Phase 2 study of letrozole, abemaciclib, and metformin in estrogen receptor (ER)–positive recurrent endometrial cancer (EC).

Abstract

5513 Background: Preclinical studies have demonstrated synergism with simultaneous inhibition of the estrogen receptor (ER), CDK4/6 and PI3K pathways. Metformin suppresses PI3K signaling directly via activation of the AMP-activated protein kinase (AMPK) and indirectly via downregulating the insulin/IGF-1 signaling pathway. We conducted a phase 2 study of letrozole/abemaciclib/metformin in ER positive EC. Methods: Patients (pts) with recurrent ER positive (≥1% immunoreactive tumor nuclei) endometrioid EC, measurable disease, any number of prior therapies and any prior hormonal therapy but no prior CDK4/6 inhibitor received abemaciclib 150 mg PO bid, metformin 500mg PO qd and letrozole 2.5 mg PO qd until progression or unacceptable toxicity. Primary endpoints were objective response (OR) rate (ORR) and progression-free survival (PFS) rate at 6 months (PFS6). A safety lead-in was included, and target accrual was 25 pts; if there were ≥6 ORs or ≥9 pts without disease progression or death at 6 months, letrozole/abemaciclib/metformin would be considered worthy of further investigation. Correlative studies included pharmacokinetic (PK) analyses of metformin alone and in combination with letrozole/abemaciclib, molecular profiling using Oncopanel targeted NGS, and progesterone receptor (PrgR) expression by IHC. Results: As of 10/4/2024, all 25 pts received protocol therapy. Median follow up was 17 months, median number of prior lines was 2 and 18 (72%) pts had previously received hormonal therapy. Eight pts exhibited OR: 3 complete responses (CRs) and 5 partial responses (PRs),ORR 32% (95% CI 14.9% to 53.5%). Sixteen (64%) pts had stable disease (SD) and 1(4%) pt progressive disease (PD) as best response. Kaplan Meier estimate of PFS6 was 69.7% and median PFS was beyond 19.3 months. Most common G3+ treatment-related toxicities were G3 neutropenia (24%) and G3 fatigue (16%). No pts discontinued therapy because of toxicity. PK analyses demonstrated that metformin plasma concentrations were ~3-fold higher when combined with letrozole/abemaciclib compared to metformin monotherapy. Molecular profiling showed no objective responses in TP53 mutated ECs and no objective responses in pts with NSMP ECs with RB1 or CCNE1 alterations; median PFS was only 3.8 months in these tumors. All objective responses were observed in pts with NSMP ECs without RB1 and CCNE1 alterations; these pts exhibited an ORR of 50% and PFS6 of 87.5%. There were no MMRD and no POLE -mutated tumors. Responses were observed regardless of PrgR expression. Conclusions: Addition of metformin (at plasma concentrations sufficient to inhibit the PI3K pathway) to letrozole/abemaciclib is feasible and safe, and appears to induce deeper responses (including complete responses) and more prolonged PFS than letrozole/abemaciclib alone. NSMP tumors without RB1 and CCNE1 alterations derive the most benefit from this regimen. Clinical trial information: NCT03675893 .