Four-year Follow-up of a Phase-3 Prospective Randomized Trial of Partial Gland Ablation with Vascular- Targeted Phototherapy versus Active Surveillance for Low-risk Prostate Cancer

Introduction: The prospective trial PCM301 randomized 413 men with low-risk prostate cancer to partial gland ablation with vascular-targeted photodynamic therapy (VTP; n=207) or active surveillance (AS; n=206). Two-year outcomes were reported previously (Lancet Oncology 18: 181, 2017). Herein we report four-year outcomes and post-hoc analyses. Materials & Methods: Four-year outcomes evaluated rates of crossover to radical therapy (RT), and metastasis-free, cancer-specific and overall survival rates. Post-hoc analyses evaluated overall biopsy results, progression in Gleason Grade Group (GG), spatial location (apex, mid-gland, base), as well as ‘in-field’ (VTP treated lobe, or lobe with the index cancer in the AS cohort) and ‘out-of-field’ results. Results: Three and 4-year follow-up was available in 69% and 64% of patients, respectively. Subjects in the VTP arm had lower rates of cross-over to RT at 2 years (7% vs 33%), 3 years (14% vs 44%) and 4 years (24% vs 53%) compared to AS (HR=0.31, 95% CI=0.21-0.45; p<0.001). Four-year metastasisfree (99% vs 99%), cancer-specific (100% vs 100%) and overall survival (98% vs 99%) rates were similar between cohorts. At 2 years, rates of disease progression were lower in subjects randomized to VTP for overall progression (HR=0.35; 95%CI = 0.25-0.48) and for grade progression (HR=0.42; 95%CI = 0.290.59). Reductions were even more pronounced when focusing on ‘in-field’ biopsy results for overall disease progression (HR=0.21; 95%CI = 0.14-0.31) and grade progression (HR=0.25; 95%CI = 0.16-0.40). Overall absence of cancer on biopsy also favored the VTP cohort (50% vs 14%; RR=3.48, 95% CI=2.444.98; p<0.0001). On post-hoc analysis, fewer VTP subjects had an ‘in-field’ positive biopsy (25% vs 65%; RR=0.38, 95% CI=0.30-0.50; p<0.0001) and ‘in-field’ progression to GG >1 (10% vs 34%; RR=0.30, 95% CI=0.19-0.47; p<0.0001). After VTP treatment, the rate of negative biopsies was similar in the apex (72%), mid-gland (78%), and base (74%), indicating that treatment was equally effective in each area. Conclusions: In this prospective randomized trial of men with low risk prostate cancer, VTP significantly reduced the subsequent finding of GG 2 or higher cancer on biopsy relative to AS. Additionally, fewer men in the VTP cohort crossed over to RT, a clinically meaningful benefit that lowers treatment-related morbidity, with stable risk reduction over 4 years. Introduction Given the low probability of mortality from low-risk prostate cancer (PCa), therapeutic choices should reflect a balance between cancer control and quality-of-life. Current guidelines recommend active surveillance (AS) as the preferred treatment option. In practice, however, treatment algorithms are more complex, influenced by the substantial rates of reclassification or progression to higher grade or larger volume cancer over time, and by patient choice. Low-risk PCa does not always remain indolent and modern surveillance strategies are imperfect at detecting its biological transition to more threatening disease. In published cohorts of men on AS, some 25-60% cross-over to radical therapy (RT; radical prostatectomy, radiation therapy) within 5-10 years, exposing them to substantial treatmentrelated morbidity. Over the last decade, various technologies have been developed for focal ablation of the prostate (e.g., cryotherapy, high intensity focused ultrasound (HIFU), vascular-targeted photodynamic therapy (VTP)) with the aim of reducing cancer progression while preserving quality of life. The first multi-center, phase 3, prospective randomized trial evaluating a focal ablative treatment for localized PCa was recently published. Conducted in 47 centers in 10 European countries, CLIN1001 PCM301 is the only successfully-accrued randomized trial evaluating a novel treatment for localized PCa. VTP employs padeliporfin di-potassium (TOOKAD®), a stable, bacteriochlorophyll-derived photo-sensitizer. When excited by near-infrared light (753nm), TOOKAD generates superoxide and hydroxyl radicals, which initiate a cascade of events leading to rapid vascular occlusion and subsequent coagulative necrosis of the targeted prostate tissue. For prostate VTP, intravenously injected TOOKAD is excited within the prostate gland with laser light delivered via trans-perineally placed light diffusers. PCM301 randomized 413 men with low-risk PCa to VTP or AS. The study included patients with low-risk PCa (PSA ≤ 10 ng/ml, Gleason score 6 or less (Gleason Grade Group [GG] 1) and clinical stage T2a or less), having 2-3 positive cores with maximum cancer core length (MCCL) ≤5mm or 1 positive core with MCCL between 3 and 5mm, prostate volume 25-70 cc, and at least 10 years life expectancy. Demographics, patient characteristics and exclusion criteria are detailed in the original publication. PCM301 met both its co-primary as well as secondary end-points. Briefly, at 2 years, in the intent-totreat (ITT) analysis, VTP was superior to AS for absence of cancer on end-of-study biopsy (p <0.001) and lower risk of cancer progression (p<0.001). In the protocol, progression was defined as meeting any of the following criteria: GG >1, >3 positive cores, MCCL >5mm, persistent PSA elevation (>10 ng/ml on 3 consecutive measures), clinical stage T3 or higher, or development of metastasis. At 2 years cross-over to RT was lower in the VTP arm (p <0.001). These positive trial results led to the approval of TOOKAD in 2017 by the European Medicines Agency’s Committee for Medicinal Products for Human Use (CHMP) for the treatment of unilateral low-risk, but not very low-risk, localized PCa. For subjects participating in PCM301, the decision to cross-over to RT was largely based upon pathologic results of per-protocol biopsies at 12 and 24 months after randomization. Since the coprimary endpoints of PCM301 were assessed at 24 months, patients who crossed over to RT after that time point were not captured in the initial report. The current follow-up study presents additional data up to 4 years from randomization, as well as post-hoc analyses of 2-year biopsy outcomes and reasons for RT. Materials and Methods The initial PCM 301 protocol and extended follow up (PCM301 5FU) was approved by the IRB at study initiation. To assess the intermediate results of the planned 5-year period of extended follow up, the dataset was frozen on August 30, 2017. After 24 months, study participants initially randomized to VTP or AS were managed by their physician following a ‘local standard-of-care’ principle. Management decisions, including the need for biopsy, were at the discretion of individual physicians and patients in each center. Using an electronic secure web application, data collection was performed per usual practice at each center for indicators of PCa progression, additional treatment, morbidity and patientreported quality of life. Information about each patient’s vital status was specifically sought for all patients originally randomized, including those lost to follow-up. Results at 4 years were calculated for cumulative risk of and reasons for cross over to RT, and for metastasis-free, cancer-specific and overall survival rates. Additionally, a post-hoc analysis of biopsy results during PMC301 was conducted. We assessed per patient progression in GG status to GG 2 or higher, the location and grade of positive biopsy results both ‘in-field’ (within the VTP treated lobe, or, for AS, within the lobe containing the largest, index cancer) and ‘out-of-field’ (in the untreated lobe in the VTP cohort and the contralateral lobe in the AS cohort). Statistical analyses were done with SAS version 9.3. All randomised participants were included in the efficacy analyses according to assigned treatment (intention-totreat, or ITT, groups). Missing data were not imputed. Time-to-progression was compared between the two treatment groups by the logrank test and quantified using a Cox proportional hazards regression model to derive hazard ratios at 2 years. Treatment group, age, number of positive cores, prostate volume, and disease status at baseline were used as covariates. Biopsy results at 2 years were analyzed with censoring of patients at time of M12 biopsy when M24 biopsy was missing. A chi-square test and risk ratios were used to compare proportions of participants in the two treatment groups according to the biopsy outcome. Time to initiation of RT was estimated by the Kaplan-Meier method. Comparison between the two treatment groups was done using the log-rank test and quantified with a univariable Cox proportional hazards regression model to derive hazard ratio at 4 years. Results The VTP cohort had a lower risk of cross-over to RT than the AS cohort at 2 years (7% vs 33%), 3 years (14% vs 44%) and 4 years (24% vs 53%) (HR=0.31, 95% CI=0.21-0.45; p<0.001) (Figure 1). The absolute risk reduction of RT by VTP at 2, 3 and 4 years was 26%, 30% and 29%, respectively, compared to AS. Triggers for RT were similar in the VTP and AS arms: increase in grade to GG 2 or higher (68% of 25 cross-overs in the VTP cohort vs 49% of 77 in the AS cohort), increase in cancer volume without change in grade (16% of VTP vs 32% of AS), and patient choice (16% vs 18%) (Table 1). By 3 years, there was no significant difference in the proportion of men who crossed over to RT without evidence of progression on biopsy: 4 of 25 (16%) after VTP versus 14 of 77 men (18%) in the AS arm (RR 1.14, CI 0.41-3.14). Four-year metastasis-free (99% vs 99%), cancer-specific (100% vs 100%) and overall survival (98% vs 99%) were similar between cohorts. Rates of progression (Figure 2), and of progression in grade (Figure 3), were separately analysed post-hoc for the whole gland and the ‘in-field’ lobe and are reported at 2 years since biopsies were not mandated beyond that time point. Progression rates were significantly lower in the VTP than the AS cohort for overall progression (HR=0.35; 95%CI = 0.25-0.48) and progression in grade