iattia relaxation times in normal and malignant tissuaand
their subnllvlar fraaions: Pan I
and 2. Magn Reson Imaging 1982: I
: JA. PJ' MR evaluauon of human brain respanu to therapy. S M R M
Absrr IY86; 41.
91-104.
24. R o s B. Helsper JT. Cox J. Young I, Kcmpf R. htmsarroma
8. Bloch P. The influence of hydrogen peroxide on the T I and T2 of
and othcr nmpiasms of b n c : Magnnic resonance spcclmmpy Io monwater pmtons. Procmdings of thc Radiation Research Saiq, 35th Annor therapy. Arch Su@ 1987; 1221464-1468.
nual Mcning, Atlanta, Georgia, 1987.
25. Solak CH. F m a n DM. Monitoring lactic x i d production in
9. Gore JC, Brown MS, Mizumoto CT,Annilage IM. Influence of
mted failw~ngthcmmulatedcchospaualloaliraDonucbNquc.
glycogen on water pmton relaxation times. Magn Reson Med 1986; 5
Rcvnled at thc Sociny of Magnnic Rmnancc in Medicine. 61h Annual
463-466.
M m i n g New York New York. 1987.
IO. Beail PA, Haziewwd CF,Rao PN. Nuclear magnaic resonance
Sexual Behavior, Venereal Diseases, Hygiene Practices, and
Invasive Cervical Cancer in a High-Risk Population
ROLAND0 HERRERO, MD,'.t LOUISE A. BRINTON. PHD.t WILLIAM C. REEVES, MD.*
MARIA M. BRENES. ES.* FRANCISCO TENORIO, MD.5 ROSA C. DE BRITTON. MD.11
EDUARDO GAITAN. MD.7 MARlANA GARCIA, MS.* AND WILLIAM E. RAWLS. MD#
A case-eontrol study of 759 women with Lvasive cervical cancer and 1430 controls in four LatinAmerican
countries evaluated risk in relation to sexma1behavior, histories of speck venereal diseases, and hygiene
practices. Early age at first sexual iatercoune and increasing number uf sexual partners were associated
with significantly incrmsed risk even miter adjustment for their mutual effects. Risk iirrawd to a
twofold excess among women repwtiq first intercourse at 14 to IS years of age conpied with M+
yeus. The number of steady sexual pibcrs was a more important prediietor of risk thin the number 01
nonsteady partners, puticuiarly hefore age 30, pnssihly reflecting the need for prolunged or repeated
exposum to a tnnsmissibk agent, a düierent methods of protfftioa against s e x d l y tnosmined d i v w s
or pregnancy. Reported frequency of intercourse w a s not generally associated with risk, except among
women reporting incread frequencies before 20 years of age. Histonca of gonorrhea or crab lice we*
ns.wxhted with increased risk, but hist&
ofutber venereal diwere not signifinat predictors. No
consistently increasedrisks were detected for women reporting specifir hygiene or douching habits, except
the practice of washing the genitalia inhcquently during menstruation. These results provide support for
a period of incrmsed susceptibility to areinogena during adolescence, and suggest that this my be an
importiat determinant of the high iacikaee of cervical cancer in Latin America.
Cancer 65:380-386,1990.
T
of invasive cervical cancer has been the
subject of numerous studies over the last 50 years.
However, most of the epidemiologic research has been
done in developed countries where cervical cancer incidence has declined significantly in the last three decades.'
Latin America is still a high-risk area for invasive cervical
cancer? but few studies have analyzed the role of specific
factors in this region.34
HE CAUSE
SupponedinpanbymntrpetNOICP41026andgrantROICA-42042
from the National Cancer Institute, National Institutes of Health, and
by a grant from the National C a n a r Institute of Canada.
Unidad Nacional de Canamlogin. Caja CoJtamcens de %uro
Social, San Josc, Costa Ria.
t Environmental Epidemiology Branch, National Canar Institute,
Bethexin. Miryhnd.
#Gorga# M e m o d Laboratory. Panama City. Republic of Panama.
$Hospital de Oncdogji Nacional, Instituto Mexicano de Scguridad
Social. Mexico Cily, Mexico.
I Instituto Oncologico Nacional. Panama, Republics de Panama.
1 Division de Epidemiolosia.Instituto Nacional de Canmlwa. L b
pth Colombia.
Y Mokcular V i m l w and Immunolw. Dewment of Pathology,
Mc Master Uniwnity, Hamilton, Ontario. Can&
A d d m for reprint% Rolando Hemm, MD. Environmental €pidem b k w B~-acshN i i i o d C a m Institute. Exenitiw ñ u Nonh. Room
443, Buhedi. MD 20892.
Amp<ed for puMition July 14. 1989.
Cervical cancer may represent a late sequela of a sexually
transmitted disease>' and early age at first sexual intercourse, increasing number of sexual partners, low socioeconomic status, and nonparticipation in screening programs are established risk factors? Other suggested facton
include number of pregnancies?" smoking,".'2 histones
of specific venereal diseasesi3use of oral contra~eptives,'~
and the male partner's sexual
In a previous analysis o f the same study population 10
be described in this article.I8 we noted a strong relation
between human papillomavirus (HPV) types 16/18 and
risk of cervical cancer, but the Occurrence of viral DNA
was not associated with sexual behavior. This unexpected
finding, which suggests an effect of sexual practices independent of infection or measurement of this specific
virus, renewed our interest in the relation of sexual activity
to risk, particularly with respect to possible modes Of
transmission of infectious agents.
Venereal diseases have been associated with risk ofcervical cancer?" but no specific infection transmitted venereally has been identified consistently as a causal agent.
In addition, the possible role of personal hygiene in tho
cause of cervical cancer has been s ~ g g e s t e d ,but
~ Ud
available studies yield contradictory results.
380
*,.
,
.
No. 2
r-.
...
I
c
-._
c-
.-.
,
CERVICAL CANCER IN LATIN
AMERICA
To determine the reasons for the h@ incidence of cervical cancer in Latin America, with emphasis on sexual
characteristics,we conducted a case-control study in four
countries. We obtained detailed information on past and
current sexual behavior. hygiene practices, tobacco and
alcohol exposure, medical, reproductive, and screening
history, and presence of H P V DNA as measured by in
situ hybridization studies of ceMcal samples.
Materials and Methods
L
.
.
c
L
._I
c
....
c
...r"
.".
F-
4lY
er'loDrsb
i
ies
'
1
., 4 r
3,
-
!
to
onnd
4ACI
.edinfic
ity
of'-'
1
i
3,
;
i
i
--..
-
ervent.
beC.
he-
L.
- :
I
j
Newly diagnosed cases of invasive cervical cancer were
ascertained before treatment in four Latin American study
sites: Bogota, Colombia; Costa Rica;Mexico City, Mexico:
and Panama between January 1986 and June 1987. Eligible patients were younger than 70 years of age and had
lived in the study area for at least 6 months. The six study
hospitals were the major cancer treatment centers in the
study sites.
in Bogota and Mexico City, two age-matched (5-year
groups) hospital controls were selected for each case; in
Costa Rica and Panama, one hospital and one community
control were chosen. Potential controls with a hysterectomy or history of cancer were replaced. Hospital controls
were selected randomly from admission lists excluding
patients with neoplastic, endocrine, nutritional, psychiatric, selected circulatory, gynecologic, or smoking-related
diagnoses. In Costa Rica and Panama, hospital controls
were selected from inpatients at the primary referral hospitals that served patient areas of residency. In Bogota,
they were chosen from eight tertiary level government
hospitals and in Mexico City from three social security
hospitals serving the population from which the patients
were derived. Community controls were selected randomly from current census listings of the corresponding
patient counties of residency. Eligible controls who refused
to participate were not replaced.
A personal interview, lasting an average of 60 minutes,
was conducted with each participant to obtain detailed
information on demographic, socioeconomic, reproductive, occupational, and medical history, as well as dietary,
sexual. hygienic, and contraceptive practices. Interviewers
who were trained uniformly under strict supervision administered the questionnaire in private settings.
During the 18-month study period, 766 eligible patients
were identified and an interview was obtained from 759
patients (99.1%). Of 1532 eligible hospital controls, 1467
were interviewed (95.8%). Nonresponse among patients
and controls was due to death (three patients, zero controls), refusal (zero patients, 41 controls), language and
hearing problems (two patients, ten controls), mental incompetence (one patient, eight controls), and change of
residency (one patient, six controls). Histologic information was available for 728 patients (96%). of whom
Herrero et ai.
38 I
92% had squamous carcinomas and 8% had adcnocarcinomas. Thirty-seven (2.5%)controls reported no prcv¡ous
sexual experience. compared with zero patients. The 37
virgin controls were excluded from analysis in order to
adjust simultaneously for number of sexual partners and
age at first intercourse. The final group consisted of 759
patients and 1430 controls.
To assess the presence of HPV, a ceMcal swab was
obtained from 755 patients (99.6%) and 1325 controls
(95.4%). We used cotton tipped applicators to collect cell
samples from the neoplastic lesion (patients) and the cervical os (controls). Swabs were eluted in phosphate buffered saline, frozen as soon as possible, and maintained
at -2OOC until tested. Assays were conducted by the filter
in situ hybridization method as reported in detail elsewhere.'* All autoradiographswere studied independently
by three observers blinded to case-control status, and
specimens that were recorded as positive by at least two
observers were considered positive.
To estimate the risk of invasive cervical cancer associated with various factors, odds ratios were calculated as
approximations of relative risks (RR). Unconditional legistic regression was used to adjust for potential confounding variables," deriving maximum likelihood estimates of RR and 95% confidence intervals (CI). Tests for
trends in the logistic analyses were obtained by categorizing the exposure variable and treating the scored variable
as continuous. The results of conditional logistic
regressionz2were similar to the unconditional analysis,
and the latter have been chosen for ease of presentation.
Results
Patients and controls were similar regarding age, race,
and religion. The mean age was 46.5 years for both patients and controls. Most of the participants were mestizos
(65.7% of patients, 64.9% of controls) or white (30% of
patients, 29.8% of controls), and Catholic (87.1% of the
patients, 89.4% of controls). No major differences were
seen by calculating the RR with the use of only hospital
or community controls; therefore, both types of controls
were combined in the analysis.
Apart from the sexual variables, the most important
risk factors identified in the current study were the following (Table I): detection of H P V types 16/18 DNA,
increasing interval since last Pap smear, increasing number of pregnancies, and a smaller number of household
fadities (the socioeconomic status estimator related most
strongly to risk in this study). interval since last Pap test,
number of pregnancies, and socioeconomic status were
correlated with the sexual behavior variables and exerted
confounding effects. The occurrence of H P V DNA, although associated with the highest relative risk, was not
correlated with sexual behavior and, when included in
1
't
A*#‘
CANCERJamaty I5 1990
382
TABLEI, Major Risk Factors for Cervical Cancer
and Awciated Relative Risks
CWS
Controls
RR*
273
448
38
833
392
205
1.0
3.8
0.6
(3.1-4.7)
(0.44.9)
123
109
45
66
372
44
384
345
84
I39
409
69
I .o
0.9
1.6
1.5
3.0
2.1
(0.7-1.3)
(1.0-2.5)
(1.0-2.2)
(2.3-4.0)
(1.3-3.4)
0-1
2-3
4-5
6f
33
I I7
I62
447
O
148
332
330
619
I .o
1.7
2.5
3.2
(1.1-2.8)
(1.5-39)
(2.0-5.0)
6
4-5
220
322
217
654
511
265
1.0
1.5
I .6
(1.2-1.9)
I I .2-2.0)
PresenceofHPV 16/18
DNA
Negative
Positive
Unknown
interval sinn last Pap
smear
<2 yr
2-3 yr
4-5 yr
6 f yr
Never
Unknown
Number of pregnancies
Unknown
Numbn of household
facilitiat
s3
I
958C1
Adjusted for all fadors shown. number of sexual partners, and age
at first sexual intncoune.
t Includeselafnaty. toikt inside the how, radio, TV XI, refrigerator,
Vol. 65
Nearly 4090of patients reported only one lifetime sexual
partner, compared with 52% of the controls. Patients had
an average of 2.3 sexual partners compared with 2.0 for
controls. A w a t e r number oflifetime sexual partnerswas
associated with risk; women reporting six or more partners
had an adjusted risk of 1.7 compared with monogamous
women (P= 0.0007).
The total number of events of intercourse was calculated from the reported frequencies in specific decades,
and no associations with risk were seen. Frequency of
sexual intercourse was associated with increased risk only
among women reporting a frequency of seven or more
times a week before age 20 (adjusted RR, 1.5; 95% CI,
1.I to 2.2). Anal intercourse was associated with a RR of
1.5 (95% CI, 0.9 to 2.6) for women reporting the practice
once in a lifetime and 1.9 (95% CI, 1.3 to 2.6) for those
reporting it more than once. Sexual intercourse during
menstrual periods showed a slightly increased risk that
was reduced after adjustment.
The number of steady sexual partners (relationships
lasting more than 3 months) was a more important preTABLE
2.
Relative Risks o í Invasive Cervical Cancer A~ocialed
With Specific Scrual Practices
and stove.
.
*
the regresion model,produced only minor alterations in
our estimates. Smoking” and previous use of oral contraceptives were not associated significantly with risk of
cervical cancer in this population.
Early first intercourse was common; 34% of the patients
and 22% of the controls reported initiation of sexual activity before age 16. The average age at first coitus was
17.4 years for patients and 18.8 years for controls. Younger age at first intercourse was associated with a significantly increased risk of cervical cancer. Maximum risk
occurred among women reporting first coitus between 14
and I5 yearsofage(adjusted RR, 1.8; 95% C1,l.3 to 2.4)
compared with 20 years or more (Table 2). Women reporting first intercourse before age 14were not at increased
risk after adjustment for confounding variables. Identical
associations with age at first intercourse were seen with
restriction of the analysis to monogamous women. Furthermore, for thio group of monogamous women additional adjustment for the number of their husbands’ partners (a risk factor in this study population reported
elsewhere”) was feasihle, but did not alter the risk estimates associated with early intercourse (data not shown).
The interval between menarche and first sexuai intercourse was calculated, and smaller intervals were asociated with an increase in risk, but the effect disappeared
afier adjustment for age at fint coitus.
Cases Controls
Age at first sexual
internurse (yr)
20+
18-19
16-17
14-15
<I4
Unknown
P for trend
Numbn oilifelime
P X U ~p r i n r n
I
2-3
4-5
6+
Unknown
P for trend
Frequency of anal
internunet
Never
Once
More than once
Unknown
P for trend
Sexual intmOurx during
mennnial period
No
Ya
Unknown
170
139
189
217
42
2
521
284
303
233
84
5
303
340
65
SI
783
489
92
64
O
631
2
35
88
5
1248
44
103
35
655
104
O
1283
145
2
RR*
1.0
1.5
1.9
2.9
1.5
RRt
1.0
1.2
1.3
1.8
1.1
954bCl
(0.9-1.7)
(1.0-1.8)
(1.3-2.4)
(0.7-1.7)
<0.0001 0.007
1.0
1.8
1.8
2.1
1.0
1.6
1.4
1.7
(1.3-2.0)
íl.0-2.1)
(1.1-2.7)
<0.0001 0.0007
1.0
1.6
1.7
1.0
1.5
1.9
(0.9-2.6)
(1.3-2.6)
0.0002 0.0003
1.0
1.4
1.0
1.3
(0.9-1.7)
Adjuued for v.
t Adjwted for p~c.number o i x r u i l partners, age al hrn intercou*~
prrrenx of HPV 16/18 DNA. inlerval J i n a lp<l Pap smear, number Of
Prcyinck and number of household facilitia
t Only women reunting at leut one study pltnn included (morn
than 3 monthah
O
C
O
O
Billms
Check
Chargt
Masiei
Account P
Signature
Nanw Addmu
city -
o
p l e
-
(M
O New
o
CAW
OAnrk
ohmfil
0 hn*S
CERVICAL
C A N C E R IN LATINAMERICA
No. 2
TABLE3. Rditivc Risks* of Invasive Cmrpl Cinar Accordin8 10
Number of Stcsdv and N o n n d v Sexual R n n e n
Steady u a u d
mnnen
I
2
Nonsteady sexual panncn
O
1-2
z3
I .o
(3031783)
l.7t
1.2
(84175)
1.5
I .4
(25/41)
1.1
Numbers in parentheses are caxs/mntmls (unknowns excluded).
Adjusted for age, age at finl internurse, pmence of HPV 16/18
D N A , intmal P n a kt Pap smear,number of pregnancies. and number
of household facilities.
t 95% CI = 1.3-2.3.
t9S%Cl = 1.5-3.0.
695% CI = 1.0-3.8.
dictor of risk than the number of nonsteady partners (Table 3). Risk increased with the number of steady partners
in each category of nonsteady partners. However, for each
category of steady sexual partners the risk did not rise
with a greater number of nonsteady partners. The total
number of sexual partners at specific decades of life was
TABLF4
Relatwc Risks oflnvavvc Cervical Cancer Associated
With Histoncr of Specific VcncMJ D8-s
-
Any venereal diNo
Yes
Gonorrhea
No
Yes
Unknown
Syphilis
No
Yes
Unknown
Venereal warn
No
YCS
Unknown
GeniIal herpes
No
Cases
Controls
RR*
95% CI
676
83
1333
97
I .o
I .8
(1.2-2.5)
727
30
2
1408
18
I .o
743
1407
21
2
745
13
1392
36
2
15
I
I
4
3.3
I .o
I .3
I .o
0.9
I .o
YCS
730
28
1405
23
2
2.2
Ya
744
12
1415
1.o
Unknown
Other venereal
di(nonwecified)
No
Unknown
(0.6-2.7)
(0.5-1.8)
(0.1-6.5)
YCS
Unknown
Crab lice
No
(1.7-6.4)
I
-
2
13
7
1.4
(1,242)
(0.6-3.5)
__
Adjusted for age, number of sexual partners, age at lint intmourse,
pmence of HPV 16/18 DNA, interval Pna last Pap smear, number of
presnancics and number of household facilities.
-
383
Herrero et al.
not related to risk, but a greater number of sieady patiners
before age 30 was associated with increasing risk (data
not shown).
The risk assOciated with having had a venereal disease
was 1.8 (95% CI, 1.2 to 2.5) (Table 4). The two specific
diseases ass0ciaied with risk were gonorrhea (adjusted RR,
3.3) and crab lice (RR, 2.2). A history of these two diseases
was approximately ten times more common among
women with six or more sexual partnersthan among monogamous women. Histories of syphilis, genital warts,
herpes, and other (unspecified) venereal diseases were not
associated with significant increases in risk among the few
women reporting these diseases. Women reporting a history of more than one venereal disease were at the same
risk as those reporting only one.
A slightly increased risk was associated with bathing
less than once a day (Table 5). An increased risk (RR,
I .7;95% CI, I . I to 2.6) was noted for women who washed
their genitals more than twice a day. In contrast, women
reporting frequent washing of their genitals during men-
TmLf 5
Relative Risks of Invsslve Cenical Cancer Associated
With Spccihc Hygunc Practices
Bathing
z once a day
<once aday
Unknown
Frequency of washing
genital areat
Never
5 once a day
Twice a day
> twice a day
Unknown
P for trend
Frequency of washing
pnital area
during menstmal
period
<once a week
1-7 tima a week
> once a day
Unknown
P for trend
Use of sanitary napkins
during menstmation
No
Yes
Unknown
Use oftampons during
menstruation
No
Ya
Unknown
RR'
Cases
Controls
534
225
1114
315
I .o
20 I
343
I45
47 I
627
248
80
4
I .o
O
70
O
95
190
474
O
I
139
293
997
I
308
450
508
92 1
I
745
1392
31
1
13
I
I
1.3
1.1
1.1
I .7
95% CI
(1.0-1.6)
(0.9-1.5)
(0.8-1.5)
(1.1-2.6)
0.04
1.0
0.9
0.6
(0.6-1.3)
(0.5-0.9)
O.wO7
Io
.
I .o
(0.8-1.3)
I .o
0.9
(0.5-1.9)
AdjUaed for age, number of sexual parinera, age at ñnl intercourse,
pmence ofHPV 16/18 DNA, interval piace last Pap smear,number of
pregnancies and number of household facilities.
t Separate occasions from gmeral bathing.
CANCER
January I5 1990
384
stnial periods were at significantly decreased risk. Those
women reporting this practice more than once a day had
an adjusted RR of 0.6 (95% Ci, 0.5 to 0.9) compared with
less than once a week. No particular feminine hygiene
product used during menstrual periods was assoCiated with
risk, including tampons, sanitary napkins, cloths, and
various other products.
A marg¡nally significant 40% risk was seen for women
reporting regular use of vaginal douches for 2 months or
more compared with women who did not douche, but
there were no consistent trends for age at initiation, frequency, or duration of douching (Table 6).
Diseussion
in concordance with the notion that coitus is a prerequisite for the development of cervical carcinoma, none of
the patients reportedthemselves as virgins, compared with
2.5% of age-matched controls. Early age at first sexual
intercourse was associated with increased risk of invasive
cervical cancer, independent of number of sexual partners
and other confounding factors. The highest risks were seen
among women reporting first intercourse at ages 14 and
15, whereas those who initiated sexual activity before 14
TABLE6.
Relative Risks of Invasive Cervical Cancer Associated
With Douching Practices
CaseS
Controls
RR’
513
131
I14
960
270
198
I .o
513
29
55
29
I
960
54
80
63
I .o
I.3
SI3
14
38
51
I
960
40
86
71
I
513
43
38
38
23
960
55
59
40
40
40
4
95% CI
___
U r of vaginal douche
for 2 months or
more
Never used
Not regularly
Regularly
Agc nvtcd using?
Never used
231
21-30
<21
Unknown
P for trend
Frequency of use?
Never u x d
<I a w k
1-2 a w k
>2 a week
Unknown
P for iwnd
Months of regular u s ?
Never used
SI2
KI
20
121-240
a240
Unknown
P for trend
10
O
I
1.1
I .4
I .8
I .o
(0.8-1.5)
(I.1-1.9)
(0.8-2.3)
(I.2-2.8)
(0.6-1.6)
0.31
I .o
1.8
1.1
1.6
(1.0-3.3)
(0.7-1.8)
(I.0-2.4)
0.04
1.0
1.7
1.7
1.5
0.7
(1.0-2.7)
(1.0-2.7)
(0.8-2.8)
(0.3-1.4)
0.35
Adjwcd (or .A n u m b oíxaud pnnm,a8c at ñni inicrmursc.
prcsnce of HPV 16/18 DNA, inmil since laü Plp smear, numba of
prrlnmnciea 8nd number of household frilitia
t Nonmguh usn cacludcd.
Vol. 65
years of age were not at increased risk. This finding may
be a reflection of social or behavioral charactexistics of
this group of early starters, which we were not able to
determine.
Most epidemiologic studies of cervical cancer have reported increased risk for women who initiated sexual activity during adolescence?c28 and it has been hypothesized that the adolescent cervix is particularly susceptible
to the effect of coitwrelated carcinogens?9sMAreas of
atypical metaplasia are seen among sexually active adolescents more often than among virgins.” However, it is
not clear if the effect of age at first intercourse is due to a
particular vulnerability of the young cervix or to the fact
that women who start sexual activity earlier have longer
exposure to coitus-related
in an attempt to
evaluate this issue, risks were considered in relation to
age at diagnosis and latency periods. To the extent that
such an analysis was possible in an age-matched study,
we found no evidence of an effect of longerexposures that
could explain the association with age at first intercourse,
and the effect of the latter was evident regardless of age
at diagnosis.
Number of lifetime sexual partners was associated with
a significant trend of increasing risk, with women who
reported six or more partners having an adjusted RR of
1.7 compared with monogamous women. However, risk
was elevated for 2 to 3 partners, primarily reflecting a
difference between monogamous women and nonmonogamous women. This finding is in agreement with other
report^,^'*"-'^ and has been interpreted as evidence for
an important role of sexually transmitted agents. Two
previous studies of preinvasive and invasive cervical neoplasia did not show an independent effect of early intercourse after adjustment for number ofsexual partne13.”5.’~
However, two recent studies showed that independent effects of both factors do
Our results support
independent effects of early ages at first intercourse and
numerous sexual partners.
The number of steady partners (relationships lasting
more than 3 months) dated more to risk than the number
of nonsteady partners, and this effect was more apparent
for the persons who had multiple steady partners at young
ages. Possible explanations for this finding include the
need for more pmlongd or repeatad exposure to a partner
who carries a transmissible agent. Alternatively, it could
be related to more frequent use. of condoms with nonsteady rather than steady partners, as suggested by a recent
study of the transmission of human immunodeficiencY
virus (HIV).* Although we found that use of condoms
was associated with a slight reduction in risk and that
usage was more frapuent among women with n o n s t d y
partners, this did not explain the difference in effect between steady and nonsteady partnea AdditiOndlY, we
studied whether differential use.of other methods Of con-
No.
tmc
rest
m
wit1
F
rese
rele
and
inte
rep<
por
S
pro!
to c
infe
infe
mis
pen
thai
plai
mis
onl:
sub
for
C
gen
epit
trar
oft
fon
HOT
effe
for
the
üth,
the¡
S
the
Am
ass<
WiLl
rep
Pro
it p
Cral
like
I
rela
SOCi
i
1
i
Prai
gen
re01
gen
.-_-
, ..I^c-
...
No. 2
CERVICAL CANCER I N L A T I N
AMERICA
*
Herrero el al.
385
trsception or ftepuency of intercourse could explain the
sociated with reduced risk. in concordance with other
results, but then remained a stronger effect of steady
studia,I0 the type offeminine hygiene products used durpartnm.A recent studyloalso failed to find an assoCiation
ing menstrual penods was not associated specifically with
with nonsteady partners.
risk.
Frequency of intercourse has been studied by several
Vaginal douching, which alters the vaginal milieuzbor
-hers,
most of whom have concluded that it is not
causes initation," has long been reported as a risk factor
. ~ ~study,
, ~ ~ ~ for
~ ~cervical cancer.u A recent study did not find any rerelevant to risk of cervical ~ a n c e r . ~in~ this
and in concordance with other authors,""
frequency of
lation,1° whereas another showed i n d n g risk with freintercourse was a significant risk factor only for women
quency and duration of use of various douching prepareporting frequent intercourse before age 20,again sup
rations, excluding water or vinegar." The current study
portingthe hypothesis of a vulnerable period.
does not show any consistent associations of douching
Several possible carcinogenic mechanisms have been
practices with risk of invasive cervical cancer.
proposed to explain the relation of risk of cervical cancer
The participation rates in this study were almost loogb
to coital factors. The one that is accepted most is that an
for both patients and controls. Although questions arise
infectious agent, possibly HPV, is transmitted from an
regarding recall biases assoCiated with.the sexual variables,
infected male to his partner. Under this premise, iftransthe correlation of number of sexual partners with reported
mission occurs with a single exposure, we would not exhistories of venereal diseases provided an indirect form
pect the number of steady partners to be more important
of validation. Furthermore, the interviewers were inthan total number of sexual partners. An alternative exstructed to stress the confidentiality of the information
planation is that the infectious agent is not always transand to administer the interviews in an appropriate setting.
misible, because of cyclic shedding, and is transmitted
in conclusion, age at first sexual intercourse and numonly after repeated exposure. A recent study on female
ber of steady sexual partners exerted independent effects
subjects" suggests that there is such a cyclic pattern
on risk of cervical cancer. The high incidence of cervical
for HPV.
cancer in Latin America is probably influenced by the
Other hypotheses consider the possibility of carcinolack of adequate cytologic screening programs and the
genic constituents of the human semen; it has been sughigh prevalence of early sexual activity. This factor may
gested that basic proteins from human semen can alter
also explain partially the high incidence ofcervical cancer
epithelial and subepithelial cells and induce neoplastic
in low Socioeconomic groups in developed countries, as
transformation.4 A recent study showed that protamine
suggested by a recent study!9 Furthermore, effects with
of human and animal origin produces neoplastic trans- the number of constant sexual partners before 30 years
formation of human cervical epithelial cells in v i m . 4 7 of age and frequent intercourse at early ages suggest inHowever, the general lack of evidence in regard to an
creased susceptibility of the cervix during adolescence,
effect of frequency of intercourse provides little support
and emphasize the need for additional research on the
for this mechanism. The evidence of increased risk with
means of transmission of putative causal agents.
the practice of anal intercourse may be a reflection of
REFERENCES
other aspects related to sexual behavior of the women or
1. Lkvesa SS. DaCnPtiveepidemiology of canmof the uterine anix.
their partners, or to hygiene practices.
Obstefrics and Gynerdogy 1984; 63605412.
Specific venereal diseases are usually correlated with
2. Restrepo HE, Gonzalw 1. RobeN E, Litvak J. Epidcmiologia y
the incidence and mortality From cervicalcancer! in Latin
mntml del cancer del cuello uterino en America iatina y el Caribe. Bo/
OfSanifPanam 1987; 102578-593.
America, gonorrhea and crab lice were the two diseases
3. Z u l w L. Cancer de ceMx y w faom de riq+: Ea\udio de
associated with cervical cancer, whereas herpes, genital
cpy>s y m n m l a . Rev Colombiana Obsrer Ginad 1982; 23:219-232.
warts, and syphilis were not related, although few women
4. Carreras O. Algunos aspectos egidcmiologicos del &noma
cervimutcnno. R e Cuhna Obsrei Ginmd 1981; 7:376-383.
reportedthese diseases. No specific mechanisms have been
5. Reeves WC, Brena MM, LkBritton RC, Valdes PF, ioplin CFB.
Proposed for a direct carcinogenic effect ofgonorrhea and
Cervical Cancer in the Republic of Panama. Am J Epidemid 1984; I19
it probably represents general venereal exposure, as do
7 14-724.
6. Duque E, Cuello C, Aristirabal N, Haenvel W, ñotero S, Coma
crab lice. In addition,these two diseases are probably most
P. PIcmaliiant lesions of the
in women of Cali, Colombia. J
likely to be remembered by patients.
Nod Cancer Ins1 1979; 63953-963.
It has been suggested that hygiene practices may be
1. KcJJler 11. Human aMcal canar as a venereal di-.
Cancer
Res 1976: 36783-791.
related to risk of cervical cancer:'
but no consistent as8. Beral V. Cancer of the cervix: A sexually transmitted infection?
sociations with risk have been established for specific
Lnncel 1974: 1:1037-1040.
practices!'"
In our study, the frequency of washing the
9. Brinton LA, Fraumeni JF. Epidemiology of uterineces-vkd-.
JChron Dis 1986: 391051-1065.
genitals was associated with increasing risk, which may
IO. Brinton LA, Hamman RF,Huagins G R n al. Sexual and repro
reflect inrreased sexual activity. Frequent washing of the
ductive risk fnctors for invasive squamous all cmical eamer. 3 Nafl
gental area during menstrual periods, however, was asCancer Ins1 1987; 7923-30.
386
CANCERJanuary IS 1990
I I . Winkebtan W.CigarrttcsmoLingandcanceroftheutrrinecervix.
BanbwyRep 1986; 23329-341.
12. World Health O~nizationllnternationalAgency for Research
on Cancer. Epdemiological studies of cancer in humans. In: IARC
Monograph on the Evaluation of the Carcinogenic Risk of Chemic&
to Humans Tobaao Smoking. Lyon. France: International Agency for
Rexarch on Cancer, 1986; 199-308.
13. Fcncglio CM, Fcrenny A. Etiologic factors in cervical neoplasia.
Semin Onml1982: 9349-372.
14. Piper JM. Oral contraceptiva and cervical cancer. Gynecol Onml
1985;22:l-14.
15.' Lunzunegui MV, King MC, Coria CF, Charlet J. Male influenas
on cervical canal risk. Am J Epidemiol 1986; 123:302-307.
16. Buckley JD, Harris RW, Doll R, Vesey MP. Williams FT.Casecontrol study ofthe husbands ofwomen with dysplasia or carcinoma of
the cervix uteri. Lancer 1981;21010-1014.
17. Brinton LA. Reeves WC, Brenes MM el al. The male factor in
the etiology ofcervical cancer among sexually monogamous women. Int
J Cancer 1989: M199-203.
18. Reeves WC, Brinton LA, Garcia M et al. Human papillomavirus
infection and cervical cancer in Latin America. N Engl J Med 1989
320143741
19. Guijon FB.PnnskevasM, BNnham R. The arsociation of vxually
transmitted dwith cnvicalintraepithelialneoplasia: A caScontrol
study. Am JObsler Gynecd 1985; 151:185-190.
20. Jayant K. Notani PN, Gadre VV, Gulati SS, Shah PR. Personal
hygiene in groups with varied cervical cancer rates. Indian J Cancer
1987: 2447-52.
2 I. Breslow NE, Day NE. Statistical methods in cancer m a r c h The
analysis of cay«intml studies. IARC Monogr Eva1 Carcinog Risk Chem
Hum 1980 1:192-246.
22. Lubin JH. A computer program for the analysis of matched casecontrol studies. CompulBiomedRes 1981; 14138-143.
23. Herrero R. Brinton LA, Reeves WC el a/. Invasive cervical cancer
and smokingin Latin America. JNal1Cancer Ins1 1989 81:205-211.
24. T e m s M, Wilson F, Smith H. Spring E, Nelson JH. The relationship ofcoitus to carcinoma o f the cervix. Am J Public Healrh 1967;
57840447.
25. Stewan HL. Dunham W, Casper J el a/. Epidemiology ofcancers
of uterine cervix and corpus, breast. and ovary in Israel and New York
City. J Nail Cancer Inst 1966 37 1-95.
26. Graham S. Schotz W. Epidemiology of cancer of the cervix in
Buffalo, New York. J Nafl Cancer Inst 1979: 6323-27.
27. Fujimoto I, Nemoto H, Fukuda K, Masubuchi S, Masuhuchi K.
Epidemiologic study of carcinoma in siiu of the cervix. J Repmd Med
1985;30535-540.
28. La Vnrhia C. Francerhi S, DecarliA n a/. Sexual factors, venercal
diseases, and the risk of intraepithelial and invasive cervical neoplasia.
Cancer 1986; 58935-941.
29. Rotkin ID, Cameron JR. Clusters of variables influencing risk of
cervical cancer. Cancer 1968: 21:663-671.
Vol. 65
30. Rotkin ID. Relation of adolescent coitus to cervical cancer risk.
JAMA 1962; 179486491,
3I. Singer A. The uterine cervix from adolmnce to the menopause.
Br Jobstel Gynaecol 1975;8281-99.
32. La Vecchia C. The epidemiology
of cervical neoplasia. Biomed
.~
Pharmacolher 1985;39425433.
33. Marlin CE. Marital and coital facton in cervical cancer. Am J
Public H d I h 1967: 57803-814.
34. Tems M, Fítzpatrick W. Nelson JH. Comparative epidemiology
of i n w i v e carcinoma of the cervix, carcinoma in situ. and cervical d y s
plash. Am J Epidemiol 1980 112253-257.
35. Harris LW, Brinton LA, Cowdell R H el d.Chancteristicd of
women with dysplasia or carcinoma in silu of the cervix uteri. Br J
Cancer 1980 42359-369.
36. Herity B, Murphy IF. üaly L, Moriany MI. Studies of the epidemiology of preinvasive and invasive carcinoma of the uterine cervix
in Ireland. I r Med J 1985; 78177-182.
37. dc GraaiT J, Stolte LAM, Janssens J. Marriage and childbeanng
in relation to cervical cancer. Eur J Obslel G y n d Reprcd Biol 1977:
7307-312.
a
38. Faill E, Simmons ME. Kamcert JB. Factors associated with h
andlow riskofcervical neoplasia. JNalICancerImI 1981:6663l-á36.
39. Raves WC. Brinton LA, Brenes MM, Quiroz E, Rawls WE,
DcBrittonRC. CaJe control study of cervical cancer in H m m Province,
Republic of Panama. In1 J Cancer 198% 365560.
40. Centen for Disease Control. Antibcdy to human immunodehciency virus in female prostitutes. M M W R 1987: 36157-161.
41. Rotkin ID. A comparison review of key epidemiological studies
in cervical cancer related to current m h e s for transmissible agents.
CancerRes 197); 331353-1367.
42. Rotkin ID. Adolescent coitus and cervical cancer: Auociations
ofrelated events with increased risk. Cancer Res 1967: 27:603-617.
43. Peters RK, Thomas D,Hagan Do.Mack T M , Hendenon BE.
Risk factors for invasive cervical cancer among latinas and non-latinas
in Los AngelesCounty. JNall Cancer Ins1 1986; 711063-1077.
44. Tems M, Oalmann MC. Carcinoma of the m i x : An epidemiologicstudy. JAMA 1960: 174:1847-1851.
45. Reeves WC. Arwemena JR. Garcia M el al. Genital human pap
illomaviws infection in Panama City prostitutes. JInfecl Dis 1989: IM):
599-603.
46. Singer A, Reid BL, Coppleson M. A hypothesis: The role of a
high-risk male in the etiology of cervical carcinoma Am J Obsta Cynm/
1976 126110-115.
47. French PW. Coppleson M. Reid EL. Effects ofprotamine
on human cervical epithelial cells and BHK 21 cells in vitro. I R Sw
Med 1987:80434-437.
48. Smith FR. Etiologic factors in carcinoma of the cervix. Am J
Obstei Gynecol 1931;21:18-25.
49. Mant D, Vessey M, Loudon N. Social clari differences in sexual
tehaviour and cervical cancer. Communily Med 1988: 1052-56.
Si
rth
Phil
SW
shoi
to c
The
not
Si
carb
x 11
be9
a co
title
theii
in tt
whe
auth
ackr
of 1k
on a
nghi
plus
T
t
mat1
and
Stan
relat
iCuF
lenh
c:
Mar
Amr
LiPF
BE&
MD
to tt
editi
con
(UM
plica
Fa
shou
whic
nam
and
thesi
by I:
man
tion!
its fi
No. 4
visable.
stcad.
Mitomycin C Adjuvant Chemotherapy After Wertheim's
Hysterectomy for Stage IB Cervical Cancer
V. SIVANESARATNAM. MBBS. FRCOG. FICS. FACS,'t
is today an
accepted method for the management of Stage IB
and early Stage IIA carcinoma of the ceMx. It is partic-
ularly appropriate in the young patient in whom many
years of ovarian and coital function can be preserved.
However, then are a group of patients undergoing such
therapy who might already have exirapelvic micrometastases and thus are at "high risk"of developing recurrences.
These patients include those with a large primary tumor,l3
those with clinically undetected parametrial tumor extension,' thost with lymphatic/vascular channel permeation
in the ceMx,"' and those with histologic evidence of metsstases to peivic lymph nodes!.'
Although the use of adjuvant pelvic irradiation does
decrease the incidence of local pelvic recurrence,the development of distant metastaJcs gives no imptovement
in the overall survival." The use ofsysiemic adjuvant chemotherapy in this select group of high-risk cases thus a p
pcars to be logical.
Fmm tbe Dcpr<mcnts of Wmetria and Gynvcolooy and Wathoiow. Fruhy of Medicine,Unimsity of M.Ly* KWL Lumpur.
t Rofaoi. Had sad Senior Consulunt.
. -o-lor
M o t & ' V.Sivamuatnun, FRCOO. F
a
l FACS. Had
and Senia Conudmnl, Department of Oimmria idGynwsolqy.
F d i y of Medicine. Univemiiy of MiLm 59100 KWL Lumpur.
Accepted fo?publiation March 7, 1989.
aminatie
puted t<
AND P. JAYALAKSHMI, MBBS. MPATH. MRCPATHSS
hhta undergoing d
i
a
l a m g i d tre8tmeat for Stmgc IB dI
U cervical arcloomi are i t hlgh risk
of develop& I d rewrmce and/or dlitint metashses when ame M owre of the fobwing fxtors are
present: p n s o m of whseitk pelVic lymph Warr 8 we piimUy po.Vtb, full-tbklmu hlmolhdoa
of the cervix, cllnially undetected PyimCMiI extensbo, iad Iymphitie/vueiihr chinael perwitloo in
the cervix by hmDlcella. cireimmioith cervix 8 p p M to be kbivisp iikc 8 systemic dirare. w o r e ,
syatemk wiaurea shnuld be mnsidmd its ther8py. The 8 i t h a report the ini(hl eXperknce 4 t h tbe
use of mitomycin C M 8 sh#ie 8gent 8djuvMt in 16 patknta with
is eudaomi of the cervix who
had o o d Wertheim
~
ndK.lbyaterectomy dwere tbn&t to be in tbis high-tisk group. Fourteen
of the patients i
n
dlve and Ires of d
h after duntiws of follov-irp mng¡ng from 16 to 38 months,
the discisclrrc surVinl8t 8 medhn f o l h - u p or 29 nwtb be¡llg 87.5%. @IC ptkd q u u c d ditinuation of adjivnat chemotherapy b u s e of severe marrow toxidtr, hnwever, in view of the prrreaec
of a multipk risk factors, pelvieimdIntloo WM glvea inatead. She dkd 13 molitb Inter from disseminited
disease. A second pitieat died 6 moatha Inter from congestive ardir Idlure.
Cancer 64:r)8-800,1989.
ERTHEIM'S RADICAL HYSTERECTOMY
WP;
smears,
This report describes our preliminary results with the
use of mitomycin C as a single agent adjunct to surgery
on the above high-risk cases.
Materids and Methods
Sixteen consecutive patients who had undergone a
Weriheim radicalhysterectomy at the University Hospital
Kuala Lumpur, Malaysia, for Stage 1B carcinoma of the
ceMx were found &r surgery to have one or more of
ihe foUowiw high-risk f&om (I) metastatic pelvic nodes,
(2) lymphatic/vascular channel permeation, (3) h
i
l
l
thickness tumor invasion of the ceMx, and (4) micm
ropic parametrial tumor extension. The pmperative
evaluation of the patients included chest radiograph, intravenous urography, and where neceSSary a cystoscopic
examination. A full intraabdominal exploration was performed in all patients before commencementofdefinitive
surgery. None had palpably enlarged or suspicious paraaortic nodea The peritoneal washingc in all cases showed
no malignant cells.
Chemotherapy was initiated between 7 and IO days
&r surgery. Mitomycin C at a dosage of IS mum' was
administered intravenously at intervals of 3 weeks for five
course%Aü but one patient completed the scheduled bve
course% The latter developed scvm marrow depression
after two courses, making continuation of therapy inad798
Four
cervix :
(Table
15 o f ü
in dian
tastases
Histolc
vadede
V
d
but on'
sion w
Fou
dence
These
from 1
tient d
had fL
lymph
pearec
The o
multii
vasior
idly
vic nc
them1
seven
fusior
13 m(
meta!
AI
for I
Pmsi
sever
these
chao
and a
and 1
TI
Pletii
aRer
vere
fic
VaSC
h w
end
--r
--
..
NO. 4
CHEMOTHERAPY
AFTERHYSTERECTOMY
SivaneJaratnam el
al.
d b l e . She was then given whole pelvic irradiation in-
stead.
~ 1 patients
1
were closely followed with vaginal vault
mam, colposcopy, abdominal and pelvic ultrasound examination, chest radiograph, and where ncceSSary computed tomography (Cr)scan evaluation.
Results
Fourteen patients had squamous cell carcinoma of the
a& and two others had adenosquamous carcinoma
(Table I). The primary tumor size was less than 3 cm in
15 of them. In the remaining patient it measured 4 cm
in diameter. In the three patients with pelvic node meit was bilateral in one patient, involvingsix nodes.
Histologic examination showed that the tumor had invaded almost the full thickness ofthe ceMx in ten patients.
Vascular/lymphatic channel permeation was seen in all
but one and microscopic evidence of parametrial extension was present in another.
Fourteen of the 16 patients are alive and without evidence of disease after a median follow-up of 29 months.
These patients have been followed for periods ranging
from 16 to 38 months. Two other patients died. One patient died 6 months after surgery from cardiac failure; she
had full-thickness invasion of the cervix by tumor and
lymphatic channel permeation by tumor cells, and a p
peared to have tolerated the five courses reasonably well.
The other, who had an adenosquamous carcinoma, had
multiple high-risk factors, viz., full-thickness tumor invasion of the ceMx, lymphatic channel permeation, clinically undetected parametrial extension, and bilateral pelvic node melastases (six positive nodes). Mitomycin C
therapy was discontinued after two courses because of
severe bone marrow depression. After appropriate transfusions, she was given whole pelvic irradiation. She died
13 months later from disseminated disease (lung and brain
metastases).
A majority of patients experienced nauses and vomiting
for 1 to 2 days after administration of drugs. Myelosup
pression was present in four patients (25%)resulting in a
severe drop in hemoglobin and platelet counts; one of
these patients developed petechial hemorrhages. These
changes were observed after two cou~sesin one patient
and after the fifth course in three others. AU required blood
and platelet transfusion.
Three patients complained of breathlessness on completion of therapy. In one patient this occurred 3 months
after completion of therapy and was accompanied by severe anemia, thrombocytopenia,and bilateral pleural effusion. She was thought to have mitomycin C-induced
vasculitis with pulmonary infarction. AAer prolonged
hospitalization of 3 months she appeatrd to have m v ered from the toxic effecis of the drug. Another had facial
-
T d
(n = 16)
-
~
199
-
Rcnimd
(n I)
-
Dead
(n 2)
HUtola~ictypc
Squimous ctU arcinom
Lule allkemtiauing
i u g c cell nonkcntinizing
Adcnosquunout arcinom
Rimuy tumor a k @os)
<2 cm
2-3 cm
>3 cm
pelvic mode metpRlsQ
UniliterPI
BilitrrpI
Full thickness invasion of ihe m i x
(histologic)
Lymphatic channel permeation
Vascular permeation
ppramcuial extension
Duntioo o l lollow-up (mo)*
<IO
10-15
16-20
.
. -~
21-25
25-30
31-35
35-40
* Median suMval.29 mo.
puffiness and difficulty in breathing. A chesi radiograph
showed mediastinal widening. A CT scan demonstrated
a lobulated mass in the anteriormediastinum.At anifxior
mediastinotomy, the thymus gland was eniarged and a
thymectomy was performed.She remains well. In the third
patient the breathlessness was transient. Moderate alopecia occurred in two patients. Transient elevation of
m m glutamic oxaloacaetic transaminase (SGOT)and
serum glutamic pyruvic transaminase (SGPT) levels was
present in one patient.
Diseussioa
SuMval in patients with carcinoma of the cervix mated
by radical surgeq is influenced by several factors. A large.
primary growth, undetectad paramarial tumor extension,
the presence of tumor in lymphatic/vascular channels of
the cervix, and histologicevidence of pelvic node metaE
tases are features that influence prognosis in such patients.'"
Although the overall 5-year survival rates after radical
surgery or radiotherapy for Stage IB disease is 75% to
85%,89 when metastases to pelvic nodes are present this
drops to 48.5% if between one and four peivic nodes ace
positive, and to 19% if five or more nodes are involved.
Where undetected parametrialextension exists,the 5-year
CANCER Augurt I5 1989
suniVd
ir only 508.1 m e presence oflymphaticlvmh
pnnution
the survival rats to 60% to 70%:''
l.lter. in our view, is an ominous findingeven in the
o~~
ofpelvic node metastases; even with postoperative
pelvic imdiation in such cases many have been reported
IO
recur.'O
When whole pelvic irradiation is added to radical sur-
wry there is a high risk of morbidity and mortality." The
use of postoperative pelvic irradiation does decrease the
incidence of local pelvic recurrence, but the development
of distant metastases is not prevented, resulting in no improvement in overaü s ~ M v d . 8 ~
Gynecologic oncologists have today accepted cytotoxic
chemotherapy as a useful adjunct to surgery in the management of ovarian cancer. Its use in the management of
those patients undergoing radical hysterectomy for early
cervical cancer is not yet established.Cervical cancer can
no longer be considered to be a pelvic disease spreading
slowly by local infiltration and lymphatic spread; from a
smallprimary lesion spreadto pelvic and upper abdominal
lymphatics have been seen, and frequently, vascular
channel tumor permeation has been observed, suggesting
that spread by the blood stream is possible. It is thus seen
to be behaving as a systemic disease, a pattern of spread
that requires systemic measures. In view of the poorprognosis when the above high-risk factors are present, it is
fair to assume the possible presence of micrometastases
not only within the pelvis, but at extrapelvic sites as well
at the time of surgery. Adjuvant chemotherapy in such a
situation can be expected to increase cure rate over what
can be achieved with local therapy alone.
We had recently reported our experience with cisplatin
in combination with bleomycin and vinblastine used as
an adjunct to radical surgery in these high-risk patients
and found a disease-free survival rate of 86.4% at a median
follow-up of 23 months." Cisplatin combination drugs
are, however, costly. Mitomycin C, a cheaper drug, used
as a single agent or in combination with bleomycin, has
been used in the therapy of advanced or recurrent squamous cell carcinoma of the cervix. Responses of up to
22% and 93% have been rep~rted."~"In the current study
the use of mitomycin C as a single agent adjuvant in I6
patients has shown a disease-free survival rate of 87.5%
'at a median follow-up of 29 months, a rate similar to
survival ratea reported in patients with negative pelvic
In one of the two patients who died, adjuvant
chemotherapy was discontinued after two courses, and
despite full pelvic irradiation the patient died from pulmonary and brain metastases, supporting our earlier
comment on the drawback ofadjuvant pelvic irradiation.
The other patient died from acute congestive cardiac faiiun 6 months later. A few casm of severe congestive cardiac
Vol. 64
failure have been reported after mitomycin C adminii
tration in patients previously treated with
Our patient ülushates that mitomycin C by itself canrarely
have cardiotoxic effects. Doroshowl* had demonstrated
that mitomycin C may enhance superoxide and hydrogen
peroxide formation in the rat heart. This finding might
explain the toxic effect of mitomycin on the heart. Although such therapy is not without risks, the preliminary
results do show the usefulness of the drug as an adjunct
to radical surgery in patients with high-risk factors present
and merits further study.
i
i
1
i
vi
~
REFERENCES
I.
PivaMS,ChungWS.~osticsignifimaofarvicnllnionsize
and pclvic node metatasa in cervical &noma
Obsiet Gyncol1975;
507-5 IO.
2. Buckley CH, Bepnh CS, Fox H. Pathological prognostic indicaton
io cervical m a r with paRiculnr rdmna to pticnts under the age of
40 y e a n BrJ Obsiet Gynmd 1988; 9547-56.
3. Sidhu O, Koss Uj.Barbcr HRK. Relation of histological factors
to the m p o n s of Stage 1 cpidumoid Caronoma of the c e M x to surgical
matment Obsrn Gynecol 1970; 35329-339.
4. F r i a OH, Wnom L. Blood 4invasionin mar of the d x .
Cancer 1962; 151269-1274.
5. üarber HRK, Sommm SC, Ronrrdam H et al. Vascular invasion
as a prognostic factor in Stage IB canar of the
Obsrei G y m d
1976; 52343-348.
6. Martinbeau PW, Kjomad KE, ivcrscn T. Stage IB &noma of
the cervix. The [email protected] Radium Hospitll: ti. Results when pelvic
nodes involved. Obsfet Gynml 1982; tWZI5-218.
7. Fulla AF, EUiot N, KoJoloff C el al. Lymph node metaRara from
carcinoma of the d x : Stage IB and I i A implications for pmgnosU
and treatment GynccdOncd 1982; 13164-171.
8. HaLins WJ, Ford JH Ir, Lutz M H et al. Radical hyitemtomy
and d v i c IymphSdcKMmy for the managementofearly invasivecanm
of the cervix. G y n d Oncd 1976; 4278-290.
9. Morley OW, Seksi IC. Radical plvic surgery venus radiation for
Stage I m i n o m a of the arvix (cxclusivc of microcarcinoma). Am J
Obsiet G y d 1976; 126785-789,
IO. Shingleton HM,Orr JW. Rimary surgical and combined matmcnt. in: Sinpcr A,Jordan J. & Canar of the Cervix: Diagnosis and
Treatment. London: Churchill Livingnone, 1983; 76-100.
I I. Dirhc S. Radiotherapy of cervical canar. Clin Obsief C y m d
1985; 12203-227.
12. Sivanannlnam V, Scn OK, Jayalakshmi. Adjuvant cytotoxic
chemotherapy follOmng Wmheim radiealhysterrctomy for ceMcal
canar. A u f NZ JObsret Gynecol1987; 22231-233.
13. n i i g p n T,Vana RB. 8.lduc-i L, Blesing J. Chemotherapy in
the management of advanad or rrcumnt d c a l canar and endomnrislrcamri. Cancer 1981; 48658-665.
14. Miyamota T. T i w K Y, Watinpbe M, Tensimi T. E W v e n a r
of a ysvcntiPl combinationof bleomycin and mitomydn-C on an advanced emial mm.Cancer 1978; 41:403-414.
IS. mu a,ch.oiYs, su sc. Flu@& of utmne ceMcal mm
with extensive lymph node mclmkses Am J Obsfn Gpwcd 1972; I 1 4
954-962.
16. Crcceh RH. Caulino RB. Shah MK e) u/.An eü&w low dose
mitomycin regimenfor h m o d and chemotherapy refractory ptimu
withm*utiticbrrrdaaxr.Conen1983;5I:IOK1040.
17. B u d u AU, Legba s$ Tashima CK et d.Mriuaynn and mitomycin C POiaiMc synerjcmic toxicity. C a m T r a Rep 1978; 6 1
1005-1008.
18. Domsbow IH. Mitomyeid enhatmi rupmride and hydpemxideformationintherathmn. JPlilm<imlExpTher 1981:218:
206-211.
&.
T2:
chemoth
tentidy
neurotoi
which n
festation
such as 1
pnesthe
toxicity
include
Buctuat
nomic 4
adverte
and syn
neurop
sureme
ASSe
&mat
-
No. :
-
pho
chi1
Sub
con
on t
in :
rins
chlr
met
the!
emi
43"
for
The Demonstration of Human Papillomavirus 7 6 Genomes
in the Nuclei of Genital Cancers Using Two Different
Methods of in Situ Hybridization
ELKE-INGRID GRUOENDORF-CONEN, MD,' AND SABINE CREMER. MW
Biopsy specimens of 16 invasive genital cancers (two vuivai arcinomas, two archmu of the vngitu,
and 12 cervix d w m u ) were examined for the presence and dlitrlbntion of human paplbmivina
(HPV) DNA by in r i m hybrldiutioa with 'H-likkd and bbünylited DNA probes of HPV 6, HPV 11,
and HPV 16. None of the tumors reacted with HPV 6 or HPV 11. Using in situ hybridbation with 'Hiabeled DNA probes nine of the 16 nnms &ive positive results with HPV 16. Only three of the nine
wen positive for HPV 16 by in situ hybridbatioa with biotinylited probes. Cnrrently, the method of in
situ hyiwidlrrtlon with commerciai blotlnyiated probes is kss sensitive than in situ hybridlrrtiw with
'H-libeled HPV DNA.
Cancer 65:23&241,1990.
I
N RECENT YEARS, mounting evidence has associated
human papillomaviruses (HPV) with genital cancer
and its precursors. Human papillomaviruses have been
detected in more than 80% of biopsy specimens from genital cancers.' Although different types of H P V have been
observed in Benita1 cancerous and precancerous lesions,
H P V 16 and H P V LE are most oAen associated with highgrade neoplasias. This implies that these types may have
a high oncogenic potential.
The use of cloned D N A in hybridization experiments
proved to be the most sensitive method for detecting virus
infection in clinical specimens. When cellular morphologic type is to be correlated with the presence of viral
nucleic acid, in siiu hybridization on tissue sections is
indispensable. This technique permits localization of virus-infected cells within a biopsy sample.'.'
Although labeling of the H P V DNA probe with radioactive nucleotides is still the method of choice, major progress has been made in the incorporation of biotinylated
nucleotides into the DNA strands, followed by detection
with enzyme-coupled avidin and a staining reaction.
However, opinions are contradictory with regard to the
sensitivity of in situ hybnduation with biotin-labeled
-
Fmm UK *apr<mnit of h a t o l o g y al the Medid Faculty of the
R W " . Arhcn: and thc tr>cpr<ment oiGynssology at the ONUdt
HOapiUl ofthe Mcdial S c h d Hannowr. Hannover, Federal R-MK
of ocnnuiy.
Supported by a DeuWhe Fonchungq#meiMeM grant No. Gr.641.
Mdrar br nrpnitc U b i @ GniBmdoñconm. MD, HiuIkünik
der Msduinixim FakulUi, an der RWTH Aachrn. F'auwekStnBe. D
5100 A.ebm. Fedenl RepuMK of Germany.
AcecDIsd for publition July 21, 1989.
P
in €
bat1
Was
stai
and
Of<
tot
In :
Hu.
S
probes as compared with the sensitivity of methods using
isotope-labeled p r ~ b e s . ~In
' this study, two different in
sifu hybridization methods using 'H-labeled or biotinlabeled DNA probes are compared in disclosing H P V
DNA sequences in frozen tissue samples of female genital
cancers.
to s
for
eth:
que
buf
Materials and Methods
con
W e determined the presence and localization of viral
DNA in female genital lesions using different methods of
in situ hybridization with labeled DNA of H P V 6, H P V
I I, and H P V 16. Sixteen genital cancers were studied.
The deep-frozen tumor tissues were received from the
Department of Gynaecology at Hannover. Federal Republic of Germany. W e examined 12cervix carcinomas.
two vulva carcinomas, and two carcinomas of the vagina.
For comparison of the different in situ hybridization techniques, parallel sections of each tissue sample were s u b
jected to both procedures. We performed method I using
HPV 6, HPV 1I , and H P V 16 DNA nick translated with
'H-TTP as probes. The different HPV clones were generously provided by Dr.Lutz Gissmann/Heidelbeg. For
method 2 we used the "PathoGene" DNA probe assay
for identification of human papillomavirus types 6/1 I and
16 in tissue sections (Enzo Diagnostics, Inc., New York).
coli
eac
Ute
diu
tail
anc
inc
soli
roc
mn
I
ish
on!
for
3-a
in 1
an<
slic
her
In Situ Hybridization With 'H-Labeled
Human Papilloma Virus DNA
Frozen d o n s ofabout 5 wn were mounied on slides.
dried in air, and fixed in methanokglacial acetic acid 3: I
(v/v) for 3 minutes. ARer heat denaturation in boiling
of <
I1
tes
L
WJO2.
HPV 16 IN GENITALCANCERS
saline (PBS) for I 5 pconds slides were
Ph-.hate-buffered
.
gilled in iced water and fixed in methanol for 3 minutes.
Subsequently 50 hl of denaturated labeled D N A solution
eonmining IO' cpm per slide and a coverslip were placed
on the d o n s which were incubated for 20 hours at 65OC
in a moist chamber. After incubation, coverslips were
rinsed with 2x ssc (ix ssc contains 0.15 mol/l sodium
chloride (NaCI) and 0.01 5 mol/l sodium citrate) fixed in
methanol for 3 minutes and air dried. The slides were
then dipped into a 1: I dilution of Kodak nuclear track
emulsion (Eastman Kodak Co., Rochester, NY)kept at
4 3 T . AAer having been dried in air, they were exposed
for 20 days at -20°C.
After exposure, the slides were developed for 4 minutes
in Eukobrom, developing was stopped in an acetic acid
bath for 2 minutes, then normal photographic fixation,
washing, and airdrying followed. The sections were
stained with hematoxylin and eosin, mounted in Eukitt,
and covered with a glass slip. Labeling appears in the form
of dark grains, dispersed or in clusters, strongly limited
to the nuclei of epithelial cells.
In Situ Hybridizaiion WirhBioiin-Lnbeled
Human Papilloma Virus DNA
Sections of about 6 pm of frozen specimen were applied
to adhesive pretreated slides, fixed by soaking in acetone
for 10 minutes, and airdried. After dehydration in
ethanol, slides were dried at room temperature. Subsequently 1drop ofdenaturated biotinylated HPV D N A in
buffered NaCI/ethylenediaminetetraacetic acid (EDTA)
containing 40% formamide, 8% hybridization enhancer,
coloring and a coverslip were placed on the sections of
each slide. The slides were then incubated for 4 to 5 minutes at 92°C and for 25 minutes at 37OC. Coverslipswere
discarded and 0.5 ml ofa post hybridization solution containing 20 mmol/l phosphate buffer, 260 mmol/l NaCI,
and 40%formamide were applied to every specimen and
incubated for 10 minutes at 37°C. The posthybridization
solution was then tapped off and slides were washed at
room temperature with IO mmol/l PBS containing 5
mmol/l EDTA.
A total of 0.5 ml of streptavidin-biotinylatedhorseradish peroxidase complex in buffered NaCl was dropped
onto each specimen and slides were incubated at 37°C
for 15 minutes. After washing, 0.5 ml of a mixture of 2%
3-amino-9-ethylwbazole (AEC) and hydrogen peroxide
in 0.05 mol/l acetate buffer was applied to each specimen
and slides were incubated at 37°C for 15 minutes. The
slides were then washed in buffer and counterstainedwith
hematoxylin.
The appearance of a red precipitate within the nuclei
of epithelial cells indicates the presence of either HPV 6/
I I or HPV 16, or both, in the biopsy specimen being
tested.
239
[email protected]/-Conenand Cremer
Results
We examined biopsy spechens of 16 women with
genital cancer for the presence of HPV D N A in these
lesions using two different methods of in siru hybridization. All cases studied represented invasive, poorly differentiated squamous cell carcinomas. The results of in
siiu hybridization are shown in Table I .
Nine of the 16 invasive cancm gave positive results
with HPV 16 using an in siru hybridization technique
with tritium labeled D N A probes of HPV 6, I I and 16.
Three of the nine were positive for HPV 16 using in siiu
hybridizationwith biotinylated probes. These tissue specimens gave especially strong positive signals with radioactively labeled HPV 16 DNA. None ofthe tumors reacted
with HPV 6 or I1 DNA.
The specific hybridization signals were seen to varying
extents within nuclei of cells in the tumor strands. in some
cases,positive labeling was confined to only a few nuclei
which were grouped within the invasive tumor tissue (Fig.
I). The intensity of nuclear staining varied considerably
from cell to cell within these HPV I&@tive cell clusters.
Tumor strands of other cases showed extensive areas in
which nearly all tumor cells contained nuclei with strong
positivesignals(Figs. 2 and 3). Morphologicallythese areas
could not be differentiated from those without labeling.
They did not contain conspicuous cytoplasm or cytoplasmic halos similar to the cells that stain in condylomas
or in bowenoid papulosis.
Diseussion
The well-established close relationship between HPV
and genital cancers has rendered the in siiu hybridization
TABU I.
HPV In Siru HybndUation RnulU in Genital Cancers
In siru hvbridization with
Biotinylatcd
'H DNA
Age
Sample
(yr)
HI
H2
H3
H4
HS
H6
H7
H8
H9
HI0
HI I
HI2
HI3
HI4
HI5
HI6
43
42
51
65
88
83
57
36
41
47
78
42
41
78
DNA
Cancer
diagnosis
HPV
HPV
6/11
HPV
16
HPV
6/11
cmix
cervix
cervix
Cmix
cmix
cervix
+
++
O
O
-
O
O
O
-
O
O
+-
O
O
O
O
O
O
O
-
Vagina
Cmix
Vulva
CeMx
cmix
CeMx
CeMx
Vulva
cervix
Vagina
H P V human papillornavinis.
16
+
+
+-
+
+-
+-
O
O
O
O
+
-
O
O
O
-
O
O
-
+
-
-
O
O
O
O
O
O
O
O
O
240
CANCERJanuary 15 1990
Vol. 65
RG. 1. Case H9: InvaUvc vulval
carcinoma showing hybridization
with 'H-labeled HPV 16 DNA. Cell
clusten with paaitively labled nwki
lie within the indiUnentiatal tumor
tissue (X160).
methods increasingly imporiant in the study of both the
distribution o f HPV DNA sequences within the tumor
tissue and the identification of cells whose nuclei contain
the HPV DNA. Nevertheless, several problems are associated with in silu hybridization techniques, including
significant time-consumption, the use of radioactivity in
the case of in situ hybridization with tritium labeled
probes, and lower sensitivity with biotinylated probes.
In the current study, frozen tissue of 16 invasive cancers
of the female lower genital tract were subjected to in situ
hybridization with 'H-labeled probes and biotinylated
probes. Using HPV 16 DNA nick translated with 'H-TTP,
clearly labeled nuclei could be detected within nine of the
cancers studied. Interestingly, the HPV-infected areas did
not differ morphologically from those with lack oflabelin&
These findings in invasive genital carcinomas deviate from
observations made by other authors in early cervical neoplasia. Crum el al.' found HPV sequences in areas of
cervical intraepithelial neoplasms (CIN) which contained
obvious maturation signs as conspicuous cytoplasm or
cytoplasmic halos, similar to cells that stain in condylomas. This might reflect differences in the replication and
expression of HPV 16 DNA in invasive cancers in comparison to its precursors as CIN.
Recent examinations of formalin-fixed and paraffinembedded tissue sections have shown that insitu hybrid-
No. 1
Fi,
nom
izati<
DN/
ira1
Par
ex¡'
DE
est;
Th
pre
ma
stu
we
cia
tai
Wii
Wb
aci
fez
th;
irn
iz2
ce
thg
Flc?.
noma c
ization
DNA. I
annsi:
H7: Invasive Enrovagina. In situ hybndth 'H-labeled HPV 16
it of the tumor cclh show
kled nuclei (Xlóü).
Cye
Y
fYi
ce
tis
ca
si,
ex
NO. 2
HPV 16 IN GENITALCANCERS
[email protected]$Conen and Cremer
24 I
w.
3. C a s H7: Invasive carcinoma ofthe vagina. In siri: hyhrida t i o n with biotinylated HPV 16
DNA (X160).
re&--or
ization with radioactive and biotinylated probes are comparable in sensitivity,although contradictory opinions also
exist? The method of ir! siiir hybridization with 'H-labeled
DNA. which requires frozen tissue sections, has been well
established in our laboratory over a period of IO years.
The theoretical sensitivity of this method is such that the
presence of approximately 20 viral genomes (approximately IO grains) should be detectable? In our current
study we compared the results of this method with those
we obtained with biotin-labeled probes using a commercial detection kit. Only three of the nine HPV 16-containing tumors were positive with in siiu hybridization
with biotinylated probes. These were the tissue specimens
which gave especially strong positive signals with radioactively labeled HPV 16 DNA. However, the morphologic
features in biotin-treated sections were better preserved
than in those which were 'H-DNA treated. This is a very
important observation, for it is the aim of in situ hybndization to correlate the presence of viral sequences with
cell morphologic type. In the case ofbiotin-treatedsections
the lack of radioactive background was especially satisfying. Positive signals were restricted to the nuclei oftumor
cells, whereas the nuclei of the surrounding connective
tissue and of infiltrating lymphocytes were negative.
Currently, the method of in sirir hybridization with
commercial biotinylated probes is less sensitive than in
sirir hybridization with 'H-labeled HPV DNA. Further
experimentation focused on improved sensitivity of the
avidin-biotin- based in siru hybridization techniques are
required. A s rapid, less complicated and nonradioactive
methods, these should have increasing application in further studies of HPV-dependent cancers.
REFERENCES
I . Zur Hauxn H. Papillomavinixs in human cancer. Cancer 1987;
591692-1696.
2. Gnirwndorf-Conen El. In siru hybridization with papillomavirus
D N A in genital Inions. In: Peto R, zur Hausen H, eds. Viral Etiology
ofCervical Canm(Banbury Rw~2
iI). Cold Spring Harbor, N Y : Cold
Spring Harbor Laboratory, 1986 724.
3. Ortrow R, Manias D, Clark B, Mragaki T, Twiggs L, Faras A.
Detection ofpapillomannirspecific sequences in human genital tumon
by in situ hybridization. Banbury Rep 1986; 21:253-258.
4. Cnim CP, N.@ N, Lcvine RU,Silvmein S. In silir hybridization
analysis of HPV 16 D N A sequences in early amcal neoplasia. Am J
Parhol 1986 123174-182.
5. Gissmann L. Dun1 M,Oltmdorf T, Daterilz K. Human papiilomaviruses and cervical cancer. In: Steinberg BM.Brandsma JL, Taichman LB, eds. Cancer Cells: V. Papillomaviruses. Cold Spring Harbor,
N Y : Cold Spring Harbor Laboratory, 1987; 275-280.
6. Syjanen S, PaMnen P, Syjanen K. CompIlnson of in siru D N A
hybridization pmlocols using "Slabeled and biotin-labeled probes in
detection ofhuman papillomavinis D N A xquennr. In: Steinberg BM,
Brandsma JL. Taichman LB. eds. Cancer Cells: V. Papillomaviruses.
Cold Spring Harbor, Ny.Cold Spring Harbor Laboratory, 1987; 329336.
7. Wells M, GriffithsS. Lewis F. Bird CC. Demonstration of human
papillomavinis t y p s in paraffinpromsedtissue from buman anogenital
laions by in siru D N A hybridization. JPorhol 1987; 15277-82.
8 . GnisrendoñE-I, zur Hausen H. W i z a t i o n of viral D N A replication in sections of human warts by nucleic acid hybridization with
complementary R N A of human papilloma virus type 1,Arch Dermarol
Re5 1979: 26455-57.
Detection of Squamous Cell Carcinoma Antigen in Normal
Squamous Epithelia and in Squamous Cell Carcinomas of the
Uterine Cervix
GERD CROMBACH, MD.' ANTON SCHARL, MD.' MA'ITHIAS VIERBUCHEN, MD.t HANNELORE WÜRL, PIID.'
AND ACHIM BOLTE, MD'
I
Torigoe isolated the tumor antigen
TA-4 from a cervical squamous cell carcinoma.' TA4 is a glycoprotein with a molecular Sue of 4 8 . W daltons
which is located in the cytoplasm of normal squamous
epithelia and squamous cell carcinomas of the uterine
cervix.'.' wuamous cell carcinoma (SCC) antigen is one
of 14 subfractions of TA-4, purif~edfrom liver metastam
of a cervical squamous ceü carcinoma.' The clinical value
of TA-4 and SCC antigen as serum tumor markers for
cervical cancer hasbeen demonstrated in numerous studies,2.S-15 The sensitivity is 44% to 67% inprimary and 67%
to 100% in recumnt cervical squamous cell carcinomas.
%Nm lCVCk Of both antigens Efl& the extent Of the
carcinoma and the prognssion or e o n of the tumor
during follow-up. However, neither antigen is specific for
cervical squamous cell carcinomas. Elmied serum levels
arc alu, found in 24% to 53% of patients with squamous
cell carcinomas of the head and neck, esophagus and
N 1977, Kat0 and
and in 8% to 42% of patients with adenocarcinomas of the uterus, ovary, and
Until now, detailed studies on the distribution of SCC
antigen in diffmnt body tissues have not bcen publied,
probably due to the lack of a suitable antibody for immunohistochemical analysis. The current study evaluates
the specifici*. of SCC antigen and the factors influencing
the release of the antigen into the circulation. Antigen
concentrations wm simultaneously mcasurcd in tissue
extracts and sera of patients with normal epithelia 01different carcinomas of the female gcnitai tract. Serum concentrationsof SCC antigen in patients with cervical squamous cell carcinomas were correlated to the ciinical stage
and to the degree of histologic differentiation of the tumors.
M i t d iadMethods
One hundrcd tifiy-sevm t i m e &mens were obtained
from
108 patinis undewoing s q q for nonmaiignant
From the aprtmcnis of Wlsctne
' indOynmlosyandtPathoiogy.
gyna~ologicdisorders (myoma, endometriosis, desceasus,
Univmity of Colagne. F d d RcpuMK of Gnminy.
The authors thank D o n s p*enfor teehnieiluriauia.
fibroadenoma) (n = 45) or for malignant gynacologic tuAddrcss for reprints: GCNICmmbach. MD.UnivmitW+Frauenklinik
K(lln,KerpnerStr. ) 4 , D S O O O K M n 4 I , F c d e n l R ~ M i c o f ~ a n y . mors (n = 63). In 21 of the 45 women with benign lesions,
tissue sections were simultaneously pnpand from two or
Aax!ptd for publication o*& 24.1988.
1337
CANCERApril I 1989
1338
specimens were stored at - 7 O T until used. A volume of
SCC
Wmo cp
I
f.
1MWz)-
Vol. 63
4- +
I
"
*r
1OW-
:
I
I
100-
..
I
i
r
I
10,
c
I
Nmrds<rrmiNmn*
- c H
-s
glh.Lm
Cuckana.
mh.Lm
ni45
"-22
n- 24
r
..
...
MeraS"Sh0ni.
n- 41
FIG. I. squamous all caminoms antigen concentrations (ns/mr a l l
protein) in the cytosol ofnornul epithdu and &nomar ofthe vulva,
vagina, uterine &r, endomeuium. ovary, and breast (-: median
value).
more sites of the genital tract. The following biopsy
specimenswere taken: (I) normal squamous epithelia of
the exocewix (n = 31), vagina (n = 7), and bnast skin (n
= 5); (2) normal glandular epithelia of the endocervix (n
= 17) and endometrium (n = 17); (3) normal stroma of
the arVicplwall (n = 17); (4) squamous cell carcinomas
of the uterine cervix (n = 15), vulva (n = 5). and vapina
(n = 2); and (5) adenocarcinomas of the endocervix (n
= 4), endometrium (n = 17). ovary (n = lo), and breast
(a = IO).
Tissue specimens were divided into two parts for histologic examination and for CytoJOl preparation. Fresh
biopsy specimens wen cooled on ice immediately after
excision, cut and weighed, puivcrized under liquid nitrogen (winga microdismembrator, obtained from BraunMelningen, FRG), extracted with a four-fold amount of
phorphite buffer (0.01 mom, pH 7, containing 10% of
giycerol) and centrifuged for 45 minutes at IOOOOO X g.
The cytosol was separated from the Wel and the Lipid
l a w . The totai pmtcin content waa determined axodq
to iowry d d.'' and was wmcted for scmm albumin
contamination by radial immunodiñusion (usinsM-Partisen h
t
e
s from Behnnswerke, Marb~ug,FRG). C y t d
0.1 mlpunordiiutedcytosolwasuscdforeachasay.
Serum samples were obtained from 82 of the abovementioned 108 patients, from 93 women with primary
squamous ceU carcinomas and from 13 patients with recurrent squamous cell carcinomas of the uterine cervix.
AU sera wm stored at -2O'C until used. Diagnosis was
conürmed by histologic examination in all cases. Clinicai
siaging of primary ceMcal squamous cell carcinomaswas
bascd on the criteria of the International Fo'deration of
Gynecologists and Obstetricians (FiGO). cmical squamous d carcinomas wm c M e d aswcüdüTcrcntiatcd
(Grade I), moderately di&nntiated(Grade 2). and poorly
diñerentiated (Grade 3) tumors. Criteria of c l a d i d i o n
w m the polymorphism of ceUs and nuclei, nucleus/cytoplasm ratio, mitotic activity, and kmtin formation."
The SCC concentrations in the cytosol ( W m g ceU protein [CP]) and serum (ng/ml) were determined by a double-antibody radioimmunoassay (Abbott-Diagnostics,
Wiesbaden, FRG), based on the d v i t y of a polyclonal
goat antibody. The limit of the normal range in serum
corresponded to the threshold value of 95% specificity in
healthy female controls and was 2.5 ng/ml." The coefficients of intraassay and interassay variation were 13.8%
and 15.0%. mpcctively.
The Mann-Whitney test was applied for evaluation of
signiñcant differences between individual groups of tissue
specimens or serum samples. Differences w m considered
significant when the probability of error was below 5% (P
< 0.05).
ReSnltl
The disixibution of S C C aniigm concentrations in the
cytosol of normal epithelii of the female gcnitai tract and
of gynecologic carcinomas is shown in F i i 1. The antigen values ranged from 194 to 25.033 ng/mgCP in nor-
mai squamous epithelia, from 350 to 6917 Wrng CP in
squamous cell carcinomas, from 1to 975 numg CP in
normal columnar epithelia, and from 1to 850 ng/mg CP
in adenocarcinomas. The median values in these groups
wen 4529,2477, 15, and 3 ng/wCP, rcsp&vely. The
an*n concentrationsin normal squamousepithelll and
in squamous cell carcinomas were sionifcantiy higher
than those in normai columnar epithelia and in adenocarcinomas (P 0.oooOi). The SCC antigen valua exloo0 WmgCP wm mep(uítd in 40 of43 normal
squamous epitheli and in 18 of 22 squamous cell car-
cinomas, but in none of 34 normal columnar epithelia or
of41 adenocarcinomas.
The highat SCC antigen levels wm found in the cytodofnodsquamousepithdi.ofthecxoaniix(Tabk
I). In nine of 3 I casu the valua iay above 1O.oo0 Wmg
No. 7
SCC A N T I G E N I N i H E UTERINE C Z R V i X
TABLEI.
31
7
5
194-25,033
2008-8671
2696060
Columnar epithelium
Endocervix
Endomnnum
17
17
4-975
1-57
Stmma
Cmical
17
Wall
CrOfdJUChd
I339
d.
Squamous CCU cuiinoni Anli#m Conmtmions in the cytoool oí Normal bilhclium and
Stroma (Wnu Cell W n ) and in Serum (Wml)
Squamous cpithdium
E x d x
Vana
B w &in
*
MI3
2520
2/19
I 16
o14
0.3-5.6
0.3-5.6
0.3-2.0
1.7
I .6
0.8
97
6
1/12
1/12
1.2-3.3
1.2-3.3
I .9
I .9
22
1/12
1.2-3.3
I .9
6ou)
C P all protein.
CP. The median concentration in this group was 2.3-fold
hie& than that in squamous cell carcinomas of the arvix
(Table 2). The antigen values of that groups wen Sipnificantly different (P< 0.0005). Dcspite the extremely
high antigen concentrations in normal squamous portio
epithelia, only two of 19 patients had inrrcascd SCC antigen levels in serum (3.3 and 5.6 ne/mi, rrspCtively). In
contrast,women with exvical squamous allcarcinomas
had lower cytosol concentrations,but bigher aniigen values in serum (Tables 1 and 2). Elevated serum lcvels up
to 61.9 ne/mi wen found in eight of 13 patients. Linear
rcgrcJsion d*showed no correlation between cytosol
concentrations and serum levels in cervical squamous all
carcinomas(r = O. 18). Cyiosol concentrations in normal
epithelia of the vagina and breast skin wen of the same
order of magnitude as those in squamous cell carcinomas
of the excccrvix, vulva, and vagina (Tables I and 2).
The median concentrations of SCC antigen in normal
squamous epithelia and in squamous cell carcinomas of
the exoccrvix wen five to &fold higher than thosc in
normal muma and adenocarcinomas of the endocervix
and 100-fold to 6ooo.fold higher then thosc in normal
endometrium, cervical stroma, and adenocarcinomasof
the endometrium, ovary, and breast (Tables 1and 2).
Antigen concentrations lying above the lowest values
measured in normal squamous portio epithelia (194 ne/
m g CP) and in Cmiical squamous cell carcinomas (350
w
m
g CP)wcre only found in a few specimens of normal
mucosa and of adenocarcinomasofthe endoccrvix.However, the highest concentrationsmeasured in normal and
malignant lesions of the endooemX did not c x d loo0
ne/- CF'. Elevated SCC antigen kvels in serum werc
only found in one of 12 women with normal glanduiar
epithciia and cervical stroma and in two of 34 patients
(6%) with adenocarcinomas.
Patients with primary cervical squamous cell carcinomas had elevated SCC aniigen serum levels in 61% of
-. Thc positivity rate and the m m concentrations
TABLE2. Squamous Cell Carcinoma An-
Conantntions in the Cytoml of Squamous Ccll C a r i n o m and
Adenoareinomu ( W m g Cell Rotein) and in Smim (Wml)
Cytosol
Squamous all carcinoma
cervix
Vulva
Vagina
Adenomrhoma
cervix
Endomnnum
OWrj
Blurt
Median
&nBc
(WnuCP)
twnu CP)
5
350-4578
1905-69 I7
2124-2450
2483
2520
2287
4
167-850
1-141
508
n
L.
Smim
I5
2
17
IO
IO
1-6
1-14
Elevatcd kvels
(>2.5 Wml)
8/13
015
012
114
10
0110
2
O110
I
1/10
Mdmn
Rinec
(WmU
.
1.3-6 I .9
0.5-1.8
I .3- I .4
3.7
I .o
I .4
1.3-2.7
0.5-2.0
0.5-6.0
0.6-2.4
1.9
1.1
0.8
I .o
C P cell protein.
-
-.
1
L.
(WmU
~~
~~
~
CANCER
April I 1989
1340
TABLE3.
~
Vd. 63
Incukna of Elevated Squamous Cell Carnnoma Antigen Lrvclr (>2 5 ne/ml) and Mediui Conanuruona in Serum ocpcndiq on
Tumor S t y c and M the üegm of Hiaolopic DiRmnlution of R i m w C m a l Squunous Cell Carcinompr
Histologic grading
O1
1
I1
1
1
1
1v
Toul
214
Ill
414
-
7 P (78%)
2.3
3.3
36.5
-
5/17
14/21
17/19
616
7.7
42/63 (67%)
T d
O3
G2
2.2
3.6
11.7
48.1
4.8
2/10
316
lb?
213
8/21 (38%)
1.6
3.6
21.2
6.4
9/31 (2%)
18/28 (64%)
w25 (88%)
8/9 (89%
I .9
3.5
14.5
33.1
1
I .8
FIOO: Intcmationai FcQauion of oynsolcgy and Obst*r*r
measured depended on tumor extent and on the d c p e
of Ilhtologic di&rentiatim (Table 3). The incidence of
pathologic valued increpsed from 29% in Stage I to 89%
in Stage tV, and the median d u e s
from 1.9 ng/ml
to 33.1 ng/ml. Antigen kVcL in Stage III/lV wen significantly di&nnt from those in Stages I and I1 (P< 0.01).
Patierits with well-diffmntiatcd or moderateiy
entiatcd carcinomasbad paihologic antigen levels in 78%
and 67% of cases, rcspeCtivciy. The positivity rate in
women with poorly di&mitisted tumors was only 38%.
ThcsenrmconmtrationsinGradc 1 andGrade2tumon
wen genificantly higher thanthoac in<jrade 3 carcinomas
(P 0.05). The median valua were hiest in Gradc 1
(7.7 &mi) and 10- in Grade 3 (1.8 Wml) tumors.
Similar rcsults were found in women with recurrent ccívical squamous cell carcinomas.Elevated rrmm concentrations were measured in all eight patients with Gradc 2
tumon, but only in two of five women with Grade 3 carcinomas.However, the value$üom patients with rrcurrmt
Grade 2 tumon wen not sigdicautly di&rrnt üom t h e
with Grade 3 carcinomas (P> 0.0s).
Mediaa $02 antigen co~~ccntratiom
in the cytosoi of
squamous ceü carcinomas wen higher in Grade I and
Grade 2 tumon (n 10, t = 2522 ng/mg CP) than in
Gnde 3 csrcinomas (n = 5, i
970
CP). However,
the antigen ranpsin both groups wen similar (350-4578
and 4210 n g / q CP,nspcctiVely). SUtbtid aluation of the cytosol conccIItratiommeskd no seniñcpnt
difFerenccs (P> 0.05).
- -
r)bcoMh
High conantrations dsCC a
n
w
n were deteaed in
the @ of U
O S~UXUOUS
~
epitheli. and of quamow cell aminomne. The median concentrations were
300.foId to 1500-foM hi#herthan those in n o d &Idulu epithelia and in adeiioam'nomas (Fi I). The
L
highest median value was found in normal squamous
portio epithelia e x d u g that of cervical squamous ceU
carcinomas by a factor of 2.3. In &adular epithelia the
antigen conmtratiom wm
in normal muwsu
and in carcinomas of the endocervix. Some v d u a even
overlapl>sd with the lowa5t concentrations incammi in
squamous epithelia This fact could be attributed to the
contaminationof some endocmid iissuc apecimen with
small parts of adjacent squamous epithelia Entirely pun
tissue Prepantions of normal giandular epithciia and adenocarcinomas of the endometrium, ovary and breast
without squamous piuo contained only very low antigen
concentrations.
Similar resuits were reporttdby MonokaZiwho found
TA-4 concentrations in normal
10-fold to SO-fold
and malignant lesions of squamous portio epithelia than
in normal cndoceMcai mucos& Neither ceMcal stroma
and n o d endometrium nor ~ d n o m a ofs the
endoarvix, endom&um, and ovary had mcawmbieTA4 activitia. In contraed to the nsultc of the cumnt study,
the mean TA-4 concentrations measured in ceMcal
spuamousdcaranomaswerefi~fdd~thanthe
in normal zquamous portio epithelia. According to the
results of Bow cytometric analpis in cervical squamous
cells, the TA-4 content in CPI~CCIcells waa indeed eight
tima greater than in n o d ceüs.=
Howem, in the esophaeis aignibmtly higher SCC
a n w n concentrations were measured in n o d squamous epithelu than in squamous ceü carcinomaa." The
reason for the di&nng TA-4 and SCC antigen mtioa in
thcq studiu may be the hcterosnieity of the JycoProtnn
TA4 which coushía of 14 Subfiactiona witb isoelectric
points (IEP) [email protected] from 5.44 to 6.624 The SCC aniigen
is a ncariy pure protein with a carbohydrate content of
less than 0.6% and har the most neutral IEF' (6.62) of all
subfdoas."'2 Qto n al." dcmonssnted in an d e r
study that aeutrplTA-4 (IEP, 6.3-6.6)bth~
prrdomiannt
I
in
mr
No. 7
ScC ANTIGEN
IN THE UTERINE
CERVIX
M o a in mnmriioii.ntsquamous epithelia, whmps
-
Crombach n d.
1341
I
I
TA-4 was detenabk only ¡n di&rratiiudsquimous cells
of the intermediate hyn; and not in the immature ceils
of the basallaycr?J2In caviai squamouscell arcinornag
TA-4 was positive in more di&rentiated tumora such as
large cell kmtininng and luge cell nonkrntinlling acinomas, but never in mall cell nonkerstinizing &nomas?.’o
The results of our i n v d g d o n show that SCC antisen
is not tumor ~ 6 cHi@
. quantities arc pnicnt not only
in squamous cell carcinomp* but also in the cytoplasm
of normal squamous epithelia. The releaac of the antigcn
into the circuiation depends on infiitrative p w I h and
the mass of the tumm rcither than on the local tissue
content. Apparently, SCC an&cn is a marker of cell diffmntiation for squamous cdls. Further chuactmzati‘on
of SCC antigen requires simultaneous histologic examinations, immunohWciwmd
’
studiegandmeuwemnit
of concentrations in the cytosol and serum in a large
number of patiente with &cal
squamous allcarcinomas.Biochemical studiesare needed to elucidate the differences between TA-4, SCC antigen, and other subtions.
acidic T A 4 (IEP, 5.9-6.2) is mainly present in the tissue
Ud m u m of patienis with cervical squamous ail carcinoms%. Since it is not clm which subfraction of TA-4
was measured in the studies of Monoka” and Sasaki et
al.,=tissue mcasuiements of TA-4 and SCC antigen cannot k directly compared.
The d t s of TA-4 and SCC antigen daenninatioas
in the cytorol ur in apemeat with the data of an immunohiaxhemical study by Ueda et al.,’who detected
I
T A - ~in alln o d squamous portio epithelia, in most of
!
cervial squamous allcarcinomas and in a few normal
columnar epitheiia and adenocarcinomas of the ende
Cmu, but not in nonnal glandular Cpithdia and in carcinomas ofthe endometrium.
Desplrc the high SCC antigen conantrations in the
cyton01 of normal squamous epithelia, serum levels were
&y
always within the normal range. Women with CCIvical squamous cell carcinomashad lower tissue concentrations, but higher serum values of SCC antigcn. According to our own results and to the data of Morioka?’
t h m was no correlation between cytosol concentrations
and serum levels of SCC antigen or TA-4in cervical squaREFERENCES
mous allcarcinomas.The median serum concentrations
i
and the incidence of elevated serum levels increased with
i
I. Kat0 H,Torigoe T. RdiQnimunoupy for tumor antksn of hu1
tumor extent. niese findings indicate that the release of
man a w b l squamous cdl Minoma. Cmuer 1977; 40:1621-1628.
2. Kat0 H, Mniolu H, Arrmiti S, T o r i p T. RdioimmunoPrPy
1
SCC an@ into the circulation depends on infiltrative
for tumorantipniofhumuiem*J~quimousall&nom. CdlMd
tumor pmih and on tumor mass, but not on the local Biol1979;
I
2551-56.
anaen content in the tissue.
3. U& G. lnoue Y ,YiminliiM n a/.Immunohiaochcm*ll &monstration of tumor a n i k a TA-4 in gyaaolopic himon. lm J GynrOr
Analgous findings have been reportedfor carcinoemPahol1984.3291-298.
bryonic antigen (CEA).Them was no correlation bmvcen
4. lk&
1. Two& D d i o i m m u n o ~ c(nndwieh) m a y of s<x:
piamia and tumor concentrations of CEA in carcinomas
uitign wing monodonil i n n ¡ In: Ksto H, de üruijn HWA, Etai
W.Hnkrmui Rñ,Johnem IT.& SquimwrCdl cucimmi Anlism
of the liver, lung, gsstrointeStiaal, and female genital
in thetM
of5qiuiaaa Cdl cUnwn>..Rinedon: EaCaW
tract.*” Van Nagell et al?’ supposed that plasma CEA
Medics, 1987; 215-226.
reüects the total tumor burden (i.e..tumor CEA concen5. Km0 H,Miyawhi F, Maioiu H,Fqmo T. T o r i p T. Tumor
tration X tumor mass) rather than the tumor CEA con- ~ t i g m o f h u m u i c m i ~ . l ~ o r e i n o m : C a r r * t i o n o f e i r culiting kvds with diax prcsus C m 1979; 43385-390.
centration alone.
6. Km0 H,Mniolu H, Tairaii H, AnmikiS, T ~ ~ i s oT.eVdue of
Howver, thew conclusionscannot explain the fact that
himnuifiscn~A~)ofq~cdlcucuIormuI~~~rrtmt
10%to 20% of patients with cervical squamous cell a- ofcmicai cancer. Cancer 1982, XkI.294-12%.
1
7. Kat0 H, Moriolu H,AnrmLi S, T m u K, To~igocT. Rogioatic
cinomasS~l~andIVhavenomialSCCantigen~enimsignificance
of the tumor mlip~
TA-4 in squmow cdl &noma of
I
levels.Apparently, additional factors influence the release
the utcnne cervix. Am J W n GymccuI1983; 145350-354.
8. Keto H,Tuiui K.MorioLi H. Npú M, Nipya T, Torigac T.
of the marker, one of those being the differentiation of
Tumornnti~TA-4intbcdc<rctionofrswmnainemierlquunous
the d n o m a s . The incidence of elevated SCC antigen
all arcinom. Cancer 19% sJ:ISU-1546.
levels and the median serum values were higher in patients
9. Kat0 H,Tam¡ K. N.pyp T. NSpi M, Torigac T. Tñe we of
i
with well-diffmntiated and moderately Wemntiated
tumor antigen TA-4 for tbc m u y m n t nf squamous all &nom.
C a m DAen ñpv 1985: 8:155-.159.
cervical squamous cell carcinomas than in women with
IO. Mamo T. ShibsU K. Kirnun A,Hoahina M, Mocbiniü M. Tupoorly differentiated tumors. Mcdian cytosol conmtramoru30cistcdint¡ga.TA-4, ¡ n t b c m o n i ~ o f t h e ~ W S o f l h C l a p y
tiom in Grade 1 and Grade 2 minomas were higher
forsquunousallmmiMnudthelrrmnecmicMx. Cancer 1985:56M2308.
than in Grade 3 tumors.Similar data were found in squaI I . Oba~Y.TdokomM,KiuboSnal.BiUcenlvsfionofmcsmous cell carcinomas of the esophagus.’* These findings
s u ~ m mofthe
t
m m knldqumnow alloreinoma-relited antigen
are also conñrmed by the msults of immunohistochemical
(SCC)and its vdue following imdiitionof cancer of the utmne a M x .
Gan No Rinsho 1987: 33óü-M.
studies. In the normal squamous epitheliumof the a n i x ,
.
I
1
-___I
1342
CANCER
April 1 1989
12. Kuo ü, Modolu H. Huhimoto K n d Sccuityn and in
cliaiaiapphtbnr in: W H . de h i j n HWA, Ebut W.Habaiaia
RE, IOhnWn JT, & %lWllOUI CdlcUaMnU Anin the MMsgemat dSqumour Cell curinom. Rinceto~hcetpm Media,
1987; 1-14,
13. Cromb& O. WUn H, Bdtc A. DctcmiiMtioa ofSCC M~
in the m m dpmienn with M i n o m i ofthe cavix umi. GebunJlibé
Fmuenhtilkd 1987; 47439445.
14.
EK, You@ YM. Weiicr P h S p m r CE,M&c SE.
H&ALAnsnliutiondanumo<ndatrimnvmtyninphBO
m t h a r v i a l s q u i m w r a l l c u n ~ n n r A m J O b n n G ~ 1 9 8I57
7;
433439.
IS. ouhi T. MMM T, Y
d M,M ~ ~ l ~ ¡ z uM.
k ¡?diction dthe
r r e u ~ ~ n cofsqwmoua
e
d a t r i ~ mof
i fhe utainc cmix by momtorin< mum TA-4. Nippon Sank Fqiinka G&i
-hi
1987; 3 9
799-806.
16. Ebai W. I&nu>n IT. Tumormarken in the Mamgmat of
Squunwr Cdi curinom of the H a d , Neck and Lun, R i n c ~ n :
ExcapU Mcdla, 1987.
17. Muuolu T. Mitucda Y. Dokawa H, Wiunibc K, Mimoto S.
Tbc mcpunmnt of SCC inin aqulmwr all atriaom of the
lun& Gon No Rimho 198% 31.)14-9l8.
18. Oini M, Miom R Dinidi R üruxagnin O. SCC in*
in
paticno with crophi<t.l ca~%¡nomr
In: Kato H. dc ümün H W h Ebut
W, H&crnm
Ra.Johnson JT, CQ SquunouiCellcueinormA&en
Vol. 63
in thc Mnsgmmt of Squimau Cdl curinohm
Rimton: Excupm
M d a 1987; l3J-141.
19. Lowry DH. Roxbmw NI, Fur& Fud.u RJ. Rotan masuremat mtb the Mn
m<cnt. J Bid Chon 1951: 193265-
275.
mad
a. FSrmay A cucinomiindotba~uUainthimaroftbcc+Nu.
In: ühu.ustan A. rd. R t h o l ofthe
~
Fmiilc 0cnit.l Tnct, ed 2. New
Yo* S ~ ~ ~ I I E1982;
C I , I8CuT
21. Monolu H. Tumor ar&m CrA-4) daqumnoua d d n o m
in tu8ic dlltriñition and ¡la rrbtpiubipto m m TA-4mn0cntmtion.s.
Asia Oceania J ObsIet G+
1980; 691-97.
22. Snaki K,Na#i M, W H . to rig^^ T, Nlpminc Y, T i u u i b i
M. Flow cyiom&c Uuiyliro f i u m o r a n ~
TA-4 in aMalaquamom
cella. Gam 1984,73703-706.
23. Kat0 H, Nlpiya T,T o w T. Hcmqnwty
’ Of8hiWUlTA-4 daqtumour d Muionu in relation to in ippmwcin the
circulation. C a n 1984,15433435.
dvity
24. Khoo SK,Wuncr NL, Lion.c.ianoanbryonie
e
oftiinucexVM;Aquliltititiinhdyof~and~~
plism*Nmoticliva.nomuldultandfetilmpnrImJCanm1973;
11:681-687.
25. Van N.acU IR Donildroii ES,Wood Eo.Slurkey RM. oddciibcr#DM. n i e grqnonic &¡kana
of Minaembyonic inin
the pLm and tumm of paticnu with cndometruld n i d n o m a
A m JObsrn G y n d 1977; 12áM8-313.
TECHNICAL NOTES
I!'
I
L
Sensitivity of the Cytologic
Diagnosis of Cervical Condyloma
in Comparison With HPV-DNA
Hybridization Studies
*
. ., .
., -,
Achim Schneider, M.D., Oabriele Melnhrirdt, C.T.. Ethel-Michele
DeVilliers. M.D.. end Lutr Gi8smann. Ph.0.
The [email protected] diagnosis of nrvicai condyloma is bawd on
witrrio dwrloprd o w thr Inst 10 yrars. Ir has now brcomr
possiblr to documrnt the prcsencr of humon popilioma virus
(HPV) DNA dirrciiy in cervical y b s by thr highly srnsitive
irchnique of DNA filter hybridiraiion in situ. The purpose of
this article is to evaluotr critically thr rmpirirollr established
c~tologic
criteria of condyloma b j amparing them with HPVDXA hybridirationstudies in the ramr matrriol. The rrsults of
this study indicotr that Wassic'*kuilorytatis and djskrrorocytosis arr not highly slnsitiw criteriafor the prrsencr of HPV
infrction, identfiing only 15% of the HPV-DXA-positive
cars mrrectly. In on attempt IO improw the srnsitivity of the
[email protected] diagnosis of HPVinfrctionr. a panrl of nine 'honclassic" criterio was rvaluated. Thr fiw most valuable s i p s were
'kiild koilocyiosis." mild dyskrratwytosis. " hyperchromotic
nuclei. bi- and multinuclration, and & a n d cytoplasm. Using
ihese critrria in eombination. storistically discriminant 0 ~ 1 j sis muld correctly idrntify 84% of the HPV-positive
p u p . Diagn Cytopataoi 1987;3250-255.
Kry Words: Human
loma
papillonu virus; DNA, Cervical wndy
Molecular-bioiogical methods using DNA hybridization
are the most sensitive techniques with which to identify
human papilloma virus (HPV) infections. DNA fragments of ccnain H P V types could k recognized with this
technique in up to 60% of cervical cancer biopsies that
had been completely negative by light and electron
microscopy as well as by immunoperoxidase techniques
using a groupspecific antigen against papilloma virusReceived AIILUSI
22.1986. Acap<ed F e h a r y 28. 1987.
From the Pithologs Liborators d the Dcpnmeni or Obstetrics and
Gynsdop)..University d Ulm. West Germany. 8nd ihc ücpanment or
Vim Rcrurch. Gmnin Gnoa R a a K h Cmtcr. Heidelberg, U'UI
Ganuny.
SuPWId by the Dcutvhe farhunppnwinvhsft (Gi 128/I-i).
Addrrpint r e q u a í u)Achim Scbnaidcr, M.D.. DcpPnment of
GyneodSs. University of Uim. Priitwiusr. 43, D-79 Uim. U'SI
Germany.
250
Diognosiir Cytq>oihologv.I'ofolJ. .Ko3. Sepentber IV87
es?-' Condyloma of the cervix in its exophytic and fiat
form is the mort coinmon manifestation of HPV infection
of the female genital tract. By using generally accepted
criteria, condyloma is diagnosed in I%-Z%of
asymptomatic female patients of a nonrisk population by cytologic
techniques.'-'
The sensitivity of the cytologic diagnosis of an H P V
infection based on the generally accepted criteria has
never been established. The cytologic criteria were dewloped by assuming that an H P V infection presents in
morphologicallysimilar fashion in histologic sections as in
cytologic smears. The purpose of this study was to test the
validity of this assumption by relating the "classic"
cytomorphologic criteria currently in use to HPV-DNA
hybridization studies. A trial was made to improve the
sensitivity of cytology for the detection of condyloma by
using a panel of "nonclassic" cytologic signs.
Material and Metsods
Cmical smears were taken during routine gynecologic
examination of 1,052 women visiting the Outpatient
Clinic of the Department of Gynecology and Obstetrics,
University of Ulm. between July I. 1984, and June 30.
1985. All patients were asymptomatic; none had a previous history of abnormal cytology. Smears were taken
from ihe ecto and endocerrix by mtton-tipped applicator. Half of the material obtained was smeared on a glass
slide, processed for conventional cytologic examination.
and classified according to establiñhed cytomorphologic
criteria? The other half was submitted for filter hybridiuition in situ in order to detect HPV-DNA. The details
of this test haw been described pmiously." The filters
were hybridized uith HPV-I 1 (which a l r o d e t ~ HPV-6
s
under the conditions used) and separately with a mixture
of HPV-I6 and HPV-18. Papanicolaou-stained smears
from I O0 patients positive for H P V - D N A were selccted
e l U 7 IOAKUSIK114 M W 1 C A L M U I H U S . I ~
,'...
I
fa rrrcrrening. Anoibcr I00 smea~s obtained from
pticnu mqativc for HPV-DNA and matched for age
were rclcetcd for comporison. Thee 200 smun are the
buir of this study.
Cells with nuclear atypia. defined as enlargemcni of
tbe aucleus o m b i d with an irregular chromatin pattern. w m absent in all smears that were part of this
siudy. Tbacsmtpn were rescreened retrospcctivcly, with
knowledge of the HPV-DNA rault, for the presence of
the "cluric" cytomorphologic signs of koilocytosis (Fig.
IA) and dyskeratocytosis (Fig. ZA). In addition. nine
nonelassic cytologic signs were cval~~aled.
,
!
* I.
L,,.
CERVICAL CONDYLOMA
4
A
Mild koilocytosis isdistinguished from its classic counand a less
well-defined border between the central cavity and the
peripheral cytoplasm (Fig. 1B).
2. Mild dyskeratocytosis is diñerentiated from classic
dyskeratocytosis by size and shape of the nucleus (Fig.
28). Threcdimcnsional arrangement and orangeo-
cerpart by the absence of nuclear atypia
Fig. Z (A) C h i c dyikentoeylais m ria suprhci.1 qwmour alls
with hypcrcbranitk. slightly cnlr)ed nuclei. mall allulir size. and
m n p p h i k cylophnic niin (x400).- Q)Mild n c n c h i c dyrkeraurytair in a shea of uipibeil quamow alis with -1-sized.
rclulir nuclei, nan~wphilic cytoplasm. and tbrccdimcnrionil
irnnpmcnt (XUX)).
6
-1
c
El
philic cytoplasmic stain are additional features of this
cytologic sign.
3. Cleared cytoplasm (Fig. 3) consists of an evident lack
of stainable matter between nucleus and cell border.
This diagnosis was only made in the absence of bacterial cytolysis as Doederlcin bacteria can cause a similar phenomenon.
4. Keratohyalin and keratobyalin-like granules (Fig. 4)
consist of basophilic or e<ninophilic staining condensations varying in size between 1and 10pm. Frequently.
the nucleus is missing in thee cells. Ocfasionally, the
entire cytoplasmic matter has condensed intogranuler,
making the remaining cytoplasm appear empty
("measle cells").
5. Condensation of filaments (Fig. 5) was Seen as cracklike fissura in the cytoplasm with faint stainability of
the cytoplasm.
DiogmsrirCyroparhdol). Vd3. N03.SrptmbrrIW7
251
c
SCHNEIDER ET AL.
'I
I
Noir
t
c
sips
nuck
firm
HP\
exce
prev
was
A
,.gI,
?V
nona
& 5. Condensed iilimcnu in four supr(kiil4iunrw dlr (x4üü).
-
@
6. Spindle cells (Fe.6) represent a more pronounced
form of dyskentocyta and are distinguished by a
regular chromatin pattern from similar cells found in
keratinirinp squamous carcinoma.
7. Nuclear hy&rchromatism consisted of increased
stainability in the absence of irregularities of the
chromatin and nuclear membrane, discriminating this
criterion from nuclear atypia.
8. The presence of bi- or multinucleation (Fin.
. - 4). is a
nonclassic sign.
9. The presence of perinuclear halos is a nonclassic sign.
i
-
c
*
Statistical analysis for each criterion was done in a
two-dimensional contingency table according to the Bonferroni inequality. A probability of error of Ins than a
half percent bad to k found in order to yield statistical
neg:
sign
chrr
plar
significance of 5%. A discriminant analysis was p r formed to establish a combination of the most valuable
cytomorphologic signs for @e identification of HPVinfected smears.
Will
acc
Results
O f a total sample of 1.052 patients, 100were positive for
H P V - D N A (9.5%). Thirty-eight smears were infected
with HPV-6/11.45 with HPV-16/18, and 17 with HPV6/i 1and HPV-16/18. The mean age of the patimu was
32 yr (range, 18-73). Contraceptive methods used and
vaginal flora were similarly distributed in the HPVpositive and HPV-negative groups.
The classic cytomorphologic signs of koilocytosis and/
or dyskeratocytosis could be identified in I S of 100
HPV-positive smears and three of 100 HPV-negative
samples (Table I).
For the nonclassic criteria, statistical analysis con-
nun
of
1
j
I
t
t
CERVICAL CONDYLOMA
T.uI I. Comtiim dcytomoiphdqicCriteria and HPV-DNA
Ik<rñminIWHPV.bi~kind
1WHPV-Nq.liwCuar
Hpv-perlriuCuII
HIV-nrWiVr w e 8
15
14
3
8
/%I
cludc utnr
Nond.ssic u p s
fa)
Nom: Clwis sips arc koibqia¡c and dy*enioycair: mnelauic
include mild kakc)lo*r. mild dy*mtocyt&
byprchronuiic
wki.ü-a multinnckiiim. and ckired cytoplasm.
firmed a significant melation between the detection of
HPV-DNA and each fytomorphologic sign with the
exception of spindle cells. This was due to the overall low
prevalence and not to nonspecificity of this criterion. as it
was not found in HPV-negative smears (Fig. 7). All other
nonclayic criteria could 81x1be observed in specimens
negative for HPV-DNA. The nonclassic cytomorphologic
signs of mild koilccytoris. mild dyrkeratocyWis. hyperchromatic nucki, bi-or multinudeation. and cleared cytoplum showed the most significant statistical correlation
with HPV infection. When these five signs were combined
according to their validity, discriminant analysis could
*.
7. Fnqucncy of variwr mwpholqicM
I M in& t h
identify 84% of the HPV-positive ases rnd 92%of the
HPV-negative smmn corre*ly (Table I).
The toul number of cytomoiphdogic signs in HPVDNA-positive smears compared with HPV-negative
smears is shown in Fig. 8, which indicates a significant
diñerence between the groups. In the HPV-positive
group. rmean of 4.8 cytomorphologiccriteria was found
in uch smear. whereas 2.0 signs were detected in the
HPV-negative group on the average. All smears with six
or more criteria were positive for HPV-DNA, whereas
about 905ó of the c a w with lers than t h r a criteria were
negative for HPV-DNA (Fig. 8). Due 10 the low number
of cases analyzed, statistical evaluation could not show
any correlation between the different HPV typcs, HPV6/11 or HPV-16/18. and a specific constellation of
cytopathic futures.
DiJruUioii
!Squamous epithelium forms the matrix for replication
and persistence of H W . The process of diñerentiation
into superficial epithelial cells is essential for the production of HPV particles.' The expression and distribution of
IO
the prcsu= of HW-DNA by sihcr bybr¡diz&sn in UIU.
Diapawir
I
Q m p a h d o g y . Vd 3. No 3. Septwabu 1987
253
d
r-
,/
-_'XHNEIDER ET AL.
r-
gnor
w e
v-
m.8. Number of cytcinorpholqic criteria in relition
O
I
the presence of HPV-DNA by ñltn hybridintion in situ.
i
-
cytoplasmic products are influenced by viral interaction
resulting in a recalled cytopathic virus effect. which may
L k &pecific for certain viruses as shown for cutaneous
HPV kaiohs! For genital H P V infections. koilocytosis
and dyskeratocytosis were considered to represent
L
, i pathogmmmic cytologic chang6 as HPV panicles or
Ppilioau vim antigen could k identified in thae cell
in addition, other criteria have k c n reported in
ibc literature: cytoplasmic granules,'+15 muitinucieation,'' perinuclear halo formation,"" and vacuolization
of tbe cytoplasm." On the basis of our HPV studies. we
P&fined these and additional cytologic criteria as nonclasa k signs and found a good correlation with the presence of
HPV when using the five most,valuable signs; this was in
contrast to a poor detection rate of 15% for the classic
c signs (Table I).
When comparing the detection rate of each nonclassic
dterion in the HPV-positive and HPV-negative group. a
d a r a i t diñerenee can k observed (Fig. 8). N o sign
r c mto
~ be pathognomonic for subclinical HPV infection;
all of them kside spindle cells are also found in HPVnegative casa (Fig. 8). This may partly k due to a
L.nkty of dirordcn causing rimilar cytopathic phenomena. Keratohyalin-like granules may represent a HPVp'
I y p specific protein product as shown for HPV-Ii, induced m y m i a s ? On the other hand, keratohyalin
c
-
..-
..-
I
.C^
2%
DiatmsiIc C y i o o ~ o wV. d J. No J. Srprmbrr lVd7
granules, which arc morphologically identical to kcrat*
hyalin-like granules, are found in any abnormal keratin¡ration p r o c w c a w d by nonvira1 agents, e.&. by mechan¡cal irritation. Aggregation of tonofilament bundles either
focally or in the peripheral cytoplasm is a common feature
" an identical dfcft can k
of HPV-infected E C I ~ ' . ~ ~but
evoked by several other infeaions.
It is conceivable that the specificity of these*noncl&hdc
signs is higher than may be concluded from the data of
this study. Applying a molc sensitive method for HPVDNA detection, such as Southern blot analysis, about
twice as many samples are positive for HPV-DNA compared with filter hybridization in situ." Thus, some of the
HPV-negative sman positive for classic (Table I) and
nonclassic cytologic criteria may be Icen as positive for
H P V - D N A using another hybridization method.
Four or more of the cytomorphologic sigm had to k
present in the HPV-positive group and more than w e n
criteria absent in the HPV-negative casa to yield a good
correlation with the mult of HPV-DNA detection (Fig.
8). This was confirmed by discriminant analysis using the
results of the five mort valuable signs to identify an
acceptable rate of cases correctly (Table I). W e conclude
from our data that ihc classic cytologic criteria, koilocytosir and dyskeratocytorir are not very sensitive signs for
the detection of cervical condyloma. Subclinical H P V
I
`