Background and purpose: Biochemical recurrence after prostatectomy is commonly treated with salvage radiotherapy (SRT). In this prospective observational study we investigated the PSA decay rate, determined by predefined serial PSA measurements during SRT, as a predictor for treatment outcome. Materials and methods: Between 2013 and 2016, 214 patients were included in the study. The prescribed dose to the prostate bed was 70 Gy in 35 fractions (7 weeks) without hormonal treatment. PSA was measured weekly during SRT. Assuming first order kinetics, a PSA decay-rate constant (k) was calculated for 196 eligible patients. The ability of k to predict disease progression was compared with known clinical prediction parameters using Cox regression, logistic regression and ROC analyses. Disease progression was defined as continuously rising PSA after SRT, PSA increase by !0.2 ng/ml above nadir after SRT, hormonal treatment or clinical progression. Results: After a median follow up of 4.7 years the estimated failure-free survival at 5 years was 56%. The PSA decay-rate constant (k) was found to be the strongest predictor of disease progression in both uniand multivariable analyses. Conclusion:The addition of k to established clinical variables significantly improves the possibility to predict treatment outcome after SRT and could be used to personalize future therapies.

No abstract

Background Treatment adaptation based on tumour biomarker response during radiotherapy of prostate cancer, could be used for both escalation and de-escalation of radiation doses and volumes. To execute an adaptation involving extension of treatment volumes during radiation can however be restricted by the doses already delivered. The aim of this work was to develop a treatment planning method that addresses this challenge. Material and methods A volumetric-modulated-arc-therapy (VMAT) planning method with sequential plan-on-plan optimization was developed for a prospective phase II trial including 100 patients on salvage radiotherapy (SRT) for prostate cancer recurrence. A treatment adaptation was performed after five weeks of SRT based on prostate-specific antigen response during this phase of the treatment. This involved extension of treatment volumes for non-responders (n = 64) to include pelvic lymph nodes and boost to 68 Gallium-Prostate-Specific-Membrane-Antigen-Positron-Emission-Tomography positive lesions. This method was evolved by introducing an EQD2 (equivalent dose in 2.0 Gy fractions) correction of the base plan for improved dose coverage. Results All dose-volume criteria for target coverage were met for the non-responders when based on physical dose. An EQD2 correction of the base plan for non-responders, implemented for the final 29 patients, led to a statistically significant improvement in dose coverage as compared to the 35 patients treated without EQD2 correction. Conclusions This is to our knowledge the only study presented on biomarker-guided sequential VMAT radiotherapy using a plan-on-plan technique in the pelvis. By using a biologically adapted technique an improved target coverage was achieved without compromising doses to organs at risk.