Association Between Treatment Modality and Probability of Second Malignancy in Localized Prostate Cancer Patients.

Abstract

PURPOSE/OBJECTIVE(S)\nProstate radiotherapy has been associated with an increased risk of developing a second malignancy (SM). Relative SM probabilities following stereotactic body radiotherapy (SBRT) or moderately hypofractionated radiation (HF-IMRT) remain unknown. We hypothesized that the probability of SM following SBRT or HF-IMRT would differ compared with conventionally fractionated intensity modulated radiotherapy (CF-IMRT).\n\n\nMATERIALS/METHODS\nCohort analysis of men from a nationally representative database with localized prostate cancer with at least 60 months of follow-up compared SM probability amongst men receiving either radical prostatectomy (RP), CF-IMRT, HF-IMRT, brachytherapy (BT), or SBRT, using multivariable logistic models. Sensitivity analyses were conducted to evaluate differences in SM stratified by year of treatment and follow up length. Propensity score-adjusted pairwise assessments of modalities were also performed.\n\n\nRESULTS\nFor 303,432 patients, median follow-up was 9.08 years (IQR 7.01 - 11.21), and median age was 64 years (IQR 58-69). Crude rates of SM by treatment modality were: 5.2% for RP, 8.9% for CF-IMRT, 7.7% for HF-IMRT, 7.9% for BT, and 5.3% for SBRT. Adjusted odds ratios (AOR) for development of SM (referent: RP) were: CF-IMRT vs. RP 1.20 (95% CI 1.14-1.25, P < 0.001), HF-IMRT vs. RP 1.25 (95% CI 1.01-1.55, P\u202f=\u202f0.045), BT vs. RP 1.11 (95% CI 1.07-1.16, P < 0.001), and SBRT vs. RP 0.95 (95% CI 0.81-1.12, P\u202f=\u202f0.567). Sensitivity analyses confirmed higher probability of SM comparing CF-IMRT to RP (P < 0.01 for all) and no statistically significant difference comparing SBRT to RP at all year and follow-up length groupings (P\u202f=\u202fNS for all). On propensity score-adjusted analysis, SBRT was associated with lower probability of SM compared to CF-IMRT (AOR 0.78, 95% CI 0.66-0.93, P\u202f=\u202f0.005); no significant difference was found when SBRT was compared to RP (AOR 0.86, 95% CI 0.73-1.03, P\u202f=\u202f0.102).\n\n\nCONCLUSION\nConventionally fractionated intensity-modulated radiotherapy, moderately hypofractionated intensity-modulated radiotherapy, and brachytherapy, but not stereotactic body radiotherapy, were associated with increased probability of a second malignancy compared to radical prostatectomy. Our findings underscore the need for assessment of second malignancy risk in prospective studies of SBRT.