Cancer Treatment Reviews Spontaneous (1996) 22, 395-423 regression of cancer Rose J. Papac Section of Medical Oncology, Yale University School 333 Cedar Street, New Haven, CT 06520, U.S.A. of Medicine, introduction Spontaneous regression of cancer is one of the most fascinating phenomena observed in medicine. It is generally regarded as inexplicable, although there are now some laboratory studies of regressed or regressing tumours, as well as new theoretical possibilities about mechanisms. In this review, the historical background, clinical features and possible mechanisms are discussed. The clinical aspects of the malignancies most often reported to undergo spontaneous regression, namely renal cell carcinoma, neuroblastoma, carcinoma of the breast, malignant melanoma and leukaemias/lymphomas, are reviewed. The prevalent view regarding the mechanism for spontaneous regression is the involvement of immunological factors in the host. Other mechanisms include hormonal changes, tumour necrosis, trauma and changes in blood supply. Recent. reports suggest other mechanisms such as apoptosis and differentiation to a benign tumour. Decreased telomerase activity has been reported in neuroblastomas that regress, and hypomethylation of DNA in retinoblastoma is a possibility. A role for cytokines and/or growth factors is also discussed. The significance of spontaneous regression is the demonstration of endogenous control of neoplastic growth. Spontaneous regression of cancer is defined as the complete or partial disappearance of malignant tumour in the absence of therapy that is capable of inducing anti-neoplastic effects. Most patients ultimately relapse, so it is not usually associated with cure of the malignant disease. The existence of spontaneous regression is often questioned; indeed, review of the literature reveals reported cases which do not represent malignant disease, instances in which documentation of metastases is questionable and some reported cases in which therapy may have played a role. In 1966, Everson and Cole published a classic monograph on this topic which included 176 welldocumented cases of spontaneous regression of cancer published from 1900 until 1964, and listed criteria for the diagnosis (1). These are: documented histologic regression of biopsy-proven metastases; radiologic regression of 0305-7372/96/060395+29 0 1996 $12.0010 395 W.B. Saunders Company Ltd 396 R. J. PAPAC presumptive neoplastic disease; and regression of metastatic tumour following therapy generally deemed ineffective. The actual incidence of spontaneous regression is unknown but clinical trials which include an untreated cohort of cancer patients provide some data in several malignant diseases. There is the possibility that the phenomenon occurs more frequently than can be recognized clinically. Over a 30-year interval, the author has observed eight cases and is aware of two more. These have been reported briefly and are the stimulus for a review of this topic (2). Recent literature, as a result of the advances in the biology of cancer, provides greater insight into potential mechanisms of spontaneous regression. Historical background Boyd suggested that tumours that regressed spontaneously could be called Saint Peregrine tumours (3). Near the end of the 13th century, Saint Peregrine, a young priest, developed a large bone tumour requiring amputation. On the night preceding surgery, he prayed intensely and, it is alleged, awoke without any trace of the tumour. In 1345, aged 80 years, he died without recurrence of the tumour. Anecdotal cases of spontaneous regression have appeared in the medical literature since the early 19th century (4). The first case of complete regression of malignant melanoma was reported in 1899 (5). In 1901, William Osler published an article entitled ‘The medical aspects of carcinoma of the breast with a note on the spontaneous disappearance of secondary growths’ (6). The second of Osler’s cases was desperately ill, bed-ridden with metastatic involvement of the spine, right eye and opposite breast. Two years later when Osler returned from England, he noted ‘she drove a mile and a half to the station to meet me and drove me to the station on my return’. Osler commented that the phenomena ‘presented by these cases are among the most remarkable which we witness in the practice of medicine.. . and illustrate the uncertainty of prognosis and the truth of the statement that no condition, however desperate, is quite hopeless’. The first large series of cases of spontaneous regression of cancer was reported by Rohdenberg in 1918 (4). Of a total of 302 collected cases, only 70 were reports in which either temporary or permanent regression of tumour growth was subjected to rigid scrutiny; the others could have been diagnostic errors. The greatest number of regressions followed incomplete removal of the tumour or an acute febrile process. The spontaneous regression of neuroblastoma was recognized and reported by Cushing and Wollbach in 1927 (7). Subsequent reports by Dargeon, Bodian, Koop eta/. and Farber provided illustrative cases (8-l 1). Spontaneous regression of pulmonary metastases from renal cell carcinoma was first reported by Bumpus in 1928 (12). In 1966, Boyd and Everson and Cole published monographs on the spontaneous regression of cancer (1, 3). In Everson and Cole’s review of the published cases from 1900 until 1964, cases of leukaemia and lymphoma were excluded due to the variability in natural history of these diseases (1). The SPONTANEOUS Table 1. Frequency of site Miscellaneous Lymphoma Leukaemia Neuroblastoma Retinoblastoma Kaposi’s sarcoma Haemangioendothelior Hodgkin’s disease Lymphosarcoma Malignoma of sclera Sympathoblastoma Osteoblastoma Merkel’s cell tumour Primary of cancer in spontaneous (n=211) ma unknown regression, 397 1960-87 (n=564) (n=116) 68 22 16 4 2 2 1 (n=8:) 69 6 4 4 1 (n=lO) (n=34) Head and neck Adenoid cystic Larynx Pharynx Other squamous carcinomas 10 10 10 3 1 Endocrine Adrenal Thyroid (n=2) Brain Glioma Astrocytoma lntrasellar (n=4) tumour with OF CANCER Genitourinary/breast Hypernephroma Breast Testis Bladder Urethra Uterus Endometrium Ovary Skin Malignant melanoma Bowen’s disease (vulva) Basal cell carcinoma Bowen’s disease (penis) Epithelioma 68 53 41 33 7 2 2 Gastrointestinal Colon/rectum Stomach Liver Pancreas Intestine Reprinted REGRESSION permission from 1 1 Respiratory Lung Bronchial 2 1 1 Soft tissue and Chondrosarcoma Osteosarcoma Sarcoma Challis and Stam (n= carcinoma 13) 5 4 1 cell 3 (n=25) 18 7 bone (n=5) 1 3 1 (14). majority of cases of spontaneous regression occurred in fourtumour types; renal cell carcinoma, choriocarcinoma, neuroblastoma and malignant melanoma. Boyd reported 61 cases of spontaneous regression, but did not report instances of regression in lymphomas or leukaemias (3). The two most common tumour types reported were retinoblastoma and carcinoma of the breast; there were no cases of renal cell cancer nor choriocarcinoma. In 1974, the National Cancer Institute sponsored a conference on spontaneous regression of cancer which presented a broad perspective of the clinical occurrence and possible mechanisms (13). In addition to solid tumours, lymphomas and leukaemias were included in the analysis. In 1990, Challis and Stam published a review of cases from 1900 until 1987 (14). In comparing frequency of regression by site of cancer from 1900 to 1965 with the period of 1966-87, lymphoma is reported in the latter group as often as renal cell carcinoma (Table 1). Challis and Stam comment in particular on the paucity of discussion in the literature regarding mechanisms. In 1993, O’Regan and Hirshberg published an annotated bibliography of spontaneous regression which encompasses benign as well as malignant disorders (15). It is noteworthy that over half the cases were reported following the review of Everson and Cole (I). 398 Fi. J. PAPAC Table Number 2. Incidence of spontaneous of patients Number 1139 571 73 91 110 3 4 5 4 4 regression of regressions in selected (%) (0.3) (0.8) (7) (4.4) (3.6) series of renal cell carcinoma Reference Bloom (25) Snow and Schellhammer Oliver et a/. (27) Marcus et al. (23) de Riese et al. (19) (26) Clinical features Collectively, from the review of O’Regan and Hirshberg, the five most common tumours types to undergo spontaneous regression in order of number of reports are renal cell carcinoma, lymphoma and leukaemias, neuroblastoma, carcinoma of the breast and melanoma (15). The circumstances under which spontaneous regression occurs differs in various neoplastic diseases, so it is of interest to review the clinical aspects of spontaneous regression in these diseases. Renal cell carcinoma Renal cell carcinoma accounts for the largest number of patients with spontaneous regression, although the histologic confirmation of metastatic disease is in the range of 20-25% (16-18). In 1986, Kavoussi et al. reported that 20% of all cases of spontaneous regression of renal cell carcinoma had acceptable histologic confirmation (16). In recent reports, there is a trend towards better radiologic and histologic confirmation of metastatic disease. Since 1990, there have been six reports in the English literature, and of the 10 cases cited, nine met the criteria for spontaneous regression. Five of the nine cases had biopsies of metastases (17-23). Merz et a/. observed ‘pseudoregression’ of pulmonary lesions in a chest radiograph of a patient with renal cell carcinoma. Computerized tomography (CT) demonstrated disappearance of the mass, presumed to be an artifact (24). In general, the recently reported cases have more stringent standards for radiologic diagnosis of metastatic disease, namely CT, magnetic resonance imaging (MRI) and/or lung tomograms, as noted in a recent report from the National Cancer Institute of cases of untreated renal cell carcinoma patients who developed spontaneous regression prior to the institution of therapy (23). The incidence of spontaneous regression in renal cell carcinoma is low (Table 2) (19,23, 25, 27). DeRiese et al. reported four cases of spontaneous regression in 110 cases of metastatic renal cell carcinoma. One of their cases was treated with tamoxifen to complete disappearance of tumour so that the role of therapy is not entirely excluded; nevertheless, the patient remained disease-free for 5 years which is a most unusual outcome for tamoxifen treatment of metastatic renal cell carcinoma (19). Males predominate in reports of spontaneous regression of renal cell SPONTANEOUS REGRESSION OF CANCER 399 carcinoma by a ratio of 3:l. The duration of regressions has ranged from 3 months to 20 years. In over 90% of cases, pulmonary metastases are the site of spontaneous regression (28). There are four reports of well-documented regression of osseous lesions (22). Regression of other sites such as skin, liver, intestine and node1 areas has also occurred. Most frequently, spontaneous regression has followed nephrectomy, although regressions are reported to occur prior to a planned nephrectomy and in the absence of surgery for the primary tumour (1, 13, 14, 23, 28, 29). In his review of 67 reported cases of spontaneous regression of renal cell carcinoma (as well as two of his own cases), Fairbanks noted that 57 cases had had nephrectomy; five experienced regression prior to nephrectomy (28). Everson and Cole noted that 22 of 31 renal cell carcinoma cases experienced regression following nephrectomy (I). There are a few instances in which regression followed radiotherapy to the primary tumour or to a dominant tumour site such as a large groin mass (28, 30). Vascular embolization of the tumour is reported to be associated with spontaneous regression of metastatic disease, although it has been followed by nephrectomy, and the contribution of the embolization, as distinct from the nephrectomy, is unclear (31, 32). One of the cases reported by Marcus et al. experienced a subphrenic abscess in the postoperative period (23). There is a single instance of tumour regression following plasma infusion from a family member who had experienced a spontaneous regression (33). Post mortem examination in patients who have demonstrated spontaneous regression of pulmonary metastases often shows no evidence of residual tumour. In cases who have not had nephrectomy, a residual renal mass, described as fibrotic with necrosis or some degenerating cells resembling tumour, is reported (34). Malignant melanoma Cutaneous melanomas are reported to undergo spontaneous regression in up to 30% of cases (35-37). In thin lesions, the occurrence ranges from 29 to 67% of patients. In a recent review of 563 primary cutaneous melanomas, Blessing and McLaren observed regression in 46% of thin lesions (less than 1.5mm), 32% of intermediate (1.5-3 mm) lesions, and 9% of thick lesions, i.e. >3 mm (37). Regression was observed more frequently in superficial spreading lesions and those on the trunk or lower limb. From the Sydney melanoma unit, 28 patients with a thin primary lesion and concurrent regional node involvement showed regression in all 28 lesions (38). Patients with Stage I disease had over 60% incidence of regression. Of 103 patients with thin lesions followed at the University of Illinois, 29% showed evidence of partial regression (39). The clinical appearance of regression is suggested by the development of white, grey-white or pink-white areas which are areas of depigmentation (40). Besides vitiligo, other cutaneous changes associated with regression in malignant melanoma include inflammatory changes and dermatomyositis (41, 42). The histologic appearance of regression has been characterized as 400 R. .I. PAPAC degeneration of melanoma cells with a dense lymphohistiocytic infiltrate, pigment-laden macrophages and, in late stages, increased vascularity in the papillary and reticular dermis with an atrophic epidermis (37, 38, 43). Fibrosis in the papillary dermis is also described. The clinical significance of regression in a primary melanoma is uncertain, although some authors think it is associated with a poor outcome. Shaw et a/. observed recurrent melanoma in as many Stage I patients who had regression as those without evidence of regression (38). Sondergaard and Hou-Jensen reviewed 496 cases of Stage I primary cutaneous melanoma, and found that the IO-year survival was 95% for patients without regression, in contrast to 79% for patients who had experienced regression (36). Ronan et al. observed no metastases in a group of 73 patients without regression of a thin primary melanoma, while 20% of patients who developed regression of a thin primary melanoma developed fatal visceral metastatic disease in 3-48 months (39). Blessing and McLaren found that seven of 110 patients with thin melanomas exhibiting regression developed metastases; this was not statistically significant (37). Cooper and Wanebo found that thin melanomas of the extremity were biologically favourable and that foci of regression did not have any impact on outcome (43). A study by Paladugo and Yonemoto concluded that regressive changes indicated a propensity for metastatic disease in thin malignant melanomas; this analysis included only 11 lesions with regression as compared to 25 without regression (44). Multiple logistic regression analysis involving 1015 patients with primary cutaneous melanoma failed to identify regression as a significant prognostic factor (35). Melanomas may initially present as metastatic disease without an identifiable primary site (45, 46). In such cases, which constitute about 5-10% of all cases of metastatic melanoma, the assumption is made that the primary site has undergone complete regression. Nathanson has suggested that the criteria for spontaneous regression in such instances should be: (I) a clinical history of a pigmented lesion situated in an area drained by tumour-involved lymph nodes; (2) the absence of any other primary lesion; and (3) the presence of atypical pigmented or depigmented change in the skin at the site of the untreated possible primary lesion-histologic changes in the excised site of such a possible primary lesion which support the clinical evidence of regression of primary malignant melanoma; and (4) the absence, histologically, or a configuration or primary malignant melanoma with or without the presence of melanoma cells in atypical distribution in the dermis or subcutaneous tissue at the site of the lesion (40). In 1976, Nathanson reported 32 such cases in the medical literature and added one of his own experience (40). There have since been published more than twice the number of cases cited by Nathanson (44-49). Regression of metastatic lesions from malignant melanoma is uncommon; the majority of cases involve nodal or subcutaneous deposits (48, 49). From eight different series of patients which involved 4541 cases with metastatic melanoma, Nathanson found 0.22% had spontaneous regression, although in individual reports, it varied from 0.08 to 0.71% (40). Visceral deposits are rarely reported to undergo regression. When this develops, regression has most SPONTANEOUS REGRESSION OF CANCER 401 frequently been reported in pulmonary metastases; other sites include liver, brain, intestines and even osseous lesions (40, 50). Regression of ocular melanomas is rarely reported, but histological changes are similarto those described for cutaneous melanomas (51,521. The occurrence of regression of metastatic lesions from an ocular primary melanoma is about 10% (40). The majority of patients who developed regression of primary melanoma are men; in regression involving metastatic disease, there is an almost equal representation of the sexes (46, 47, 49). Nathanson suggested that about 40% of patients who had spontaneous regression of metastatic disease would be cured (40). This was based upon the proportion of patients free of disease or dead of other causes at 5 years follow-up. The phenomenon of late recurrence in malignant melanoma suggests that a better definition of cure would be death without disease. In the author’s experience, a dramatic instance of spontaneous regression of metastatic melanoma was of 8 years duration with relapse followed by death. Similar cases of spontaneous regression followed by relapse at intervals from 6 to 12 years following spontaneous regression are reported (47, 50). Few cases have had autopsy confirmation of disappearance of metastases. There have been several clinical settings in which regression of malignant melanoma have occurred (1,401. For the majority of cases, regression has been linked to a surgical procedure to the primary or a metastatic lesion. In some instances, infection in proximity to the tumour has been noted. Irradiation of a metastatic site remote from the regression lesion has been cited as a possible factor. Several instances of spontaneous regression developed following termination of pregnancy. Regression has been observed following blood transfusion, plasma infusion from a patient who had had a spontaneous regression, and following anti-rabies vaccine. Neuroblastoma Neuroblastoma is cited as a tumour with well-documented occurrence of spontaneous regression, although there is considerable debate regarding the actual incidence. Evans et a/. found an 8% incidence (53). Mass screening data from a Danish survey found 2% of cases of neuroblastoma had well-documented spontaneous regression (54). However, the incidence may well be higher since the survey found an increased incidence from 1943 until 1980with an unchanged mortality rate. Moreover, the incidental finding of neuroblastoma at autopsy in infants is much more frequent than the clinical detection (54). The patients likely to develop spontaneous regression of neuroblastoma are predominantly infants (53-55). In Evans et al.‘s review of this topic, 65% of the cases were less than 6 months of age when the diagnosis was established (53). The clinical presentation of patients who have developed spontaneous regression is mainly Stages II and IVS (53, 55, 56). Stage II patients often have a partial resection of the tumour which may be followed by regression of the residual disease. Stage IVS is of interest since this was described to identify patients who had disseminated disease but a favourable prognosis (57). Patients typically have involvement of the skin, liver and bone marrow, and are frequently 402 R. J. PAPAC under the age of 1 year. Recent reports suggest that Stage IVS is a more heterogenous disease than generally recognized (58). The site of the primary tumour does not seem to be a significant factor, although the size may be important since the regressions in Stage II seem associated with little residual tumour, and the IVS cases of regression have had small primary tumours. Regression rarely develops in patients with osseous metastases. Biologic features of the tumour have been examined to provide data which may assist in determining whether it is possible to define a type of neuroblastoma which will undergo spontaneous regression. N-myc amplification, expression of the TRK gene, DNA ploidy, the presence of chromosome Ip abnormalities, the numbers of argyrophilic nucleolar organizer regions, serum neuron-specific enolase and ferritin levels have had prognostic importance (59-62). There is not a consistent pattern of these biologic features that describes patients who experience spontaneous regression, although the lack of N-myc gene expression or amplification is considered favourable, as are high levels of expression of the TRK gene, a low ferritin and serum neuron-specific enolase, and a hyperdiploid karyotype. A consistent pattern of argyrophilic nucleolar organizer regions, determined by a silver colloid tissue staining procedure, is reported to suggest a prediliction for spontaneous regression. The histologic findings in regression of neuroblastoma include total disappearance of the tumour, a fibrotic scar on maturation to a benign ganglioneuroma, an observation described by Cushing and Wollbach (7). In the review by Everson and Cole of 29 cases, five showed maturation to ganglioneuroma (1). A subsequent case report described a patient who had subcutaneous nodules which were biopsied serially. At 4.5 months of age, neuroblastoma was present; biopsies at 16, 36 and 76 months of age showed ganglioneuroma and, in some areas, only fibrotic scar. At 3 years of age, there was a persistent left suprarenal mass which was removed, showing no viable tumour but central necrosis and calcification, demonstrating a spectrum of histologic findings in the same patient (63). There are reports in which waxing and waning of palpable tumour without therapy are documented with ultimate biopsy evidence of transformation to ganglioneuroma (53, 63). Survivals of patients who have had spontaneous regression vary but are generally of long duration (53, 54, 64). In a recent study of Stage IVS neuroblastoma, eight of 18 patients without life-threatening symptoms were followed until disease progression developed (60). Eight patients developed spontaneous regression without treatment and of these eight, one patient relapsed 10 months later and died despite therapy. The other seven cases were alive at follow-up ranging from 26 to 76 months. While relapse may develop following spontaneous regression, it is considered a rare development (63). Carcinoma of the breast In the anecdotal reports of spontaneous regression in the 19th century, breast cancer was cited (4). One involved a patient whose breast cancer showed change relative to the onset of menses, and another patient whose tumour decreased at the menopause. In the review of Everson and Cole, there were SPONTANEOUS Table Number 3. Incidence of spontaneous of patients 140 31 83 Number 18 (12.9) 5 (16.3) 19 (23) REGRESSION OF CANCER regression in low-grade of regressions (%) 403 lymphomas Reference Gattiker et al. (70) Mead et al. (7 1) Horning et al. (72) only six cases of breast carcinoma (1). Lewison observed 12 cases of partial or temporary regressions of breast cancer over a 25-year period (65). In 1982, Ross et al. reported follow-up of a striking instance of spontaneous regression of breast carcinoma, and reviewed all reports published in English since 1900 (66). Only 13 met the criteria of Everson and Cole (1). By 1987, Challis and Stam found a total of 22 reported cases (14). The spontaneous regression is most frequently related to onset of menses or to menopause, and in some cases to development of ovarian metastases (1, 66). Regression of lung metastases occurred in one case following mastectomy (67). It is of interest that Ross et al. noted that 11 of the 13 cases were pre-menopausal when the spontaneous remission occurred (66). Spontaneous regression has involved primary breast tumours which were untreated and observed to regress, metastatic lung, subcutaneous and osseous lesions, malignant pleural effusions and ascites (I, 6, 65-68). Regression involving an untreated primary breast carcinoma has been reported for a duration of 17 years (68). In most cases, spontaneous regression of breast cancer is followed by relapse of the disease (65). Haematoiogic malignancies The haematologic malignancies, although cited by Rohdenberg, were not included in several subsequent compilations of cases of spontaneous regression (I, 3, 4). In 1976, Wiernik tabulated the cases of lymphoma, leukaemia and plasma cell dyscrasias, and presented an analysis of clinical features (69). In the 1990 review of Challis and Stam, lymphomas were as frequently reported as cases of renal cell carcinoma (I 4). There were 53 cases of leukaemia, a number exceeding that of neuroblastoma and breast carcinoma. Other haematologic malignancies, namely multiple myeloma, Waldenstrom’s macroglobulinemia and myelodysplastic syndromes, are reported to undergo spontaneous remission, but the majority of cases of remissions of haematologic malignancy are lymphoma and leukaemia cases (69). Malignant lymphoma. Since 1980, three series of low-grade lymphomas have documented the incidence of spontaneous regression in these lymphomas (Table 3) (70-72). Gattiker eta/. observed 18 of 140 patients who had regressions associated with either no treatment or no treatment for 4 or more years in association with stable disease that then regressed (70). Of these, 4.7% were complete regressions. In a randomized trial of low-grade lymphomas in which patients with no treatment were compared to those with therapy, spontaneous regression was noted by Mead et al. in five of 31 patients who were untreated; 404 R. J. PAPAC two had complete remissions (71). Horning and Rosenberg reported that 23% of low-grade lymphomas in a Stanford series developed spontaneous regression (72). In the low-grade lymphomas, untreated cases have been observed to demonstrate fluctuations in nodal size and/or organomegaly. The duration of spontaneous regression reported by Mead et al. ranged from 13 to 36+ months; in the retrospective review of Gattiker et al., the median duration of regression in 11 patients who had either complete or partial regression was 1.2 years with a range of 2 months to 12 years (70, 71). Of the five patients who had complete regressions, all but one relapsed. The survival of patients who experienced regressions was significantly longer than that of patients without such regression. In the Stanford experience, the median duration of regression varied from 4 to 72+ months with median duration from 6 to 15 months, depending upon the histology (72). The sample size in the follicular, mixed and small lymphocytic types is very small (three cases in each group). It is noteworthy that only six of the 19 patients with regression in this series showed subsequent disease progression, and of these, only three required therapy. The disease sites which demonstrated spontaneous regression in low-grade lymphomas include both peripheral and visceral nodal masses, skin lesions, liver, spleen, pleural effusion and ascites (70-72). In two cases, there was disappearance of peripheral blood and bone marrow lymphocytosis. Histologic features of cases who had spontaneous regression were assessed in five patients by Strickler et al. and compared to six patients with the same type of lymphoma who had progressive disease (73). In patients with spontaneous regression, there were significantly more T-helper cells in the host infiltrate than in the control patients. Otherwise there were no significant differences in cytotoxic/suppressor T-cells, macrophages, or other lymphoid populations determined by immunophenotyping. In high-grade lymphomas, the occurrence of spontaneous regression is reported less frequently (70). There are a number of isolated case reports of large cell diffuse lymphoma that are quite typical of the disease, namely involvement of peripheral and visceral nodal areas with hepatosplenomegaly and an aggressive clinical course (74-76). Spontaneous regression has generally developed in the setting of a febrile illness. In one patient, remission developed twice upon treatment with vitamin C in high dosage with a survival of 17 years from diagnosis (77). The occurrence of spontaneous regression in high-grade lymphomas is noted in several histologic types and also in specific disease sites. From a study of 41 patients with Ki-1 positive large cell lymphoma, three adults develped spontaneous regression of 9,27 and 120 months duration (78). In two instances, regression involved cutaneous lesions; in the third patient, nodal disease regressed. Several case reports of spontaneous regression of this lymphoma have also been reported; one in a paediatric patient and another in a patient with AIDS (79-81). Burkitt’s lymphoma has been reported to undergo spontaneous regression (82-84). Burkitt and Kyalwazi observed four cases; two were complete regressions and two were partial (82). One of the latter received medication from a witch doctor. One of Burkitt’s cases was included in a review by Ziegler SPONTANEOUS REGRESSION OF CANCER 405 who added two additional cases (83). One remission was noted to be of 8 years duration, and two cases experienced a 1 year remission. One remission followed measles rash and measles vaccine. In a recent case report of Burkitt’s lymphoma, remission developed in association with herpes zoster infection (84). Other instances of regression of high-grade lymphoma include a small noncleaved cell, non-Burkitt’s lymphoma involving tonsil and cervical nodes in a 12-year-old boy following biopsy with a 3-year follow-up (85). Similarly, regression of diffuse large cell lymphoma involving the tonsillar region and left neck occurred in a 54-year-old woman prior to the institution of treatment (74). This regression persisted for up to 4 years on follow-up. Cases of gastric lymphoma, large cell type, are reported in which endoscopic biopsy revealed the diagnosis, and subsequent gastric resection showed no evidence of tumour (86). The only therapy was the administration of the H-2 histamine antagonist cimetidine, suggesting that the eradication of Helicobacter pylori was not a factor in these regressions. Several reports of spontaneous regression of intra-cerebral lymphoma include instances of CT scan evidence of change in size of masses that ultimately progressed and, on biopsy or at autopsy, were found to be lymphoma (87, 88). The regressions are generally of l-4 months duration and partial in extent. Whether these represent true regression or changes due to associated necrosis, infarction or haemorrhage is unclear. These cases were not in AIDS-related lymphomas. There are, however, several cases of regression of lymphomas occurring in AIDS patients (80, 89-91). In these cases, infections were present at the time of spontaneous regression. Other instances of regression of highgrade lymphoma have included angiocentric cutaneous T-cell lymphoma (characterized by immunophenotyping) with a complete regression after therapy was declined (92). Adult T-cell leukaemia/lymphoma is another aggressive disease that has shown spontaneous regression (93-95). In the cases cited, the diagnosis has been well established by immunophenotyping, and in two instances by study of HTLV-1 proviral DNA. In a series from Japan of 82 patients, three (3.7%) experienced spontaneous regression of 3, 12 and 72 months duration (94). An interesting feature of this report was the gene analysis of HTLV-1 proviral DNA and T-cell receptor re-arrangement studies. In one patient, the monoclonal integrated band of HTLV-1 proviral DNA and re-arranged band of TCT-P gene were at the same positions before regression and after recurrence; in the other patient, they were at different positions before regression and after recurrence. The development of spontaneous regression in lymphoma is sometimes of long duration and has most frequently followed a febrile illness, sometimes a viral infection, and occasionally has developed after biopsy (69). Leukaemia. Spontaneous remission of both chronic and acute leukaemias has been described (69). Reports prior to the past three decades often do not meet either the standards for diagnosis or for remission of the diseases. Nevertheless, there appear to be some common clinical findings in the occurrence of remissions in these diseases. In chronic lymphocytic leukaemia, the incidence of spontaneous regression is about 1% (96-98). Wiernik observed that one-third of the cases were associated 406 R. J. PAPAC with a new primary carcinoma (69). In more recent reports, this has not been noted. A more frequent association has been viral infection antedating spontaneous regression (97, 98). The development of spontaneous regression in chronic lymphocytic leukaemia has been characterized as a gradual clinical improvement in adenopathy, hepatosplenomegaly and peripheral blood findings (96). The documentation of bone marrow remission was not present in all cases despite normalization of other features, so that a partial remission ensued. In some instances, clonal remission developed from 9 months to 1.6 years following the clinical remission (96, 98). T-cell subsets also became normal in some but not all patients (99). The duration of reported remissions ranges from 1 to over 11 years. In chronic myelogenous leukaemia (CML), there are several reports of spontaneous remission (100-104). One case report is clearly consistent with a spontaneous remission, since the patient was untreated and developed cytogenetic regression with a marked decrease in the number of Philadelphia chromosome positive metaphases (from 100 to 37%) over a g-year follow-up period, and also demonstrated complete disappearance of an associated trisomy 8 abnormality (103). In two cases, reported as spontaneous regressions, disappearance of the Philadelphia chromosome followed therapy with busulphan in one case and with hydroxyurea in the other patient, neither having been treated to marrow hypoplasia (101, 104). This is a most unusual outcome with these treatments; hence, it is possible that these cases could represent spontaneous regression but the effect of treatment cannot be discounted. Cases of acute leukaemia reported by Southam et al. in 1951 to have spontaneous regression were generally preceded by bacterial infection (105). Collected from the literature were 57 cases, not all of whom had bone marrow documentation. Clinical observation included the occurrence of remissions in patients who were leucopenic as well as those who had severe pyogenic infections. In some instances, the transfusion of blood products was implicated in the genesis of spontaneous remission. The remissions were generally brief, from 4 weeks to 4 months. In childhood acute lymphoblastic leukaemia, Diamond and Luhby reviewed 300 cases and found spontaneous remission in 4.5% of patients, all of whom had preceding cytopenias and severe infection (106). Remissions of infantile acute leukaemia are described although there is difficulty in establishing the diagnosis of leukaemia in infants (107). Demonstration of an abnormal karyotype or immunophenotype has been useful. There are reports of infantile acute leukaemia with a leukaemic blood picture, no therapy, spontaneous remission and later relapse (108, 109). There are very few cases of spontaneous remission in acute lymphoblastic leukaemia in adults (69, 105). One report includes a 57-year-old man and another a 22-year-old woman who developed remission without cytotoxic treatment. Remissions without therapy are described in acute T-cell leukaemia/ lymphoma (93-95). One patient with prolymphocytic leukaemia who experienced spontaneous remission has been reported (110). In adult acute myeloid leukaemia, Wiernik found five cases who entered remission without specific therapy (69). Common features were severe bacterial SPONTANEOUS REGRESSION OF CANCER 407 infection and brief duration of remission, the longest being 5 months. Following this, there have been additional reports of at least 12 cases of spontaneous regression in acute myeloid leukaemia, generally brief in duration, the median being 4 months (111-116). Interestingly, three of these cases had acute promyelocytic leukaemia (112, 113, 115). Cytogenetic studies were done in eight cases; in three, there were no abnormalities (114-116). Two cases had documented cytogenetic as well as haematologic remission. In all but four cases, the initial course was complicated by severe pyogenic infection. In one patient, acute promyelocytic leukaemia was diagnosed at 32 weeks of pregnancy (115). The patient underwent Caesarean section and, within hours, her platelet count which had been 30,00Ocu/mm rose to 121,000cu/mm and remained normal. Bone marrow examination showed complete remission 12 days later. There are four reported instances of remission without relapse of acute leukaemia (2, 11 I). One of these cases bears a remarkable similarity to a patient whom the present author has followed. Enck reported a 28-year-old man admitted with a diagnosis of acute promyelocytic leukaemia, clinical evidence of bleeding and without haematologic improvement after 6 weeks of therapy (112). He was admitted febrile in a near terminal state, anti-leukaemia therapy was discontinued, he was treated for infection and improved dramatically with normalization of the bone marrow. He was alive and well 8 years later. The patient from the present author’s study, a 52-year-old male, was admitted febrile without documented infection, cytopenic with 90% myeloid blasts in the marrow, and received prednisone and 6-mercaptopurine. He was initially treated with 50 mg qid, but inadvertently this was changed to 50 mg qd after 2 weeks. He entered complete remission within 10 days and has not relapsed in 24 years (2). In these leukaemia patients, the dose and duration of treatment were unlikely to result in sustained remission of 8 and 24 years. Spontaneous remission seems likely, although the possibility exists that the anti-leukaemic agents exerted effectiveness through some anomalous mechanism. Miscellaneous tumours Choriocarcinoma, seldom reported in the past few decades, was one of the most frequently reported tumours to undergo spontaneous regression in the review of Everson and Cole (1). With the advent of curative therapy for this tumour type, it is unlikely that any patient would remain untreated. Spontaneous regression is of interest from several points of view. The tumour is regarded as a grafted neoplasm in the host since it arises from placental tissue; regression may have an immunological basis (117). The very high incidence of total regression of the primary tumour in the presence of metastatic disease is remarkable. Boyd’s review, as well as subsequent reports, suggests that retinoblastoma is among the tumours most frequently noted to develop spontaneous regression (3). Most often this has occurred with bilateral tumours, following enucleation of one eye (118). It is noteworthy that Everson and Cole’s review listed relatively few of R. J. PAPAC 408 Table 4. Mechanisms suggested basis for spontaneous of cancer Immunological mechanisms Hormonal mechanisms Elimination of a carcinogen Induction of differentiation Tumour necrosis Angiogenesis inhibition Psychologic mechanisms Oncogenes, growth factors, Apoptosis Epigenetic mechanisms as the regression cytokines the commonly occurring tumours, such as lung, breast, prostate and gastrointestinal tumours, among those reported to develop spontaneous regression (1). In fact, the frequently reported tumours to regress spontaneously were relatively uncommon tumours. In the review by Challis and Lamb, there are increased numbers of some of the common tumour types, mainly lung and gastrointestinal (14). However, there is still a disproportionate occurrence of reports of spontaneous regressions in rare tumours such as Merkel cell tumour, hepatocellular carcinoma, non-AIDS-related Kaposi’s sarcoma and adrenal cortical carcinoma (118-126). Almost every type of cancer has been reported to undergo spontaneous regression (14, 15). Mechanisms. Most reports of spontaneous regression describe the occurrence but do not provide a discussion regarding possible causative mechanisms. In the review of Everson and Cole, and in subsequent articles, Cole referred to immunological mechanisms as the most important factor in spontaneous regression (I, 127, 128). This is a generally prevalent view, although there is now an expanded list of potential mechanisms (Table 4). Other causative factors cited by Everson and Cole include operative trauma, hormones and the elimination of a carcinogen (I). In their tabulation of cases, Challis and Stam found that subsequent reports listed maturation and differentiation of tumours, necrosis and psychological factors as mechanisms of spontaneous regression (14). Stoll, in a discussion of new insights into spontaneous regression, suggested that natural factors and chemical agents could induce regression (118). The specific natural factors cited were oncogenes, growth factors and cytokines. The chemical agents mentioned by Stall include methotrexate in choriocarcinoma, and compounds capable of inducing differentiation such as retinoids, interferon, and some dietary components. In view of current knowledge of the biology of neoplasia, additional mechanisms proposed as the basis for spontaneous regression include epigenetic changes and induction of apoptosis. While most discussions of causative factors in spontaneous regression are speculative, recent literature includes some laboratory studies of regressed lesions as well as comparison of cellular infiltrates from regressed and proliferative lesions (73, 129). A discussion of the suggested mechanisms and supportive data follows. SPONTANEOUS Immunological REGRESSION OF CANCER 409 mechanisms Immunologic factors almost certainly play a role in some instances of spontaneous regression of tumours (118,130). The increased incidence of some tumours in immunosuppressed individuals, and regression following reduction of immunosuppressive agents, suggests a role for immunologic factors (131). Examples of this phenomenon are the occurrence of lymphoma in transplant patients and in AIDS cases, the regression of Kaposi’s sarcoma when corticosteroid therapy was withdrawn, and the reversibility of lymphomas developing with methotrexate therapy for rheumatoid arthritis when methotrexate was stopped (118, 124, 132). In the observed regressions of renal cell carcinoma and malignant melanoma, immune mechanisms are often invoked as the mechanism of regression (40, 118). In both tumours, regressions have occurred following plasma infusion from patients who have experienced a regression, suggesting that humoral factors may play a role (33, 127). Also, in both tumour types, cytokines, namely interferon and interleukin 2 (IL-21, exert antitumour effects (133, 134). IL-2 is presumed to activate T-lymphocytes, natural killer (NK) cells, lymphocyteactivated killer cells (LAK) and tumour infiltrating lymphocyte (TIL) cells as a mechanism of action; interferons are capable of multiple immunomodulatory effects involving monocytes, macrophages and B-cells, as well as induction of IL-2 receptors (135, 136). In a patient with renal cell carcinoma, in whom biopsy-proven lung metastases showed complete resolution without treatment, assays of LAK activity, and mixed lymphocyte cytotoxicity, were obtained (18). These immune parameters were within the normal range. While specific immunologic factors are not yet identified in the regression of renal cell carcinoma, the effects of nephrectomy are attributed to immune mechanisms, assuming that removal of bulk tumour enables or stimulates the body’s immune system to control residual disease (118, 127). In melanoma there is evidence for immune-mediated host responses in spontaneous regression. Tefany et al. reported immunocytochemical comparisons between regressing and non-regressing melanomas (129). These indicate increased helper T-cells with increased expression of the IL-2 receptor infiltrating the regressing primary melanoma without changes in T-suppressor cells. A number of leucocyte antigens have been assessed including antigenpresenting cells and leukocyte adhesion molecules, but no difference was observed in regressing as compared to non-regressing lesions. Melanoma tumour associated antigens, such as GD3, glycoprotein 75 antigen, glycoprotein 110 and GD2, were expressed at a higher percentage in regressing lesions, but not to a significant level. Mackensen eta/. characterized lymphocytes of a regressive melanoma lesion, demonstrating amplification of cytotoxic T-lymphocytes that were reactive against the autologous tumour (137). From the lymphocytes infiltrating the regressive melanoma lesion, a series of lymphocyte clones were analysed for the T-cell receptor gene expression. A unique cell expressing a VP 13.1/J6 1.1 gene segment was found expressed in the cloned cells. When compared to the uncultured tumour tissue, the specific cytotoxic T-lymphocytes were selected and amplified in vivo at the lesion site. 410 R. J. PAPAC In melanoma patients with regional nodal metastases, several studies have shown suppressor T-cell activity against induction of cytotoxic T-cells in lymph nodes containing metastases (38, 138). It is suggested that migration of helper T-cells from these lymph nodes may boost cytotoxic T-cell responses at the primary regressing melanoma site. These studies in melanoma are consistent with a cell-mediated antitumour effect and suggest that an effector T-cell has contributed to the regression. The effector cell may be a cytokine secreting cell. Cytokines per se have not been cited in the spontaneous regression of melanoma, although there is evidence that interleukin 6 (IL-6) may function as an antagonist or agonist in melanoma growth (139). The expression of IL-8 by melanoma cells is also demonstrated to influence the biologic behaviour of melanoma (140). The role of immunologic factors in regressions observed in basal cell carcinoma has been reported recently (141). lmmunocytochemical comparison of proliferative and regressive lesions was carried out. The regressive lesions showed increased numbers of helper T-cells, increased expression of the IL-2 receptor and no change in T-suppressor cells. These results are similar to those reported in cutaneous melanomas with regression. Experimental evidence for an immune mechanism in the spontaneous regression of neuroblastoma is suggested by Bolande, who reported maternal pregnancy serum possessed an IgM natural antibody binding to neuroblastoma cells and capable of amplifying the response to the lytic action of serum complement (142). Arguments against the immunologic mechanism are the observation that neuroblastoma is not observed in immunodeficient states, and the fact that the role of immune mechanisms is not well established in childhood malignancies (143). In the spontaneous regression of choriocarcinoma, immune mechanisms are considered to be significant since the neoplasm may be regarded as a grafted tumour whose regression may represent graft rejection (117). There are four reported cases of small cell lung carcinoma in whom spontaneous regression occurred in association with neuropathy, which was associated in three of the four cases with antineuronal antibodies (144, 145). Two cases had anti-Hu and atypical antineuronal antibody. One case had only anti-Hu antibody. The autoantibodies reactive with neurons are also reactive with small cell lung carcinoma, suggesting that the regression is possibly antibody or cell mediated. The spontaneous regression in cases in lymphoma and leukaemia is often attributed to immune mechanisms (69, 118). Characteristics of the cellular infiltrate in a regressed lymphoma are similar to those noted in regressed cutaneous tumours (73, 129). Since the occurrence of regressions in these conditions is often in the setting of febrile illnesses, bacterial or viral, cytokines associated with host responses to infections could mediate the regressions. Coley’s toxin which contained bacterial endotoxin was associated with occasional instances of tumour regression (146). It is possible that tumour necrosis factor, which is cytotoxic for some tumour cells, is an important cytokine for instances of fever-related regressions (147). The regression-of low-grade gastric mucosa-associated lymphoid tumour (MALT) lymphomas with eradication of Helicobacter pylori suggests that these SPONTANEOUS REGRESSION OF CANCER 411 lymphomas represent a response to an antigenic stimulus from the H. pylori (148). It is conceivable that other lymphomas might be antigen-driven, and regress following dimunition in antigen exposure or the development of an anti body. Hormonal mechanisms Hormonal mechanisms are most likely the mediators of spontaneous regression in breast carcinoma (1, 65, 66). The evidence is circumstantial since the regressions have developed mainly at the time of menopause, but also at the onset of menses or with the development of ovarian metastases, presumed to induce an endogenous oophorectomy. There are several cases of complete remissions of malignant melanoma reported during pregnancy and following delivery (I, 40). Some melanoma cells express hormonal receptors which led to some clinical therapeutic trials of hormonal therapy which were generally disappointing (149-151). There are three case reports of remission of acute leukaemia following termination of pregnancy (15,152,153). The mechanism of regression is unclear but it is suggested that the leukaemic cells may have expressed hormonal receptors, so that with hormonal alterations at the termination of pregnancy, the leukaemia regressed (118). Oestrogen receptors have been found in some human myeloid leukaemia cell lines, and oestrogen, with colony stimulating factor, has been demonstrated to have stimulatory effects on growth of some leukaemia cell lines (154). Five cases of regression of metastases from ovarian carcinoma following oophorectomy are cited by Everson and Cole (I). There are also rare instances of regression of metastases of sarcomas with termination of pregnancy (I). Elimination of a carcinogen The elimination of a carcinogen is suggested in instances of bladder cancer when transplantation of the ureters resulted in regression of the bladder tumour (1). It has been postulated that a carcinogen present in the urine was eliminated. To date, the carcinogen has not been identified. However, in instances of pre-invasive lesions such as those noted in the bronchial tree of heavy smokers, cessation of smoking has led to the disappearance of pre-invasive changes (I 18). Induction of differentiation Differentiation, a mechanism by which malignant cells develop a non-malignant phenotype, is postuled to occur in several types of neoplastic disease. Retinoblastoma, neuroblastoma, choriocarcinoma, embryonal cell carcinoma of the testis and leukaemia are malignancies in which differentiation is possibly a major factor in spontaneous regression (118, 155). The development of ganglioblastoma, a benign tumour, is noted in a minority of cases of regressed neuroblastoma (7,53). In tissue cultures of neuroblastoma cells, differentiation can be induced with exposure to nerve growth factor, 412 R. J. PAPAC serum-free media, cyclic nucleotides, retinoic acid and some cytotoxic agents (53, 118). In embryonal carcinoma and teratocarcinoma of the testis, differentiation to a mature teratoma has generally developed following treatment with chemotherapy, radiotherapy or both (156). A case report of a patient with pulmonary metastases who refused therapy and developed spontaneous regression describes the finding of a mature teratoma in the testicular site of the original primary tumour (157). Spontaneous regression of primary testicular tumours is presumed in cases with metastatic disease in whom the primary site cannot be identified and intra-abdominal metastases are evident. Cultured human teratocarcinoma cells provide in vitro demonstration of extensive capability to differentiate into several cellular lineages (158). In choriocarcinoma, Friedman and Skehan reported morphological differentiation of human choriocarcinoma cells in culture by methotrexate (159). The mechanism of the differentiating effect is felt related to possible depletion of several different folate co-factors or differential sensitivity of varied target sites. It is possible that differentiation by depletion of folate co-factors could occur in instances of spontaneous regression. In the acute leukaemias, laboratory studies support induction of differentiation as a mechanism of spontaneous regression. A wide variety of compounds, including vitamins, steroid hormones and cytotoxic agents in low dosage, demonstrate the capability of inducing differentiation in vitro (160). The effects of trans-retinoic acid in acute promyelocytic leukaemia are an example of in viva differentiation (155). Laboratory studies indicate that the suppression of the leukaemic phenotype with differentiation does not restore all the normal controls, a finding that may be the basis for the generally brief duration of spontaneous remissions in acute leukaemia (160). The potential for spontaneous regression by differentiation of lymphoma cells is suggested by a case report of a patient with follicular lymphoma who developed a febrile illness with high IL-6 serum levels, following autologous bone marrow transplant and IL-3 therapy (161). The patient developed plasmacytosis and paraproteinaemia. Immunophenotyping, cytogenetics and Southern blotting indicated a clonal relationship of the plasma cells to the original lymphoma. This was interpreted as evidence that the malignant lymphoma cells were stimulated by IL-3 and IL-6 to become terminally differentiated plasma cells. Turnout- necrosis and angiogenesis inhibition Compromise of the blood supply to a tumour or necrosis of tumour tissue is listed as a mechanism for spontaneous regression. The regressions following surgery or irradiation to a tumour site could be related to this mechanism (14). The survival of tumours is dependent not only on an adequate blood supply but on the production of new vessels. Angiogenesis as an essential component of the neoplastic process is a focused area of research (162). A number of cytokines are known to inhibit this process, namely TNF CI and transforming growth factor fi which could play a role in spontaneous regressions (163). SPONTANEOUS Psychologic REGRESSION OF CANCER 413 mechanisms Psychological mechanisms in spontaneous regression are generally regarded with sceptisism since the phenomenon lends itself to misuse by faith healers. Psychological methods are reported by several authors but corroborating studies are lacking (164-166). While psychological interventions benefit many cancer patients, these techniques do not induce regression of cancer in reproducible fashion. Psycho-neuro-immunological mechanisms are suggested by some authors on the basis of interactions between immune, endocrine and nervous systems (165, 167). The finding of opioid receptors on macrophages and lymphocytes suggests that psychological factors could influence the immune system and modify cancer growth through the cell-mediated reactions involving macrophages and T-lymphocytes (167, 168). While there may be some links in these processes, it is now a remote and highly theoretical construct. Oncogenes and growth factors In human neoplasia, the expression of oncogenes, the inactivation, mutation or loss of suppressor genes and the function of growth factors are regarded as central to the neoplastic state (169-172). The expression of certain genes promotes cell survival which is enhanced by trophic growth factors. Genes for suppression of malignancy, as well as inhibitory growth factors, have roles in regulation of cell growth and differentiation (171, 172). Study of chronic myeloid leukaemia (CML) cells in vitro indicates that the balance of stimulatory and inhibitory factors is shifted, as compared to normal haematopoietic cells (173). In normal haematopoiesis, two endogenous inhibitors, transforming growth factor p (TGF /I) and macrophage inflammatory protein-la (MIPa), maintain primitive cells in a reversible state. In cells of CML, the primitive cells have a selective unresponsiveness to MlPcl such that the fusion gene BCL-ABL is affecting only some of the pathways involved in the complex signalling for cell cycle control of haematopoiesis. The demonstration of retention of some control provides further emphasis for the concept of existence of mechanisms for reversal of the malignant phenotype. While there are no specific examples to document the role of inhibitory growth factors in spontaneous regression nor for change in oncogene expression, these possibilities have been cited (169, 170). In CML, expressing BCL-ABL augments sensitivity of haematopoietic cells to growth-factor mediated signals of differentiation, suggesting that CML progenitors could be eliminated in response to haemopoietic growth factors (174). Additionally, as noted earlier, remission in acute leukaemias has been associated at times with viral infections. Recent studies indicate that in association with viral infection, interferon CI exerts myelosuppressive effects indirectly mediated by inhibition of the paracrine production of haemopoietic growth factors (175). An indirect effect of growth factors in spontaneous regression was suggested in a case report of spontaneous regression in metastatic lung cancer following myxedema coma (176). The authors note that the hypothyroid state has been associated with reduced rates of neoplastic growth. The thyroid hormone (T3) is a potent 414 R. J. PAPAC stimulator of growth factors, notably epidermal growth factor (frequently expressed in lung cancer), nerve growth factor and insulin-like growth factor. Presumably, severe deficiency of the hormone by altering autocrine growth factors could influence programmed cell death, apoptosis. Apoprosis A mechanism which has received attention for spontaneous regression in recent literature is the occurrence of apoptosis or programmed cell death (177). In tumouregenesis, the protein expressed by the BCL-2 gene, as well as other proteins and environmental factors, inhibits programmed cell death (177, 178). A developmental time switch for activation of apoptosis has been proposed by Pritchard and Hickman as the basis for spontaneous regression in neuroblastoma (143). Biological variation in time of apoptosis could explain the regression. This mechanism is amenable to investigation by measurements of DNA fragmentation and in situ end labelling. In regression of renal cell carcinoma, Pansera has suggested that apoptosis occurs as a re-expression of embryonic cell death, a normal occurrence in embryogenesis of the kidney (179). During kidney development, the pronephros and mesonephros regress completely so that these regressive phenomena are instances of cell death during development. Re-expression of cell death in renal cell carcinoma may be a manifestation of embryonic characteristics, as sometimes seen in neoplasia. The role of apoptosis in spontaneous remission of a mantle zone lymphoma is described in a recent report by Kaufman et al. (180). An untreated patient experienced 15 cycles of remission and disease activity. At stages of the disease when remission was present, analysis of the peripheral blood lymphocytes showed a small population of lymphocytes, IgM and CD5+. During relapse, two B-cell clones with differently re-arranged heavy chain genes were detected. Spontaneous apoptosis was detected by DNA fragmentation of cultured cells obtained during relapse of the disease. Viability of cells from the acute disease phase was low with 50% of cells non-viable in 16 h as compared to 90% of cells viable at 3 days during remission. In this case, no molecular alterations were detected in BCL-2 gene nor ~53, a tumour suppressor gene that functions as an inducer of apoptosis and which can be a regulator of BCL-2. The authors suggest that the neoplastic cells, in response to loss of a T-cell-mediated survival signal or cytokine-mediated regulation, underwent apoptosis followed by clinical remission. Epigeneric mechanisms Epigenetic change may play a role in the spontaneous regression of some retinoblastoma tumours (181). Abnormalities of DNA methylation are associated with the incipient stages of tumour development. The abnormalities include overall genomic hypomethylation, regional hypermethylation and increased capability for hypermethylation (182). Regional hypermethylation may involve CpG islands which are cytosine-guanine rich regions of the genome found in the promoter area of many genes, normally free of methylation, except for the SPONTANEOUS REGRESSION OF CANCER 415 genes involved in X-inactivation or gene imprinting (183). Greger et a/., using methylation-sensitive restriction enzymes and a cloned DNA probe for the unmethylated CpG island at the 5’ end of the retinoblastoma gene (a tumour suppressor gene), obtained evidence of hypermethylation of this gene in a sporadic retinoblastoma tumour (181). There were, however, two sites of the CpG island not methylated, possibly reflecting demethylation. Since the maintenance methylase does not replicate methylation patterns during DNA synthesis with absolute fidelity, it is surmised that loss of methylation could lead to re-activation of the retinoblastoma gene and suppression of tumour growth. There is no evidence for inactivation of the retinoblastoma gene by hypermethylation, but there is a tentative suggestion that loss of hypermethylation may be involved in the spontaneous regression of some retinoblastomas. Telomerase, a ribonucleoprotein polymerase, functions to maintain the essential genetic elements of telomeres, the eukaryotic ends of chromosomes (184). With successive replications, telomere shortening develops in senescent aging with suppression of telomerase activity (185). Activation of telomerase, however, is regarded as essential to cell immortalization and most cancers. In a survey of 12 tumour tissue types, 90 of 101 were positive for telomerase activity (186). Some adjacent tissues also demonstrated positivity, suggesting involvement not evident by pathologic examination. A recent study of telomerase activity in neuroblastoma is of particular interest. Yokoyama et a/. studied 79 untreated cases and found that 16 cases with high telomerase activity had a poor prognosis, and 60 patients with low levels had a favourable outcome (187). Those patients whose tumours did not show telomerase activity underwent spontaneous regression, suggesting repression of telomerase activity as a possible mechanism for regression. Discussion Observations of spontaneous regression of cancer are, as described by Osler, one of the most remarkable phenomena observed in medicine (6). There is hope that elucidating the mechanism could lead to replication of the process and a better method of treating cancer. With regard to the occurrence of spontaneous regression, the observation of waxing and waning of tumour growth, particularly in neuroblastoma and lymphomas, suggests that it may not be an uncommon phenomenon in some tumours. Indeed, the reports of a 23% incidence in low-grade lymphomas supports this (72). The rather disproportionate occurrence of spontaneous regression in uncommon tumours suggests that these tumours may be uncommon because regression develops frequently, perhaps when these tumours are clinically unrecognized. A corollary of this consideration is that the commonly occurring tumours are those with much less capability of spontaneous regression; hence their frequent occurrence. From review of the clinical features of spontaneous regression, there seems to be a prediliction for regression to develop frequently in some sites, notably 416 R. J. PAPAC lung and in cutaneous sites, whether the tumour is a primary skin neoplasm or a secondary cutaneous deposit, as in lymphoma and neuroblastoma. The site predilection in regression of tumours suggests that micro-environmental factors play a role in the regression. While there are studies indicative of involvement of immunologic mechanisms in the process in cutaneous tumours, growth factors are probably significant as well. In melanoma growth, IL-6 clearly plays a role; it has not been assessed in regressive lesions. Treatment of the primary tumour by surgery or radiation has been associated with regression in three of the four most common tumour types reported to undergo spontaneous regression. The mechanism by which regression of metastatic lesions disappear following removal of a primary tumour is unclear. This is generally attributed to ‘host factors’ considered to be immune-mediated (1,141. Treatment of the primary tumour, however, is not essential for regression to develop, suggesting that reduction in cell numbers or tumour mass is not crucial to the process of spontaneous regression. It is possible that cytokines and/or autocrine growth factors, produced by the primary tumour and promoting proliferation and metastases, could be depleted or reduced with removal of the primary neoplasm with consequent re-establishment of growth controls and spontaneous regression of metastatic deposits. Ultimately in cases of spontaneous regression, the tumour is no longer evident, suggesting that either apoptosis or differentiation to a benign neoplasm occurs. Apoptosis is well documented in the case report of a mantle cell lymphoma demonstrating cycles of spontaneous regression; Pritchard and Hickman suggest apoptosis as a possible mechanism in the spontaneous regression of neuroblastoma (143, 180). Differentiation to a benign neoplasm is documented to occur in patients with neuroblastoma and testicular germ cell tumours; it certainly seems a possible mechanism in acute leukaemias. An intriguing issue in spontaneous regression is whether a common mechanism is essential forthe processto occur orwhether multiple mechanisms lead to spontaneous regression. Immunologic factors and hormonal changes are clinically significant in the occurrence of spontaneous remissions, and would seem either directly or indirectly to promote the process. Dietary factors, cytokines and/or growth factors conceivably could play a role in differentiation. The discovery of endogenous inhibitors of angiogenesis offers yet another possible contributing factor in the process of spontaneous regression. The epigenetic phenomena, demethylation and telomerase inactivation or inhibition are of particular interest since these could be determined in regressing tumours. Endeavours to study spontaneous regression with histologic or immunologic analyses is a comparatively recent development. The most comprehensive studies are in cutaneous melanoma and in lymphoma. The studies in retinoblastoma and neuroblastoma are of exceptional interest, and provide an impetus for further work to verify and confirm the data. A single mechanism for spontaneous regression remains a possibility, however. Variability in expression of the malignant phenotype may be assumed in some settings. An example is noted in the patients who have prolonged remissions of malignant disease, followed by relapse, observed particularly in patients with carcinoma arising in breast, kidney, occasionally melanoma as well as lymphomas. Molecular biology provides evidence of occult disease in SPONTANEOUS REGRESSION OF CANCER 417 instances in which relapse or clinically evident disease does not develop. Suppression of the malignant phenotype may occur as a consequence of modifiers intrinsic to the host (genetic, immunologic, angiogenesis inhibitors, etc.) or external agents (differentiating agents). Reversibility of the malignant process in spontaneous regression could, therefore, evolve from several initiating events-but the ultimate reestablishment of controls could involve a single crucial process. Perhaps, as suggested by Thomas, ‘In the end, when all the basic facts are in . . . it will turn out that all forms of cancer, in whatever organs and of whatever cell types, are a single disease, caused by a single central controlling mechanism gone wrong’ (188). The significance of spontaneous regression of cancer is that it provides evidence for the existence of endogenous control of neoplastic growth. Perhaps efforts to define and apply methods of re-establishing or strengthening growth control could yield greater benefits in the quest for effective methods of cancer treatment. Acknowledgements The interest and encouragement of Dr. Mary Keohane-Pelliccia, C. Hirshberg and Dr. J. J. Fischer in the preparation of this manuscript is deeply appreciated. References 1. Everson, T. C. & Cole, W. H. (1966) Spontaneous Regression of Cancer: Philadelphia, PA: W. B. Saunders. 2. Papac, R. J. (1990) Spontaneous regression of cancer. Corm. Med. 54: 179-182. 3. Boyd, W. (1966) The Spontaneous Regression of Cancer. Springfield, IL: Charles C. Thomas. 4. Rohdenburg, G. L. (1918) Fluctuations in the growth of malignant tumors in man, with especial reference to spontaneous regression. J. Cancer Res. 3: 192-221. 5. Bennett, W. H. (1899) Some peculiarities in the behavior of certain malignant and innocent growths. Lancet 1: 3-7. 6. Osler, W. (1901) The medical aspects of carcinoma of the breast, with a note on the spontaneous disappearance of secondary growths. Am. Med. 17-19: 63-66. 7. Cushina. H. & Wollbach. S. B. (1927) The transformation of a malianant oaravertebral sympat-hicoblastoma into’s benign ganglioneuroma. Am. J. fatho/. 3: 203-216.’ 8. Bodian, M. (1959) Neuroblastoma. Pediatr. C/in. North Am. 6: 449-472. 9. Dargeon, H. W. (1962) Neuroblastoma. J. f’ediatr. 61: 456-471. 10. Koop C. E., Kiesewetter, W. B. & Horn, A. C. (1955) Neuroblastoma in childhood. Surg. 38: 272-278. 11. Farber, S. (1940) Neuroblastoma. Amr. J. Dis. Child 60: 749-50. 12. Bumpus, H. C. (1928) The apparent disappearance of pulmonary metastasis in a case of hypernephroma following nephrectomy. J. Ural. 20: 185-191. 13. Natl. Cancer Inst. Monogr. Spontaneous regression of cancer. 1976; 44. 14. Challis, G. B. & Stam, H. J. (1990) The spontaneous regression of cancer. A review of cases from 1900 to 1987. Acta Oncol. 29: 545-549. 15. O’Regan, B. & Hirschberg, C. (1993) Spontaneous Remission. An Annotated Bibliography. Sausalito, CA: Institute of Noetic Sciences. 16. Kavoussi, L. R., Levine, S. R., Kadmon, D. et a/. (1986) Regression of metastatic renal cell carcinoma: A case report and literature review. J. Ural. 135: 1005-1007. 17. Vozelgang, N. J., Priest, E. R. & Borden, L. (1992) Spontaneous regression of histologically proved pulmonary metastases from renal cell carcinoma: a case with a 5-year followup. J. Ural. 148: 1127-1128. 418 R. J. PAPAC 18. Abukar, Y. A., Chou, T. H. & Redman, B. G. (1994) Spontaneous remission of renal cell carcinoma: a case report and immunological correlates. J. Ural. 152: 156-157. 19. de Riese, W., Goldenberg, K., Allhoff, E. et a/. (1991) Metastatic renal cell carcinoma (RCC): Spontaneous regression, long-term survival and late recurrence. Int Ural. Nephrol. 23: 13-25. 20. Palmer, M. A., Viswanath, S. & Desmond, A. D. (1993) Spontaneous regression of metastatic renal cell carcinoma. J. R. Sot. Med. 86: 113-114. 21. Kallmeyer, J. C. & Dittrich, 0. C. (1992) Spontaneous regression of metastases in a case of bilateral renal cell carcinoma. J. Ural. 148: 138-140. 22. Kerbl, K. & Pauer, W. (1993) Spontaneous regression of osseous metastasis in renal cell carcinoma. Aust. N.Z. J. Surg 63: 901-903. 23. Marcus, S. G., Choyke, P. L., Reiter, R., et a/. (1993) Regression of metastatic renal cell carcinoma after cytoreductive nephrectomy. J. Ural. 150: 463-466. 24. Merz, V. W., Looser, C., Kraft, R. et al. (1993) Pseudoregression of pulmonary metastasis after nephrectomy for renal cell carcinoma. BI: J. Ural. 71: 751-753. 25. Bloom, H. J. G. (1973) Regression of renal cancer. Cancer32: 1066-1071. 26. Snow, R. M. & Schellhammer, P. F. (1982) Spontaneous regression of metastatic renal carcinoma. Urology 20: 177-181. 27. Oliver, R. T. D., Nethersell, A. B. W. & Bottomley, J. M. (1989) Unexplained spontaneous regression and alpha interferon as treatment for metastatic renal cell carcinoma. Br. J. Ural. 163:28-31. 28. Fairlamb, D. J. (1981) Spontaneous regression of metastases of renal cell cancer. Cancer 47: 2102-2106. 29. Chapple, C. R., Gannon, M. X., Shah, V. M. eta/. (1987) Spontaneous regression of pulmonary metastases from renal adenocarcinoma before nephrectomy. Br. J. Surg. 74: 69-70. 30. Ridings, G. R. (1971) Renal adenocarcinoma: Regression of pulmonary metastases following irradiation of primary tumor. Cancer 27: 936-938. 31. Mohr, S. J. & Whitesel, J. A. (1979) Spontaneous regression of renal cell carcinoma metastases after preoperative embolization of primary tumor and subsequent nephrectomy. Urology 14: 5-8. 32. Swanson, D. A., Johnson, D. E., von Eschenbach, A. C. et a/. (1983) Angioinfarction plus nephrectomy for metastatic renal cell carcinoma-an update. J. Ural. 130: 449-452. 33. Horn, L. & Horn, H. (1971) An immunologic approach to the treatment of cancer? Lancet 2: 466-469. 34. Goodwin, W. E., Mims, M. M., Kaufman, J. J. et al. (1967) Under what circumstances does “regression” of hypernephroma occur? In: King, Jr., G. S., ed., Renal Neoplasia. Boston, MA: Little, Brown & Co., pp. 13-39. 35. Trau, H., Kopf, A. W., Rigel, D. S. et a/. (1983) Regression in malignant melanoma: J Am Acad Dermatol8: 363-368. 36. Sondergaard K. & Hou-Jensen, K. (1985) Partial regression in thin primary cutaneous malignant melanomas clinical stage I. Virchows Arch. 408: 241-247. 37. Blessing, K. & McLaren, K. M. (1991) Histological regression in primary cutaneous melanoma: recognition, prevalence and significance. Histopathology20: 315-32. 38. Shaw, H. M., McCarthy, S. W., McCarthy, W. H. et al. (1989) Thin regressing malignant melanoma: significance of concurrent regional lymph node metastases. Histopathology 15: 257-265. 39. Ronan, S. G., Eng, A. M., Briele, H. A. et a/. (1987) Thin malignant melanomas with regression and metastases. Arch. Dermatol. 23: 1326-1330. 40. Nathanson, L. (1976) Spontaneous regression of malignant melanoma: A review of the literature on incidence, clinical features, and possible mechanisms. Nat/. Cancer Inst. Monogr. 44: 67-76. 41. Wagner, R. F. & Nathanson, L. (1986) Paraneoplastic syndromes, tumor markers and other unusual features of malignant melanoma. J. Am. Acad. Dermatol. 4: 249-256. 42. Sais, G., Marcoval, J. Juggla, A. et a/. (1994) Dermatomyositis and metastatic malignant melanoma, with complete regression of the primary lesion. Br. J. Dermatol. 130: 796-797. 43. Cooper, P. H. & Wanebo, H. J. (1985) Regression in thin malignant melanoma. Microscopic diagnosis and prognostic importance. Arch. Dermafol. 21: 1127-1131. 44. Paladugu, R. R. & Yonemoto, R. H. (1983) Biologic behavior of thin malignant melanomas with regressive changes. Arch. Surg. 118: 41-44. 45. Panagopoulos, E. & Murray, D. (1983) Metastatic malignant melanoma of unknown primary origin: A study of 30 cases. J. Surg. One. 23: 8-10. 46. Kessler, E., Schwartz, P. & Antebi, E. (1984) Spontaneous regression of primary malignant melanoma with metastases. flast. Reconstr. Surg. 74: 427-429. 47. Sroujieh, A. (1988) Spontaneous regression of intestinal malignant melanoma from an occult primary site. Cancer 62: 1247-1250. SPONTANEOUS 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. REGRESSION OF CANCER 419 Avril, M. F., Charpentier, P., Margulis, A. et al. (1992) Regression of primary melanoma with metastases. Cancer69: 1377-1381. Shai, A., Avinoach, I. & Sagi, A. (1994) Metastatic malignant melanoma with spontaneous and complete regression of the primary lesion: Case report and review of the literature. Dermatol. Surg. Oncol. 10: 342-345. Mikhail, G. R. & Gorsulowsky, D. C. (1986) Spontaneous regression of metastatic malignant melanoma. Dermatol. Surg. Oncol. 12: 497-500. Jensen, 0. A. &Anderson, S. R. (1974) Spontaneous regression of a malignant melanoma of the choroid. Acta Ophthalmol. 52: 173-182. Lambert, S. R., Char, D. H., Howes Jr. E. et a/. (1986) Spontaneous regression of a choroidal melanoma. Arch. Opthalmol. 104: 732-734. Evans, A. E., Gerson, J. & Schnaufer, I- (1976) Spontaneous remission of neuroblastoma. Nat/. Cancer Inst. Monogr. 44: 49-54. Carlsen, N. L. T. (1990) How frequent is spontaneous remission of neuroblastomas? Implications for screening. Br. J. Cancer 61: 441-446. Carlsen, N. L. T. (1992) Neuroblastoma: Epidemiology and pattern of regression. Am. J. Pediatr. Hematol. Oncol. 14: 103-110. Kemshead, J. T., Patel, K. & Phimister, B(1992) Neuroblastoma in the very young child: biological considerations. Br. J. Cancer 66 Suppl. XVIII): S102-105. Evans, A. E., D’Angio, G. J. & Randolph, J. (1971) A proposed staging for children with neuroblastoma. Children’s Cancer Study Group A. Cancer 27: 374-378. Wilson, P. C. G., Coppes, M. J., Solh, H. eta/. (1991) Neuroblastoma Stage IV-S: A heterogeneous disease. Med. Pediatl: Oncol. 19: 467-442. Nagkagawara, A., Sasazuki, T., Akiyama, H. et a/. (1990) N-myc oncogene and Stage IV-S neuroblastoma. Cancer 65: 1960-1967. Brodeur, G. M. & Nakagawara, A. (1992) Molecular basis of clinical heterogeneity in neuroblastoma. Am. J. PediatK Hematol. Oncol. 14: 111-116. Nakagawara, A., Arima-Nakagawara, M., Scavarda, N. J. et a/. (1993) Association between high levels of expression of the TRK gene and favorable outcome in human neuroblastoma. N. Eng. J. Med. 326: 847-854. Shimotake, T., Iwai, N., Tokiwa, K. et a/. (1994) Increased numbers of argyrophilic nucleolar organizer regions between primary and metastatic sites predict tumor progression in Stage IV and IIV-S neuroblastoma. Cancer 73: 3103-3107. Haas, D., Ablin, A. R., Miller, C. et a/. (1988) Complete pathologic maturation and regression of Stage IVS neuroblastoma without treatment. Cancer 62: 818-825. Suarez, A., Hartmann, O., Vassal, G. et a/. (1991) Treatment of Stage IV-S neuroblastoma: A study of 34 cases treated between 1982 and 1987. Med. Pediak Oncol. 19: 473-477. Lewison, E. F. (1976) Spontaneous regression of breast cancer. Nat/. Cancer Inst. Monogr. 44: 23-26. Ross, M. B., Buzdar, A. U., Hortobagyi, G. et a/. (1982) Spontaneous regression of breast carcinoma: Follow-up report and literature review. J. Surg. Oncol. 19: 22-24. Dao, T. L. (1962) Regression of pulmonary metastases of a breast cancer. Arch. Surg. 84: 574-577. Smithers, D. W. (1952) Cancer of the breast and menopause. C/in. Radio/. 4: 89-96. Wiernik, P. H. (1976) Spontaneous regression of hematologic cancers. Nat/. Cancer Inst. Monog,: 44: 35-38. Gattiker, H. H., Wiltshaw, E. & Galton, D. A. G. (1980) Spontaneous regression in non-Hodgkin’s lymphoma. Cancer 45: 2627-2632. Mead, G. M., Macbeth, F. R., Ryall, H. et a/. (1984) A report on a prospective trial of no initial therapy in patients with asymptomatic favourable prognosis non-Hodgkin’s lymphoma. Hematol. Oncol. 2: 179-188. Horning, S. 81 Rosenberg, S. A. (1984) The natural history of initially untreated low-grade nonHodgkin’s lymphoma. N. Eng. J. Med. 311: 1471-1475. Strickler, J. G., Copenhaver, C. M., Rojas, V. A. et a/. (1988) Comparison of “host infiltrates” in patients with follicular lymphoma with and without spontaneous regression. Am. J. C/in. Pathol. 90: 257-261. Grem, J. L., Hafez, G. R., Brandenburg, J. H. et a/. (1986) Spontaneous remission in diffuse large cell lymphoma. Cancer 57: 2042-2044. Markman, M. (1983) Chronic pneumococcal infection complicating progressive lymphoma. JAMA 249: 352-353. Tamura J., Jinbo, T., Take, H. et a/. (1993) Spontaneous regression in B cell, diffuse large cell type non-Hodgkin’s lymphoma. Acta Haematol. 90: 46-47. Campbell, A., Jack, T. & Cameron, E. (1991) Reticulum cell sarcoma: Two complete 420 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. R. J. PAPAC “spontaneous” regressions in response to high-dose ascorbic acid therapy. Oncology 48: 495-497. Chott, A., Kaserer, K., Augustin, I. et a/. (1990) Ki-1 positive large cell lymphoma. A clinicopathologic study of 41 cases. Am. J. Surg. Path. 14: 439-448. Bittencourt, A. L., Mendonca, N., de Freitas, L. A. et al. (1992) Ki-1 large cell lymphoma with regressing lesions in a child. Pediatr. Dermatol. 9: 117-122. Diekman, M. J., Bresser, P., Noordun, L. A. eta/. (1992) Spontaneous regression of Ki-1 positive T cell non-Hodgkin’s lymphoma in a patient with HIV infection. Br. J. Haematol. 82: 477-478. Motley, R. J., Jasani, B., Ford, A. M. eta/. (1992) Regressing atypical histiocytosis, a regressing cutaneous phase of Ki-1 positive anaplastic large cell lymphoma. Immunocytochemical, nucleic acid and cytogenetic studies of a new case in view of current opinion. Cancer 70: 476-483. Burkitt, D. P. & Kyalwazi, S. K. (1967) Spontaneous remission of African lymphoma. Br. J. Surg. 21: 14-16. Ziegler, J. L. (1976) Spontaneous remission in Burkitt’s lymphoma. Nat/. Cancer lnst Monogr: 44: 61-65. McClain, K., Warkentin, P. & Kay, N. (1985) Spontaneous remission of Burkitt’s lymphoma associated with herpes zoster infection. Am. J. fediarr. Hemato/./Onco/. 7: 9-14. Poppema, S., Postma, L., Brinker, M. et a/. (1988) Spontaneous regression of a small noncleaved malignant lymphoma (Non-Burkitt’s lymphoblastic lymphoma). Cancer 82: 791-794. Strauchen, J. A., Mdran, C., Goldsmith, M. et a/. (1987) Spontaneous regression of gastric lymphoma. Cancer 60: 1872-1875. Sugita, Y., Shigemori, M., Yuge, T. et a/. (1988) Spontaneous regression of primary malignant intracranial lymphoma. Surg. Neural. 30: 148-152. Weingarten, K. L., Zimmerman, R. D. & Leeds, N. E. (1983) Spontaneous regression of intracerebral lymphoma. Radiology 49: 721-724. Daniels. D.. Lowdell. C. P. & Glaser. M. G. (1992) The soontaneous rearession of lvmohoma , . in AIDS’. C&I Oncoli 3: 196-197. Terriff, B. A., Harrison, P. & Holden, J. K. (1992) Apparent spontaneous regression of AIDSrelated CNS lymphoma mimicking resolving toxoplasmosis. J. Acquit. immune Defic. Syndr: 5: 953-954. Karnad, A. B., Jaffar, A. & Lands, R. H. (1992) Spontaneous regression of acquired immune deficiency syndrome related, high grade extranodal non-Hodgkin’s lymphoma. Cancer 69: 1856-1857. Mayou, S. C., Anstey, A., Not-ton, A. J. et al. (1991) Angiocentric T-cell lymphoma in skin. J R Sot Med 84: 683-684. Schnitzer, B., Lovett, E. J. & Kahn, L. E. (1983) Adult T-cell leukaemia with spontaneous remission. Lancet 2: 1030. Shimamoto, Y., Kikuchi, M., Funai, N. et al. (1993) Spontaneous regression in adult T-cell leukemia/lymphoma. Cancer 72: 735-740. Taniauchi. S.. Yamasaki. K.. Shibuva. T. et a/. (1993) Soontaneous remission of acute adult Tcell kukaemia with chr~m&omal’abnormality infiltrating to skin and liver. BI: J. Haematol. 85: 413-414. Han, T. & Sokal, J. (1971) Spontaneous remission of leukaemic lymphoproliferative disease. Cancer 27: 586-585. Ribera, J. M., Vinolals, N., Urbano-lspizua, A. et a/. (1987) “Spontaneous” complete remissions in chronic lymphocytic leukemia: report of three cases and review of the literature. Blood Cells 12: 471-479. Han, T. (1987) Spontaneous remission in chronic lymphocytic leukemia: An update. Blood Cells 12: 481-483. Buchi, G., Termine, G., Zappala, C. et a/. Spontaneous complete remission of CLL: Report of a case studied with monoclonal antibodies. Acta Haematol. 70: 198-201. Singer, C. R. J., McDonald, G. A. & Douglas, A. S. (1984) Twenty-five year survival of chronic granulocytic leukaemia with spontaneous karyotype conversion. Br. J. Haematol. 57: 309-313. Wodzinski, M. A., Potter, A. M. & Lawrence, A. C. K. (1989) Prolonged survival in chronic granulocytic leukaemia associated with loss of the Philadelphia chromosome. /3r. J. Haematol. 71: 296-297. Provan, A. B., Majer, R. V., Herbert, A., et a/. Spontaneous remission of chronic myeloid leukaemia with loss of the Philadelphia chromosome. 6c J. Haematol. 78: 578-579. Smadja, S. N., Krulik, M., Audebett, A. A. et a/. (1986) Spontaneous regression of cytogenetic and hematologic anomalies in Ph 1 positive chronic myelogenous leukaemia. Br. J. Haematol. 83: 257-262. Fegan, C., Morgan, G. & Whittaker, J. A. (1989) Spontaneous remission in a patient with chronic myeloid leukaemia. Br. J. Haematol. 72: 594-595. Southam, L. M., Craver, L. F., Dargeon, H. W. et a/. (1951) A study of natural history of acute SPONTANEOUS 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 125. 126. 127. 128. 129. 130. 131. 132. 133. 134. 135. REGRESSION OF CANCER 421 leukemia with special reference to the duration of the disease and the occurrence of remissions. Cancer 4: 39-49. Diamond, L. K. & Luhby, A. L. (1951) Pattern of “spontaneous” remissions in leukemia of childhood observed in 26 of 300 cases. Am. J. Med. 10: 238-239. Penchansky, L., Wollman, M. R., Gartner, J. C. et a/. (1993) Spontaneous remission of infantile acute nonlymphocytic leukemia for 11 years in a child with normal karyotype. Cancer 71: 1928-1930. Lampkin, B. C., Peipon, J. J., Price, K. K. et a/. (1985) Spontaneous remission of a presumed congenital acute nonlymphoblastic leukemia (ANLL) in a karyotypically normal neonate. Am. J. PediatK Hematol. Oncol. 7: 346-351. Van Eys, J. & Flexner, J. M. (1969) Transient spontaneous remission in a case of untreated congenital leukemia. Am. J. Dis: Child 118: 507-514. Blecher, T. E. (1986) ‘Spontaneous’complete remission in a case of prolymphocytic leukaemia. Br. J. Haematol. 163: 395-398. Paul, R., Remes, K., Lakkala, T. eta/. (1994) Spontaneous remission in acute myeloid leukaemia. Br. J. Haematol. 86: 210-212. Enck, R. E. (1985) Spontaneous complete remission in acute promyelocytic leukemia. N.Y. State J. Med. 85: 662. Raza, A., Gill, L. M., Rakowski, I. et a/. (1985) Spontaneous remission in acute nonlymphocytic leukemia. N.Y: State J. of Med. 85: 269-270. Ifrah, N., James, J. M., Viguie, F. et a/. (1985) Spontaneous remission in adult acute leukemia. Cancer 56: 1187-I 190. de Mayolo, J. A., Ahn, Y. S., Temple, D. eta/. (1989) Spontaneous remission of acute leukemia after the termination of pregnancy. Cancer 63: 1621-1623. Jehn, U. W. & Mempel, M. A. (1986) Spontaneous remission of acute myeloid leukemia. A report of a case and brief review of the literature. Blut 52: 165-168. Hertz, R. (1976) Spontaneous regression in choriocarcinoma and related gestational trophoblastic neoplasms. Nat/. Cancer Inst. Monogr. 44: 59-66. Stall, 6. A. (1992) Spontaneous regression of cancer: new insights. Biotherapy 4: 23-30. O’Rourke, M. G. E. & Bell, J. R. (1986) Merkel cell tumor with spontaneous regression. J. Dermatol. Surg. Oncol. 12: 994-996. Duncan, W. C. & Tschen, J. A. (1993) Spontaneous regression of Merkel cell (neuroendocrine) carcinoma of the skin. J. Am. Acad. Dermatol. 29: 653-654. Kayashima, K., Ono, T., Johno, M. et a/. (1991) Spontaneous regression in Merkel cell (neuroendocrine) carcinoma of the skin. Arch. Dermatol. 127: 550-553. Ayres, R. C., Boertson, D. A., Dewbury, K. C. et a/. (1990) Spontaneous regression of hepatocellular carcinoma. Gut 31: 722-772. Chien, R. N., Chen, R. J. & Liaw, Y. F. (1992) Spontaneous regression of hepatocellular carcinoma. Am. J. Gastroenterol. 87: 903-905. Tebbe, B., Mayer da Silva, A., Garbe, C. et a/. (1991) Genetically determined coincidence of Kaposi’s sarcoma and psoriasis in an HIV negative patient after prednisolone treatment. Spontaneous regression 8 months after discontinuing therapy. Int. J. Dermatol. 30: 114-120. Blayney, D. W., Ito, J. I. &Jensen, F. C. (1986) Spontaneous remission of Kaposi’s sarcoma in an HTLV-III negative homosexual man. Cancer 58: 1583-1584. Gilchrest, N. L., Espiner, E. A., Donald, R. A. et a/. (1983) Disappearance of‘a pulmonary metastasis after removal of an adrenal carcinoma. N.Z. Med. J. 96: 469-470. Cole, W. H. (1974) Spontaneous regression of cancer. The metabolic triumph of the host? Ann. N.Y Acad. Sci. 230: 111-141. Cole, W. H. (1981) Efforts to explain spontaneous regression of cancer. J. Surg. Oncol. 17: 201-209. Tefany, F. J., Barneston, R., Sta, C. et a/. (1991) lmmunocytochemical analysis of the cellular infiltrate in primary regressing and non-regressing malignant melanoma. J. invest, Dermatol. 97: 197-202. Baker, H. W. (1986) Biologic control of cancer. Arch. Surg. 121: 1237-1241. Penn, I. (1990) Cancers complicating organ transplantation. N. Eng. J. Med. 323: 1767-1769. Kamel, 0. W., van de Rijn, M., Weiss, L. M. et a/. (1993) Reversible lymphomas associated with Epstein-Barr virus occurring during methotrexate therapy for rheumatoid arthritis and dermatomyositis. N. Eng. J. Med. 28: 1317-1321. Krown, S. E. (1986) lnterferons and interferon inducers in cancer treatment. Semin. Oncol. 13: 207-217. Rosenberg, S. A. (1986) The adoptive immunotherapy of cancer using the transfer of activated lymphoid cells and interleukin-2. Semin. Oncol. 13: 200-206. Hawkins, M. J. (1993) Interleukin-2 antitumor and effector cell response. Semin. Oncol. 20 (SuppI. 9): 52-59. 422 136. 137. 138. 139. 140. 141. 142. 143. 144. 145. 146. 147. 148. 149. 150. 151. 152. 153. 154. 155. 156. 157. 158. 159. 160. 161. 162. 163. 164. 165. R. J. PAPAC Kirkwood, J. M. & Ernstoff, M. S. (1984) lnterferons in the treatment of human cancer. J. C/in. Oncol. 2: 336-345. Mackensen, A., Ferradini, L., Caracelain, G. et al. (1993) Evidence for in situ amplification of cytotoxic T-lymphocytes with antitumor activity in a human regressive melanoma. Cancer Res. 53: 3569-3573. Hoon, D. B., Bowker, R. J. & Cochran, A. J. (1987) Suppressor cell activity in melanomadraining lymph nodes. Cancer Res. 47: 1529-33. Lu, C., Vickers, M. F. & Kerel, FL S. (1992) Interleukin-6: a fibroblast-derived growth inhibitor of human melanoma cells from early but not advanced stages of tumour progression. froc. Nat/. Acad. Sci. U.S.A. 89: 9215-9219. Gutman, M., Singh, R. K., Zie, K. et a/. (1995) Regulation of interleukin-8 expression in human melanoma cells by the organ environment. Cancer Res. 55: 2470-2475. Hunt, M. J., Halliday, G. M., Weedon, D. et a/. (1994) Regression in basal cell carcinoma: an immunohistochemical analysis. Br. J. Dermatol. ,130: 1-8. Bolande, R. P. The spontaneous regression, of neuroblastoma: Experimental evidence for a natural host immunity. Pathol. Annu. 1: 187-199. Pritchard, J. & Hickman, J. A. Why does stage 4s neuroblastoma regress spontaneously? Lancet 44: 23-26. Darnell, R. B. & DeAngelis, L. M. (1993) Regression of small-cell lung carcinoma in patients with paraneoplastic neuronal antibodies. Lancet 341: 21-22. Zaheer, W., Friedland, M. L., Cooper, E. B. et al. (1993) Spontaneous regression of small cell carcinoma of lung associated with severe neuropathy. Cancer Invest. 1: 306-309. Nauts, H. C., Swift, W. E. & Coley, B. L. (1946) The treatment of malignant tumors by bacterial toxins, as developed by the late William B. Coley, MD, reviewed in the light of modern research. Cancer Res. 6: 205216. Balkwill Fr., Naylor, M. S. & Malik, S. (1990) Tumor necrosis factor as an anticancer agent. Eur. J. Cancer 26: 641-644. Hussell, T. W., Isaacson, P. H., Crabtree, J. E. et al. 1993) The response of cells from low-grade gastric lymphomas of mucosa-associated lymphoid tissue to Helicobacter pylori. Lancet 342: 571-574. Fisher, R. I., Neifeld, J. P. & Lippman, M. E. (1976) Oestrogen receptors in human malignant melanoma. Lancet 2: 337-339. Walker, M. J. Role of hormones and growth factors in melanomas. Semin. Oncol. 15: 512-523. Papac, R. J. (1983) Kirkwood, J. (1983) High dose tamoxifen in malignant melanoma. Cancer Treat. Rep 67: 1049. Birge, R. F., Jenks, A. L. & Davis, S. K. (1949) Spontaneous remission in acute leukemia: Report of a case complicated by eclampsia. JAMA 140: 589-592. Cantini, E. & James, B. (1984) Acute myelogenous leukemia in pregnancy. South Med. J. 77: 1050-1052. Taetle, R. & Guittarad, J. P. (1982) Modulation of leukaemia blast colony growth by steroid hormones. Br. J. Haematol. 50: 247-255. Castaigne, S., Chomienne, C., Daniel, M. T. et al. (1990) All trans retinoic acid as differentiation therapy for acute promyelocytic leukemia. I. Clinical results. Blood 76: 1704-1709. Hong, W. K., Wittes, R. E., Hajdu, S. T. et al. (1977) The evolution of mature teratoma from malignant testicular tumors. Cancer 40: 2987-2992. Schofield, J. E. (1947) Teratoma of testis: Spontaneous disappearance of lung metastases. BI: Med. J. 1: 411. Kurie, J. M., Bosl, G. J. & Dmitrovsky, E. D. (1992) The genetic and biologic aspects of treatment response and resistance in male germ cell cancer. Semin. Oncol. 19: 197-205. Friedman, S. J. & Skehan, P. (1979) Morphological differentiation of human choriocarcinoma cells induced by methotrexate. Cancer Res. 39: 1960-1967. Sachs, L. (1993) The cellular and molecular environment in leukemia. Blood Cells 19: 709-726. Kramer, M. H. H., Kluin, P. M., Wijburg, E. R. eta/. (1995) Differentiation of follicular lymphoma cells after autologous bone marrow transplantation and haematopoietic growth factor treatment. Lancet 345: 488-490. Folkman, J. (1990) What is the evidence that tumors are angiogenesis dependent? J. Nat/. Cancer Inst. 82: 4-6. Broder, S. & Karp, J. E. (1995) Progress against cancer. J. Cancer Res. C/in. Oncol. 121: 633-647. Meares, A. (1983) Psychological mechanisms in the regression of cancer. Med. J. Aust. 1: 583-584. Weinstock, C. (1984) Further evidence on psychobiological aspects of cancer. Int. J. Psychosom. 31:20-22. SPONTANEOUS 166. 167. 168. 169. 170. 171. 172. 173. 174. 175. 176. 177. 178. 179. 180. 181. 182. 183. 184. 185. 186. 187. 188. REGRESSION OF CANCER 423 Mahrer, A. R. (1980) The treatment of cancer through experiential psychotherapy. Psychotherapy: Theory, Res. Pratt. 17: 335-342. Levv. S.. Herberman. R.. Liooman. M. et al. (1987) Correlation of stress factors with sustained depression of NK ceil activity and predicted prognosis in patients with breast cancer. J. C/in. Oncol. 5: 345-353. Ader, R. & Cohen, N. (1985) CNS-immune system instruction. Conditioning phenomenon. Behav. Brain Sci. 8: 379-394. Nowell, P. (1976) The clonal evolution of tumor cell populations. Science 194: 23-28. Nicolson, G. L. (1987) Tumor cell instability, diversification and progression to the metastatic phenotype: From oncogene to oncofetal expression. Cancer Res. 47: 1473-1487. Klein, G. (1987) The approaching era of the tumor suppressor genes. Science238: 1539-1545. Alexandrow, M. G. & Moses, H. L. (1995) Transforming growth factor beta and cell cycle regulation. Cancer Res. 55: 1452-1457. Eaves, C. & Eaves, A. (1994) Differential manipulation of normal and chronic myeloid leukemia stem cell proliferation in vitro. Blood Cells 20: 83-95. Bedi, A., Griffin, C. A., Barber, J. P. eta/. (1994) Growth factor-mediated terminal differentiation of chronic myeloid leukemia. Cancer Res. 54: 5535-5538. Aman, M. J., Keller, U., Derigs, G. eta/. (1994) Regulation of cytokine expression by interferonalpha in human bone marrow stromal cells: Inhibition of hemato-poietic growth factors and induction of interleukin-I receptor antagonist. Blood84: 4142-4150. Hercbergs, A. & Leith, J. T. (1993) Spontaneous remission of metastatic lung cancer following myxedema coma-an apoptosis-related phenomenon? J. Nat/. Cancer Inst. 85: 1342-1343. Hoffman, B. & Liebermann, D. A. (1994) Molecular controls of apoptosis: differentiation/growth arrest primary response genes, proto-oncogenes, and tumor suppressor genes as positive and negative modulators. Oncogene 9: 1807-1812. Korsmeyer, S. J. (1992) Bcl-2 initiates a new category of oncogene: regulators of cell death. Blood 80: 879-886. Pansera, F. (1992) Regression in renal carcinoma as re-expression of cell death in kidney development. Perspect. Biol. Med. 35: 416-440. Kaufman, Y., Many, A., Rechavi, G. et al. (1995) Brief report: Lymphoma with recurrent cycles of spontaneous remission and relapse-possible role of apoptosis. N. Eng. J. Med. 332: 507-510. Greger, V., Passarge, E., Hopping, W. et al. (1989) Epigenetic changes may contribute to the formation and spontaneous regression of retinoblastoma. Hum. Genet, 83: 155-158. Baylin, S. B., Makos, M., Wu, J. et a/. (1991) Abnormal patterns of DNA methylation in human neoplasia: potential consequences for tumor progression. Cancer Cells 3: 383-390. Bird, A. (1992) The essentials of DNA methylation. Cell 170: 5-8. Blackburn, E. ll. (1991) Structure and function of telomeres. Nature 350: 569-573. Levy, M. Z., Allsopp, R. C., Futacher, A. B. et a/. (1992) Telomere end-replication problems and cell aging. J. lvlol. Biol. 225: 951-960. Kim, N. W., Piatyszek, M. A., Prowse, K. R. et a/. (1994) Specific association of human telomerase activity with immortal cells and cancer. Science 266: 2011-2015. Yokoyama, K., Matsuura, T., Piatyszek, M. A. et al. (1995) Correlating telomerase activity levels with human neuroblastoma outcomes. Nature Med. 11: 249-255. Thomas, L. (1983) The Youngest Science: Notes of a Medicine-Watcher: New York, NY: Viking Press, p. 202.
© Copyright 2018