Preexisting Cancer in Transplant Candidates

Transplant candidates with a history of cancer pose a unique dilemma for clinicians and patients themselves. On the one hand, it is well known that there is a risk of cancer recurrence after transplantation,1 and those outcomes are generally poor.2 Early studies suggested that a prolonged period of waiting (ie, >5 years) from the time of cancer diagnosis to transplantation could substantially reduce the risk of cancer recurrence.1 This led to our current practice of minimum waiting times before patients with cancer could proceed with transplantation.3,4 For patients with end-stage kidney disease, this means prolonged exposure to the known deleterious effects of dialysis. For patients with end-stage heart, lung and liver failure prolonged waiting before transplantation may not even be a realistic option. An improvement in cancer survival in the general population5 has led some groups to question the current waiting periods before transplantation.6

In this issue of Transplantation, Acuna et al7 report on outcomes of solid-organ transplant recipients with a history of preexisting malignancy. This was a carefully done population-based study involving 6 transplant programs in the province of Ontario, Canada. Using a propensity score matched cohort study design, they identified 443 transplant recipients with pretransplant cancer and 886 matched transplant recipients without pretransplant malignancy. Approximately 70% of patients received a kidney transplant with the remainder being liver, heart, and lung transplant recipients. Consistent with previous literature, they found that patients with a history of pretransplant malignancy had an increased risk of death compared to those without prior cancer. Although they did not have information on cancer stage, they classified malignancies as high risk (lung, breast, melanoma, gastrointestinal, hematologic, or gynecologic) or low risk (thyroid, prostate, bladder, kidney, oropharynx, or testis,) based on the known risk of recurrence in transplant patients. Patients with a history of a high-risk malignancy had an increased risk of death regardless of whether they were transplanted within 5 years of cancer diagnosis (hazard ratio [HR], 1.47; 95% confidence interval [CI], 1.14-1.90) or beyond 5 years (HR, 1.79; 95% CI, 1.40-2.67). The same was not true for patients with low-risk malignancies. If these patients were transplanted within 5 years of cancer diagnosis, the overall survival was actually better than patients without a history of cancer (HR, 0.72; 95% CI, 0.52-0.99). If, however, they were transplanted more than 5 years after cancer diagnosis, they had an increased risk of death (HR, 1.76; 95% CI, 1.27-2.49) similar to high-risk patients.

Using competing risk methodology, they went on to show that patients with pretransplant malignancy had an increased risk of both cancer-specific (HR, 1.85; 95% CI, 1.20-2.86) and noncancer death (HR, 1.29; 95% CI, 1.08-1.54). In addition, patients who waited more than 5 years from malignancy diagnosis to transplantation had an increased risk of noncancer death. When stratified by type of pretransplant cancer, only those with high-risk malignancies were at increased risk for cancer-specific mortality (HR, 3.16; 95% CI, 1.98-5.06). Overall, 22% of patients with preexisting cancer had a recurrence. High-risk malignancies and those who waited more than 5 years from cancer to transplantation had an increased risk of recurrence.

This study was well done but unfortunately has limitations common to analyses using administrative data. Specifically, the study lacked granular patient-level data, such as transplant immunosuppression, functional status, tumour markers/histology, cancer stage, and response to treatment, among others. Patients with a short time from cancer to transplantation most likely had many “good” characteristics that we unfortunately cannot define from this type of study. Similarly, those that had to wait a longer period before transplantation likely had more “bad” characteristics (eg, aggressive histology).

What should we make of these data? The authors provide no firm recommendations for clinicians to change current practice. The data do suggest, however, that there are subgroups of patients with preexisting cancer that will do quite well with minimal wait and may in fact be harmed by prolonged delay to transplantation. The difficult part, to date, has been to correctly identify these patients. The low-risk and high-risk groupings used in this study are not, in our opinion, detailed enough to make clinical decisions. For example, prognosis after potentially curative treatment of a large, symptomatic renal cell carcinoma is certainly not “low risk.”8 What is needed is a much more refined definition of what constitutes a low-risk transplant candidate with a prior history of malignancy. Advances in cancer tumor biomarkers have improved the ability to prognosticate significantly.9 Commercially available genomic profiling assays are now available and have been used to determine the risk of cancer recurrence in kidney transplant candidates with breast cancer.10 In the cases reported, a low-risk score allowed the patients to proceed with transplantation without the traditional 2- to 5-year waiting period.10 Similar genomic profiling assays are being developed for ductal carcinoma in situ of the breast as well as for colon and prostate cancer.10

As the use of more precise cancer biomarkers evolves, we in the transplant community need to use this information to make informed decisions about our patients. Along with traditional measures of risk such as cancer type, tumor size, histologic grade, and local/regional spread, these novel biomarkers will hopefully stratify patients into more defined groups of risk for disease recurrence for the common malignancies as we now have for breast cancer. Rather than coarsely lumping patients into a predetermined wait time based solely on cancer type and stage as we currently do, we are hopeful that practice will evolve such that we will use a more individualized prognosis for our transplant candidates with previous malignancy. Given the pace of change in the field of biomarkers and prognosis, guidance and close collaboration with our colleagues in oncology will be essential.

The article by Acuna et al7 adds to the body of evidence, prompting us to reexamine current recommendations on cancer-free wait time prior to transplantation. The concept that the risk of recurrence (and consequent premature cancer-related death posttransplant) should be weighed against the risk of a prolonged wait before transplant (and the ensuing risk of noncancer death both pretransplant and posttransplant) is an important one. Acuna et al have, through the use of competing risks analysis, demonstrated the magnitude of such risks. Our ultimate goal is to define exactly how much waiting time, if any, is required for specific cancers in individual patients before we can safely proceed with transplantation. Further data are clearly required to address this question, and as treatment strategies and the prognosis for patients with cancer changes the answer is likely to evolve over time. Although it may be premature to adopt the policy promulgated in Norway, where patients are listed for kidney transplantation within a year of potentially curative cancer therapy,6 consideration of shorter wait times than advocated by current guidelines appears to be reasonable. Due for release in 2018, the Kidney Disease: Improving Global Outcomes (KDIGO) practice guideline on the evaluation of the kidney transplant candidate will include a thorough review of the evidence and provide updated guidance to clinicians regarding the difficult issue of cancer and transplantation.