In our ultimate goal of determining whether exercise could have a potential role in reducing prostate cancer risk, we thank Dr. Kaufman for his letter to the editor concerning our article.[1] However, there are several points that need clarification.

First, we did not conclude that exercise had “a protective effect” for the cohort or the subgroups based on race, because our study merely examined associations between exercise and the risk of prostate cancer, and it was not designed to establish causality. Second, we acknowledged in our article that our report presents preliminary results in a small study sample from a single Veterans Affairs hospital and, thus, will require external validation involving larger cohorts from different settings. This would be a far better way to validate the relation between exercise and prostate cancer risk. This includes determining whether the disparity of the association by race holds true and establishing more generalizable estimates for the magnitude of these associations. Of note, the results from heterogeneity tests that were mentioned in the preceding letter can be unreliable when power is an issue.[2] In addition, subgroup effects may well exist despite negative heterogeneity test results.[3] Third, there is no evidence to support the idea that exercise has an impact on digital rectal examination (DRE) findings, as suggested by the commenter, and whether exercise has an impact on serum prostate-specific antigen (PSA) results is inconclusive.[4] It is noteworthy that it seemed more clinically relevant to examine whether the relation between exercise and prostate cancer risk was independent of PSA results and DRE findings, especially for designing future studies. Indeed, if quantified exercise is proven to be an independent predictor of overall and/or high-grade prostate cancer, then it may aid in prebiopsy risk assessment and may even be included later in predictive nomograms designed for clinical use.[5]

We reported the relation between quantified exercise and prostate cancer risk as we have observed them among white and black veterans. We exercised caution to be certain that we would not overstate our conclusions, and we reiterate that future investigations are needed before any clinical recommendations may be formulated based on our findings. The preliminary nature of our report and the limitations inherent to such a study should not deter further investigations of exercise as a modifiable risk factor with a potential benefit against overall and/or aggressive prostate cancer and as a possible source of racial disparity that future discoveries could potentially mitigate.

  • Lionel L. Bañez, MD1

  • Abhay A. Singh, MD2

  • 1Veterans Affairs Medical Center, Durham, North Carolina

  • 2Duke Prostate Center and Division of Urology, Duke University Medical Center, Durham, North Carolina


  1. Top of page