Cancer Risk After Organ Transplantation

Abstract

To the Editor: Dr Engels and colleagues performed a largescale, population-based linkage study of cancer risk after solid organ transplantation. The findings demonstrated an increased risk for a wide range of cancers after transplantation. However, the inclusion in the study of cancers prevalent at transplantation may have upwardly biased some risk estimates, rendering comparisons with other immunosuppressed populations problematic. Most recent population-based cohort studies of cancer risk in solid organ transplant recipients excluded cancers occurring in the 30-day period after transplantation, including prior publications using the Scientific Registry of Transplant Recipients. Excluding cancers occurring 30 days after transplantation is necessary to avoid the inadvertent inclusion of cancers that were present at transplantation. In the Engels et al study, follow-up time commenced at transplantation, and the results showed high risks of some cancers within the first 6 months, most notably liver cancer in liver transplant recipients, lung cancer in lung cancer recipients, and kidney cancer in kidney transplant recipients. The authors acknowledged that the observed risk estimates for these tumors were biased upward due to the inclusion of prevalent cancers diagnosed histopathologically in the explanted organ after transplantation. Despite this, the biased estimates for these cancers are reported in Tables 2 and 3 and in the abstract of the article. In addition to excluding the first 30 days follow-up, patients whose indication for transplantation is cancer must be excluded from the corresponding site-specific cancer risk analyses. Essentially this only applies to patients undergoing liver transplantation for primary liver cancer. After transplantation, such patients cannot contribute linked “incident” cancers of that type or person-years at risk for that cancer. This is also the case for patients with a pretransplantation history of cancer identified via linkage with cancer registry records. Engels et al should reanalyze their data excluding such patients.