Background: Biochemical recurrence (BCR) after definitive radiotherapy (RT) in localized prostate cancer (PC) is a clinically relevant event that impacts long-term management and prognosis. However, the prognostic value of certain biopsy-derived pathological parameters remains underexplored in RT-treated cohorts.
Methods: We retrospectively
[...] Read more.
Background: Biochemical recurrence (BCR) after definitive radiotherapy (RT) in localized prostate cancer (PC) is a clinically relevant event that impacts long-term management and prognosis. However, the prognostic value of certain biopsy-derived pathological parameters remains underexplored in RT-treated cohorts.
Methods: We retrospectively analyzed 444 patients with localized PC treated with external beam radiotherapy (with or without androgen deprivation therapy) between 2013 and 2019. Clinical, radiological, and detailed histopathological data, including Gleason score, perineural invasion, and the number and proportion of positive biopsy cores—were collected. Logistic regression models were used to identify predictors of BCR.
Results: After a median follow-up of 72 months, 11.7% of patients developed BCR. In multivariable analysis, higher PSA at diagnosis (
p = 0.05), higher Gleason score (ISUP ≥ 4;
p = 0.036), and greater tumor burden in biopsy cores—quantified as both the number and proportion of positive cores per lobe and overall (
p < 0.05)—were independently associated with BCR. Perineural invasion showed a univariable association (
p = 0.036), though it did not remain significant after adjustment. Overall, 19.2% (10/52) of recurrences were diagnosed beyond five years post-treatment, underscoring the need for prolonged follow-up.
Conclusions: PSA at diagnosis, the extent of tumor involvement in diagnostic biopsies, and ISUP grade group ≥ 4 (Gleason score ≥ 8) were identified as independent predictors of biochemical recurrence after RT in localized prostate cancer. Lower Gleason categories and perineural invasion showed only a trend toward significance in the multivariable analysis, suggesting that their predictive effect may be attenuated by other covariates.
Full article
