Abstract

1098 Background: New therapies are urgently needed for aTNBC. The PI3K/AKT/mTOR pathway and its link to chemoresistance warrant further investigation in pre-treated TNBC patients. The pan-AKT inhibitor IPA, alone or combined, has shown promise for treating advanced solid tumors. This study assessed the toxicity and efficacy of IPA with non-taxane CT in aTNBC pts. Methods: PATHFINDER (NCT04464174) is a multicenter, open-label, non-comparative, phase IIa trial with a safety run-in phase. Taxane-pretreated aTNBC pts with 1-2 prior CT lines for advanced disease and ECOG 0-1 were eligible. Based on prior treatment and available slots, pts received 21-day cycles of IPA (oral, 400 mg/d for 14 d) either with capecitabine (CA) (oral, 2 g/m2/d for 14 d) (arm A); eribulin (E) (IV, 1.23 mg/m2 on d1 and d8) (arm B); or carboplatin plus gemcitabine (CG) (IV, AUC 5 on d1 and 1000 mg/m2 on d1 and d8, respectively) (arm C) until disease progression (PD) or unacceptable toxicity. Initially, the safety run-in phase included 3 pts/arm to determine phase IIa doses based on the toxicity observed after 2 cycles. The primary endpoint was incidence of adverse events (AEs) as per NCI-CTCAE v.5.0. Key secondary endpoints included progression-free survival (PFS), overall survival (OS), and objective response rate (ORR). Analysis was exploratory with no hypothesis testing. Results: From September 2020 to October 2022, 54 pts were assigned to arms A (n = 22), B (n = 25), and C (n = 7). In arm C, 4 additional pts received a lower dose level of IPA (300 mg/d) in the safety run-in phase. Based on the toxicity profile, this arm did not proceed with the expansion phase. At data cutoff (10 th November 2023), median follow-up was 13.8 (2.4-35.5, arm A) and 11.2 (0.2-36.9, arm B) months (mo). All pts discontinued treatment, mainly because of PD (74.1%). Discontinuations due to AEs were 4.5% and 4% in arms A and B, respectively. Incidence of treatment-related (1) any grade (G) TEAEs was 81.8% and 88% and (2) SAEs was 4.5% and 20% for arms A and B, respectively. G≥3 TEAEs occurring in ≥10% of pts are shown (Table). No treatment-related deaths were reported. Median PFS was 2.7 (95%CI, 1.5-4.1) and 3.8 (95%CI, 1.5-9.6) mo; median OS was 15.5 (95%CI, 11.8-19.2) and 11.5 (95%CI, 8.8-25.1) mo; and ORR was 9.1% and 36% for arms A and B, respectively. Conclusions: Combining IPA with CA and E presented an acceptable safety profile. However, adding IPA to CG was not tolerable. E + IPA showed promising efficacy regardless of PI3K/AKT status that merits further evaluation. Clinical trial information: NCT04464174 . [Table: see text]