- Email: [email protected]
Cancers in Renal Transplant Recipients
Cancers in Renal Transplant Recipients Israel Pennt Data regarding posttransplant cancers are reviewed from the Cincinnati Transplant Tumor Registry (CTTR) and from the literature. The CTTR has data on 9,688 types of cancer that developed in 9,032 renal allograft recipients. The predominant tumors are lymphomas and Iymphoproliferations (PTLD). carcinomas of the skin and lips, carcinomas of the vulva and perineum, in situ carcinomas of the cervix of the uterus, Kaposi's sarcoma (KS), hepatocellular carcinomas, renal carcinomas, and various sarcomas (excluding KS). Prominent features of the PTLD cases are their high incidence, frequent involvement of extranodal sites, a marked predilection for the brain, and frequent involvement of the allograft by tumor. Skin cancers also present unusual features, a remarkable high frequency of KS, reversal of the ratio of basal cell carcinomas to squamous cell carcinomas that is seen in the general population, the young age of the patients, the high incidence of multiple tumors (in 44% of patients), and the aggressive behavior of some squamous cell carcinomas. Cancers of the vulva and perineum occur at a much younger age than in the general population and are preceded by a history of condyloma acuminatum in over 57% of cases. Reduction or cessation of immunosuppressive therapy is of value in some patients with PTLD or KS but carries the risk of allograft rejection. © 2000 by the National Kidney Foundation, Inc. Index Words: Transplantation; immunosuppression; cancer; lymphoma; Kaposi's sarcoma.
R
enal transplant physicians and surgeons must always be on the lookout for complications of the immunosuppressive therapy that is used to prevent and treat rejection. A major problem is a great variety of infections. In addition, each immunosuppressive drug has its own unique toxic side effects. A third problem is an increased incidence of certain cancers. This report is based on material collected by the Cincinnati Transplant Tumor Registry (CTIR).1-3 The registry was started in the fall of 1968 and gathers information from transplant centers throughout the world. Until July 1999, the registry had data on 11,966 different types of neoplasms that occurred de novo after transplantation in 11,239 solid organ allograft recipients. Excluded from this study are 2,278 malignancies that occurred in 2,207 recipients of nonrenal organs. In this report we describe the more common malignancies seen in renal allograft recipients and their clinical features and management (Table 1).
34% women, in keeping with the 2:1 ratio of male to female patients who undergo renal transplantation. 1-3
Time of Appearance of Cancers
Age and Gender of Patients
The incidence of malignancies increased with the length of follow-up posttransplantation.1-3 An Australasian study of 6,596 cadaveric renal allograft recipients showed that the percent probability of a neoplasm developing 24 years after transplantation was 66% for skin tumors, 27% for nonskin cancers, and 72% for any type of tumor.4 These exceptional figures must be interpreted with caution because most malignancies were skin cancers (which are very common in Australia), and the number of 24-year survivors was small. Nevertheless, they emphasize the need to follow up with transplant patients indefinitely. Review of the CTTR database showed that neoplasms occurred a relatively short time after transplantation, with Kaposi's sarcoma (KS) appearing at an average of 21 (median, 13) months posttransplantation, lymphomas at an average of 35 (median, 13) months, skin
The tumors occurred in a relatively young group of patients, whose average age at the time of transplantation was 43 years (range, 8 days to 80 years).1-3 At the time of transplantation, 40% percent were younger than 40 years. The average age of the patients at the time of diagnosis of their neoplasms was 48 years. Sixty-six percent of patients were men and
From the Department of Surgery, Transplantation Division, University of Cincinnati Medical Center, Cincinnati, OH. tDeceased. Supported in part by a grant from Department of Veterans Affairs. © 2000 by the National Kidney Foundation, Inc. 1073-4449/00/0702-0004$3.00/0 doi:10.1053/rr.2000.5269
Advances in Renal Replacement Therapy, Vol 7, No 2 (April), 2000: pp 147-156
147
148
Israel Penn
Table 1. 9,688 De Novo Cancers in Renal Allograft Recipients
Type of Neoplasm Cancers of skin and lips PTLD Carcinomas of the lung Kaposi's sarcoma Carcinomas of uterus (cervix, 343; body, 59; unspecified, 4) Carcinomas of the kidney (host kidney, 326; allograft kidney, 45; unspecified, 21; both, 1) Carcinomas of colon and rectum Carcinomas of the breast Carcinomas of the vulva, perineum, penis, scrotum Carcinomas of the head and neck (excluding thyroid, parathyroid, and eye) Carcinomas of urinary bladder Metastatic carcinoma (primary site unknown) Carcinomas of prostate gland Leukemias Hepatobiliary carcinomas Carcinomas of thyroid gland Cancers of stomach Sarcomas (excluding Kaposi's sarcoma) Testicular carcinomas Ovarian cancers Carcinomas of pancreas Miscellaneous tumors
No. of Tumors* 3,897 1,108 515 422 406
393 342 330 272 269 236 217 174 174 170 129
125
117
75 68 67 182
'There were 9,032 patients of whom 616 (6.8%) had 2 or more distinct tumor types involving different organ systems. Of these, 38 patients each had 3 separate types of cancer and 1 had 4.
tumors at an average of 66 (median, 51) months, whereas vulvar and perineal carcinomas appeared at the longest time posttransplantation, at an average of 114 (median, 113) months. I -3If all malignancies were considered, the average time of their appearance was 63 (median, 47) months.
Types of Posttransplant Cancers Overall there was a 3- to 5-fold increased risk of neoplasia compared with age-matched controls in the general population. s The incidence of skin cancer was increased 4- to 2l-fold, the highest rise being in areas of the world with high sunshine exposure.1-8 Apart from cutaneous malignancies, renal allograft recipients did not experience an increased incidence of the cancers that are common in the general
population (carcinomas of the lung, breast, prostate, colon, and invasive carcinomas of the uterine cervix).1-3 Instead, they developed a variety of mostly unusual tumors. Epidemiological studies showed increases of 28- to 49-fold of posttransplant lymphoproliferative disease (PTLD-compared with lymphomas in the general population),5 29-fold of lip carcinomas,6 400- to SOO-fold of KS,9 100-fold of vulvar and anal carcinomas,6 20- to 38-fold of hepatocellular carcinomas,5,lO 14- to l6-fold of in situ uterine cervical carcinomas,l,1l-12 and small increases in sarcomas (excluding KS) 13,14 and renal carcinomas. 15 Cancers of the Skin and Lips Cancers of the skin and lips were the most common malignancies in the CTTR comprising 40% of all cancers. Of the 3,897 patients, 3,468 (89%) had skin tumors, 195 (5%) had lip lesions, and 234 (6%) had cancers of the skin and lips. Skin cancers occurred on sun-exposed areas, particularly of the head, neck, and upper extremities.1-3 Light-skinned individuals with blue eyes and blond or red hair were particularly prone to these neoplasms. Human leukocyte antigens (HLA), which play an important role in host defense against certain malignancies, may have contributed to the development of skin cancer in some patients. 7 An association between HLA-B mismatching and HLA homozygosity and the development of skin cancer, reported in Dutch patients,16 could not be confirmed in a study of Australian patients. 17 The risk of skin cancer rose with the degree of sunshine exposure. I -8 In regions with limited exposure, there was a 4- to 7-fold increase, but in areas with copious sunshine there was an almost 2l-fold increase over the already high incidence observed in the local population. Almost all the increase was in squamous cell carcinomas (SCCs). However, exposure to sunshine was not the only etiologic factor. A surprisingly high incidence of SCCs was recorded from regions with relatively little sunlight in Canada, Sweden, and Great Britain, and possibly may have been related to malignant change in papillomavirus-induced warts, under the influence of immunosuppression, sunlight, and possibly other factors.1- 3,8 For
Cancers in Renal Transplant Recipients
example, in a British study of 291 renal transplant recipients, 59% had cutaneous warts, and 22% had nonmelanoma skin cancers. 8 However, the role of papillomavirus in causing skin cancers in transplant patients currently is the subject of controversy. The prevalence of cutaneous malignancies increases with the length of follow-up after transplantation, as demonstrated by an Australasian study of 6,596 cadaveric renal allograft recipients who experienced a linear increase in the prevalence of skin cancer, reaching 66% at 24 years posttransplantation. 4 Similarly, a Dutch study showed a 10% prevalence of nonmelanoma skin cancers in renal transplant recipients 10 years after transplantation, which rose to 40% after 20 years? In renal transplant recipients skin cancers showed several unusual features compared with their counterparts in the general population. 14,7,8 Basal cell carcinomas (BCCs) outnumber SCCs in the general population by 5 to 1, but the opposite occurred in transplant recipients, in whom SCCs outnumbered BCCs by 1.8 to 1. In the population at large, nonmelanoma skin cancers occur predominantly in people in their 60s and 70s, whereas affected transplant patients' average ages were 30 years younger.1-3 The incidence of multiple skin cancers (present in at least 44%) was remarkably high and was comparable to that seen only in individuals in the general population who live in areas of abundant sunshine. Some patients each had more than 100 skin tumors. Malignant melanomas comprised 4.8% of cutaneous neoplasms, in contrast to 2.7% of skin cancers in the general population of the United States. In the general population most lymph node metastases and deaths from skin cancers are caused by melanomas. In contrast, SCCs were much more aggressive in transplant patients than the general population and accounted for the majority of lymph node metastases and deaths from skin cancer.1-3 Thus, 5.8% of patients with skin or lip tumors in the CTTR had lymph node metastases. Of these 73% were from SCCs and only 18% from melanomas. Similarly, 5% of patients died of skin or lip cancers, with 60% of deaths being from SCC and only 30% from melanomas. 1-3 Of 39 patients with Merkel's cell tumors (a neuroendocrine tumor of the skin) 62% had lymph node
149
metastases, and 49% died of their malignancies. Lymphomas and Lymphoproliferations These tumors were second only to skin cancers in their frequency in transplant patients, but are much more important because they are the major cause of cancer-related mortality and morbidity. Since we described the first lymphomas in 1968P a great deal of confusion has arisen about the nomenclature of these lesions. In consequence, the nonspecific term posttransplant lymphoproliferative disease (PTLD) has been used widely.19,2o This covers a very broad spectrum of disorders ranging from benign hyperplasia at one end of the spectrum to frankly malignant lymphomas at the other extreme. Less than 3% of cases of PTLD in the CTTR were Hodgkin's disease, whereas it comprises 10% of lymphomas in the general population. Similarly, plasmacytoma and myeloma comprised less than 4% of PTLDs in the CTTR, whereas they comprise 19% of lymphomas in the general population. Of PTLDs in the CTTR that were studied immunologically, 85% were of B-cell origin, 15% were of T-cell origin, and rare cases were of null cell origin or were combined B- and T-cell lymphomas. 1-3 There are several risk factors for the development of PTLD. Intense immunosuppression is a major factor.1- 3,19-21 Often 3, 4, or even 5 immunosuppressive agents are administered over a short time period. Whenever a new immunosuppressive agent is introduced there is a "learning curve" while we discover how to use it in appropriate doses, especially when it is added to combinations of other immunosuppressive medications.1-3 Nonrenal allograft recipients (such as heart or lungs) are much more prone to PTLD than renal recipients. Heavy immunosuppressive therapy often is used in the former group to reverse rejection to save their lives, whereas with severe rejection of kidney allografts physicians have the option to discontinue immunosuppression and return the patients to dialysis therapy.1-3 These findings are reinforced by reports that whereas PTLD occurs in less than 1% of renal recipients, it involves 3% of heart, 3% of liver, 8% of lung, and 19% of intestinal recipients. 19 More than 90% of PTLDs are positive for EpsteinBarr virus, including some T-cell PTLDs.1-3,19 A
150
Israel Penn
risk factor for development of PTLD is seronegative status at the time of the transplantation. 21 Some investigators suggest that high viral loads of EBV in the peripheral blood may warn of the development PTLD.21 The CTTR data show that pediatric organ transplant recipients had a disproportionately high incidence of PTLD compared with adults. 1-3 A major reason is that primary EBV infections are more common in children than in adults, and in immunosuppressed patients they are likely to result in PTLD. Another reason why PTLD is more common in pediatric recipients is that children have more lymphoid tissue that may undergo malignant change when exposed to the appropriate stimuli. PTLD cases in the CTTR involved multiple organs or sites in 53% of cases, whereas 47% were confined to a single organ or site. Although palpable lymph nodes were present in many patients, PTLD differed from nonHodgkin's lymphoma in the general population in that extranodal disease was much more common, occurring in 70% of patients. Surprisingly, one of the most frequent extranodal sites was the central nervous system, which was involved in 20% of patients with PTLD.1-3,22 The brain was usually involved, whereas the spinal cord was rarely affected. Another remarkable finding was the frequency of either macroscopic or microscopic allograft involvement, which occurred in 23% patients with PTLD.1-3 In some individuals with renal allografts, the lymphomatous infiltrate was misdiagnosed as rejection when results of biopsies, done because of graft dysfunction, were studied microscopically. This resulted in erroneous therapeutic decisions as immunosuppressive therapy was intensified, whereas a major treatment of PTLD is reduction of dosage. 23 The clinical presentation of PTLD was remarkably variable. 1-3,19 Some patients were completely asymptomatic or had features resembling infectious mononucleosis. Other presenting features included fever, night sweats, upper respiratory infection, weight loss, diarrhea, abdominal pain, lymphadenopathy, and tonsillitis. Tonsillar enlargement sometimes was so extreme that emergency tracheostomy was necessary. Because the gastrointestinal tract was frequently involved, a quite common
presentation was acute perforation of an intestinal lesion causing peritonitis. Less commonly, gastrointestinal bleeding or intestinal obstruction was the initial feature. At times patients presented with lung lesions, or a renal mass, or a mass in another organ. Occasionally, the presentation imitated allograft rejection, and the diagnosis was made by finding an atypical lymphoblastic infiltrate in a biopsy specimen. Some patients with widespread PTLD presented with a confusing picture of disseminated sepsis and multiple organ failure, and the diagnosis was made only at autopsy examination. A possible CNS lymphoma should be suspected whenever a renal allograft recipient has neurological symptoms. 1-3,22 A thorough workup is necessary and may include examination of the cerebrospinal fluid, computerized axial tomography, magnetic resonance imaging, and single-photon emission computed tomography (SPECT). Such tests help to exclude other causes of neurological symptoms in these patients such as hypertensive encephalopathy, meningitis, brain abscess, or intracranial bleeding. In the CTTR series 15% of patients died without treatment, either because the diagnosis was missed or was made too late for effective therapy to be started. In 45% of these patients the diagnosis was only made at autopsy examination. Overall, regardless of whether patients were treated, 44% died of PTLD, and 20% died of other causes, although PTLD may have contributed to the deaths of some of these patients. Of treated patients, 38% had complete remissions. Most patients responded to multimodality therapy. However, in 17% of these recipients, the only treatment was reduction or cessation of immunosuppressive therapy/-3 and in 12% the only treatment was chemotherapy. Increasingly, 2 distinct types of PTLD are being recognized. 24-26 The first, and by far the biggest, group occurs early after transplantation, usually within the first and, to a lesser extent, in the second posttransplant year. The disease usually is associated with EBV infection, and children, in particular, are affected. Reduction of immunosuppressive therapy and administration of antiviral agents are 2 major treatments.
Cancers in Renal Transplant Recipients
The second type occurs 25 to 300.5 months posttransplantation. Most patients were adults, but a few were children. The lesions were often EBV negative, were invariably clonal, and some showed rearrangement of the c-myc oncogene. 25 Reduction of immunosuppressive therapy was ineffective in patients with monomorphic lesions, and patients did not respond well to chemotherapy because of infectious complications and resistant disease. 25 The etiology of these PTLDs probably is very different from early-occurring EBV-positive PTLDs.26 Kaposi's Sarcoma The frequent development of KS among renal allograft recipients (Table 1) contrasts markedly with its occurrence in the general population of the United States (before the AIDS epidemic started) when it comprised only 0.02% to 0.07% of all neoplasms.1-3,13,14 The high frequency of KS in this worldwide collection of patients is comparable to that seen in tropical Africa, where it occurs most commonly, and comprises 3% to 9% of all cancers. It is remarkable that the number of transplant patients in the CTTR with KS (n = 422) exceeded those with carcinomas of the colorectum (n = 342), breast (n = 330), or prostate (n = 174; Table 1). Apart from people with AIDS, who frequently have KS, there probably is no other large series, except possibly in tropical Africa, in which the numbers of patients with KS exceed those who have these common malignancies. KS affected men more than women in a 3:1 ratio, a figure far less than the 9:1 to 15:1 ratio seen with KS in the general population. I -3,13 Only a small percentage tested positive for HIy'14 Transplant-related KS was rare in children. KS occurred most frequently in transplant patients who were Arab, black, Italian, Jewish, Turkish, or Greek.1 4 Two studies support these findings. KS occurred in 1.6% of 820 Italian renal transplant recipients 27 and was the most common tumor in renal allograft recipients in Saudi Arabia, comprising 76% of all cancers.28 In the CTTR series 59% of patients with KS had nonvisceral involvement that was confined to the skin, conjunctiva, or oropharyngolaryngeal mucosa, and 41% had visceral disease that involved mainly the gastrointestinal
151
tract, lungs, and lymph nodes but also affected other organs.1-3,13 Of patients with nonvisceral disease, the lesions were confined to the skin in 98% of patients and to the mouth or oropharynx in 2%. The patients with visceral disease had no skin involvement in 23% of cases, but 4% had oral involvement, which provided a readily accessible site for biopsy and diagnosis of the disease. Of those with nonvisceral disease, 55% had complete remissions after treatment. Interestingly, 36% of these remissions occurred when the only treatment was a drastic reduction of immunosuppressive therapy. In patients with visceral disease, only 30% of patients had complete remissions. However, 57% of these remissions occurred in response to reduction or cessation of immunosuppressive therapy only. Fifty-three percent of patients with visceral KS died, of whom 75% died of the neoplasm. Of 39 renal allograft recipients in whom renal function was recorded after reduction or cessation of immunosuppressive therapy, 21 lost their allografts to rejection, 2 had impaired function, and 16 retained stable function.13 Renal Carcinomas
A notable finding was that 24% of patients in the CTTR series had incidentally discovered renal cancers, mostly renal cell carcinomas. These were discovered during workup for other disorders, at nephrectomy for hypertension or other reasons, during operation for some other disease, or at autopsy examination. Unlike most other posttransplant cancers that arose as complications of immunosuppressive therapy many renal carcinomas were related to the recipient's underlying kidney disease. I -3,15 Most neoplasms developed in their own diseased kidneys, although 45 of the 393 tumors (11 %) appeared in renal allografts from 2 to 258 (average 85) months after transplantation. Nine of the 45 tumors (20%) were diagnosed within 2 years of transplantation. It is possible that they may have been present in the allograft at the time of transplantation but were sufficiently small to be noticed. 15 Two predisposing causes of renal carcinomas could be identified. Analgesic nephropathy was the underlying indication for transplantation in 29 of the 393 (7%) patients. This disorder is known to cause cancers, mostly
152
Israel Penn
transitional cell carcinomas, in various parts of the urinary tract. This is borne out in the CTTR series in which 17 of 29 (59%) patients with analgesia-related renal carcinomas had similar neoplasms elsewhere in the urinary tract.1 5 Another predisposing cause of tumors in renal transplant recipients is acquired cystic disease (ACD) of their own diseased kidneys. It occurs in 30% to 95% of patients receiving long-term hemodialysis, and is complicated by renal adenocarcinoma, which is increased 30- to 40-fold over its incidence in the general population.15 With a successfully functioning transplant the ACD tends to regress, and theoretically the risk of carcinoma developing is reduced. However, cases of persistence of ACD and development of renal cell carcinoma have been reported in patients with successfully functioning renal allografts.1 5 The precise incidence of ACD-related carcinomas in renal transplant recipients is not known, but at least 18 patients in the CTTR had this underlying disease. Carcinomas of the Vulva and Perineum This group of neoplasms includes carcinomas of the vulva, perineum, scrotum, penis, perianal skin, or anus (Table 1).1-3 Women outnumbered men in a ratio of 2.5:1 in contrast with most other posttransplant malignancies in which men outnumbered women by more than 2:1. One-third of patients had in situ lesions.l-3 A disturbing feature is that recipients with invasive lesions were much younger (average age, 42 years) than their counterparts in the general population whose average age usually is between 50 and 70 years. Of 136 patients in whom information was available 78 (57%) had a history of condyloma acuminatum (genital warts), which must be regarded as a premalignant lesion. In women, multicentric lesions were quite common. These often involved several sites in the vulva, perianal area or anus, but sometimes the cervix or vagina as well.1-3 Although many patients with neoplasms of the vulva and perineum responded well to local or extensive excisions of their lesions, 12% died of their malignancies despite abdominoperineal resections or radical vulvectomies.
Carcinomas of the Cervix Carcinomas of the cervix occurred in 10% of women with posttransplantcancers (Table 1).1-3 At least 70% of patients had in situ lesions. The CTTR database shows a negligible increase in the incidence of in situ uterine cervical carcinoma compared with the general population. This is surprising in view of 2 epidemiological studies that showed a 14- to 16-fold increased incidence in renal allograft recipients. This suggests that many cases are being missed. To avoid this error, every postadolescent female kidney transplant recipient should have regular pelvic examinations and cervical smears to search for this malignancy as well as carcinomas of the vulva and perineum. 1-3 Hepatobiliary Tumors Two epidemiological studies showed a 20- to 38-fold increased incidence of hepatobiliary tumors compared with controls. 5,10 Most cases (70%) in the CTTR (Table 1) were hepatomas, and a substantial number of patients gave a preceding history of hepatitis B infection.lO Since hepatitis C screening has become available, increasing numbers of recipients with a preceding history of this viral infection are being reported. Sarcomas (Excluding KS) The majority of sarcomas involved the soft tissues or visceral organs, whereas cartilage or bone involvement was uncommon.1-3,13 The major types were fibrous histiocytoma, leiomyosarcoma, hemangiosarcoma, fibrosarcoma, (including cases of dermato fibrosarcoma), rhabdomyosarcoma, mesothelioma, liposarcoma, and synovial sarcoma. Other Cancers The Nordic Transplant Registry29 and the Australian and New Zealand Transplant Registry30 have reported an increased incidence of a broad variety of neoplasms other than those described above. These may represent regional variations in cancer incidence. However, some of the calculated increases are based on very small numbers of patients. 31 Whether there is an increased incidence of other malignancies will require epidemiological studies of large numbers of renal allograft recipients. 31
Cancers in Renal Transplant Recipients
Possible Causes of Posttransplant Neoplasms Posttransplant neoplasms probably result from a complex interplay of multiple factors that are discussed in detail elsewhere.1-3,13,14 Severely depressed immunity may impair the body's ability to eliminate cancer cells induced by various carcinogens. 1-3 Chronic antigenic stimulation caused by the foreign antigens of transplanted organs, repeated infections, or transfusions of blood or blood products may overstimulate a partially depressed immune system and lead to PTLD.1-3 Alternatively, impaired feedback mechanisms may fail to control the extent of immune reactions and lead to unrestrained lymphoid proliferation and PTLD. Furthermore, once this loss of regulation occurs, the defensive ability of the immune system is weakened, and other nonlymphoid malignancies may appear. 1-3 Nalesnik and Starzp2 believe that host-donor microchimerism may be an overlooked factor in the development of PTLD. Activation of oncogenic viruses in some immunosuppressed patients is highly likely.1-3 EBV is strongly implicated in causing lymphomas in primary immunodeficiency diseases, in patients with AIDS and PTLD in organ allograft recipients.1-3,19,20 It also may playa role in the development of some smooth muscle tumors that develop posttransplantation33 and in some cases of Hodgkin's disease.1-3 Certain papillomaviruses playa role in the etiology of carcinomas of the vulva, perineum, uterine cervix, and anus. There is much controversy whether papillomavirus play a role in the development of some skin cancers. 1-3,8 Hepatitis Band C viruses are known to give rise to hepatomas. lO Human herpes virus type 8 (HHV-8), also known as Kaposi's sarcomaassociated virus, may playa key role in the development of KS.34 Some immunosuppressive agents such as azathioprine, cyclophosphamide, and cyclosporme may directly damage DNA and cause malignancies. 1-3 Immunosuppressive agents may enhance the effects of other carcinogens, such as sunlight in causing skin cancers or papillomavirus in causing carcinomas of the uterine cervix. 1-3 Genetic factors may affect susceptibility to malignancy by affecting car-
153
cinogen metabolism, level of interferon secretion, response to virus infections, or regulation of the immune response by the major histocompatibility system. 1-3 For example, several studies have linked various HLA groups either to increased susceptibility or resistance to the development of KS. However, a CTTR study of HLA-A and HLA-B typing in 135 patients and HLA-DR typing in 67 recipients with KS showed no significant differences when the patients' ethnic backgrounds were taken into consideration. Fifty-six percent of the patients were Arab, black, Italian, Greek, or Jewish.1-3,13 The role of HLA in predisposing to non-KS skin cancers is mentioned above in the section about cutaneous malignancies.
Prophylactic Measures Measures to avoid the development of cancers include minimizing immunosuppressive therapy to a level that is compatible with good allograft function, while, at all times, realizing the risk of rejection from too little immunosuppressive therapy. Additionally, hepatitis B vaccination in nonimmune individuals may prevent HBV-related hepatomas. Avoidance of excess sun exposure, wearing protective clothing, and using sunscreen may prevent skin cancers. Use of barrier methods of contraception may prevent the development of condyloma acuminatum, which may decrease the occurrence of carcinomas of the vulva, perineum, and uterine cervix. And finally, although the currently available antiviral agents are virustatic and not virucidal, and act only on linear EBV and not the circular (latent) virus, peritransplant or perirejection administration of ganciclovir or acyclovir or related drugs may decrease the development of PTLD.1-3 One study showed that ganciclovir or acyclovir given preemptively, during administration of antilymphocyte agents, reduced the incidence of PTLD to 1 in 198 consecutive recipients, compared with an historic control group in which this disorder developed in 7 of 179 recipients. 21 Antiviral prophaxis is especially important when transplanting organs from EBV-positive donors into EBV-negative recipients, most of whom are children. 21
154
Israel Penn
Treatment of Posttransplant Cancers When treating neoplasms in renal allograft recipients, be warned that some cancers show more aggressive behavior than similar malignancies in nontransplant patients. 35 Early cancers are curable with local treatment provided that their growth has not given them adequate vascular access. Once this has occurred, the patient's depressed immune system is believed to allow greater than normal survival of tumor cells in the bloodstream. 35 The result is more rapid neoplastic dissemination and death of the patient than would be expected in a setting of immunocompetence. Treatment of skin cancers usually involves surgical excision, although cryosurgery is used by some dermatologists for superficial BCCs and keratoses. At times, skin grafting may be necessary and resurfacing of the dorsal aspects of the hands may be necessary in patients in whom multiple lesions develop in these areas. Repeated surgical excisions and multiple cryotherapy sessions may lead to considerable scarring and disfigurement. Some dermatologists recommend reduction of immunosuppressive therapy in patients with multiple skin neoplasms,36 but no convincing results of such treatments have been reported, and patients run the risk of losing their allografts. Topical retinoids may be useful to treat solar keratoses and warts, but whether they reduce the risk of skin cancer is not clear. Systemic retinoids have been used in small numbers of renal transplant recipients to try to reduce the incidence of premalignant and malignant skin lesions. However, the results have not been impressive, and retinoids have side effects including mucocutaneous xerosis, hair loss, elevated serum lipids, hepatic intolerance, and skeletal abnormalities. 36 In situ carcinomas of the uterine cervix respond well to simple hysterectomy, cervical conization, or cryotherapy.l,3 Localized PTLD may be excised successfully or treated with radiation therapy. A significant proportion of more extensive lesions have regressed partially or completely after reduction or cessation of immunosuppressive therapy.1-3,23 In particular, in recipients with widespread, extensive, or potentially lifethreatening PTLD, all immunosuppression
should be stopped except for a minimal dose of prednisone, until all evidence of the disease has disappeared. Allograft rejection may not occur or may evolve slowly in a chronic fashion, because many of these patients have been very heavily immunosuppressed, and a long time may ensue before immunocompetence is restored. Once PTLD has regressed, immunosuppressive therapy should be resumed in small doses and then gradually increased to maintenance levels, which, however, should be smaller than those given before the appearance of PTLD. Often PTLD responds to multimodality therapy, which may include excision, radiation therapy, reduction of immunosuppression, acyclovir or ganciclovir administration to control an associated EBV infection, or treatment with interferon-a. 37 Other treatments that may be beneficial include infusion of IgG, administration of monoclonal antibodies directed against B cells,38 infusion of donor T lymphocytes,39 and administration of lymphokine activated killer (LAK) cells.40 Chemotherapy usually is reserved for patients who do not respond to other measures but can be a very effective treatment.1-3 Localized KS responds well to excision, radiation therapy, or intralesional injections of chemotherapeutic agents such as bleomycin.1,3,13,14 More extensive lesions may respond to reduction or cessation of immunosuppressive therapy. Interferon-a is another useful treatment in some patients with widespread KS. KS also responds well to chemotherapy using agents such as vincristine, vinblastine, bleomycin, or etoposide.1 3,14 Attainment of complete remissions frequently requires use of various combinations of the above therapies. Malignancies, other than those mentioned above, should be treated by standard surgical, radiation, or chemotherapeutic modalities.1,3 As mentioned above, one option for treating posttransplant neoplasms is reduction or cessation of immunosuppressive therapy.1-3,23 The value of this approach is borne out by experience with inadvertently transplanted cancers, some of which regressed completely after cessation of immunosuppressive therapy and removal of a renal allograft. 2,41 As mentioned above, cessation or reduction of immunosuppressive therapy, when used by itself, resulted
Cancers in Renal Transplant Recipients
in a substantial number of complete remissions of PTLDl-3,22 and KS.1-3,13,14 However, such treatment has rarely caused regression of epithelial tumors. A drawback of this treatment is that it may precipitate allograft rejection, which may result in the return of renal allograft recipients to dialysis therapy, but nonrenal allograft recipients may die of this complication. For example, this treatment caused impaired function or allograft loss from rejection in 21 of 39 renal recipients treated for KSP Similarly, in a series of 14 renal recipients whose PTLD was treated (among other methods) by reduction or cessation of immunosuppression, 8 of 12 survivors lost their allografts. 41
In patients requiring systemic chemotherapy of widespread malignancies we must realize that most agents depress the bone marrow.1-3 It is, therefore, prudent to stop or reduce the administration of azathioprine, cyclophosphamide, or mycophenolate mofetil dosage during such treatment to avoid severe bone marrow depression. Because most cytotoxic drugs have immunosuppressive side effects, satisfactory allograft function may persist for prolonged periods. Treatment with prednisone may be continued because it is an important component of many cancer chemotherapy protocols. Because many patients, particularly with PTLD, are already heavily immunosuppressed, chemotherapeutic agents should be used with caution because some patients have died of overwhelming infections after their use. The administration of granulocyte colony stimulating factor (G-CSF) may be life saving in such individuals. When using cytotoxic therapy in renal allograft recipients, one should avoid, if possible, the use of nephrotoxic agents, such as cisplatin. Alpha-interferon has been used to treat some patients with KS, PTLDs, or other neoplasms. 1-3,37 Interferon is a potent immune modulator that increases membrane expression of class I antigens of the major histocompatibility complex, T-cell mediated cytotoxicity, and natural killer cell function; therefore, it may stimulate rejection. However, conflicting findings have been reported after its use in renal allograft recipients. A review of the literature suggests that small doses may be safe, but large doses may precipitate rejection.
155
Acknowledgment The author thanks numerous colleagues working in transplant centers throughout the world who have generously contributed data concerning their patients to the Cincinnati Transplant Tumor Registry.
References 1. Penn I: Why do immunosuppressed patients develop cancer? in Pimentel E (ed): CRC Critical Reviews in Oncogenesis. Boca Raton, FL, CRe, 1989, pp 27-52 2. Penn I: The problem of cancer in organ transplant recipients: An overview. Transplant Sci 4:23-32,1994 3. Penn I: Malignancy after immunosuppressive therapy: How can the risk be reduced? Clin Immunother 9:207-218,1995 4. Sheil AGR, Disney APS, Mathew TH, et al: De novo malignancy emerges as a major cause of morbidity and late failure in renal transplantation. Transplant Proc 25:1383-1384,1993 5. Kinlen LJ: Incidence of cancer in rheumatoid arthritis and other disorders after immunosuppressive treatment. AmJ Med 78:44-49, 1985 (Suppl1A) 6. Blohme 1, Brynger H: Malignant disease in renal transplant patients. Transplantation 39:23-35, 1985 7. Bouwes Bavinick JNB, Vermeer BJ, Van der Woude FL, et al: Relation between skin cancer and HLA antigens in renal transplant recipients. N Engl J Med 325:884887,1991 8. Barr BB, Benton Ee, McLaren K, et al: Human papilloma virus infection and skin cancer in renal allograft recipients. Lancet 1:124-129, 1989 9. Harwood AR, Osoba D, Hofstader SL, et al: Kaposi's sarcoma in recipients of renal transplants. Am J Med 67:759-765,1979 10. Schrbter GPJ, Weil R III, Penn 1, et al: Hepatocellular carcinoma associated with chronic hepatitis B virus infection after kidney transplantation. Lancet 2:381382, 1982 (letter) 11. Porreco R, Penn I, Droegemueller W, et al: Gynecologic malignancies in immunosuppressed organ homograft recipients. Obstet GynecoI45:359-364, 1975 12. Sillman F, Stanek A, Sedlis A, et al: The relationship between human papillomavirus and lower genital intraepithelial neoplasia in immunosuppressed women. Am J Obstet Gynecol150:300-308, 1984 13. Penn I: Sarcomas in organ allograft reCipients. Transplantation 60:1485-1491,1995 14. Penn I: Kaposi's sarcoma in transplant recipients. Transplantation 64:669-673,1997 15. Penn I: Primary kidney tumors before and after renal transplantation. Transplantation 59:480-485,1995 16. Bouwes Bavinck IN: Epidemiological aspects of immunosuppression: Role of exposure to sunlight and human papilloma virus on the development of skin cancer. Hum Exp Toxicol14:98, 1995 17. Bouwes Bavinck IN, Claas FJJ, Hardie D, et al: Relation between HLA Antigens and skin cancer in renal transplant recipients in Queensland, Australia. J Invest Dermatol1008:708-711, 1997
156
Israel Penn
18. Penn I, Hammond W, Brettschneider L, et al: Malignant lymphomas in transplantation patients. Transplant Proc 1:106-112, 1969 19. Nalesnik MA, Starzl TE: Epstein-Barr virus, infectious mononucleosis, and posttransplant lymphoproliferative disorders. Transplant Sci 4:61-79,1994 20. Hanto DW: Classification of Epstein-Barr virusassociated posttransplant lymphoproliferative diseases: implications for understanding their pathogenesis and developing rational treatment strategies. Annual Rev Med 46:381-394,1995 21. McDiarmid SV, Jordon S, Lee G, et al: Prevention and pre-emptive therapy of post-transplant lymphoproliferative disease in pediatric liver recipients. Transplantation 12:1604-1611, 1998 22. Penn I, Porat G: Central nervous system lymphomas in organ allograft recipients. Transplantation 59:240244,1995 23. Starzl TE, Nalesnik MA, Porter KA, et al: Reversibility of lymphomas and lymphoproliferative lesions developing under cyclosporine-steroid therapy. Lancet 1:583-587,1984 24. Nalesnik MA: Clinicopathologic features of posttransplant lymphoproliferative disorders. Ann Transplant 2:33-40,1997 25. Dotti G, Fiocchi R, Motta T, et al: EBV-negative lymphoproliferative disorders in long term survivors after heart, kidney and liver transplant. Transplantation (in press) 26. Penn I: Some problems with posttransplantation lymphoproliferative disease. Transplantation (in press) 27. Montagnino G, Bencini PL, Tarantino A, et al: Clinical features and course of Kaposi's sarcoma in kidney transplant patients: Report of 13 cases. Am J Nephrol 14:121-126,1994 28. Al-Sulaiman MH, AI-Khader AA: Kaposi's sarcoma in renal transplant recipients. Transplant Science 4:46-60, 1994 29. Birkeland SA, Storm HH, Lamm LU, et al: Cancer risk after renal transplantation in the Nordiic countries 1964-1986. Int J Cancer 60:183-189, 1995
30. Sheil AGR, Disney APS, Mathew TH, et al: De novo malignancy emerges as a major cause of morbidity and late failure in renal transplantation. Transplant Proc 25:1383-1384, 1993 31. Penn I: Posttransplant malignancies. Transplantation Proc 31:1260-1262, 1999 32. Nalesnik MA, Starzl TE: On the crossroad between tolerance and posttransplant lymphomas. Curr Opinion Org Transplant 2:30-35,1997 33. Lee ES, Locker J, Nalesnik M, et al: The association of Epstein-Barr virus with smooth-muscle tumors occurring after organ transplantation. N Engl J Med 332:1925,1995 34. Kedda M-A, Margolius L, Kew Me, et al: Kaposi's sarcoma-associated herpesvirus in Kaposi's sarcoma occurring in immunosuppressed renal transplant recipients. Clin Transplant 10:429-431, 1996 35. Barrett WL, First R, Aron BS, et al: Clinical course of malignancies in renal transplant recipients. Cancer 72:2186-2189,1993 36. Euvrard S, Kanitakis J, Pouteil-Noble C, et al: Skin cancers in organ transplant recipients. Ann Transplant 22:28-32,1997 37. Davis CL, Wood BL, Sabath DE, et al: Interferon-alpha treatment of posttransplant Iymphoproliferative disorder in recipients of solid organ transplants. Transplantation 66:1770-1779, 1998 38. Benkerrou M, Durandy A, Fischer A: Therapy for transplant-related Iymphoproliferative disease. Hematol Oncol Clin North Am 7:467-475, 1993 39. Papadopoulos EB, Ladanyi M, Emanuel D, et al: Infusions of donor leukocytes to treat Epstein-Barr virus associated Iymphoproliferative disorders after allogenic bone marrow transplantation. N Engl J Med 331:679-680,1994 40. Nalesnik MA, Rao AS, Zeevi A, et al: Autologous Iymphokine-activated killer cell therapy of Iymphoproliferative disorders arising in organ transplant recipients. Transplant Proc 29:1905-1906, 1997 41. Penn I: Neoplasia: An example of plasticity of the immune response. Transplant Proc 28:2089-2093, 1996
R
enal transplant physicians and surgeons must always be on the lookout for complications of the immunosuppressive therapy that is used to prevent and treat rejection. A major problem is a great variety of infections. In addition, each immunosuppressive drug has its own unique toxic side effects. A third problem is an increased incidence of certain cancers. This report is based on material collected by the Cincinnati Transplant Tumor Registry (CTIR).1-3 The registry was started in the fall of 1968 and gathers information from transplant centers throughout the world. Until July 1999, the registry had data on 11,966 different types of neoplasms that occurred de novo after transplantation in 11,239 solid organ allograft recipients. Excluded from this study are 2,278 malignancies that occurred in 2,207 recipients of nonrenal organs. In this report we describe the more common malignancies seen in renal allograft recipients and their clinical features and management (Table 1).
34% women, in keeping with the 2:1 ratio of male to female patients who undergo renal transplantation. 1-3
Time of Appearance of Cancers
Age and Gender of Patients
The incidence of malignancies increased with the length of follow-up posttransplantation.1-3 An Australasian study of 6,596 cadaveric renal allograft recipients showed that the percent probability of a neoplasm developing 24 years after transplantation was 66% for skin tumors, 27% for nonskin cancers, and 72% for any type of tumor.4 These exceptional figures must be interpreted with caution because most malignancies were skin cancers (which are very common in Australia), and the number of 24-year survivors was small. Nevertheless, they emphasize the need to follow up with transplant patients indefinitely. Review of the CTTR database showed that neoplasms occurred a relatively short time after transplantation, with Kaposi's sarcoma (KS) appearing at an average of 21 (median, 13) months posttransplantation, lymphomas at an average of 35 (median, 13) months, skin
The tumors occurred in a relatively young group of patients, whose average age at the time of transplantation was 43 years (range, 8 days to 80 years).1-3 At the time of transplantation, 40% percent were younger than 40 years. The average age of the patients at the time of diagnosis of their neoplasms was 48 years. Sixty-six percent of patients were men and
From the Department of Surgery, Transplantation Division, University of Cincinnati Medical Center, Cincinnati, OH. tDeceased. Supported in part by a grant from Department of Veterans Affairs. © 2000 by the National Kidney Foundation, Inc. 1073-4449/00/0702-0004$3.00/0 doi:10.1053/rr.2000.5269
Advances in Renal Replacement Therapy, Vol 7, No 2 (April), 2000: pp 147-156
147
148
Israel Penn
Table 1. 9,688 De Novo Cancers in Renal Allograft Recipients
Type of Neoplasm Cancers of skin and lips PTLD Carcinomas of the lung Kaposi's sarcoma Carcinomas of uterus (cervix, 343; body, 59; unspecified, 4) Carcinomas of the kidney (host kidney, 326; allograft kidney, 45; unspecified, 21; both, 1) Carcinomas of colon and rectum Carcinomas of the breast Carcinomas of the vulva, perineum, penis, scrotum Carcinomas of the head and neck (excluding thyroid, parathyroid, and eye) Carcinomas of urinary bladder Metastatic carcinoma (primary site unknown) Carcinomas of prostate gland Leukemias Hepatobiliary carcinomas Carcinomas of thyroid gland Cancers of stomach Sarcomas (excluding Kaposi's sarcoma) Testicular carcinomas Ovarian cancers Carcinomas of pancreas Miscellaneous tumors
No. of Tumors* 3,897 1,108 515 422 406
393 342 330 272 269 236 217 174 174 170 129
125
117
75 68 67 182
'There were 9,032 patients of whom 616 (6.8%) had 2 or more distinct tumor types involving different organ systems. Of these, 38 patients each had 3 separate types of cancer and 1 had 4.
tumors at an average of 66 (median, 51) months, whereas vulvar and perineal carcinomas appeared at the longest time posttransplantation, at an average of 114 (median, 113) months. I -3If all malignancies were considered, the average time of their appearance was 63 (median, 47) months.
Types of Posttransplant Cancers Overall there was a 3- to 5-fold increased risk of neoplasia compared with age-matched controls in the general population. s The incidence of skin cancer was increased 4- to 2l-fold, the highest rise being in areas of the world with high sunshine exposure.1-8 Apart from cutaneous malignancies, renal allograft recipients did not experience an increased incidence of the cancers that are common in the general
population (carcinomas of the lung, breast, prostate, colon, and invasive carcinomas of the uterine cervix).1-3 Instead, they developed a variety of mostly unusual tumors. Epidemiological studies showed increases of 28- to 49-fold of posttransplant lymphoproliferative disease (PTLD-compared with lymphomas in the general population),5 29-fold of lip carcinomas,6 400- to SOO-fold of KS,9 100-fold of vulvar and anal carcinomas,6 20- to 38-fold of hepatocellular carcinomas,5,lO 14- to l6-fold of in situ uterine cervical carcinomas,l,1l-12 and small increases in sarcomas (excluding KS) 13,14 and renal carcinomas. 15 Cancers of the Skin and Lips Cancers of the skin and lips were the most common malignancies in the CTTR comprising 40% of all cancers. Of the 3,897 patients, 3,468 (89%) had skin tumors, 195 (5%) had lip lesions, and 234 (6%) had cancers of the skin and lips. Skin cancers occurred on sun-exposed areas, particularly of the head, neck, and upper extremities.1-3 Light-skinned individuals with blue eyes and blond or red hair were particularly prone to these neoplasms. Human leukocyte antigens (HLA), which play an important role in host defense against certain malignancies, may have contributed to the development of skin cancer in some patients. 7 An association between HLA-B mismatching and HLA homozygosity and the development of skin cancer, reported in Dutch patients,16 could not be confirmed in a study of Australian patients. 17 The risk of skin cancer rose with the degree of sunshine exposure. I -8 In regions with limited exposure, there was a 4- to 7-fold increase, but in areas with copious sunshine there was an almost 2l-fold increase over the already high incidence observed in the local population. Almost all the increase was in squamous cell carcinomas (SCCs). However, exposure to sunshine was not the only etiologic factor. A surprisingly high incidence of SCCs was recorded from regions with relatively little sunlight in Canada, Sweden, and Great Britain, and possibly may have been related to malignant change in papillomavirus-induced warts, under the influence of immunosuppression, sunlight, and possibly other factors.1- 3,8 For
Cancers in Renal Transplant Recipients
example, in a British study of 291 renal transplant recipients, 59% had cutaneous warts, and 22% had nonmelanoma skin cancers. 8 However, the role of papillomavirus in causing skin cancers in transplant patients currently is the subject of controversy. The prevalence of cutaneous malignancies increases with the length of follow-up after transplantation, as demonstrated by an Australasian study of 6,596 cadaveric renal allograft recipients who experienced a linear increase in the prevalence of skin cancer, reaching 66% at 24 years posttransplantation. 4 Similarly, a Dutch study showed a 10% prevalence of nonmelanoma skin cancers in renal transplant recipients 10 years after transplantation, which rose to 40% after 20 years? In renal transplant recipients skin cancers showed several unusual features compared with their counterparts in the general population. 14,7,8 Basal cell carcinomas (BCCs) outnumber SCCs in the general population by 5 to 1, but the opposite occurred in transplant recipients, in whom SCCs outnumbered BCCs by 1.8 to 1. In the population at large, nonmelanoma skin cancers occur predominantly in people in their 60s and 70s, whereas affected transplant patients' average ages were 30 years younger.1-3 The incidence of multiple skin cancers (present in at least 44%) was remarkably high and was comparable to that seen only in individuals in the general population who live in areas of abundant sunshine. Some patients each had more than 100 skin tumors. Malignant melanomas comprised 4.8% of cutaneous neoplasms, in contrast to 2.7% of skin cancers in the general population of the United States. In the general population most lymph node metastases and deaths from skin cancers are caused by melanomas. In contrast, SCCs were much more aggressive in transplant patients than the general population and accounted for the majority of lymph node metastases and deaths from skin cancer.1-3 Thus, 5.8% of patients with skin or lip tumors in the CTTR had lymph node metastases. Of these 73% were from SCCs and only 18% from melanomas. Similarly, 5% of patients died of skin or lip cancers, with 60% of deaths being from SCC and only 30% from melanomas. 1-3 Of 39 patients with Merkel's cell tumors (a neuroendocrine tumor of the skin) 62% had lymph node
149
metastases, and 49% died of their malignancies. Lymphomas and Lymphoproliferations These tumors were second only to skin cancers in their frequency in transplant patients, but are much more important because they are the major cause of cancer-related mortality and morbidity. Since we described the first lymphomas in 1968P a great deal of confusion has arisen about the nomenclature of these lesions. In consequence, the nonspecific term posttransplant lymphoproliferative disease (PTLD) has been used widely.19,2o This covers a very broad spectrum of disorders ranging from benign hyperplasia at one end of the spectrum to frankly malignant lymphomas at the other extreme. Less than 3% of cases of PTLD in the CTTR were Hodgkin's disease, whereas it comprises 10% of lymphomas in the general population. Similarly, plasmacytoma and myeloma comprised less than 4% of PTLDs in the CTTR, whereas they comprise 19% of lymphomas in the general population. Of PTLDs in the CTTR that were studied immunologically, 85% were of B-cell origin, 15% were of T-cell origin, and rare cases were of null cell origin or were combined B- and T-cell lymphomas. 1-3 There are several risk factors for the development of PTLD. Intense immunosuppression is a major factor.1- 3,19-21 Often 3, 4, or even 5 immunosuppressive agents are administered over a short time period. Whenever a new immunosuppressive agent is introduced there is a "learning curve" while we discover how to use it in appropriate doses, especially when it is added to combinations of other immunosuppressive medications.1-3 Nonrenal allograft recipients (such as heart or lungs) are much more prone to PTLD than renal recipients. Heavy immunosuppressive therapy often is used in the former group to reverse rejection to save their lives, whereas with severe rejection of kidney allografts physicians have the option to discontinue immunosuppression and return the patients to dialysis therapy.1-3 These findings are reinforced by reports that whereas PTLD occurs in less than 1% of renal recipients, it involves 3% of heart, 3% of liver, 8% of lung, and 19% of intestinal recipients. 19 More than 90% of PTLDs are positive for EpsteinBarr virus, including some T-cell PTLDs.1-3,19 A
150
Israel Penn
risk factor for development of PTLD is seronegative status at the time of the transplantation. 21 Some investigators suggest that high viral loads of EBV in the peripheral blood may warn of the development PTLD.21 The CTTR data show that pediatric organ transplant recipients had a disproportionately high incidence of PTLD compared with adults. 1-3 A major reason is that primary EBV infections are more common in children than in adults, and in immunosuppressed patients they are likely to result in PTLD. Another reason why PTLD is more common in pediatric recipients is that children have more lymphoid tissue that may undergo malignant change when exposed to the appropriate stimuli. PTLD cases in the CTTR involved multiple organs or sites in 53% of cases, whereas 47% were confined to a single organ or site. Although palpable lymph nodes were present in many patients, PTLD differed from nonHodgkin's lymphoma in the general population in that extranodal disease was much more common, occurring in 70% of patients. Surprisingly, one of the most frequent extranodal sites was the central nervous system, which was involved in 20% of patients with PTLD.1-3,22 The brain was usually involved, whereas the spinal cord was rarely affected. Another remarkable finding was the frequency of either macroscopic or microscopic allograft involvement, which occurred in 23% patients with PTLD.1-3 In some individuals with renal allografts, the lymphomatous infiltrate was misdiagnosed as rejection when results of biopsies, done because of graft dysfunction, were studied microscopically. This resulted in erroneous therapeutic decisions as immunosuppressive therapy was intensified, whereas a major treatment of PTLD is reduction of dosage. 23 The clinical presentation of PTLD was remarkably variable. 1-3,19 Some patients were completely asymptomatic or had features resembling infectious mononucleosis. Other presenting features included fever, night sweats, upper respiratory infection, weight loss, diarrhea, abdominal pain, lymphadenopathy, and tonsillitis. Tonsillar enlargement sometimes was so extreme that emergency tracheostomy was necessary. Because the gastrointestinal tract was frequently involved, a quite common
presentation was acute perforation of an intestinal lesion causing peritonitis. Less commonly, gastrointestinal bleeding or intestinal obstruction was the initial feature. At times patients presented with lung lesions, or a renal mass, or a mass in another organ. Occasionally, the presentation imitated allograft rejection, and the diagnosis was made by finding an atypical lymphoblastic infiltrate in a biopsy specimen. Some patients with widespread PTLD presented with a confusing picture of disseminated sepsis and multiple organ failure, and the diagnosis was made only at autopsy examination. A possible CNS lymphoma should be suspected whenever a renal allograft recipient has neurological symptoms. 1-3,22 A thorough workup is necessary and may include examination of the cerebrospinal fluid, computerized axial tomography, magnetic resonance imaging, and single-photon emission computed tomography (SPECT). Such tests help to exclude other causes of neurological symptoms in these patients such as hypertensive encephalopathy, meningitis, brain abscess, or intracranial bleeding. In the CTTR series 15% of patients died without treatment, either because the diagnosis was missed or was made too late for effective therapy to be started. In 45% of these patients the diagnosis was only made at autopsy examination. Overall, regardless of whether patients were treated, 44% died of PTLD, and 20% died of other causes, although PTLD may have contributed to the deaths of some of these patients. Of treated patients, 38% had complete remissions. Most patients responded to multimodality therapy. However, in 17% of these recipients, the only treatment was reduction or cessation of immunosuppressive therapy/-3 and in 12% the only treatment was chemotherapy. Increasingly, 2 distinct types of PTLD are being recognized. 24-26 The first, and by far the biggest, group occurs early after transplantation, usually within the first and, to a lesser extent, in the second posttransplant year. The disease usually is associated with EBV infection, and children, in particular, are affected. Reduction of immunosuppressive therapy and administration of antiviral agents are 2 major treatments.
Cancers in Renal Transplant Recipients
The second type occurs 25 to 300.5 months posttransplantation. Most patients were adults, but a few were children. The lesions were often EBV negative, were invariably clonal, and some showed rearrangement of the c-myc oncogene. 25 Reduction of immunosuppressive therapy was ineffective in patients with monomorphic lesions, and patients did not respond well to chemotherapy because of infectious complications and resistant disease. 25 The etiology of these PTLDs probably is very different from early-occurring EBV-positive PTLDs.26 Kaposi's Sarcoma The frequent development of KS among renal allograft recipients (Table 1) contrasts markedly with its occurrence in the general population of the United States (before the AIDS epidemic started) when it comprised only 0.02% to 0.07% of all neoplasms.1-3,13,14 The high frequency of KS in this worldwide collection of patients is comparable to that seen in tropical Africa, where it occurs most commonly, and comprises 3% to 9% of all cancers. It is remarkable that the number of transplant patients in the CTTR with KS (n = 422) exceeded those with carcinomas of the colorectum (n = 342), breast (n = 330), or prostate (n = 174; Table 1). Apart from people with AIDS, who frequently have KS, there probably is no other large series, except possibly in tropical Africa, in which the numbers of patients with KS exceed those who have these common malignancies. KS affected men more than women in a 3:1 ratio, a figure far less than the 9:1 to 15:1 ratio seen with KS in the general population. I -3,13 Only a small percentage tested positive for HIy'14 Transplant-related KS was rare in children. KS occurred most frequently in transplant patients who were Arab, black, Italian, Jewish, Turkish, or Greek.1 4 Two studies support these findings. KS occurred in 1.6% of 820 Italian renal transplant recipients 27 and was the most common tumor in renal allograft recipients in Saudi Arabia, comprising 76% of all cancers.28 In the CTTR series 59% of patients with KS had nonvisceral involvement that was confined to the skin, conjunctiva, or oropharyngolaryngeal mucosa, and 41% had visceral disease that involved mainly the gastrointestinal
151
tract, lungs, and lymph nodes but also affected other organs.1-3,13 Of patients with nonvisceral disease, the lesions were confined to the skin in 98% of patients and to the mouth or oropharynx in 2%. The patients with visceral disease had no skin involvement in 23% of cases, but 4% had oral involvement, which provided a readily accessible site for biopsy and diagnosis of the disease. Of those with nonvisceral disease, 55% had complete remissions after treatment. Interestingly, 36% of these remissions occurred when the only treatment was a drastic reduction of immunosuppressive therapy. In patients with visceral disease, only 30% of patients had complete remissions. However, 57% of these remissions occurred in response to reduction or cessation of immunosuppressive therapy only. Fifty-three percent of patients with visceral KS died, of whom 75% died of the neoplasm. Of 39 renal allograft recipients in whom renal function was recorded after reduction or cessation of immunosuppressive therapy, 21 lost their allografts to rejection, 2 had impaired function, and 16 retained stable function.13 Renal Carcinomas
A notable finding was that 24% of patients in the CTTR series had incidentally discovered renal cancers, mostly renal cell carcinomas. These were discovered during workup for other disorders, at nephrectomy for hypertension or other reasons, during operation for some other disease, or at autopsy examination. Unlike most other posttransplant cancers that arose as complications of immunosuppressive therapy many renal carcinomas were related to the recipient's underlying kidney disease. I -3,15 Most neoplasms developed in their own diseased kidneys, although 45 of the 393 tumors (11 %) appeared in renal allografts from 2 to 258 (average 85) months after transplantation. Nine of the 45 tumors (20%) were diagnosed within 2 years of transplantation. It is possible that they may have been present in the allograft at the time of transplantation but were sufficiently small to be noticed. 15 Two predisposing causes of renal carcinomas could be identified. Analgesic nephropathy was the underlying indication for transplantation in 29 of the 393 (7%) patients. This disorder is known to cause cancers, mostly
152
Israel Penn
transitional cell carcinomas, in various parts of the urinary tract. This is borne out in the CTTR series in which 17 of 29 (59%) patients with analgesia-related renal carcinomas had similar neoplasms elsewhere in the urinary tract.1 5 Another predisposing cause of tumors in renal transplant recipients is acquired cystic disease (ACD) of their own diseased kidneys. It occurs in 30% to 95% of patients receiving long-term hemodialysis, and is complicated by renal adenocarcinoma, which is increased 30- to 40-fold over its incidence in the general population.15 With a successfully functioning transplant the ACD tends to regress, and theoretically the risk of carcinoma developing is reduced. However, cases of persistence of ACD and development of renal cell carcinoma have been reported in patients with successfully functioning renal allografts.1 5 The precise incidence of ACD-related carcinomas in renal transplant recipients is not known, but at least 18 patients in the CTTR had this underlying disease. Carcinomas of the Vulva and Perineum This group of neoplasms includes carcinomas of the vulva, perineum, scrotum, penis, perianal skin, or anus (Table 1).1-3 Women outnumbered men in a ratio of 2.5:1 in contrast with most other posttransplant malignancies in which men outnumbered women by more than 2:1. One-third of patients had in situ lesions.l-3 A disturbing feature is that recipients with invasive lesions were much younger (average age, 42 years) than their counterparts in the general population whose average age usually is between 50 and 70 years. Of 136 patients in whom information was available 78 (57%) had a history of condyloma acuminatum (genital warts), which must be regarded as a premalignant lesion. In women, multicentric lesions were quite common. These often involved several sites in the vulva, perianal area or anus, but sometimes the cervix or vagina as well.1-3 Although many patients with neoplasms of the vulva and perineum responded well to local or extensive excisions of their lesions, 12% died of their malignancies despite abdominoperineal resections or radical vulvectomies.
Carcinomas of the Cervix Carcinomas of the cervix occurred in 10% of women with posttransplantcancers (Table 1).1-3 At least 70% of patients had in situ lesions. The CTTR database shows a negligible increase in the incidence of in situ uterine cervical carcinoma compared with the general population. This is surprising in view of 2 epidemiological studies that showed a 14- to 16-fold increased incidence in renal allograft recipients. This suggests that many cases are being missed. To avoid this error, every postadolescent female kidney transplant recipient should have regular pelvic examinations and cervical smears to search for this malignancy as well as carcinomas of the vulva and perineum. 1-3 Hepatobiliary Tumors Two epidemiological studies showed a 20- to 38-fold increased incidence of hepatobiliary tumors compared with controls. 5,10 Most cases (70%) in the CTTR (Table 1) were hepatomas, and a substantial number of patients gave a preceding history of hepatitis B infection.lO Since hepatitis C screening has become available, increasing numbers of recipients with a preceding history of this viral infection are being reported. Sarcomas (Excluding KS) The majority of sarcomas involved the soft tissues or visceral organs, whereas cartilage or bone involvement was uncommon.1-3,13 The major types were fibrous histiocytoma, leiomyosarcoma, hemangiosarcoma, fibrosarcoma, (including cases of dermato fibrosarcoma), rhabdomyosarcoma, mesothelioma, liposarcoma, and synovial sarcoma. Other Cancers The Nordic Transplant Registry29 and the Australian and New Zealand Transplant Registry30 have reported an increased incidence of a broad variety of neoplasms other than those described above. These may represent regional variations in cancer incidence. However, some of the calculated increases are based on very small numbers of patients. 31 Whether there is an increased incidence of other malignancies will require epidemiological studies of large numbers of renal allograft recipients. 31
Cancers in Renal Transplant Recipients
Possible Causes of Posttransplant Neoplasms Posttransplant neoplasms probably result from a complex interplay of multiple factors that are discussed in detail elsewhere.1-3,13,14 Severely depressed immunity may impair the body's ability to eliminate cancer cells induced by various carcinogens. 1-3 Chronic antigenic stimulation caused by the foreign antigens of transplanted organs, repeated infections, or transfusions of blood or blood products may overstimulate a partially depressed immune system and lead to PTLD.1-3 Alternatively, impaired feedback mechanisms may fail to control the extent of immune reactions and lead to unrestrained lymphoid proliferation and PTLD. Furthermore, once this loss of regulation occurs, the defensive ability of the immune system is weakened, and other nonlymphoid malignancies may appear. 1-3 Nalesnik and Starzp2 believe that host-donor microchimerism may be an overlooked factor in the development of PTLD. Activation of oncogenic viruses in some immunosuppressed patients is highly likely.1-3 EBV is strongly implicated in causing lymphomas in primary immunodeficiency diseases, in patients with AIDS and PTLD in organ allograft recipients.1-3,19,20 It also may playa role in the development of some smooth muscle tumors that develop posttransplantation33 and in some cases of Hodgkin's disease.1-3 Certain papillomaviruses playa role in the etiology of carcinomas of the vulva, perineum, uterine cervix, and anus. There is much controversy whether papillomavirus play a role in the development of some skin cancers. 1-3,8 Hepatitis Band C viruses are known to give rise to hepatomas. lO Human herpes virus type 8 (HHV-8), also known as Kaposi's sarcomaassociated virus, may playa key role in the development of KS.34 Some immunosuppressive agents such as azathioprine, cyclophosphamide, and cyclosporme may directly damage DNA and cause malignancies. 1-3 Immunosuppressive agents may enhance the effects of other carcinogens, such as sunlight in causing skin cancers or papillomavirus in causing carcinomas of the uterine cervix. 1-3 Genetic factors may affect susceptibility to malignancy by affecting car-
153
cinogen metabolism, level of interferon secretion, response to virus infections, or regulation of the immune response by the major histocompatibility system. 1-3 For example, several studies have linked various HLA groups either to increased susceptibility or resistance to the development of KS. However, a CTTR study of HLA-A and HLA-B typing in 135 patients and HLA-DR typing in 67 recipients with KS showed no significant differences when the patients' ethnic backgrounds were taken into consideration. Fifty-six percent of the patients were Arab, black, Italian, Greek, or Jewish.1-3,13 The role of HLA in predisposing to non-KS skin cancers is mentioned above in the section about cutaneous malignancies.
Prophylactic Measures Measures to avoid the development of cancers include minimizing immunosuppressive therapy to a level that is compatible with good allograft function, while, at all times, realizing the risk of rejection from too little immunosuppressive therapy. Additionally, hepatitis B vaccination in nonimmune individuals may prevent HBV-related hepatomas. Avoidance of excess sun exposure, wearing protective clothing, and using sunscreen may prevent skin cancers. Use of barrier methods of contraception may prevent the development of condyloma acuminatum, which may decrease the occurrence of carcinomas of the vulva, perineum, and uterine cervix. And finally, although the currently available antiviral agents are virustatic and not virucidal, and act only on linear EBV and not the circular (latent) virus, peritransplant or perirejection administration of ganciclovir or acyclovir or related drugs may decrease the development of PTLD.1-3 One study showed that ganciclovir or acyclovir given preemptively, during administration of antilymphocyte agents, reduced the incidence of PTLD to 1 in 198 consecutive recipients, compared with an historic control group in which this disorder developed in 7 of 179 recipients. 21 Antiviral prophaxis is especially important when transplanting organs from EBV-positive donors into EBV-negative recipients, most of whom are children. 21
154
Israel Penn
Treatment of Posttransplant Cancers When treating neoplasms in renal allograft recipients, be warned that some cancers show more aggressive behavior than similar malignancies in nontransplant patients. 35 Early cancers are curable with local treatment provided that their growth has not given them adequate vascular access. Once this has occurred, the patient's depressed immune system is believed to allow greater than normal survival of tumor cells in the bloodstream. 35 The result is more rapid neoplastic dissemination and death of the patient than would be expected in a setting of immunocompetence. Treatment of skin cancers usually involves surgical excision, although cryosurgery is used by some dermatologists for superficial BCCs and keratoses. At times, skin grafting may be necessary and resurfacing of the dorsal aspects of the hands may be necessary in patients in whom multiple lesions develop in these areas. Repeated surgical excisions and multiple cryotherapy sessions may lead to considerable scarring and disfigurement. Some dermatologists recommend reduction of immunosuppressive therapy in patients with multiple skin neoplasms,36 but no convincing results of such treatments have been reported, and patients run the risk of losing their allografts. Topical retinoids may be useful to treat solar keratoses and warts, but whether they reduce the risk of skin cancer is not clear. Systemic retinoids have been used in small numbers of renal transplant recipients to try to reduce the incidence of premalignant and malignant skin lesions. However, the results have not been impressive, and retinoids have side effects including mucocutaneous xerosis, hair loss, elevated serum lipids, hepatic intolerance, and skeletal abnormalities. 36 In situ carcinomas of the uterine cervix respond well to simple hysterectomy, cervical conization, or cryotherapy.l,3 Localized PTLD may be excised successfully or treated with radiation therapy. A significant proportion of more extensive lesions have regressed partially or completely after reduction or cessation of immunosuppressive therapy.1-3,23 In particular, in recipients with widespread, extensive, or potentially lifethreatening PTLD, all immunosuppression
should be stopped except for a minimal dose of prednisone, until all evidence of the disease has disappeared. Allograft rejection may not occur or may evolve slowly in a chronic fashion, because many of these patients have been very heavily immunosuppressed, and a long time may ensue before immunocompetence is restored. Once PTLD has regressed, immunosuppressive therapy should be resumed in small doses and then gradually increased to maintenance levels, which, however, should be smaller than those given before the appearance of PTLD. Often PTLD responds to multimodality therapy, which may include excision, radiation therapy, reduction of immunosuppression, acyclovir or ganciclovir administration to control an associated EBV infection, or treatment with interferon-a. 37 Other treatments that may be beneficial include infusion of IgG, administration of monoclonal antibodies directed against B cells,38 infusion of donor T lymphocytes,39 and administration of lymphokine activated killer (LAK) cells.40 Chemotherapy usually is reserved for patients who do not respond to other measures but can be a very effective treatment.1-3 Localized KS responds well to excision, radiation therapy, or intralesional injections of chemotherapeutic agents such as bleomycin.1,3,13,14 More extensive lesions may respond to reduction or cessation of immunosuppressive therapy. Interferon-a is another useful treatment in some patients with widespread KS. KS also responds well to chemotherapy using agents such as vincristine, vinblastine, bleomycin, or etoposide.1 3,14 Attainment of complete remissions frequently requires use of various combinations of the above therapies. Malignancies, other than those mentioned above, should be treated by standard surgical, radiation, or chemotherapeutic modalities.1,3 As mentioned above, one option for treating posttransplant neoplasms is reduction or cessation of immunosuppressive therapy.1-3,23 The value of this approach is borne out by experience with inadvertently transplanted cancers, some of which regressed completely after cessation of immunosuppressive therapy and removal of a renal allograft. 2,41 As mentioned above, cessation or reduction of immunosuppressive therapy, when used by itself, resulted
Cancers in Renal Transplant Recipients
in a substantial number of complete remissions of PTLDl-3,22 and KS.1-3,13,14 However, such treatment has rarely caused regression of epithelial tumors. A drawback of this treatment is that it may precipitate allograft rejection, which may result in the return of renal allograft recipients to dialysis therapy, but nonrenal allograft recipients may die of this complication. For example, this treatment caused impaired function or allograft loss from rejection in 21 of 39 renal recipients treated for KSP Similarly, in a series of 14 renal recipients whose PTLD was treated (among other methods) by reduction or cessation of immunosuppression, 8 of 12 survivors lost their allografts. 41
In patients requiring systemic chemotherapy of widespread malignancies we must realize that most agents depress the bone marrow.1-3 It is, therefore, prudent to stop or reduce the administration of azathioprine, cyclophosphamide, or mycophenolate mofetil dosage during such treatment to avoid severe bone marrow depression. Because most cytotoxic drugs have immunosuppressive side effects, satisfactory allograft function may persist for prolonged periods. Treatment with prednisone may be continued because it is an important component of many cancer chemotherapy protocols. Because many patients, particularly with PTLD, are already heavily immunosuppressed, chemotherapeutic agents should be used with caution because some patients have died of overwhelming infections after their use. The administration of granulocyte colony stimulating factor (G-CSF) may be life saving in such individuals. When using cytotoxic therapy in renal allograft recipients, one should avoid, if possible, the use of nephrotoxic agents, such as cisplatin. Alpha-interferon has been used to treat some patients with KS, PTLDs, or other neoplasms. 1-3,37 Interferon is a potent immune modulator that increases membrane expression of class I antigens of the major histocompatibility complex, T-cell mediated cytotoxicity, and natural killer cell function; therefore, it may stimulate rejection. However, conflicting findings have been reported after its use in renal allograft recipients. A review of the literature suggests that small doses may be safe, but large doses may precipitate rejection.
155
Acknowledgment The author thanks numerous colleagues working in transplant centers throughout the world who have generously contributed data concerning their patients to the Cincinnati Transplant Tumor Registry.
References 1. Penn I: Why do immunosuppressed patients develop cancer? in Pimentel E (ed): CRC Critical Reviews in Oncogenesis. Boca Raton, FL, CRe, 1989, pp 27-52 2. Penn I: The problem of cancer in organ transplant recipients: An overview. Transplant Sci 4:23-32,1994 3. Penn I: Malignancy after immunosuppressive therapy: How can the risk be reduced? Clin Immunother 9:207-218,1995 4. Sheil AGR, Disney APS, Mathew TH, et al: De novo malignancy emerges as a major cause of morbidity and late failure in renal transplantation. Transplant Proc 25:1383-1384,1993 5. Kinlen LJ: Incidence of cancer in rheumatoid arthritis and other disorders after immunosuppressive treatment. AmJ Med 78:44-49, 1985 (Suppl1A) 6. Blohme 1, Brynger H: Malignant disease in renal transplant patients. Transplantation 39:23-35, 1985 7. Bouwes Bavinick JNB, Vermeer BJ, Van der Woude FL, et al: Relation between skin cancer and HLA antigens in renal transplant recipients. N Engl J Med 325:884887,1991 8. Barr BB, Benton Ee, McLaren K, et al: Human papilloma virus infection and skin cancer in renal allograft recipients. Lancet 1:124-129, 1989 9. Harwood AR, Osoba D, Hofstader SL, et al: Kaposi's sarcoma in recipients of renal transplants. Am J Med 67:759-765,1979 10. Schrbter GPJ, Weil R III, Penn 1, et al: Hepatocellular carcinoma associated with chronic hepatitis B virus infection after kidney transplantation. Lancet 2:381382, 1982 (letter) 11. Porreco R, Penn I, Droegemueller W, et al: Gynecologic malignancies in immunosuppressed organ homograft recipients. Obstet GynecoI45:359-364, 1975 12. Sillman F, Stanek A, Sedlis A, et al: The relationship between human papillomavirus and lower genital intraepithelial neoplasia in immunosuppressed women. Am J Obstet Gynecol150:300-308, 1984 13. Penn I: Sarcomas in organ allograft reCipients. Transplantation 60:1485-1491,1995 14. Penn I: Kaposi's sarcoma in transplant recipients. Transplantation 64:669-673,1997 15. Penn I: Primary kidney tumors before and after renal transplantation. Transplantation 59:480-485,1995 16. Bouwes Bavinck IN: Epidemiological aspects of immunosuppression: Role of exposure to sunlight and human papilloma virus on the development of skin cancer. Hum Exp Toxicol14:98, 1995 17. Bouwes Bavinck IN, Claas FJJ, Hardie D, et al: Relation between HLA Antigens and skin cancer in renal transplant recipients in Queensland, Australia. J Invest Dermatol1008:708-711, 1997
156
Israel Penn
18. Penn I, Hammond W, Brettschneider L, et al: Malignant lymphomas in transplantation patients. Transplant Proc 1:106-112, 1969 19. Nalesnik MA, Starzl TE: Epstein-Barr virus, infectious mononucleosis, and posttransplant lymphoproliferative disorders. Transplant Sci 4:61-79,1994 20. Hanto DW: Classification of Epstein-Barr virusassociated posttransplant lymphoproliferative diseases: implications for understanding their pathogenesis and developing rational treatment strategies. Annual Rev Med 46:381-394,1995 21. McDiarmid SV, Jordon S, Lee G, et al: Prevention and pre-emptive therapy of post-transplant lymphoproliferative disease in pediatric liver recipients. Transplantation 12:1604-1611, 1998 22. Penn I, Porat G: Central nervous system lymphomas in organ allograft recipients. Transplantation 59:240244,1995 23. Starzl TE, Nalesnik MA, Porter KA, et al: Reversibility of lymphomas and lymphoproliferative lesions developing under cyclosporine-steroid therapy. Lancet 1:583-587,1984 24. Nalesnik MA: Clinicopathologic features of posttransplant lymphoproliferative disorders. Ann Transplant 2:33-40,1997 25. Dotti G, Fiocchi R, Motta T, et al: EBV-negative lymphoproliferative disorders in long term survivors after heart, kidney and liver transplant. Transplantation (in press) 26. Penn I: Some problems with posttransplantation lymphoproliferative disease. Transplantation (in press) 27. Montagnino G, Bencini PL, Tarantino A, et al: Clinical features and course of Kaposi's sarcoma in kidney transplant patients: Report of 13 cases. Am J Nephrol 14:121-126,1994 28. Al-Sulaiman MH, AI-Khader AA: Kaposi's sarcoma in renal transplant recipients. Transplant Science 4:46-60, 1994 29. Birkeland SA, Storm HH, Lamm LU, et al: Cancer risk after renal transplantation in the Nordiic countries 1964-1986. Int J Cancer 60:183-189, 1995
30. Sheil AGR, Disney APS, Mathew TH, et al: De novo malignancy emerges as a major cause of morbidity and late failure in renal transplantation. Transplant Proc 25:1383-1384, 1993 31. Penn I: Posttransplant malignancies. Transplantation Proc 31:1260-1262, 1999 32. Nalesnik MA, Starzl TE: On the crossroad between tolerance and posttransplant lymphomas. Curr Opinion Org Transplant 2:30-35,1997 33. Lee ES, Locker J, Nalesnik M, et al: The association of Epstein-Barr virus with smooth-muscle tumors occurring after organ transplantation. N Engl J Med 332:1925,1995 34. Kedda M-A, Margolius L, Kew Me, et al: Kaposi's sarcoma-associated herpesvirus in Kaposi's sarcoma occurring in immunosuppressed renal transplant recipients. Clin Transplant 10:429-431, 1996 35. Barrett WL, First R, Aron BS, et al: Clinical course of malignancies in renal transplant recipients. Cancer 72:2186-2189,1993 36. Euvrard S, Kanitakis J, Pouteil-Noble C, et al: Skin cancers in organ transplant recipients. Ann Transplant 22:28-32,1997 37. Davis CL, Wood BL, Sabath DE, et al: Interferon-alpha treatment of posttransplant Iymphoproliferative disorder in recipients of solid organ transplants. Transplantation 66:1770-1779, 1998 38. Benkerrou M, Durandy A, Fischer A: Therapy for transplant-related Iymphoproliferative disease. Hematol Oncol Clin North Am 7:467-475, 1993 39. Papadopoulos EB, Ladanyi M, Emanuel D, et al: Infusions of donor leukocytes to treat Epstein-Barr virus associated Iymphoproliferative disorders after allogenic bone marrow transplantation. N Engl J Med 331:679-680,1994 40. Nalesnik MA, Rao AS, Zeevi A, et al: Autologous Iymphokine-activated killer cell therapy of Iymphoproliferative disorders arising in organ transplant recipients. Transplant Proc 29:1905-1906, 1997 41. Penn I: Neoplasia: An example of plasticity of the immune response. Transplant Proc 28:2089-2093, 1996