Secondary Logo

Journal Logo

De Novo Cancers Arising in Organ Transplant Recipients are Associated With Adverse Outcomes Compared With the General Population

Miao, Yun1,2; Everly, Jason J.1; Gross, Thomas G.3; Tevar, Amit D.1; First, M Roy4; Alloway, Rita R.5; Woodle, E Steve1,6

doi: 10.1097/TP.0b013e3181a238f6
Clinical and Translational Research
Free
SDC

Background. Transplant recipients are at increased risk of malignancy; however, the influence of transplantation on cancer outcomes has not been rigorously defined. The purpose of this study was to examine the influence of transplantation on the outcomes of individual cancers.

Methods. De novo nonsmall cell lung cancer, colon cancer, breast cancer, prostate cancer, bladder cancer, renal cell cancer (RCC), and malignant melanoma data in 635 adult (>18 years of age) transplant recipients (from the Israel Penn International Transplant Tumor Registry) were compared with data from 1,282,984 adults in the general population (from the Surveillance, Epidemiology, and End Results database).

Results. Compared with the general population, transplant patients were more likely to have early stage (AJCC stage 0–II) RCC, but more advanced (AJCC stage >II) colon cancer, breast cancer, bladder cancer, and malignant melanoma. Compared with the general population, disease-specific survival was worse in the transplant population for colon cancer (all stages), nonsmall cell lung cancer (stage II), breast cancer (stage III), prostate cancer (stage II, III, and IV), bladder cancer (stage III), and RCC (stage IV). Multivariate analyses demonstrated transplantation to be a negative risk factor for survival for each cancer studied, and transplantation and cancer stage at diagnosis to be the most profound negative survival predictors.

Conclusions. These analyses indicate that, for several common cancers, transplant patients experience worse outcomes than the general population. The data also suggest that cancers in transplant recipients are more aggressive biologically at the time of diagnosis.

1Division of Transplantation, Department of Surgery, University of Cincinnati College of Medicine, Cincinnati, OH.

2Department of Transplantation, Nan Fang Hospital, Southern Medical University, Guangzhou, Guangdong, People’s Republic of China.

3Division of Hematology and Oncology, Department of Pediatrics, The Ohio State University College of Medicine, Columbus, OH.

4Research & Development, Astellas Pharma US, Inc., Deerfield, IL.

5Division of Hypertension/Nephrology, Department of Internal Medicine, University of Cincinnati College of Medicine, Cincinnati, OH.

6Address correspondence to: E. Steve Woodle, M.D., Division of Transplantation, Department of Surgery, University of Cincinnati, 231 Albert Sabin Way, ML 558, Cincinnati, OH 45267-0558.

E-mail: woodlees@uc.edu

Received 18 November 2008. Revision requested 8 December 2008.

Accepted 29 December 2008.

Previous studies have quantitated the increased risk of cancer after transplantation (1–3). These studies have demonstrated that the greatest increase in risk is observed for virally driven cancers such as posttransplant lymphoproliferative disorders (Epstein-Barr virus), Kaposi sarcoma (human herpes virus-8), and vulvar cancer (human papillomavirus). In contrast, lower risks have been observed for other more common cancers that are not virally driven, including breast (4–9), colon (10–16), prostate (17–22), and lung cancer (23–25). Defining these risks and is requisite for preventing and managing of posttransplant malignancies.

Formulation of evidence-based approaches for prevention, screening, and management of cancers in organ transplant recipients requires additional data including: temporal aspects of cancer risk, outcomes of individual cancers, disease-specific survival, cancer stage at diagnosis, and factors that influence death risk. In addition, application of therapies developed in the general population to cancers that develop in immunosuppressed patients requires comparison of data between transplant and general populations. To our knowledge, studies focused on cancer outcomes are limited, and studies with comparisons to outcomes in the general population are nonexistent. The purpose of this study was to define these cancer-related data in transplant recipients and compare these results to the general population, and in doing so, to gain insight into outcomes of cancer therapy and the effect of immunosuppression on the biologic behavior of cancer.

Several registries contain data on cancers in transplant recipients; however, these data are generally not detailed, and contain considerably fewer subjects than general population-based cancer databases. In contrast, the Israel Penn International Transplant Tumor Registry (IPITTR) is unique in that it combines large numbers of transplant patients with cancer (more than 19,000 subjects at present) with a high degree of detailed information on individual cancers and patients. The following de novo cancers (i.e., cancers that first arose after transplantation) were selected for analysis as they provided comparatively large numbers of transplant recipients: nonsmall cell lung cancer (NSCLC), colon cancer, breast cancer, prostate cancer, bladder cancer, renal cell cancer (RCC), and malignant melanoma.

Back to Top | Article Outline

METHODS

Database

Transplant patients with cancer were identified in the IPITTR. The IPITTR is an international registry comprised primarily of patients from the United States. Patients are identified retrospectively and data collected prospectively. Malignancy data from the general population were obtained from the Surveillance, Epidemiology, and End Results (SEER) database. The SEER registry consists of a group of population-based tumor registries covering approximately 26% of the US population. Within SEER database, case listing session for SEER 17 Regs Limited-Use database was used. Any case with complete staging information but no AJCC stage catalog was restaged using AJCC criteria.

Back to Top | Article Outline

Patient Selection Criteria

Patient selection criteria in both databases were identical except for time span. IPITTR data were collected from 1980 to 2007, and SEER data were selected during 1988 to 2004. Cancers analyzed included NSCLC (lung and main bronchus), colon cancer (colon, cecum, and appendix), breast cancer (breast and nipple), prostate cancer (prostate gland), bladder cancer (bladder, bladder neck, ureteric orifice, and urachus), RCC (kidney), and malignant melanoma (skin). The analyses were restricted to patients older than 18 years and cases with incomplete staging information were excluded. Because there were in situ lesion data for breast cancer and bladder cancer in IPITTR, corresponding data from SEER were also collected for this study. In situ lesion data for NSCLC, colon cancer, prostate cancer, RCC, and malignant melanoma were not included. A total of 1749 IPITTR patients (246 NSCLC, 245 colon cancer, 295 breast cancer, 368 prostate cancer, 396 bladder cancer, 31 RCC, and 168 malignant melanoma) and 807,334 patients in the general population (406,835 NSCLC, 18,263 colon cancer, 48,494 breast cancer, 277,751 prostate cancer, 27, 682 bladder cancer, 8,661 RCC, and 19,648 malignant melanoma) were excluded from analysis because of missing cancer-stage information. Data on race were limited to White, African Americans, and others as these were the only three race categories in the SEER database. And race data in both populations included those patients whose race was reported.

Back to Top | Article Outline

Statistical Analysis

SPSS 15.0 (Chicago) was used to compile data and for statistical analysis. Student’s t-tests were used for continuous variables comparison. Chi-square and Mann-Whitney U tests were used for categorical variables. Kaplan-Meier tests were performed for survival analyses. Cox proportional hazards model was created for adjusting with other mixed factors to determine if transplantation is an independent factor for survival of patients with de novo malignancies. All tests of statistical significance are at the 0.05 level.

Back to Top | Article Outline

RESULTS

Demographic and transplant data are presented in Table 1. Immunosuppression types at time of cancer diagnosis in transplant patients are presented in Table 2. Data from a total of 635 transplant patients and 1,282,984 patients from the general population were analyzed. Males comprised a greater proportion of transplant recipients with NSCLC, colon cancer, and malignant melanoma. With respect to race, transplant recipients with each cancer type had a similar proportion of Whites except RCC. For each cancer type, transplant recipients were most commonly kidney transplant recipients, with heart transplant recipients being the second most frequent transplant type for several tumors including NSCLC, colon, prostate, bladder, and malignant melanoma. Follow-up was shorter in transplant recipients for each tumor type.

TABLE 1

TABLE 1

TABLE 1

TABLE 1

TABLE 1

TABLE 1

TABLE 1

TABLE 1

TABLE 2

TABLE 2

Initial analysis of mean and median age at cancer diagnosis demonstrated striking differences for each cancer with transplant patients being diagnosed several years earlier (6–13 years) than the general population for each tumor type (Fig. 1A). Transplant candidates are often excluded because of advanced age, which could account for the observed differences in age. Therefore, age distribution was analyzed for patients less than 60 years of age. This analysis demonstrated no difference in age at diagnosis between groups for all cancers studied except RCC (Fig. 1B).

FIGURE 1.

FIGURE 1.

FIGURE 1.

FIGURE 1.

Cancer stage at diagnosis is presented in Figure 2. Transplant patients were diagnosed at earlier stages of RCC. Cancer stage distributions were similar between transplant recipients and general population for colon cancer. NSCLC, breast cancer, prostate cancer, bladder cancer, and malignant melanoma were diagnosed at more advanced stages in transplant recipients.

FIGURE 2.

FIGURE 2.

Treatment data are presented in Table 1. For each tumor type, the proportions of patients receiving immunosuppression reduction or discontinuation were low. It is possible that these low numbers may represent a substantial degree of underreporting. As expected, for all cancer types, surgery was the most common treatment modality for localized stage I and II and most stage III cancers. Some differences in treatment were noted between the transplant and SEER populations, particularly with respect to radiation therapy, however, the absolute magnitude of differences were often small. In general, although statistically significant differences were noted between transplant and SEER populations with respect to therapies, the approach to treating most stages for each cancer were similar.

Stage-specific and disease-specific survival data are presented in Figure 3. For most stages for each tumor type, disease-specific survival by Kaplan-Meier was lower in transplant patients. Cox proportional hazards analysis of factors that influence disease-specific survival are presented in Figure 4. For each cancer, the two factors that had the greatest negative effect on survival were stage at diagnosis and transplantation. Prostate and bladder cancer demonstrated a greater effect of transplantation than cancer stage, whereas all other tumor types studied demonstrated a greater effect of stage at diagnosis. Surgery was observed to extent a positive effect on survival.

FIGURE 3.

FIGURE 3.

FIGURE 3.

FIGURE 3.

FIGURE 4.

FIGURE 4.

Back to Top | Article Outline

DISCUSSION

The overall evidence in this study supports the concept that cancers arising in transplant recipients are associated with worse outcomes and an apparent increase in biologic aggressiveness. Cancers arise in transplant recipients’ surprisingly early after transplantation, ranging from a median of 39 (NSCLC) to 77 months (RCC). Expressed differently, among the seven cancers studied, the longest median time from transplant to cancer diagnosis was observed with RCC; however, even this time (6 years and 5 months) is impressively short. These observations indicate that transplant recipients experience the greatest risk of developing (most) cancer(s) within the first 6 years after transplantation. A substantial proportion of cancer risk in transplant recipients is, therefore, incurred after a relatively brief period of immunosuppression. It is possible that the early appearance of cancer in the posttransplant period may result from the more intense immunosuppression, and raises the possibility that cancer risk may decline as immunosuppression is reduced over time.

Evaluation of patients at all ages for age at cancer diagnosis demonstrated that transplant patients are diagnosed at an earlier age than in the general population. This observation may have occurred because patients older than 60 years are not infrequently excluded as transplant candidates, thereby inducing an age limitation on the transplant population. When the analysis was restricted to patients younger than 60 years at the time of cancer diagnosis, all cancers studied (except RCC) showed almost identical age distributions for age at cancer diagnosis in the transplant and general populations.

Analysis of the extent of cancer spread at the time of diagnosis was of particular importance as it was the primary measure by which a possible effect of immunosuppression on cancer behavior could be assessed. These analyses demonstrated that several cancers (breast cancer, bladder cancer, and malignant melanoma) were diagnosed at a higher stage in transplant recipients. The increase in cancer stage at diagnosis for malignant melanoma was surprising, as it occurred despite a widespread recognition of the need for enhanced screening for skin cancers in transplant recipients. In addition, these observations for other cancers occurred despite the greater degree of physician contact generally given to transplant patients. These data argue strongly for increasing education of physicians and transplant recipients regarding cancer screening. Of interest, only a single cancer, RCC, was associated with a shift toward earlier cancer stage at diagnosis. A potential explanation for this observation may be the increased exposure of renal failure patients and transplant recipients to abdominal imaging studies (CT scan, magnetic resonance imaging, and ultrasound). In addition, end stage kidneys with renal cystic lesions are at increased cancer risk.

Stage-stratified, disease-specific survival was worse in transplant recipients for one or more stages in several cancers including: colon cancer (all stages), NSCLC (stage II), breast cancer (stage III), prostate cancer (stage II, III, and IV), bladder cancer (stage III), and RCC (stage IV). Stage-stratified, disease-specific survival in other stages for the cancers studied was often numerically lower for transplant recipients, but did not reach statistical significance. This may have been due to the relatively small numbers of subjects analyzed for each cancer stage.

Disease-specific survival was further analyzed by Cox proportional hazards modeling of several risk factors for survival (age, gender, cancer stage at diagnosis, surgical therapy, and transplantation). Not surprisingly, surgical therapy was a significantly positive factor that favored survival, because surgery was likely performed for early stage cancers. A significant negative effect of transplantation was observed on disease-specific survival for each cancer, which suggests a substantive effect of immunosuppression or comorbidities or both. Surprisingly, the magnitude of the negative effect of transplant status was relatively comparable to cancer stage at diagnosis, and for prostate and bladder cancer, the effect of transplantation was greater than the effect of cancer stage at diagnosis.

Why does transplantation confer such a poor prognosis for cancer outcomes? Is it that immunosuppressive therapy alters the host-tumor relationship resulting in increased biologic aggressiveness, or does the presence of increased comorbidities and reduced physical condition in transplant recipients also play a role? Functional status data could control for differences in physical condition, however, both SEER and IPITTR registries contain little functional status data. The observation of more advanced cancer stage at diagnosis in transplant recipients argues for a real effect of immunosuppressive therapy, particularly because transplant patients generally have more physician contact than the general population. This question is important, because if immunosuppressive therapy does result in increased biologic aggressiveness of cancers, immunosuppression reduction should be evaluated for its potential to improve outcomes in transplant recipients.

Several considerations were involved in the selection of individual cancers for study. From a practical perspective, the study was limited primarily by the smaller transplant population, particularly with respect to cancer stage-specific analyses. Therefore, to assure adequate numbers of patients, more commonly occurring cancers were selected for analysis. In addition, previous cancer immunotherapy studies in patients with RCC or malignant melanoma have demonstrated these cancers to be immunosensitive. The possibilities that immunosuppressive effects may be magnified in these two populations were additional considerations for their selection for study.

Differences in cancer treatment and immunosuppressive factors may influence observations when comparing transplant and general population data. For example, chemotherapy, either myelosuppressive or myeloablative, induces transient and possible longer term immunosuppression. Myeloablative chemotherapy, may cause results in the general population to resemble those in a systemically immunosuppressed population. Unfortunately, minimal chemotherapy data were available in the SEER database and were inadequate for meaningful analysis. Similarly, hormonal therapy data were also inadequate.

Transplant practitioners reduce or discontinue immunosuppression after cancer diagnosis in transplant recipients. When immunosuppression is reduced, the observed effect of immunosuppression may be diminished, thereby resulting in an underestimation of the effect of immunosuppression on the biologic behavior of cancer. In the present study, relatively small proportions of patients were reported to have had reductions in immunosuppression, indicating that underreporting may have occurred, or alternatively, that immunosuppression reduction is not frequently prescribed.

Similar proportions of transplant recipients and general population subjects underwent surgery for prostate cancer; however, transplant recipients were more likely to receive radiation therapy, suggesting that some transplant recipients may receive more intensive anticancer therapy. However, despite receiving more therapy, transplant recipients had reduced survival from prostate cancer. It is possible that additional therapy was given to treat cancer recurrence before cancer-related deaths.

Surgery represents a major component of therapy for all the cancers studied, and therefore, its application, and just as importantly, the stage at presentation, both had significant effects on survival. These observations emphasize the importance of analyzing data regarding cancer stage at presentation.

With each additional analysis in this study, the negative effect of transplant recipient status on outcomes became increasingly apparent. The substantial preponderance of data indicates that despite having known for many years that transplant recipients are at increased risk of cancer, transplant practitioners have screened, diagnosed, and treated transplant recipients for cancer without a substantive knowledge of the potentially increased biologic aggressiveness of their cancers; or alternatively poorer outcomes because of comorbidities or lower functional status. Given these data, the approach to managing cancer in transplant recipients should now be reconsidered.

Observations from the present study can be of use in defining recommendations for cancer screening in transplant recipients. Provision of precise recommendations for individual cancers is beyond the scope of this publication; however, a few general observations can be made. With respect to each individual cancer analyzed, data from the present study indicate that transplant recipients should be considered a high-risk group for each cancer, as they develop these cancers relatively early after transplantation, and transplantation represents a significant risk factor for disease-related death. Cancers that occur at younger ages and with shorter intervals from transplantation (e.g., RCC) should require earlier screening. In addition, cancers that tend to be diagnosed at more advanced stages in transplant patients should receive more intensive screening (colon cancer, prostate cancer, breast cancer, and malignant melanoma). Cancers that behave in a more malignant fashion in transplant recipients (colon cancer, prostate cancer, breast cancer, bladder cancer, malignant melanoma, and NSCLC) may be more amenable to therapy if diagnosed earlier by more aggressive screening. Therefore, we propose that transplant recipients be considered at high risk for breast, colon, prostate, and RCC with respect to screening recommendations after transplantation.

In summary, this study has shown that several of the most common cancers are associated with significantly worse outcomes in transplant recipients compared with the general population. In addition, these cancers occurred surprisingly early after transplantation, with the median time to cancer diagnosis being less than 7 years in each cancer studied.

These observations have important implications for cancer management in transplant recipients including education, screening, and therapeutic management. Carefully designed, prospective, controlled studies are needed to evaluate the ability of immunosuppression reduction to improve outcomes from cancers arising in transplant recipients. Similarly, controlled studies should be conducted to evaluate whether additional approaches, such as conversion to mammalian target of rapamycin therapy, or more aggressive cancer therapy can improve results in transplant recipients with cancer. However, caution should be noted in exposing transplant recipients to more aggressive chemotherapy, as they have high comorbidities, and may experience increased morbidity and mortality with aggressive chemotherapy than that observed in the general population.

Back to Top | Article Outline

REFERENCES

1. Villeneuve PJ, Schaubel DE, Fenton SS, et al. Cancer incidence among Canadian kidney transplant recipients. Am J Transplant 2007; 7: 941.
2. Vajdic CM, McDonald SP, McCredie MR, et al. Cancer incidence before and after kidney transplantation. JAMA 2006; 296: 2823.
3. Kasiske BL, Snyder JJ, Gilbertson DT, et al. Cancer after kidney transplantation in the United States. Am J Transplant 2004; 4: 905.
4. Yasumura T, Ohmori Y, Aikawa I, et al. Breast cancer arising de novo in recipients of kidney allograft. Jpn J Surg 1989; 19: 370.
5. Reed MJ, Ghilchik MW. Risk of breast cancer after renal or cardiac transplantation. Lancet 1995; 346: 1422.
6. Stewart T, Tsai SC, Grayson H, et al. Incidence of de-novo breast cancer in women chronically immunosuppressed after organ transplantation. Lancet 1995; 346: 796.
7. N′zi K, Asquier E, Fauchier F, et al. [Does renal transplantation increase the risk of breast diseases?]. J Radiol 2001; 82: 469.
8. Buell JF, Hanaway MJ, Trofe J, et al. De novo breast cancer in renal transplant recipients. Transplant Proc 2002; 34: 1778.
9. Oruc MT, Soran A, Jain AK, et al. De novo breast cancer in patients with liver transplantation: University of Pittsburgh’s experience and review of the literature. Liver Transpl 2004; 10: 1.
10. Stewart T, Henderson R, Grayson H, et al. Reduced incidence of rectal cancer, compared to gastric and colonic cancer, in a population of 73,076 men and women chronically immunosuppressed. Clin Cancer Res 1997; 3: 51.
11. Ebisui C, Okazaki M, Kanai T, et al. Clinicopathological study of colorectal cancers after renal transplantation. Transplant Proc 2000; 32: 1984.
12. Saidi RF, Dudrick PS, Goldman MH. Colorectal cancer after renal transplantation. Transplant Proc 2003; 35: 1410.
13. Fabia R, Levy MF, Testa G, et al. Colon carcinoma in patients undergoing liver transplantation. Am J Surg 1998; 176: 265.
14. Safadi R, Ilan Y, Eid A, et al. Solid tumors after liver transplantation. Transplant Proc 1999; 31: 1894.
15. Atassi T, Thuluvath PJ. Risk of colorectal adenoma in liver transplant recipients compared to immunocompetent control population undergoing routine screening colonoscopy. J Clin Gastroenterol 2003; 37: 72.
16. Zhu ZJ, Li L, Zhang YM. [Diagnosis and treatment of de novo malignancy after liver transplantation]. Zhonghua Zhong Liu Za Zhi 2007; 29: 237.
17. Manson AD, Landsberg DN. Prostatic carcinoma following renal transplantation. Transplant Proc 1989; 21: 3313.
18. Schloerb PR. Prostate cancer transmitted in an orthotopic heart transplant. JAMA 1997; 277: 1355.
19. Lechevallier E, Cormier L, Barrou B, et al. Prostate cancer in renal transplant recipients. Transplant Proc 2002; 34: 722.
20. Schonberger B, Giessing M, Vogler H, et al. Relevance and meaning of prostate cancer in kidney transplantation. Transplant Proc 2002; 34: 2225.
21. Whang M, Geffner S, Khachikian Z, et al. Renal transplantation and prostate cancer: Guidelines for screening and treatment. Transplant Proc 2002; 34: 3196.
22. Woodle ES, Gupta M, Buell JF, et al. Prostate cancer prior to solid organ transplantation: The Israel Penn International Transplant Tumor Registry experience. Transplant Proc 2005; 37: 958.
23. Jimenez C, Manrique A, Marques E, et al. Incidence and risk factors for the development of lung tumors after liver transplantation. Transpl Int 2007; 20: 57.
24. Jirasiritham S, Sumethkul V, Boonkasem S. Long-term survival of a lung cancer patient after kidney transplantation. Transplant Proc 2001; 33: 2043.
25. de Perrot M, Wigle DA, Pierre AF, et al. Bronchogenic carcinoma after solid organ transplantation. Ann Thorac Surg 2003; 75: 367.
Keywords:

Transplant; Malignancy; Immunosuppression

© 2009 Lippincott Williams & Wilkins, Inc.