Journal of Autoimmunity (1992) 5 (Supplement A), 363-371
Malignancy in A u t o i m m u n e Diseases
L. J. K i n l e n CRC Cancer Epidemiology Group, University of Oxford, Radcliffe Infirmary, Oxford OX2 6HE, UK
R h e u m a t o i d arthritis (RA) represents the a u t o i m m u n e disease that has been most studied in relation to malignancy. An examination of all published cohort studies has indicated a 9.7-fold increase o f n o n - H o d g k i n ' s l y m p h o m a a m o n g RA patients after i m m u n o s u p p r e s s i v e therapy, and a 2.5-fold increase in the absence of such t r e a t m e n t . Corresponding data for Sj~Jgren's s y n d r o m e point to a similar contrast. These findings are inseparable f r o m the hypothesis of i m p a i r e d i m m u n o s u r v e i l l a n c e which implies that m a l i g n a n c y is p r o m o t e d by defects in the i m m u n e system. Studies of individuals treated with i m m u u o s u p p r e s s i v e drugs, particularly to prevent graft rejection, have indicated that i m m u n o s u r v e i l l a n c e operates only against a restricted range of neoplasms. These include nonHodgkin's l y m p h o m a (NHL), s q u a m o u s cell skin cancer, Kaposi's s a r c o m a and cervical carcinoma. O t h e r states of i m m u n e i m p a i r m e n t including AIDS are also associated with m a r k e d increases of NHL. There is a striking correspondence between malignancies for which there is epidemiological or l a b o r a t o r y evidence for a virus aetiology and those that are increased by i m m u n e i m p a i r m e n t . In this respect the epidemiological evidence accords with experimental work that i m m u n o s u r v e i l l a n c e p r i m a r i l y operates against neoplasms of viral origin. It is therefore possible that a viral aetiology also underlies the excess of NHL in certain a u t o i m m u n e disorders, p a r t i c u l a r l y after i m m u n o s u p p r e s s i v e therapy.
Introduction T h e subject of malignancy in autoimmune disease is inseparable from an important m o d e r n hypothesis of cancer aetiology, namely impaired immunosurveillance. T h e concept of immunosurveillance implies that immune impairment increases the likelihood that aberrant cells, produced in the course of cell replication, will proceed to malignant clones and eventually clinical cancer, instead of being eliminated by a healthy immune system. Over the past 25 years extensive use of immunosuppressive drugs, particularly to prevent graft rejection, has provided m u c h scope for 363 0896-8411/92/0A0363+ 09 $03.00/0
© 1992AcademicPress Limited
364 L.J. Kinlen examining the consequences of i m m u n e impairment. As a result, it has become clear that immunosurveillance does not operate against all h u m a n malignancies, but only against a restricted range of neoplasms. T h e s e findings have implications for autoimmunity which has been seen as having certain similarities to an allograft, but increasingly as reflecting disturbances in the i m m u n e system itself. When, as here, the subject is to include a u t o i m m u n e disorders treated with immunosuppressive drugs, then immunosurveillance in its wider aspects has obvious relevance.
Immunosuppressive therapy Some indications of the effects of immunosuppressive drugs found in our prospective study [1, 2] are given in T a b l e 1 which shows the relative risks of certain cancers in individuals treated with immunosuppressive drugs, respectively with and without a renal transplant. It is noteworthy that although the magnitude of the excesses differ in the two groups, n o n - H o d g k i n ' s l y m p h o m a , squamous cell cancer of the skin, and p r i m a r y liver cancer are increased in both groups, indicating that these effects are not simply due to the antigenic stimulation of the graft. T h e risks of at least two other malignancies are also increased by immunosuppressive therapy, namely cervical cancer and Kaposi's sarcoma. T h e difficulty of calculating the expected n u m b e r s of cases of the former neoplasm in this closely monitored group of patients p r o m p t e d their exclusion from the table, though we have m a n y cases, mainly in the in situ stage. However, others have found a significantly greater prevalence of this neoplasm among screened transplant recipients than a m o n g a control group [3]. A significantly greater frequency of h u m a n papillovirus type 16/18 D N A in female transplant recipients than among controls has also been reported [4]. Kaposi's sarcoma in the p r e - A I D S era is represented among the neoplasms recorded in transplant recipients in our cohort study. M o r e striking is the fact that Penn was able to trace no fewer than 47 cases a m o n g transplant patients [5]. This, for so rare a neoplasm, m u s t point to an increase of several hundred fold, although an accurate estimate is not possible. Again, the graft is unlikely to be the crucial factor. Prior to the A I D S epidemic, at least 16 cases of Kaposi's sarcoma had been reported in patients without transplants who had received treatment with steroids or other immunosuppressive drugs [6-8]. For such a rare neoplasm, these case reports suggest an increased incidence, a conclusion supported by the finding that m o r e than 10% of all cases of Kaposi's sarcoma recorded in N o r w a y over a 5-year period (53) had previously received such treatment [7]. In our study of immunosuppressive drugs, the only malignancy that showed an excess only among non-transplant patients was bladder cancer, which showed an approximately four-fold increase [2]. However, this was restricted to the minority of patients who received cyclophosphamide, of which bladder cancer appears to be a specific hazard.
Rheumatoid arthritis (RA) and non-Hodgkin's l y m p h o m a (NHL) T h e c o m m o n e s t underlying disease in the non-transplant group in our study is an a u t o i m m u n e disorder, namely rheumatoid arthritis, and the relative risk for nonH o d g k i n ' s l y m p h o m a ( N H L ) among such patients was 13, compared to 8.3 a m o n g
Malignancy in a u t o i m m u n e diseases
365
T a b l e 1. Relative risk ( RR ) of cancers after immunosuppressive therapy Transplant recipients
Non-transplant
Type of cancer
RR
Obs
RR
Obs
Non-Hodgkin's lymphoma All skin cancer Squamous cell Melanoma Primary liver Bladder Other cancers Total
49.4 8.8 27.6 8.7 37.5 -1.4 3.3
42 22 8 2 3 -41 108
10.9 1.5 5.0 -9.1 3.7 1.4 1.6
6 5 3 0 1 6 47 65
T a b l e 2. Non-Hodgkin's lymphoma in cohort studies of rheumatoid arthritis without
azathioprine or cyclophosphamide NHL Reference 9 10 11 12 13 14-22 Total
Number of patients 521 489 97 46,101 202 8,336 55,746
Cancers Obs
Obs
Exp
67 36 10 1,202 21 263 1,599
3 7 1 38 2 5 56
2.60 0.29 0.08 14.18 0.26 5.26 22.67
RR=2.5*
*RR=relative risk.
the remainder. In fact, this malignancy has been associated with rheumatoid arthritis (RA) in the absence of immunosuppressive therapy. It would be surprising if immunosuppression increased the incidence of N H L only in patients without RA and if a quite difference cause explained the excess in RA patients after i m m u n o suppressive therapy. However some of the increase might well reflect the effects of the underlying disease. Details of prospective studies of cancer among rheumatoid patients are shown in Table 2 [9-22]. N o t all of these studies presented details of the expected numbers of lymphomas, and in their absence they have been conservatively estimated. In this way, an estimate of 2.5 was obtained for the relative risk of N H L in patients with rheumatoid arthritis without azathioprine or cyclophosphamide therapy. I n contrast, Table 3 shows corresponding details from studies of rheumatoid patients treated with azathioprine or cyclophosphamide. It can be seen that an almost 10-fold increase is not confined to our study [2], and is typical of the group as a whole [13, 23-27]. N o n e of the large cohort studies shown in Table 3 have involved
366
L.J. K i n l e n
Table
3. NHL in cohort studies of rheumatoid arthritis after azathioprine or
cyclophosphamide therapy NHL Reference 2 13 23 24 25 26 27 Total
N u m b e r of patients
Cancers Obs
Obs
Exp$
643 202 119 139" 25 >400 81 1,609+
36 33 29 13 2 9 14 136
4 4 2 3 0 4 3 20
0.31 0.41 0.58 0.26 0.04 0.18 0.28 2.06
RR = 9.7t
*Includes chlorambucil. #RR = relative risk. ,+Conservatively estimated from total cancers, when not given. T a b l e 4. Relative risk of non-Hodgkin's lymphoma (observed
numbers) Azathioprine or cyclophosphamide Disorder Rheumatoid arthritis* Sj6gren'sdisease#
With
Without
9.7 (20) 100.0 (2)
2.5 (56) 36.0 (5)
*From Tables 2 and 3. tFrom reference [28].
r a n d o m i z e d trials. I t is n o t p o s s i b l e t h e r e f o r e to assess w h e t h e r a n y o f this h i g h e r relative risk for N H L reflects t h e effects o f selection a n d disease severity. N e v e r t h e less, t h e s e f i n d i n g s are c o n s i s t e n t w i t h t h e view t h a t i m m u n o s u p p r e s s i v e agents such as a z a t h i o p r i n e i n c r e a s e still f u r t h e r t h e e l e v a t e d risk o f N H L a s s o c i a t e d w i t h rheumatoid arthritis.
NHL in other autoimmune
disorders
C o m p a r a b l e d a t a are u n f o r t u n a t e l y n o t available for m o s t o t h e r a u t o i m m u n e diseases, w h e t h e r o r n o t t r e a t e d w i t h i m m u n o s u p p r e s s i v e agents. H o w e v e r , t h e r e is a s u g g e s t i o n b a s e d o n small n u m b e r s t h a t a z a t h i o p r i n e f u r t h e r increases t h e well k n o w n e l e v a t e d risk o f N H L in S j 6 g r e n ' s s y n d r o m e [28]. I m m u n o s u p p r e s s i v e t h e r a p y was a s s o c i a t e d w i t h a 100-fold i n c r e a s e o f N H L ( b a s e d o n two cases) c o m p a r e d to 3 5 - f o l d in o t h e r p a t i e n t s (five cases) ( T a b l e 4). S t u d i e s o f o t h e r a u t o i m m u n e d i s o r d e r s have u s u a l l y h a d insufficient n u m b e r s o f p e r s o n - y e a r s at risk to b e a b l e to
Malignancy in autoimmune diseases
367
Table 5. Cerebral lymphomas in patients without transplants after azathioprine and
cyclophosphamide
Underlying disorder Lupus erythematosus Lupus erythematosus Multiple sclerosis Lupus erythematosus Necrotizing vasculitis Rheumatoid arthritis Thrombocytopenic purpura Lupus nephritis
Interval after start of drug (months)
CNS only
Reference
8 8 18 10 68 84 9 36
Yes Yes No Yes Yes No No Yes
30 31 32 33 34 35 36 37
indicate anything reliable about a neoplasm as relatively u n c o m m o n in the general population as N H L . A six-fold increase of N H L recorded in a cohort study of patients with chronic lymphocytic thyroiditis was mainly due to the four cases in the thyroid itself, reflecting a m u c h larger increase of thyroid lymphomas [29].
Cerebral l y m p h o m a s Additional suggestive evidence for immunosuppressive drugs increasing the risk of l y m p h o m a s in a u t o i m m u n e disorders m a y be seen in the n u m b e r s of case reports of cerebral l y m p h o m a s in such patients. It is striking that eight cases could be traced in the p r e - A I D S era, at least six of t h e m in patients with lupus or other a u t o i m m u n e diseases [30-37] (Table 5). Indeed, not m a n y more cases of cerebral l y m p h o m a were recorded in a U S population of 21 million from 1971 to 1973 by the T h i r d National Cancer Survey [38]. An excess of cerebral l y m p h o m a s in a u t o i m m u n e diseases after azathioprine or cyclophosphamide therapy would accord with the large increases recorded in transplant recipients treated with azathioprine.
NHL risk a n d severity o f i m m u n e i m p a i r m e n t An excess of N H L has been found in most states of i m m u n e i m p a i r m e n t that have been studied in h u m a n s including renal and cardiac transplants, whether treated with azathioprine or cyclosporin, A I D S and other immunodeficiency disorders. Intensity of i m m u n o s u p p r e s s i o n appears to be an important determinant of the level of l y m p h o m a risk. In an early series of cardiac transplant patients who received an intensive immunosuppressive regimen, an unusually high incidence was recorded [39]. T h e occurrence of three l y m p h o m a s in an early series of only 34 cyclosporin Atreated transplant patients is even m o r e striking for it implies a relative risk of well over 500. T h i s almost certainly reflects the effect ofcyclosporin A together with large doses of other immunosuppressive agents, combinations that have since been avoided [40]. Similarly, the occurrence of a l y m p h o m a in m o r e than one transplant patient at the injection site of anti-lymphocytic globulin suggests a local dose-
368
L.J. Kinlen
T a b l e 6. Virus-linked malignancies in relation to immune impairment Excess in transplant recipients (Tx) etc. Type of malignancy
Evidence of infectivity§
Burkitt's lymphoma Non-Hodgkin's lymphoma Squamous cell skin cancer Hepatocellular cancer Cervical cancer Penis cancer Kaposi's sarcoma Nasopharyngeal cancer Adult T-cell leukaemia Hodgkin's disease Childhood leukaemia
Tx
Other*
() + + + + 0 + 0 0 -
+ + +
+ + + + + + + + + +
0 + () () ?+ t ? + :~
*Immunosuppressivedrugs without Tx; AIDS; immunodeficiencydiseases. rAIDS. ST-cell leukaemia in ataxia telangiectasia. §Epidemiologicaland/or viral. ( ): Limited data from geographicalareas of high incidence. related effect. A decline in the incidence of post-transplant lymphomas in successive calendar periods in our study even before the advent of cyclosporin may reflect the effects of less intensive immunosuppressive therapy with the growth of medical experience. T h e above evidence for N H L risk being influenced by intensity of immune impairment may be relevant to the increased risks in rheumatoid arthritis and in Sj6gren's syndrome being further increased by immunosuppressive drugs. Significance of the excesses of cancer
Epidemiological evidence concerning immune factors suggests that they are not major factors in the aetiology of most cancers. In a few neoplasms, however, there is no doubt of their importance. N H L more than any other malignancy stands out in showing increases in different immunological disturbances, though the magnitude of the increase varies appreciably. T h e fact that certain other malignancies do not figure consistently in all states of immune impairment is not necessarily remarkable. This may reflect the combined effects of differences in relative risk (due to the type or severity of the immune impairment) and the power of the particular study. T h e immune system consists of many different and complex elements over which a delicate balance is exercised. Different states of immune impairment may involve defects acting at quite different points. It may also be noted that despite the repeated use of the term ' i m m u n e impairment', in certain situations its effective consequence may be 'immunostimulation' because of the release of elements that are normally checked, as in the B cell proliferation following T cell inhibition. Table 6 lists the h u m a n malignancies of established viral origin, together with those for which there is some suggestive epidemiological evidence of such an
Malignancy in a u t o i m m u n e diseases
369
aetiology. Also shown are those that are increased in immunodeficiency states and a striking correspondence is seen: nearly all virus-associated cancers for which there is adequate or strongly suggestive h u m a n evidence have shown an excess in association with immune impairment. Furthermore, the only malignancy seen in excess in transplant patients which is not usually regarded as viral in origin in the general p o p u lation is squamous cell cancer of the skin. T h e r e is evidence, however, that h u m a n papillomavirus infection causes skin cancers in epidermodysplasia verruciformis as well as in vulval cancers in transplant recipients. It is also noteworthy that squamous cell skin cancer may be the most highly antigenic of all malignancies in man, as it is in mice [41]. T h e evidence that a viral aetiology (and in particular the Epstein-Barr virus) underlies the excesses of N H L in states of immune impairment must raise the possibility of a similar basis for the excess in rheumatoid arthritis and in certain other autoimmune disorders. Excesses of certain malignancies not linked to viruses have however also been recorded in such patients. T h u s , besides many case reports, two prospective studies of systemic sclerosis have found increases of lung cancer that could not b e explained on the basis of smoking [42, 43]. It is possible that this is at least partly a local effect, due to the associated fibrosis and disordered lung architecture. An excess of lung cancer in the large Finnish study of R A may have a similar explanation but it also raises the possibility that men with severe arthritis smoke more than the average [12]. T h e same study also recorded excesses ofleukaemia, myeloma and Hodgkin's disease but overall no deficiency of cancer as certain early studies have claimed.
Conclusion Overall the evidence on immune impairment and cancer suggests a gratifying consistency between epidemiological and laboratory work. Evidence from both approaches suggests that it is primarily neoplasms of viral origin that are subject to immunosurveillance. Whether the increase of N H L in certain autoimmune diseases also has a viral basis has not been established, but it is one plausible explanation.
References 1. Kinlen, L. l., A. G. R. Sheil, J. Peto et al. 1979. A collaborative U.K.-Australasian study of cancer in patients treated with immunosuppressive drugs. Br. Med. J. 2: 1461-1466 2. Kinlen, L. J. 1985. Incidence of cancer in rheumatoid arthritis and other disorders after immunosuppressive treatments. Am. J. Med. 78 (suppl. 1A): 44--49 3. Halpert, R., R. G. Fruchter, A. Sedlis et al. 1986. Human papillomavirus and lower genital neoplasia in renal transplant patients. Obstet. Gynecol. 68:251-258 4. Alloub, M. I., B. B. Barr, K. M. McLaren et al. 1989. Human papillomavirus infection and cervical intraepithelial neoplasia in women with renal allografts. Br. Med. J. 298: 153-156 5. Penn, I. 1982. The occurrence of cancer in immune deficiencies. Curr. Prob. Cancer 6: 1-64 6. Gange, R. W. and E. W. Jones. 1978. Kaposi's sarcoma and immunosuppressive therapy: an appraisal. Clin. Exp. Dermatol. 3:135-146 7. Klepp, O., O. Dahl, and J. T. Stenwig. 1978. Association of Kaposi's sarcoma and prior immunosuppressive therapy. Cancer 42:2626-2630
370 L.J. Kinlen 8. Ilie, B., S. Bremner, R. Lipitz et al. 1981. Kaposi's sarcoma after steroid therapy for Pemphigus foliaceus. Dermatologica 163:455-459 9. Katusic, S., C. M. Beard, L. T. Kurland, J. W. Weis, and E. Bergstralh. 1985. Occurrence of malignant neoplasms in the Rochester, Minnesota, rheumatoid arthritis cohort. Am..7. Med. 78 (Suppl. 1A): 50-55 10. Prior, P., D. P. M. Symmons, C. F. Hawkings, D. L. Scott, and R. Brown. 1984. Cancer morbidity in rheumatoid arthritis. Ann. Rheum. Dis. 43: 128-131 11. Lewis, P., B. L. Hazleman, R. Harka, and S. Roberts. 1980. Cause of death in patients with rheumatoid arthritis with particular reference to azathioprine. Ann. Rheum. Dis. 39: 457-461 12. Isomaki, H. A., T. Hakulinen, and U. Joutsenlahti. 1978. Excess risk of lymphomas, leukemia and myeloma in patients with rheumatoid arthritis. J. Chron. Dis. 31" 691-696 13. Silman, A. J., J. Petrie, B. Hazleman, and S. J. W. Evans. 1988. Lymphoproliferative cancer and other malignancy in patients with rheumatoid arthritis treated with azathioprine: a 20 year follow up study. Ann. Rheum. Dis. 47" 988-992 14. Fries, J. F., D. Bloch, P. Spitz, and D. M. Mitchell. 1985. Cancer in rheumatoid arthritis: a prospective long-term study of mortality. Am. -7. Med. 78 (Suppl. 1A): 56-59 15. Allebeck, P., A. Ahlbom, and E. Allander. 1981. Increased mortality among persons with rheumatoid arthritis, but where RA does not appear on death certificate. Stand. J. Rheum. 10" 301-306 16. Allebeck, P. 1982. Increased mortality in rheumatoid arthritis. Stand. J. Rheum. 11" 81-86 17. Isomaki, H. A., O. Mutru, and K. Koota. 1975. Death rate and causes of death in patients with rheumatoid arthritis. Scand. -7. Rheum. 4" 205-208 18. Monson, R. R. 1976. Mortality among arthritics..7. Chron. Dis. 29" 459-467 19. Lewis, R. B., W. Castor, R. E. Knisley, and G. G. Bole. 1976. Frequency of neoplasi in systemic lupus erythematosus and rheumatoid arthritis. Arthritis Rheum. 19:1256-1260 20. Moesmann, G. 1969. Malignancy and mortality in subacute rheumatoid arthritis in old age. mcta Rheum. Scand. 15= 193-199 21. Uddin, J., A. S. Kraus, and H. G. Kelly. 1970. Survivorship and death in rheumatoid arthritis. Arthritis Rheum. 13(2): 125-130 22. Cobb, S., F. Anderson, and W. Bauer. 1953. Length of life and cause of death in rheumatoid arthritis. N. Engl. J. Med. 249:553-556 23. Baker, G. L., L. E. Kahl, B. C. Zee, B. L. Stolzer, A. K. Agarival, and T. A. Medsger. 1987. Malignancy following treatment of rheumatoid arthritis with cyclophosphamide. Am. J. Med. 83:1-9 24. Renier, J. C., C. Bregeon, C. Bonnette, M. Boasson, M. Bernat, M. Basle, J. Besson, C. Wellinger, B. Bourgeois, and N. Teisseire. 1978. Le devenir des sujets atteints de polyarthrite rhumatoide et traites par les immunodepresseurs entre 1965 et 1973 inclus. Revue du Rhumatisme 45(7-9): 453-461 25. Pinals, R. S. 1976. Azathioprine in the treatment of chronic polyarthritis: long-term results and adverse effects in 25 patients. J. Rheumatol. 3(2): 140-144 26. Love, R. R. and J. M. Sowa. 1975. Myelomonocytic leukemia following cyclophosphamide therapy of rheumatoid disease. Ann. Rheum. Dis. 34:534-535 27. Baltus, J. A. M., J. W. Boersma, A. P. Hartman, and J. P. Vandenbroucke. 1983. T h e occurrence of malignancies in patients with rheumatoid arthritis treated with cyclophosphamide: a controlled retrospective follow-up. Ann. Rheum. Dis. 42:368-373 28. Kassan, S. S., T. L. Thomas, H, M., Moutsopoulos, R. Hoover, R. P. Kimberly, D. R. Budman, J. Costa, J. L. Decker, and T. M. Chused. 1978. Increased risk of lymphoma in sicca syndrome. Ann. Int. Med. 89(6)= 888-892 29. Holm, L.-E., H. Blomgren, and T. Lowhagen. 1985. Cancer risks in patients with chronic lymphocytic thyroiditis. N. Engl.-7. Med. 312(10): 601-604 30. Lipsmeyer, E. A. 1972. Development of malignant cerebral lymphoma in a patient with systemic lupus erythematosus treated with immunosuppression. Arthritis Rheum. 5: 183-187
M a l i g n a n c y in a u t o i m m u n e d i s e a s e s
371
31. Uhl, G. S., J. E. Williams, and F. C. Arnett. 1974. Intracerebral lymphoma in a patient with central nervous system lupus on cyclophosphamide. J. Rheumatol. 1:282-286 32. Ulrich, J. a n d R . Wuthrich. 1974. Multiple sclerosis: reticular cell sarcoma ofthe nervous system in a patient treated with immunosuppressive drugs. Eur. Neurols. 12:65-69 33. Moser, K. D., C. A. Araoz, and P. L. Smith. 1972. Immunosuppressive therapy and reticulum cell sarcomas of the brain. Panminerva Med. 34(1-2): 436-440 34. Jellinger, K., P. Kothbauer, R. Weiss, and E. Sunder-Plassmann. 1979. Primary malignant lymphoma of the C N S and polyneuropathy in a patient with necrotizing vasculitis treated with immunosuppression. J. Neurol. 220:259-262 35. Neuhas, K., J. Thorhurst, O. Bertol, and B. Speck. 1976. Multiple neoplasms in a case of rheumatoid arthritis treated with azathioprine. Praxis 10:2131-2135 36. Varadachari, C., M. Paluthe, A. R. W. Climie, R. W. Weiss, and J. L. Chason. 1978. Immunoblastic sarcoma (histiocytic lymphoma) of the brain with B cell markers. J. Neurosurg. 49" 887-890 37. Lindeman, R. D., J. A. Peterson, and B. J. Matter. 1976. Long-term azathioprinecorticosteroid therapy in lupus nephritis and idiopathic nephrotic syndrome. J. Chron. Dis. 29:189-204 38. Cutler, S. J. and J. L. Young (eds). 1975. Third National Cancer Survey: Incidence Data. National Cancer Institute Monograph, U.S. Dept. of Health & Welfare 39. Anderson, J. L., C. P. Bieber, R. E. Fowles et al. 1978. Idiopathic cardiomyopathy, age and suppressor-cell dysfunction as risk determinant of lymphoma after cardiac transplantation. Lancet ii: 1174--1177 40. Calne, R. F., K. Rolles, D. J. G. White et al. 1979. Cyclosporin A initially as the only immunosuppressant in 34 recipients of cadaveric organs: 32 kidneys, 2 pancreases and 2 livers. Lancet ii: 1033-1036 41. Kripke, M. L. 1981. Immunologic mechanisms in UV radiation carcinogenesis. Adv. Cancer Res. 34:69-106 42. Peters-Golden, M., R. A. Wise, M. Hochberg et al. 1985. Incidence of lung cancer in systemic sclerosis. J. Rheumatol. 12:1136-1139 43. Roumm, A. D. and T. A. Medsger. 1985. Cancer and systemic sclerosis. Arthritis Rheum. 28" 1336-1340
Malignancy in A u t o i m m u n e Diseases
L. J. K i n l e n CRC Cancer Epidemiology Group, University of Oxford, Radcliffe Infirmary, Oxford OX2 6HE, UK
R h e u m a t o i d arthritis (RA) represents the a u t o i m m u n e disease that has been most studied in relation to malignancy. An examination of all published cohort studies has indicated a 9.7-fold increase o f n o n - H o d g k i n ' s l y m p h o m a a m o n g RA patients after i m m u n o s u p p r e s s i v e therapy, and a 2.5-fold increase in the absence of such t r e a t m e n t . Corresponding data for Sj~Jgren's s y n d r o m e point to a similar contrast. These findings are inseparable f r o m the hypothesis of i m p a i r e d i m m u n o s u r v e i l l a n c e which implies that m a l i g n a n c y is p r o m o t e d by defects in the i m m u n e system. Studies of individuals treated with i m m u u o s u p p r e s s i v e drugs, particularly to prevent graft rejection, have indicated that i m m u n o s u r v e i l l a n c e operates only against a restricted range of neoplasms. These include nonHodgkin's l y m p h o m a (NHL), s q u a m o u s cell skin cancer, Kaposi's s a r c o m a and cervical carcinoma. O t h e r states of i m m u n e i m p a i r m e n t including AIDS are also associated with m a r k e d increases of NHL. There is a striking correspondence between malignancies for which there is epidemiological or l a b o r a t o r y evidence for a virus aetiology and those that are increased by i m m u n e i m p a i r m e n t . In this respect the epidemiological evidence accords with experimental work that i m m u n o s u r v e i l l a n c e p r i m a r i l y operates against neoplasms of viral origin. It is therefore possible that a viral aetiology also underlies the excess of NHL in certain a u t o i m m u n e disorders, p a r t i c u l a r l y after i m m u n o s u p p r e s s i v e therapy.
Introduction T h e subject of malignancy in autoimmune disease is inseparable from an important m o d e r n hypothesis of cancer aetiology, namely impaired immunosurveillance. T h e concept of immunosurveillance implies that immune impairment increases the likelihood that aberrant cells, produced in the course of cell replication, will proceed to malignant clones and eventually clinical cancer, instead of being eliminated by a healthy immune system. Over the past 25 years extensive use of immunosuppressive drugs, particularly to prevent graft rejection, has provided m u c h scope for 363 0896-8411/92/0A0363+ 09 $03.00/0
© 1992AcademicPress Limited
364 L.J. Kinlen examining the consequences of i m m u n e impairment. As a result, it has become clear that immunosurveillance does not operate against all h u m a n malignancies, but only against a restricted range of neoplasms. T h e s e findings have implications for autoimmunity which has been seen as having certain similarities to an allograft, but increasingly as reflecting disturbances in the i m m u n e system itself. When, as here, the subject is to include a u t o i m m u n e disorders treated with immunosuppressive drugs, then immunosurveillance in its wider aspects has obvious relevance.
Immunosuppressive therapy Some indications of the effects of immunosuppressive drugs found in our prospective study [1, 2] are given in T a b l e 1 which shows the relative risks of certain cancers in individuals treated with immunosuppressive drugs, respectively with and without a renal transplant. It is noteworthy that although the magnitude of the excesses differ in the two groups, n o n - H o d g k i n ' s l y m p h o m a , squamous cell cancer of the skin, and p r i m a r y liver cancer are increased in both groups, indicating that these effects are not simply due to the antigenic stimulation of the graft. T h e risks of at least two other malignancies are also increased by immunosuppressive therapy, namely cervical cancer and Kaposi's sarcoma. T h e difficulty of calculating the expected n u m b e r s of cases of the former neoplasm in this closely monitored group of patients p r o m p t e d their exclusion from the table, though we have m a n y cases, mainly in the in situ stage. However, others have found a significantly greater prevalence of this neoplasm among screened transplant recipients than a m o n g a control group [3]. A significantly greater frequency of h u m a n papillovirus type 16/18 D N A in female transplant recipients than among controls has also been reported [4]. Kaposi's sarcoma in the p r e - A I D S era is represented among the neoplasms recorded in transplant recipients in our cohort study. M o r e striking is the fact that Penn was able to trace no fewer than 47 cases a m o n g transplant patients [5]. This, for so rare a neoplasm, m u s t point to an increase of several hundred fold, although an accurate estimate is not possible. Again, the graft is unlikely to be the crucial factor. Prior to the A I D S epidemic, at least 16 cases of Kaposi's sarcoma had been reported in patients without transplants who had received treatment with steroids or other immunosuppressive drugs [6-8]. For such a rare neoplasm, these case reports suggest an increased incidence, a conclusion supported by the finding that m o r e than 10% of all cases of Kaposi's sarcoma recorded in N o r w a y over a 5-year period (53) had previously received such treatment [7]. In our study of immunosuppressive drugs, the only malignancy that showed an excess only among non-transplant patients was bladder cancer, which showed an approximately four-fold increase [2]. However, this was restricted to the minority of patients who received cyclophosphamide, of which bladder cancer appears to be a specific hazard.
Rheumatoid arthritis (RA) and non-Hodgkin's l y m p h o m a (NHL) T h e c o m m o n e s t underlying disease in the non-transplant group in our study is an a u t o i m m u n e disorder, namely rheumatoid arthritis, and the relative risk for nonH o d g k i n ' s l y m p h o m a ( N H L ) among such patients was 13, compared to 8.3 a m o n g
Malignancy in a u t o i m m u n e diseases
365
T a b l e 1. Relative risk ( RR ) of cancers after immunosuppressive therapy Transplant recipients
Non-transplant
Type of cancer
RR
Obs
RR
Obs
Non-Hodgkin's lymphoma All skin cancer Squamous cell Melanoma Primary liver Bladder Other cancers Total
49.4 8.8 27.6 8.7 37.5 -1.4 3.3
42 22 8 2 3 -41 108
10.9 1.5 5.0 -9.1 3.7 1.4 1.6
6 5 3 0 1 6 47 65
T a b l e 2. Non-Hodgkin's lymphoma in cohort studies of rheumatoid arthritis without
azathioprine or cyclophosphamide NHL Reference 9 10 11 12 13 14-22 Total
Number of patients 521 489 97 46,101 202 8,336 55,746
Cancers Obs
Obs
Exp
67 36 10 1,202 21 263 1,599
3 7 1 38 2 5 56
2.60 0.29 0.08 14.18 0.26 5.26 22.67
RR=2.5*
*RR=relative risk.
the remainder. In fact, this malignancy has been associated with rheumatoid arthritis (RA) in the absence of immunosuppressive therapy. It would be surprising if immunosuppression increased the incidence of N H L only in patients without RA and if a quite difference cause explained the excess in RA patients after i m m u n o suppressive therapy. However some of the increase might well reflect the effects of the underlying disease. Details of prospective studies of cancer among rheumatoid patients are shown in Table 2 [9-22]. N o t all of these studies presented details of the expected numbers of lymphomas, and in their absence they have been conservatively estimated. In this way, an estimate of 2.5 was obtained for the relative risk of N H L in patients with rheumatoid arthritis without azathioprine or cyclophosphamide therapy. I n contrast, Table 3 shows corresponding details from studies of rheumatoid patients treated with azathioprine or cyclophosphamide. It can be seen that an almost 10-fold increase is not confined to our study [2], and is typical of the group as a whole [13, 23-27]. N o n e of the large cohort studies shown in Table 3 have involved
366
L.J. K i n l e n
Table
3. NHL in cohort studies of rheumatoid arthritis after azathioprine or
cyclophosphamide therapy NHL Reference 2 13 23 24 25 26 27 Total
N u m b e r of patients
Cancers Obs
Obs
Exp$
643 202 119 139" 25 >400 81 1,609+
36 33 29 13 2 9 14 136
4 4 2 3 0 4 3 20
0.31 0.41 0.58 0.26 0.04 0.18 0.28 2.06
RR = 9.7t
*Includes chlorambucil. #RR = relative risk. ,+Conservatively estimated from total cancers, when not given. T a b l e 4. Relative risk of non-Hodgkin's lymphoma (observed
numbers) Azathioprine or cyclophosphamide Disorder Rheumatoid arthritis* Sj6gren'sdisease#
With
Without
9.7 (20) 100.0 (2)
2.5 (56) 36.0 (5)
*From Tables 2 and 3. tFrom reference [28].
r a n d o m i z e d trials. I t is n o t p o s s i b l e t h e r e f o r e to assess w h e t h e r a n y o f this h i g h e r relative risk for N H L reflects t h e effects o f selection a n d disease severity. N e v e r t h e less, t h e s e f i n d i n g s are c o n s i s t e n t w i t h t h e view t h a t i m m u n o s u p p r e s s i v e agents such as a z a t h i o p r i n e i n c r e a s e still f u r t h e r t h e e l e v a t e d risk o f N H L a s s o c i a t e d w i t h rheumatoid arthritis.
NHL in other autoimmune
disorders
C o m p a r a b l e d a t a are u n f o r t u n a t e l y n o t available for m o s t o t h e r a u t o i m m u n e diseases, w h e t h e r o r n o t t r e a t e d w i t h i m m u n o s u p p r e s s i v e agents. H o w e v e r , t h e r e is a s u g g e s t i o n b a s e d o n small n u m b e r s t h a t a z a t h i o p r i n e f u r t h e r increases t h e well k n o w n e l e v a t e d risk o f N H L in S j 6 g r e n ' s s y n d r o m e [28]. I m m u n o s u p p r e s s i v e t h e r a p y was a s s o c i a t e d w i t h a 100-fold i n c r e a s e o f N H L ( b a s e d o n two cases) c o m p a r e d to 3 5 - f o l d in o t h e r p a t i e n t s (five cases) ( T a b l e 4). S t u d i e s o f o t h e r a u t o i m m u n e d i s o r d e r s have u s u a l l y h a d insufficient n u m b e r s o f p e r s o n - y e a r s at risk to b e a b l e to
Malignancy in autoimmune diseases
367
Table 5. Cerebral lymphomas in patients without transplants after azathioprine and
cyclophosphamide
Underlying disorder Lupus erythematosus Lupus erythematosus Multiple sclerosis Lupus erythematosus Necrotizing vasculitis Rheumatoid arthritis Thrombocytopenic purpura Lupus nephritis
Interval after start of drug (months)
CNS only
Reference
8 8 18 10 68 84 9 36
Yes Yes No Yes Yes No No Yes
30 31 32 33 34 35 36 37
indicate anything reliable about a neoplasm as relatively u n c o m m o n in the general population as N H L . A six-fold increase of N H L recorded in a cohort study of patients with chronic lymphocytic thyroiditis was mainly due to the four cases in the thyroid itself, reflecting a m u c h larger increase of thyroid lymphomas [29].
Cerebral l y m p h o m a s Additional suggestive evidence for immunosuppressive drugs increasing the risk of l y m p h o m a s in a u t o i m m u n e disorders m a y be seen in the n u m b e r s of case reports of cerebral l y m p h o m a s in such patients. It is striking that eight cases could be traced in the p r e - A I D S era, at least six of t h e m in patients with lupus or other a u t o i m m u n e diseases [30-37] (Table 5). Indeed, not m a n y more cases of cerebral l y m p h o m a were recorded in a U S population of 21 million from 1971 to 1973 by the T h i r d National Cancer Survey [38]. An excess of cerebral l y m p h o m a s in a u t o i m m u n e diseases after azathioprine or cyclophosphamide therapy would accord with the large increases recorded in transplant recipients treated with azathioprine.
NHL risk a n d severity o f i m m u n e i m p a i r m e n t An excess of N H L has been found in most states of i m m u n e i m p a i r m e n t that have been studied in h u m a n s including renal and cardiac transplants, whether treated with azathioprine or cyclosporin, A I D S and other immunodeficiency disorders. Intensity of i m m u n o s u p p r e s s i o n appears to be an important determinant of the level of l y m p h o m a risk. In an early series of cardiac transplant patients who received an intensive immunosuppressive regimen, an unusually high incidence was recorded [39]. T h e occurrence of three l y m p h o m a s in an early series of only 34 cyclosporin Atreated transplant patients is even m o r e striking for it implies a relative risk of well over 500. T h i s almost certainly reflects the effect ofcyclosporin A together with large doses of other immunosuppressive agents, combinations that have since been avoided [40]. Similarly, the occurrence of a l y m p h o m a in m o r e than one transplant patient at the injection site of anti-lymphocytic globulin suggests a local dose-
368
L.J. Kinlen
T a b l e 6. Virus-linked malignancies in relation to immune impairment Excess in transplant recipients (Tx) etc. Type of malignancy
Evidence of infectivity§
Burkitt's lymphoma Non-Hodgkin's lymphoma Squamous cell skin cancer Hepatocellular cancer Cervical cancer Penis cancer Kaposi's sarcoma Nasopharyngeal cancer Adult T-cell leukaemia Hodgkin's disease Childhood leukaemia
Tx
Other*
() + + + + 0 + 0 0 -
+ + +
+ + + + + + + + + +
0 + () () ?+ t ? + :~
*Immunosuppressivedrugs without Tx; AIDS; immunodeficiencydiseases. rAIDS. ST-cell leukaemia in ataxia telangiectasia. §Epidemiologicaland/or viral. ( ): Limited data from geographicalareas of high incidence. related effect. A decline in the incidence of post-transplant lymphomas in successive calendar periods in our study even before the advent of cyclosporin may reflect the effects of less intensive immunosuppressive therapy with the growth of medical experience. T h e above evidence for N H L risk being influenced by intensity of immune impairment may be relevant to the increased risks in rheumatoid arthritis and in Sj6gren's syndrome being further increased by immunosuppressive drugs. Significance of the excesses of cancer
Epidemiological evidence concerning immune factors suggests that they are not major factors in the aetiology of most cancers. In a few neoplasms, however, there is no doubt of their importance. N H L more than any other malignancy stands out in showing increases in different immunological disturbances, though the magnitude of the increase varies appreciably. T h e fact that certain other malignancies do not figure consistently in all states of immune impairment is not necessarily remarkable. This may reflect the combined effects of differences in relative risk (due to the type or severity of the immune impairment) and the power of the particular study. T h e immune system consists of many different and complex elements over which a delicate balance is exercised. Different states of immune impairment may involve defects acting at quite different points. It may also be noted that despite the repeated use of the term ' i m m u n e impairment', in certain situations its effective consequence may be 'immunostimulation' because of the release of elements that are normally checked, as in the B cell proliferation following T cell inhibition. Table 6 lists the h u m a n malignancies of established viral origin, together with those for which there is some suggestive epidemiological evidence of such an
Malignancy in a u t o i m m u n e diseases
369
aetiology. Also shown are those that are increased in immunodeficiency states and a striking correspondence is seen: nearly all virus-associated cancers for which there is adequate or strongly suggestive h u m a n evidence have shown an excess in association with immune impairment. Furthermore, the only malignancy seen in excess in transplant patients which is not usually regarded as viral in origin in the general p o p u lation is squamous cell cancer of the skin. T h e r e is evidence, however, that h u m a n papillomavirus infection causes skin cancers in epidermodysplasia verruciformis as well as in vulval cancers in transplant recipients. It is also noteworthy that squamous cell skin cancer may be the most highly antigenic of all malignancies in man, as it is in mice [41]. T h e evidence that a viral aetiology (and in particular the Epstein-Barr virus) underlies the excesses of N H L in states of immune impairment must raise the possibility of a similar basis for the excess in rheumatoid arthritis and in certain other autoimmune disorders. Excesses of certain malignancies not linked to viruses have however also been recorded in such patients. T h u s , besides many case reports, two prospective studies of systemic sclerosis have found increases of lung cancer that could not b e explained on the basis of smoking [42, 43]. It is possible that this is at least partly a local effect, due to the associated fibrosis and disordered lung architecture. An excess of lung cancer in the large Finnish study of R A may have a similar explanation but it also raises the possibility that men with severe arthritis smoke more than the average [12]. T h e same study also recorded excesses ofleukaemia, myeloma and Hodgkin's disease but overall no deficiency of cancer as certain early studies have claimed.
Conclusion Overall the evidence on immune impairment and cancer suggests a gratifying consistency between epidemiological and laboratory work. Evidence from both approaches suggests that it is primarily neoplasms of viral origin that are subject to immunosurveillance. Whether the increase of N H L in certain autoimmune diseases also has a viral basis has not been established, but it is one plausible explanation.
References 1. Kinlen, L. l., A. G. R. Sheil, J. Peto et al. 1979. A collaborative U.K.-Australasian study of cancer in patients treated with immunosuppressive drugs. Br. Med. J. 2: 1461-1466 2. Kinlen, L. J. 1985. Incidence of cancer in rheumatoid arthritis and other disorders after immunosuppressive treatments. Am. J. Med. 78 (suppl. 1A): 44--49 3. Halpert, R., R. G. Fruchter, A. Sedlis et al. 1986. Human papillomavirus and lower genital neoplasia in renal transplant patients. Obstet. Gynecol. 68:251-258 4. Alloub, M. I., B. B. Barr, K. M. McLaren et al. 1989. Human papillomavirus infection and cervical intraepithelial neoplasia in women with renal allografts. Br. Med. J. 298: 153-156 5. Penn, I. 1982. The occurrence of cancer in immune deficiencies. Curr. Prob. Cancer 6: 1-64 6. Gange, R. W. and E. W. Jones. 1978. Kaposi's sarcoma and immunosuppressive therapy: an appraisal. Clin. Exp. Dermatol. 3:135-146 7. Klepp, O., O. Dahl, and J. T. Stenwig. 1978. Association of Kaposi's sarcoma and prior immunosuppressive therapy. Cancer 42:2626-2630
370 L.J. Kinlen 8. Ilie, B., S. Bremner, R. Lipitz et al. 1981. Kaposi's sarcoma after steroid therapy for Pemphigus foliaceus. Dermatologica 163:455-459 9. Katusic, S., C. M. Beard, L. T. Kurland, J. W. Weis, and E. Bergstralh. 1985. Occurrence of malignant neoplasms in the Rochester, Minnesota, rheumatoid arthritis cohort. Am..7. Med. 78 (Suppl. 1A): 50-55 10. Prior, P., D. P. M. Symmons, C. F. Hawkings, D. L. Scott, and R. Brown. 1984. Cancer morbidity in rheumatoid arthritis. Ann. Rheum. Dis. 43: 128-131 11. Lewis, P., B. L. Hazleman, R. Harka, and S. Roberts. 1980. Cause of death in patients with rheumatoid arthritis with particular reference to azathioprine. Ann. Rheum. Dis. 39: 457-461 12. Isomaki, H. A., T. Hakulinen, and U. Joutsenlahti. 1978. Excess risk of lymphomas, leukemia and myeloma in patients with rheumatoid arthritis. J. Chron. Dis. 31" 691-696 13. Silman, A. J., J. Petrie, B. Hazleman, and S. J. W. Evans. 1988. Lymphoproliferative cancer and other malignancy in patients with rheumatoid arthritis treated with azathioprine: a 20 year follow up study. Ann. Rheum. Dis. 47" 988-992 14. Fries, J. F., D. Bloch, P. Spitz, and D. M. Mitchell. 1985. Cancer in rheumatoid arthritis: a prospective long-term study of mortality. Am. -7. Med. 78 (Suppl. 1A): 56-59 15. Allebeck, P., A. Ahlbom, and E. Allander. 1981. Increased mortality among persons with rheumatoid arthritis, but where RA does not appear on death certificate. Stand. J. Rheum. 10" 301-306 16. Allebeck, P. 1982. Increased mortality in rheumatoid arthritis. Stand. J. Rheum. 11" 81-86 17. Isomaki, H. A., O. Mutru, and K. Koota. 1975. Death rate and causes of death in patients with rheumatoid arthritis. Scand. -7. Rheum. 4" 205-208 18. Monson, R. R. 1976. Mortality among arthritics..7. Chron. Dis. 29" 459-467 19. Lewis, R. B., W. Castor, R. E. Knisley, and G. G. Bole. 1976. Frequency of neoplasi in systemic lupus erythematosus and rheumatoid arthritis. Arthritis Rheum. 19:1256-1260 20. Moesmann, G. 1969. Malignancy and mortality in subacute rheumatoid arthritis in old age. mcta Rheum. Scand. 15= 193-199 21. Uddin, J., A. S. Kraus, and H. G. Kelly. 1970. Survivorship and death in rheumatoid arthritis. Arthritis Rheum. 13(2): 125-130 22. Cobb, S., F. Anderson, and W. Bauer. 1953. Length of life and cause of death in rheumatoid arthritis. N. Engl. J. Med. 249:553-556 23. Baker, G. L., L. E. Kahl, B. C. Zee, B. L. Stolzer, A. K. Agarival, and T. A. Medsger. 1987. Malignancy following treatment of rheumatoid arthritis with cyclophosphamide. Am. J. Med. 83:1-9 24. Renier, J. C., C. Bregeon, C. Bonnette, M. Boasson, M. Bernat, M. Basle, J. Besson, C. Wellinger, B. Bourgeois, and N. Teisseire. 1978. Le devenir des sujets atteints de polyarthrite rhumatoide et traites par les immunodepresseurs entre 1965 et 1973 inclus. Revue du Rhumatisme 45(7-9): 453-461 25. Pinals, R. S. 1976. Azathioprine in the treatment of chronic polyarthritis: long-term results and adverse effects in 25 patients. J. Rheumatol. 3(2): 140-144 26. Love, R. R. and J. M. Sowa. 1975. Myelomonocytic leukemia following cyclophosphamide therapy of rheumatoid disease. Ann. Rheum. Dis. 34:534-535 27. Baltus, J. A. M., J. W. Boersma, A. P. Hartman, and J. P. Vandenbroucke. 1983. T h e occurrence of malignancies in patients with rheumatoid arthritis treated with cyclophosphamide: a controlled retrospective follow-up. Ann. Rheum. Dis. 42:368-373 28. Kassan, S. S., T. L. Thomas, H, M., Moutsopoulos, R. Hoover, R. P. Kimberly, D. R. Budman, J. Costa, J. L. Decker, and T. M. Chused. 1978. Increased risk of lymphoma in sicca syndrome. Ann. Int. Med. 89(6)= 888-892 29. Holm, L.-E., H. Blomgren, and T. Lowhagen. 1985. Cancer risks in patients with chronic lymphocytic thyroiditis. N. Engl.-7. Med. 312(10): 601-604 30. Lipsmeyer, E. A. 1972. Development of malignant cerebral lymphoma in a patient with systemic lupus erythematosus treated with immunosuppression. Arthritis Rheum. 5: 183-187
M a l i g n a n c y in a u t o i m m u n e d i s e a s e s
371
31. Uhl, G. S., J. E. Williams, and F. C. Arnett. 1974. Intracerebral lymphoma in a patient with central nervous system lupus on cyclophosphamide. J. Rheumatol. 1:282-286 32. Ulrich, J. a n d R . Wuthrich. 1974. Multiple sclerosis: reticular cell sarcoma ofthe nervous system in a patient treated with immunosuppressive drugs. Eur. Neurols. 12:65-69 33. Moser, K. D., C. A. Araoz, and P. L. Smith. 1972. Immunosuppressive therapy and reticulum cell sarcomas of the brain. Panminerva Med. 34(1-2): 436-440 34. Jellinger, K., P. Kothbauer, R. Weiss, and E. Sunder-Plassmann. 1979. Primary malignant lymphoma of the C N S and polyneuropathy in a patient with necrotizing vasculitis treated with immunosuppression. J. Neurol. 220:259-262 35. Neuhas, K., J. Thorhurst, O. Bertol, and B. Speck. 1976. Multiple neoplasms in a case of rheumatoid arthritis treated with azathioprine. Praxis 10:2131-2135 36. Varadachari, C., M. Paluthe, A. R. W. Climie, R. W. Weiss, and J. L. Chason. 1978. Immunoblastic sarcoma (histiocytic lymphoma) of the brain with B cell markers. J. Neurosurg. 49" 887-890 37. Lindeman, R. D., J. A. Peterson, and B. J. Matter. 1976. Long-term azathioprinecorticosteroid therapy in lupus nephritis and idiopathic nephrotic syndrome. J. Chron. Dis. 29:189-204 38. Cutler, S. J. and J. L. Young (eds). 1975. Third National Cancer Survey: Incidence Data. National Cancer Institute Monograph, U.S. Dept. of Health & Welfare 39. Anderson, J. L., C. P. Bieber, R. E. Fowles et al. 1978. Idiopathic cardiomyopathy, age and suppressor-cell dysfunction as risk determinant of lymphoma after cardiac transplantation. Lancet ii: 1174--1177 40. Calne, R. F., K. Rolles, D. J. G. White et al. 1979. Cyclosporin A initially as the only immunosuppressant in 34 recipients of cadaveric organs: 32 kidneys, 2 pancreases and 2 livers. Lancet ii: 1033-1036 41. Kripke, M. L. 1981. Immunologic mechanisms in UV radiation carcinogenesis. Adv. Cancer Res. 34:69-106 42. Peters-Golden, M., R. A. Wise, M. Hochberg et al. 1985. Incidence of lung cancer in systemic sclerosis. J. Rheumatol. 12:1136-1139 43. Roumm, A. D. and T. A. Medsger. 1985. Cancer and systemic sclerosis. Arthritis Rheum. 28" 1336-1340
