The successful renal transplant recipient faces death from cardiovascular disease, infection and cancer. In Australia, 25% of all patients surviving for 20 years after a transplant will have a cancer (nonskin); rising to 65% if one includes skin cancers. Cancer is the cause of death in 70% of those with a nonskin cancer and a major cause of morbidity in the long term for renal transplant recipients. It has been well established that the relative risk of cancer is increased in the transplant compared with the normal population. Understanding this phenomenon is thus important if we are to avoid, diagnose early, or effectively treat cancer in renal transplant patients. It may also lead to insights into the impact of immune suppression and cancer surveillance mechanisms in both the dialysis and normal populations. Kasiske et al. in this issue have now contributed to the further understanding of cancer in both the dialysed patient and in the renal transplant recipient.
All sources available for the description of risk of cancer in the transplant population have inherent weaknesses. The registries of renal transplant recipients, such as the Australia and New Zealand Registry (ANZDATA) and the Collaborative Transplant Study (CTS) are affected to varying degrees by under-reporting of cancer, while specific cancer registries relying on voluntary reporting, suffer from inaccessible denominators. Individual series and case reports provide data on the potential behavior of some cancers, but are clearly limited from an epidemiological perspective. Linkage of data from two or more trusted sources is thus attractive, especially where the sources have complementary strengths. This approach has been used in the past to validate registry data (1,2), but Kasiske and colleagues take this further by linking the USRDS with Medicare billing claims for cancer diagnoses.
The analysis confirms the high risk of cancer in the renal transplant population in the United States. The data are limited by the methodology to the first 3 years after transplantation, and to the 47% of total transplant patients who use Medicare and can thus be tracked by the billing system. The analysis shows a very high risk of many cancers compared with the normal, age-matched, US population. The 3-year cumulative incidence is also higher than shown in previous analyses, with 7.4% of patients having a skin and 7.5% a nonskin cancer by 3 years. This compares with the ANZDATA rates of approximately 10% and 3% at 3 years; and with the CTS rates of approximately 1% and 2%, respectively. Two features of this US analysis are thus the higher rate of cancer after renal transplantation compared with other reports, and the different ratio of skin to nonskin cancer (1:1 in US Medicare claimants, compared with 3:1 in the Australian sun).
The risk of cancer in the dialysis and end-stage renal failure population is of course distorted by the incidence of cancers which cause renal failure and by treatment policies which place relatively few patients with cancer on transplant waiting lists and limit the number of patients with metastatic malignant disease that are commenced on dialysis. It is important to look beyond these issues to discern whether uremia predisposes to cancer. This analysis provides confirmation of the impact of dialysis as opposed to transplantation, as the relative risk of cancer increased with the number of years on dialysis. There was also support for this effect from the limited differences in risks between the waiting list and transplanted populations as opposed to the large relative risks compared with the normal population. This concurs with other studies examining cancer risk in the dialysis population and turns some attention away from the role of immunosuppressive agents (3).
The types of cancer that show the largest relative risks for the transplant compared with the waiting list population are kidney, lymphoma, Kaposi's sarcoma and skin, but with reduction in the risk of ovarian and prostate cancer. It is tempting to offer explanatory hypotheses for each of these effects, related either to biological or nonbiological processes. Compared with the normal population, Kaposi's sarcoma, lymphoma, skin and kidney cancers stand out in the US data with more than a 15-fold increase in risk, while melanoma, leukemia, hepatobiliary, and female genital tract have a fivefold increase. Common solid tumors including gastrointestinal tract, prostate, pancreas, ovary and breast are also increased, but to a lesser degree. These data are at some variance to that seen in CTS and ANZDATA analyses where, for example, breast cancer is not significantly increased. The data have been adjusted for age, but the age-specific relative risks may be informative, with preliminary Australian data showing higher risk ratios in the younger than the older age groups.
Diagnosis and understanding of the relative and absolute risks are only the start of the journey that must be taken in order to reduce the rising impact of cancer on the quality of life and the longevity of transplanted patients. Identifying causes of the increased risks and determining appropriate renal population screening strategies combined with early treatment programs, may impact on mortality and morbidity. The different modes of action of immunosuppressive agents and the early data in randomized studies suggest that sirolimus may lead to fewer cancers (4), while longer term trial (5) and registry data (6) identify specific demographic factors and drug doses associated with increased cancer risk. These data open up the prospect of tailored therapy, targeted not only at the individual's risk of allograft rejection but also their risk of cancer.