Paclitaxel plus bevacizumab with or without atezolizumab for hormone receptor-positive/HER2-negative advanced breast cancer: JCOG1919E (AMBITION) phase III trial.

Abstract

1019 Background: Hormone receptor-positive/HER2-negative (HR+/HER2-) advanced breast cancer (aBC) is immunologically “cold,” with limited clinical benefit from immune checkpoint inhibitors. Since VEGF-mediated angiogenesis promotes an immunosuppressive tumor microenvironment, we hypothesize that inhibiting VEGF could reverse immunosuppressive state thereby augmenting response to immunotherapy. The JCOG1919E (AMBITION) phase III trial evaluates whether adding the anti-PD-L1 antibody: atezolizumab (ATZ) to paclitaxel (PTX) plus anti-VEGF antibody: bevacizumab (BEV) improves outcomes in patients (pts) with HR+/HER2- aBC. Methods: This multicenter, open-label, randomized phase III trial enrolled pts with HR+/HER2-aBC who had no prior chemotherapy for aBC and were endocrine-resistant or visceral crisis. Pts were randomized in a 1:1 ratio to PTX plus BEV (Arm A) or PTX plus BEV with ATZ (Arm B). Randomization was stratified by disease status (de novo stage IV vs. recurrent), liver metastasis (presence vs. absence), and PD-L1 expression (SP142 assay; negative [<1% on immune cells (ICs)] vs. positive [>=1%]). PTX was administered on days 1, 8, and 15 of a 28-day cycle, BEV on days 1 and 15, and ATZ on days 1 and 15 in Arm B. The primary endpoint was investigator-assessed progression-free survival (PFS). Secondary endpoints included overall survival (OS), objective response rate (ORR), PFS by blinded independent central review, and safety. Clinical trial information: NCT04732598. Results: From Sep 2021 to Sep 2023, 281 pts were randomized. At baseline, 33.1% were de novo stage IV, 68.0% had liver metastases, and 16.0% were PD-L1 positive. At a median follow-up of 30 months, median investigator-assessed PFS was 11.2 months in Arm A and 12.4 months in Arm B (hazard ratio [HR] 0.88; 95% CI, 0.67–1.15; p=0.17). Median OS was immature, but numerically longer in Arm B than in Arm A (39.1 vs. 31.2 months; HR 0.80, 95% CI, 0.58–1.11). The ORR was high and comparable between arms (71.9% vs. 73.0%). Grade ≥3 adverse events (AEs) were similar between arms. The most common immune related AEs (any grade) with ATZ were rash (17.3%), adrenal insufficiency (11.5%) and hypothyroidism (10.8%). No treatment-related deaths were observed, and no new safety signals emerged. Conclusions: PTX plus BEV with ATZ did not demonstrate a statistically significant improvement in PFS compared to PTX plus BEV as a frontline treatment for HR+/HER2- aBC, although a numerical extension in OS was observed. The safety profile of PTX plus BEV with ATZ was consistent with the known profiles. Clinical trial information: NCT04732598 . ARM A PTX+BEV (n=140) ARM B PTX+BEV+ATZ (n=141) PFS events, n 117 109 Median PFS (95% CI), months 11.2 (9.6-13.5) 12.4 (10.3-15.2) HR (95% CI) 0.876 (0.670-1.145) P-value 0.168 OS events, n 81 71 Median OS (95% CI), months 31.2 (23.9-34.7) 39.1 (30.0-47.8) HR (95% CI) 0.804 (0.584-1.108) ORR, % 71.9% 73.0%