Long-term outcomes and overall survival (OS) for zanidatamab + chemotherapy in HER2-positive (HER2+) advanced or metastatic gastroesophageal adenocarcinoma (mGEA): 4-year follow-up of a phase 2 trial.

Abstract

4013 Background: Zanidatamab (zani), a dual HER2-targeted bispecific antibody, plus chemotherapy (chemo) has previously demonstrated antitumor activity and a manageable safety profile in the first-line (1L) treatment of patients (pts) with HER2+ mGEA. Here, we report a 4-year follow-up and the first report of both median OS and translational data from this phase 2 trial. Methods: The phase 2 trial (NCT03929666) evaluated zani + chemo (mFOLFOX6, CAPOX, or FP) in the 1L treatment of mGEA. In Part 1, pts had HER2-expressing (IHC 3+ or 2+) mGEA. Pts in Part 2 had HER2+ (IHC 3+ or IHC 2+/FISH+) mGEA by central assessment. After 25 pts were treated, antidiarrheal prophylaxis was added for cycle 1. The primary endpoint was confirmed objective response rate (cORR). Secondary endpoints included duration of response (DoR), progression-free survival (PFS), OS, and safety outcomes. Plasma ctDNA samples were collected for NGS testing (Guardant360). Results: In total, 46 pts were enrolled (zani + mFOLFOX6 [n = 24], CAPOX [n = 20], or FP [n = 2]). The majority (41 [89%]) of pts had HER2+ mGEA by central confirmation (ccHER2+); 35 (76%) pts had gastric/GEJ cancer. As of July 28, 2024, the median (range) follow-up was 48 (29-59) mo; 8 pts (17%) were on zani treatment and 19 (41%) in survival follow-up. Efficacy results are shown in the Table. The median OS was 36.5 mo; longest survival time was 57.9 mo (censored without death at data cutoff). The concordance between HER2 gene amplification by centrally assessed ISH vs plasma ctDNA was 90% (18/20). Of 14 pts with matched plasma samples at baseline and on-treatment (Cycle 2, day 15), 8 had a > 90% decrease in total ctDNA levels and 2 had a decrease in HER2 copy number. Common (> 5% of pts) grade 3 or 4 treatment-related AEs (TRAEs) were diarrhea (n = 18 [39%]), hypokalemia (n = 10 [22%]), vomiting (n = 4 [9%]), and nausea (n = 3 [7%]). Grade 3 or 4 diarrhea incidence was reduced from 52% to 24% after prophylaxis implementation. No deaths occurred due to TRAEs. Conclusions: After a median 4-year follow-up, zani + chemo demonstrated clinically meaningful efficacy in the 1L treatment of HER2+ mGEA, with durable responses and a median OS > 3 years, and a manageable safety profile. Zani + chemo markedly reduced total plasma ctDNA levels early in treatment of mGEA. Clinical trial information: NCT03929666 . All pts (N = 46) ccHER2+ GEA pts (n = 41) cORR a , n (% [95% CI]) 32 (76.2 [60.5, 87.9]) 31 (83.8 [68.0, 93.8]) Median DoR b (95% CI), mo 24-mo DoR, % (95% CI) 18.7 (10.4, 44.1) 40 (22, 58) 20.4 (8.3, 44.1) 41 (22, 59) Median PFS (95% CI), mo 12.5 (8.2, 21.8) 15.2 (9.5, 33.4) Median OS (95% CI), mo 36.5 (23.6, NE) 36.5 (23.6, NE) 24-mo OS, % (95% CI) 36-mo OS, % (95% CI) 65 (49, 77) 53 (37, 67) 67 (49, 79) 53 (36, 67) TRAEs, n (%) Any 46 (100) 41 (100) Grade 3 or 4 30 (65) 26 (63) a Response evaluable (n = 42 and 37). b Complete or partial response. NE, non-estimable.