Prostate and pelvic SABR with or without a focal boost: results from a multicenter phase II prospective clinical trial Authors: Yasir Alayed, Darin Gopaul, Lucas C. Mendez, Winkle Kwan, Stanley K. Liu, William Chu, Chia-Lin Tseng, Patrick Cheung, Danny Vesprini, Hans Chung, Gerard Morton, Rohan Correa, Bindu Musunuru, Melanie Davidson, Ananth Ravi, Andrea Deabreu, Alexander Mamedov, Evan Poon, Liying Zhang, Andrew Loblaw Published: 2027 ABSTRACT PurposeGiven the complexity of SABR to the pelvis in prostate cancer (PCa), it was important to establish the safety and feasibility in a multicentric setting. We report the outcomes of a multicenter, phase II study of SABR with elective nodal irradiation (ENI) and an optional prostate focal boost.MethodsPatients had unfavorable-intermediate or high-risk PCa. All patients received 5 fractions SABR weekly delivering 25Gy ENI and 40Gy to prostate simultaneously, with an optional 50Gy focal prostate boost. All patients had ADT. Primary endpoint was acute GI and GU adverse events (AE).Results76 patients were accrued at 4 hospitals with a median follow-up of 62.8 months. 63% of patients had high-risk PCa. 41% of patients had a focal boost. Acute grade ≥ 2 GI and GU AE was 4.1% and 18%. At 5 years, 2.9% and 4.3% of patients had grade 2 GI and GU AE, and no grade 3+ GI/GU AE was observed at the last follow-up. 5-year cumulative biochemical failure was 10.5%. The median PSA nadir was 0.2 ng/mL. 84% of patients recovered their testosterone level, with a median time of 14.3 months. 2.9%, 5.9% and 4.6% of patients had a large change in QOL (SD*2) in the EPIC urinary, bowel, and sexual domains.Conclusions5-fraction SABR to prostate and pelvis for unfavorable-risk PCa was safe and feasible in a multicentric setting with or without a focal boost. There were few significant AE, encouraging PSA kinetics and biochemical control. The results are currently being validated in randomized trials. Introduction There is clear evidence of improved biochemical disease free survival (bDFS) with dose-escalated radiotherapy in prostate cancer (PCa) [1], [2]. Stereotactic ablative radiotherapy (SABR) has been tested widely in low- and intermediate-risk PCa [3], [4], [5], and the recently published PACE-B phase III trial established SABR as a new standard of care option for patients with low and favorable intermediate-risk PCa [6]. Few studies evaluated SABR in patients with unfavorable-risk PCa [7], some of which incorporated a boost to the dominant intra-prostatic lesion (DIL) seen on MRI [8], [9], [10], while others evaluated the addition of elective nodal irradiation (ENI) [11]. Given the added complexity of delivering SABR to the pelvis and prostate in unfavorable-risk PCa, it was important to establish the safety and feasibility in a multicentric setting before moving to a larger phase III trial. Here we report the outcomes of a multicenter, phase II study of 5-fraction SABR with ENI and an optional prostate focal boost. Methods and materials The trial was approved by the research ethics boards and registered on clinical trials.gov (NCT04245670). Written, informed consent was obtained in all participants. Patient selection and treatment details Between 2018 and 2020, 76 patients with unfavorable intermediate-risk (at least two of cT2c, PSA 10–20 or Gleason 7) or high-risk PCa (cT3, Gleason 8–10 or PSA >20 ng/mL) were accrued to a multicentric phase II prospective trial at 4 Canadian hospitals. Patients were excluded if they had prior pelvic radiation therapy, bleeding diathesis, hip prosthesis, castrate resistant disease, extrapelvic nodal or distant metastatic disease, poor baseline urinary function (IPSS ≥20), or inflammatory bowel disease. All patients received 5 fractions of SABR weekly delivering 25 Gy ENI. The prostate dose was either 40 Gy to the whole prostate, or 35 Gy to the prostate and up to 50 Gy to an MRI defined focal boost. Patients with unfavorable intermediate-risk disease had 6 months of ADT, while high-risk patients has 12–18 months of ADT. During treatment planning, three gold fiducial markers were implanted transperineally into the prostate, and a planning CT was performed with patients immobilized with a custom vacuum lock bag in the supine position with a comfortably full bladder and empty rectum. The nodal CTV was contoured according to RTOG guidelines [12]. The nodal CTV and the entire seminal vesicles comprised CTV1. CTV2 included the prostate only. A 6 mm margin was added around CTV1 and a 3 mm margin around CTV2 to generate PTV1 and PTV2; respectively [13]. If a boost to the DIL was planned, then the DIL was contoured on MRI and images were fused. An optimized radiotherapy plan was developed to treat CTV1 to 25 Gy in 5 fractions and CTV2 to 35 Gy in 5 fractions. PTV1 and PTV2 received 23.75 Gy and 33.25 Gy. The DIL was simultaneously boosted up to 50 Gy so long as OAR dose limits were not exceeded. If no boost was done, then the dose to CTV2 was modified to 40 Gy. OAR constraints were rectum V28Gy < 15% and V35 Gy < 2 cc, bladder V28 Gy < 15% and V39.5 Gy < 2 cc, urethra Dmax <52 Gy, small bowel and large bowel V25 Gy < 20 cc. All patients were treated with a VMAT technique. Image guidance with CBCT was used pre- and post-treatment. Study endpoints and follow-up Day 0 was the first day of SABR. Patients were assessed at baseline, weeks 2 and 4 during SABR, and at 3 and 6 months follow up. Afterwards, patients were followed every 6 months. Blood work (PSA and testosterone), quality of life (EPIC), and adverse event (AE) evaluation (using CTCAE v4.0) was performed at every follow up for the first 5 years [14], [15]. After 5 years, patients were seen annually for biochemical and survival outcomes. The primary endpoint was acute GI and GU AE. Secondary endpoints included late AE, quality of life, and biochemical disease-free survival. Sample size calculation The trial was designed as a Simon two-stage pilot study of 75 patients. The main interest was unacceptable AE rate, defined as acute grade ≥ 3 GI/GU AE more than 10% (the acute AE rate was 3% in previous trials [11]). Twenty-nine patients were enrolled in the first stage. Using the hypothesis of H0 ≤ 90% and HA ≥ 97% on the proportion of patients without AE, the trial would continue if 26 or more patients did not have grade ≥ 3 AE. An additional 46 patients were enrolled in the second stage to reach 75 total patients. The probability of early termination of the trial is 0.565 (type I error = 0.049, power = 80%). Statistical analysis Demographic and tumor characteristics were summarized using mean, standard deviation (SD), median and range for continuous variables, and proportions for categorical variables. For each domain and subscore, QOL scores were transformed to a 0–100 point scale with higher score representing better QOL. Biochemical failure (BF) was calculated from the time of first radiation treatment to failure. The cumulative incidence of BF was estimated using Nelson-Aalen curves. PSA nadir was set to be the lowest PSA value following treatment completion. The PSA response rate at 4 years (4yPSARR) was defined as PSA < 0.4 ng/mL at 4 years [16]. A minimally clinical important change (MCIC) was scored if the average EPIC QOL score (months 6–60) was >0.5 SD (mild change), > 1 SD (moderate change), or > 2 SD (large change) of baseline scores for each domain score or sub-score in all patients. All analyses were conducted using Statistical Analysis Software (SAS version 9.4 for Windows) and R package (version 3.4.2). P-values <0.05 were considered statistically significant. Results Demographic factors and basic tumor characteristics are summarized in Table 1. The median follow-up was 62.8 months. 37% and 63% had unfavorable intermediate-risk and high-risk disease, respectively. All patients had ADT as per protocol. 31 patients (41%) had an intraprostatic boost (DIL D99 was 46.4Gy). 6 patients (8%) had BF, 4 of which had high-risk PCa. The cumulative BF probability was 10.5% at 60 months (Fig. 1). The median PSA nadir was 0.2 ng/mL, with a median time to nadir of 19.6 months (Fig. 2). The 4yPSARR was 70.5%, and the cumulative BF probability was 0% in this group compared to 34% in the group that did not achieve a PSA < 0.4 ng/ml at 4 years (p = 0.0002). 84% of patients recovered their testosterone level (defined as >3 nmol/L), with a median time to testosterone recovery (TTR) of 14.3 months after receiving the last dose of ADT. One patient developed metastatic disease. 9 patients died during follow-up, one of which from prostate cancer.Fig. 1Nelson-Aalen cumulative biochemical failure probability curve.Fig. 1Fig. 2Median PSA over time.Fig. 2 Nelson-Aalen cumulative biochemical failure probability curve. Median PSA over time. The cumulative worst acute and late AE rates are summarized in Table 2. Acute grade ≥ 2 GI AE was 4.1% and grade ≥ 2 GU was 18%. Looking at the prevalence at last follow-up, 2 patients (2.9%) had grade 2 GI AE, 3 patients (4.3%) had grade 2 GU AE, and 16 patients (22.9%) had grade 2 sexual AE. No grade 3+ GI or GU AE was reported at the last follow-up, and only 1 patient had grade 3 sexual AE. When assessing cumulative worst late GI AE, 2 out of 6 patients with grade ≥ 2 GI AE had an event possibly attributed to SABR (proctitis, diarrhea). The remaining 4 patients had an AE that is less likely to be associated with SABR (constipation, nausea, hemorrhoids). The only grade 3 GI AE reported was diarrhea that improved at 12 months. Looking at cumulative worst late grade 3 GU AE, 1 patient had grade 3 hematuria and 1 patient had grade 3 scrotal pain. Out of the 10 patients with cumulative late grade 3 sexual AE, 1 had hot flashes during ADT use that improved afterwards, 6 had ED that improved after Testosterone recovery, and 2 had ED but never recovered their Testosterone to normal levels. The mean EPIC QOL change from baseline is shown in Fig. 3. At 60 months, the mean change from baseline was 1.4 for the urinary domain, 0.7 for the bowel domain, and 3.4 for the sexual domain. Regarding MCIC, 2.9%, 5.9% and 4.6% of patients had a large change in QOL in the EPIC urinary, bowel domain, and sexual domains (Table 3).Fig. 3EPIC QOL change from baseline in the urinary, bowel, sexual and hormonal domains.Fig. 3 EPIC QOL change from baseline in the urinary, bowel, sexual and hormonal domains. Discussion SABR is an effective treatment option for localized PCa, and was tested adequately in low-risk and intermediate-risk disease including the recently published PACE-B phase III trial [6]. Fewer studies evaluated SABR with a focus on higher risk disease, with no phase III trial published to date. To potentially improve outcomes in this patient group, various phase I and II studies evaluated the addition of ENI or an intra-prostatic boost [8], [9], [11]. This adds to the complexity of the treatment compared to that for favorable-risk PCa, therefore it was important to establish the safety and feasibility in a multicentric setting before moving to larger phase III trials. In this 5-year analysis of the 5STAR-PC trial, we evaluated the results of a multicenter phase II study of 5-fraction SABR with ENI and an optional intra-prostatic boost. The observed 5-year bDFS probability of 90% is comparable to that reported in PACE B (95%) and hypo-FLAME (93%) [6], [8]. A limited number of other SABR studies that included patients with high-risk PCa reported slightly worse 5-year bDFS ranging from 81% to 85% [7], [11], [17]. Improved biochemical control, approaching that reported in PACE B, could be attributed to various factors. All patients had ADT, which is proven to improve outcomes in higher-risk disease. In addition, the routine use of ENI could have potentially contributed to improved bDFS. Lastly, the optional focal boost to the macroscopic intra-prostatic tumor could also be a factor. The patterns of failure analysis of the FLAME trial showed that an increase in focal boost dose was associated with improved disease control in patients with predominantly high-risk PCa [18]. The patient- and clinician-reported QOL and AE outcomes in this trial were comparable to other reported studies. The 5-year cumulative probability of grade ≥ 2 GU AE (CTCAE) was 21%, compared to 28% in hypo-FLAME and 42% in PACE-B, and the prevalence of grade ≥ 2 GU AE at 5 years was 4% vs 12% and 8% respectively. For GI AE, the 5-year cumulative probability of grade ≥ 2 events was 9% in this study vs 14% in hypo-FLAME and 18% in PACE-B, and the prevalence at 5 years was 3% vs 4% and 2%, respectively. Late grade 3 AE were also similar between the three trials, all of which were <2%. With regards to QOL, the EPIC domain scores were all back to baseline at 5 years. Moderate to large MCIC was similar to what was reported in hypo-FLAME. In conclusion, 5-fraction SABR to the prostate and pelvis for unfavorable-risk PCa was safe and feasible in a multicentric setting with or without a focal boost. There were few significant AEs, with encouraging PSA kinetics, biochemical control and QOL. Although the trial is limited by its non-randomized nature and the optional use of focal boost, the results are in line with those reported by other phase II trials evaluating similar techniques in this patient population. The results are currently being validated in larger phase III trials. CRediT authorship contribution statement Yasir Alayed: Formal analysis, Methodology, Writing – original draft, Writing – review & editing. Darin Gopaul: Investigation, Writing – review & editing. Lucas C. Mendez: Investigation, Writing – review & editing. Winkle Kwan: Investigation, Writing – review & editing. Stanley K. Liu: Investigation, Writing – review & editing. William Chu: Investigation, Writing – review & editing. Chia-Lin Tseng: Investigation, Writing – review & editing. Patrick Cheung: Investigation, Writing – review & editing. Danny Vesprini: Investigation, Writing – review & editing. Hans Chung: Investigation, Writing – review & editing. Gerard Morton: Investigation, Writing – review & editing. Rohan Correa: Formal analysis, Writing – review & editing. Bindu Musunuru: Formal analysis, Writing – review & editing. Melanie Davidson: Formal analysis, Writing – review & editing. Ananth Ravi: Formal analysis, Writing – review & editing. Andrea Deabreu: Data curation, Formal analysis, Writing – review & editing. Alexander Mamedov: Data curation, Formal analysis, Software, Writing – review & editing. Evan Poon: Data Curation, Formal Analysis, Writing - review & editing. Liying Zhang: Data curation, Formal analysis, Methodology, Writing – review & editing. Andrew Loblaw: Conceptualization, Formal analysis, Funding acquisition, Methodology, Project administration, Resources, Supervision, Writing – review & editing. Funding statement Cure Prostate Cancer Canada. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.