Assessment of patient-reported outcomes (PROs) and longer-term adverse events (AEs) in phase I study of ²²⁵Ac-J591-PSMA for metastatic castration-resistant prostate cancer (mCRPC).

Abstract

77 Background: Prostate Specific Membrane Antigen (PSMA) is a conserved cell surface protein in PC and is used for targeted imaging and therapeutics. Antibodies circulate longer than small molecules and are less likely to reach luminal PSMA on normal organs. Here we report PROs and longer-term AEs from the dose-escalation and expansion cohorts of a first-in-human study of combined monoclonal antibody and potent alpha emitter (225Ac-J591). Methods: Eligible subjects with mCPRC were administered 225Ac-J591. Initial to maximum doses were 13.3 to 93.3 KBq/kg). AEs are reporting using CTCAE v5 and PROs, including pain (BPI-SF) and quality of life (QOL, FACT-P), and associations with PSA response were also examined. Results: A total of 32 subjects (one enrolled in both dose-escalation and expansion) were treated with a single dose of 225Ac-J591 across 7 dose levels with expansion at the level (93.3 KBq/kg, n = 16). Median age 69.5 (52-89) and PSA 149.1 (4.8-7168.4). All subjects had at least 1 AE of any grade. Most common were fatigue (31/32, 1 Gr > 2), anorexia (25/32, all Gr 1-2), and thrombocytopenia (25/32, 3 Gr 3, 2 Gr 4). Xerostomia was observed in 14/32 subjects (all Gr 1), 7 of whom had prior 177Lu-PSMA. Pain flare was reported in 43% (17/32) subjects (11 Gr 1, 6 Gr 2). 19 had evaluable PROs at baseline and efficacy visit (week 12). Pain severity (p = 0.8) and interference from pain (p = 0.4) were unchanged from baseline to 12 weeks, yet better PSA response (by percent) was associated with reduced pain severity (r = 0.7, p = 0.0023). Despite at least one AE in each subject, total FACT-P was not significantly changed after treatment (p = 0.2), but emotional well-being declined over time (15 [10.0, 18.0] v 10.0 [7.5, 13.0], p = 0.011). Reduction in median emotional well-being reached clinically important score differences. When stratified by AE, subjects with xerostomia had lower FACT-P total scores, but no difference was observed between those with and without pain. PSA response was not associated with change in QOL or subscales. Conclusions: Pain and quality of life in subjects with mCRPC did not change, on average, from baseline to 12 weeks after treatment with 225Ac-J591. This is despite preliminary evidence of clinical efficacy being accompanied by frequent, treatment emergent AEs. A promising trend toward improved pain in those with PSA response warrants further analysis. Small numbers limited statistical power for testing other subgroup associations. Additional correlations with pretreatment sites of disease, performance status, and adverse event distribution are ongoing. Assessment of changes in PROs in the follow up studies [NCT04506567] are underway. Clinical trial information: NCT03276572.