High Dose-Rate Prostate Brachytherapy as Monotherapy: a Single Fraction May Not be Enough

Abstract

Fractionated high dose-rate (HDR) brachytherapy as monotherapy provides high cancer control rates for men with low and intermediate risk prostate cancer. Based on radiobiological modelling, a single fraction of 19 Gy was predicted to provide similar outcomes. Our purpose is to report clinical and biochemical control rates with 19 Gy x 1 or 13.5 Gy x 2 in a single centre randomized trial. Eligible patients had NCCN low or intermediate risk disease, a prostate volume < 60 cc, International Prostate Symptom Score (IPSS) of 18 or less and no use of androgen deprivation therapy. Patients were randomized to receive either 19 Gy HDR as a single fraction (1F), or 27 Gy in 2 fractions one week apart (2F). Treatment was delivered using an out-patient ultrasound-based technique. Relative dosimetry was similar between arms with median prostate V100, V150, V200 and D90 of 97%, 35%, 11.4% and 110%, respectively. Follow-up with toxicity assessment, physical examination and serum prostate specific-antigen (PSA) level occurred at week 6, week 12, every 3 months for the first year, every 6 months until year 5, and annually thereafter. Biochemical failure was defined as nadir + 2 ng/ml. Rising PSA was investigated with CT, MRI, bone scan and in some cases PSMA PET. Local recurrence was confirmed by biopsy. A total of 170 patients were randomized between June 2013 and April 2015: 87 to 1F and 83 to 2F arms. Median age was 65 years, median PSA was 6.3 ng/ml and 72% had Gleason 7 cancer. NCCN low, favourable intermediate and unfavourable intermediate was present in 28%, 49% and 23%, respectively, with similar distribution between arms. Median follow-up was 51 months (range 24-68 months). PSA decreased more rapidly in the 2F arm, with median PSA at 5 years of 0.75 and 0.21 ng/ml in the 1F and 2F arms, respectively. Clinical or biochemical failure occurred in 22 patients – 20 in the 1F arm and 2 in the 2F arm, with a 5-year biochemical disease-free survival of 74.5% and 97.3%, respectively (Log Rank p=0.002). Most recurrences were local (16/20) in the 1F arm, while only one of the recurrences in the 2F arm had a local component. The 5-year cumulative incidence of local failure was 25.7% in the 1F and 2.6% in the 2F arm (p<0.0001). Of the local failures, 14 underwent salvage treatments: 7 with brachytherapy, 6 with prostatectomy and 1 with HIFU. On Cox analysis, biochemical failure was not associated with clinical stage, risk group, Gleason or baseline PSA. Distant metastases-free survival was 98% at 5 years. Despite linear quadratic considerations, single fraction 19 Gy is inferior to two fractions of 13.5 Gy, which appears to be highly effective. Single fraction 19 Gy is associated with an unacceptable rate of local failure. As recurrence is predominantly at the site of initial bulk disease, further dose escalation either focally to the dominant nodule or to the whole gland may be considered in a clinical trial.