Reply to Urinary toxicity after stereotactic body radiotherapy: The boy who cried wolf?

Abstract

We thank Kishan et al for their insightful comments regarding our analysis of the utilization of and toxicity after stereotactic body radiotherapy (SBRT) for localized prostate cancer.1 The authors raise concerns regarding the selection and interpretation of toxicity endpoints, particularly the composite endpoint of urinary incontinence. Our definitions of urinary incontinence and nonincontinence toxicity were derived from prior population-based studies evaluating radiation therapy for localized prostate cancer.2 These studies and ours used both International Classification of Diseases, Ninth Revision (ICD-9) diagnosis codes and Current Procedural Terminology, 4th Edition (CPT-4) procedural codes to capture all outcomes. In light of the aforementioned commentary, we performed a sensitivity analysis separately stratifying toxicity outcomes according to ICD-9 and CPT-4 codes (diagnostic vs therapeutic). According to ICD-9 diagnosis codes alone, SBRT has a higher rate of urinary incontinence at 2 years than conventional intensity-modulated radiation therapy (IMRT; 6.3% vs 5.9%; P < .001). Likewise, SBRT has a higher rate of urinary incontinence–related diagnostic procedures at 2 years (20.5% vs 16.7%; P < .001). When outcomes were stratified according to therapeutic procedures, there was no difference between SBRT and IMRT at 2 years (the absolute incidence was not reportable because of the cell size; P = .21). How can we interpret these findings? Kishan et al note that the absolute rates of urinary incontinence reported herein are higher than the incidence of severe toxicity reported in prior clinical trials, and they suggest that this may be an artifact of code selection. However, the use of diagnostic procedures as a surrogate for urinary incontinence–related toxicity may better capture true clinical toxicity because these procedures signal the presence of symptoms prompting an evaluation. Furthermore, even when diagnosis codes alone are examined and this results in lower absolute toxicity rates consistent with prior studies, SBRT demonstrates a higher incidence of incontinence. The authors suggest that this difference may result from a detection bias because men undergoing SBRT were often enrolled in clinical trials with rigorous protocols for the identification of toxicity. To our knowledge, there is no research to demonstrate different practice patterns concerning the diagnosis and treatment of subjects enrolled in clinical trials. Even so, a true biological effect cannot be ruled out. Although these data demonstrate a statistically significant difference in urinary toxicity after SBRT versus IMRT, the clinical significance of these findings, particularly over the long term, warrants further investigation. A higher demand for incontinence-related procedures after SBRT may be driven by greater patient bother. Furthermore, although we found that SBRT's 1-year overall costs were lower than IMRT's costs, more diagnostic procedures following SBRT may ultimately generate greater costs in the long term. Likewise, longer follow-up is needed to determine whether differences in toxicity persist over time. Ultimately, we agree that only randomized controlled data will truly elucidate the risk of toxicity after SBRT. In the interim, we aim neither to cry wolf nor to be the canary in the coal mine. Rather, we hope to shed light on the ongoing utilization and outcomes of SBRT to inform current clinical practice, in the context of our study design, until randomized data are available. No specific funding was disclosed. Jim C. Hu belongs to the Speakers' Bureau for Intuitive Surgical and Genomic Health. Joshua A. Halpern, MD, MS Department of Urology Weill Cornell Medicine New York, New York Art Sedrakyan, MD, PhD Department of Healthcare Policy and Research Weill Cornell Medicine New York, New York Wei-Chun Hsu, MS Department of Healthcare Policy and Research Weill Cornell Medicine New York, New York Jim C. Hu, MD, MPH Department of Urology Weill Cornell Medicine New York, New York