Ceralasertib (cerala) + olaparib (ola) in patients (pts) with homologous recombination repair (HRR)-deficient platinum-sensitive relapsed ovarian cancer (OC) after progression on prior PARP inhibitor (PARPi) treatment (tx).

5542 Background: Combining an ATR inhibitor (ATRi) and a PARPi may overcome acquired PARPi resistance based on preclinical and clinical data. We report a Phase 1 study (NCT02264678) of cerala (ATRi) + ola (PARPi) in pts with HRR-deficient OC who had progressed on prior PARPi tx. Methods: Pts had histologically confirmed high-grade serous/endometrioid epithelial ovarian, fallopian tube, or primary peritoneal cancer, with deleterious or suspected deleterious BRCA or HRR mutations ( BRCA m ; HRRm: RAD51C/D m , PALB2 m) or HRR-deficiency (HRD+). Pts received oral cerala 80 mg (d 1–14 every 28 d) + ola 150 mg (throughout) twice daily without/with (Cohort [Co] 1/2) intervening platinum-based tx after initial response and subsequent progression on prior PARPi tx. Primary endpoints were safety/tolerability; ORR per RECIST v1.1 and PFS were secondary endpoints. Emergence of PARPi resistance mechanisms was analyzed in archival/pre-study tx tissue samples. Results: 32 pts were treated, 30 in Co1 (7 ongoing tx at data cutoff: Mar 5, 2024) and 2 in Co2 (cohort closed early). Median age was 63.0 yrs; 68.8% (22/32) had BRCA m/HRRm and 31.3% (10/32) HRD+/non- BRCA m by local assessment. All pts had adverse events (AEs); 40.6% had grade ≥3 AEs (most common were anemia and platelet count reduced, each 21.9%); 6.3% discontinued tx due to AEs. Table shows efficacy in Co1. 24 pts had post-PARPi biopsies evaluable for genomics analysis: 3 (12.5%) had BRCA reversions (rev; 13 pts were BRCA m); 3 (12.5%) had loss of function alterations in DNA damage response (DDR) rewiring genes. Of 16 pts with post-PARPi biopsies evaluable for RAD51 foci, 81.3% (13) had RAD51 high status suggesting HRR functional proficiency as a prevalent PARPi resistance mechanism; of these 13, 61.5% (8; 80% CI, 40.2–79.9) responded to cerala + ola, while no responses occurred in 3 RAD51 low pts (80% CI, 0.0–53.6). Responders included patients with BRCA rev or TP53BP1 m, indicating cerala + ola activity in pts with BRCA rev and alterations in DDR rewiring genes. Other PARPi resistance mechanisms were rare. Conclusions: In this setting of high unmet need, cerala + ola had acceptable safety, a low discontinuation rate, and clinical activity in both BRCA m/HRRm and HRD+/non- BRCA m pts after progression on a prior PARPi. Exploratory analyses highlighted emerging PARPi resistance mechanisms; ongoing assessments of PARPi resistance to inform pt selection and novel combination strategies in post-PARPi settings will be presented. Clinical trial information: NCT02264678 . BRCA m/HRRm n=20 HRD+/non- BRCA m n=10 Total N=30 ORR, % (80% CI) 45 (29.3–61.5) 30 (11.6–55.2) 40 (27.7–53.3) Best response, n (%) Complete 2 (10) 1 (10) 3 (10) Partial 7 (35) 2 (20) 9 (30) Stable 8 (40) 4 (40) 12 (40) PFS Events, n 10 8 18 Median, months (80% CI) 7.5 (5.3–not calculable) 4.5 (1.8–5.3) 5.5 (4.7–7.5)