MacroGenics, Inc. announced results from Cohort A Part 1 of the Phase 2/3 MAHOGANY clinical trial of margetuximab. MARGENZA® (margetuximab-cmkb) is approved in HER2+ metastatic breast cancer and is being investigated as a potential first-line treatment for patients with HER2+ gastric cancer or gastroesophageal junction cancer in combination with a checkpoint inhibitor, with or without chemotherapy. The dataset is available in a poster titled Margetuximab With Retifanlimab in HER2+, PD-L1+ First-Line Unresectable/Metastatic Gastroesophageal Adenocarcinoma (GEA): MAHOGANY Cohort A (Poster #1379P) at the 2021 European Society for Medical Oncology (ESMO) Virtual Conference taking place September 16 to 21, 2021. The efficacy data and safety cutoff dates were July 19, 2021 and August 3, 2021, respectively. In Cohort A Parts 1 and 2, the efficacy and safety of combining margetuximab and retifanlimab (investigational anti-PD-1 monoclonal antibody licensed to Incyte by MacroGenics) is planned to be evaluated in approximately 100 patients whose tumors are HER2+ at the 3+ level by immunohistochemical (IHC) staining, PD-L1+ (combined positive score =1%) and non-microsatellite instability-high (non-MSI-H). A pre-specified interim analysis assessing efficacy and safety was conducted on the first 40 non-MSI-H patients enrolled in Part 1. These data support advancement to Part 2 with plans to enroll approximately 60 additional response-evaluable non-MSI-H patients. A total of 43 HER2 3+ and PD-L1+ patients were enrolled in Cohort A Part 1 and received margetuximab 15 mg/kg plus retifanlimab 375 mg/kg administered intravenously every three weeks. Twenty-five patients (58%) had gastric cancer and 18 patients (42%) had gastroesophageal junction cancer; 36 patients (84%) had metastatic disease at study entry. Anti-tumor activity was observed in patients treated with margetuximab plus retifanlimab in MAHOGANY Cohort A after the first scan. Tumor shrinkage was observed in 32 of 41 patients (78%) with at least one post-baseline target lesion measurement. Twenty-one of 40 response-evaluable patients achieved an objective response (53%, 95% confidence interval (CI): 36%-69%), including four confirmed complete responses and 17 confirmed partial responses. The number of confirmed responders by independent assessment exceeded the prespecified futility boundary for the trial, and enrollment is proceeding to Cohort A Part 2. Disease control was achieved in 29 of 40 patients (73%, CI: 56%-85%) and the median duration of response was 10.3 months (range: 2.1 14.5 months, CI: 4.6 months not evaluable (NE)). Median progression-free survival (PFS) was 6.4 months by independent assessment (CI: 6.0 months NE); median overall survival (OS) was not yet reached. At both 12 and 18 months, OS was 85% (CI: 63%-95%). Antitumor activity was comparable to historical data from the experimental arm of the Trastuzumab for Gastric Cancer (ToGA) study (trastuzumab + chemotherapy; n=294; objective response rate (ORR) of 47%; median duration of response (DOR) of 6.9 months) and initial data from the control arm (placebo + trastuzumab + chemotherapy) of the KEYNOTE-811 study (ORR of 52%; median DOR of 9.5 months). The safety analysis of all 43 patients treated with margetuximab plus retifanlimab suggests the combination was well tolerated in the study population. The most common TRAEs were fatigue (21% Grade 1-2, 0% Grade =3), infusion-related reaction (19% Grade 1-2, 0% Grade =3), rash (19% Grade 1-2, 0% Grade =3), diarrhea (16% Grade 1-2, 2% Grade 3), and pruritus (16% Grade 1-2, 0% Grade =3). A total of nine Grade 3 TRAEs were reported in eight patients (19%); no Grade 4 TRAEs were observed. Eight serious TRAEs were reported in seven patients. Infusion-related reactions considered as adverse events (AEs) of special interest occurred in six patients. Treatment-emergent AEs of Grade 3 occurred in 18 of 43 patients (42%) of patients. Three of 43 patients (7%) discontinued therapy due to immune-related AEs: renal dysfunction (Grade 3), hepatitis (Grade 3), and diabetic ketoacidosis (Grade 1); no AEs led to death. Safety data from MAHOGANY compare favorably to the experimental arm of ToGA in which overall Grade 3-4 AEs were 68% and the treatment-related mortality was 3%.1 Initial results from KEYNOTE-811 data presented at the 2021 ASCO Annual Meeting indicated that AEs of Grade 3-5 occurred in 57.1% of patients in the experimental arm (pembrolizumab + trastuzumab + chemotherapy) and in 57.4% of patients in the control arm, AEs leading to death occurred in 3.2% vs 4.6%, and AEs leading to discontinuation of any study drug occurred in 24.4% vs 25.9% of patients, respectively. Despite limitations of cross-study comparisons, regimens containing chemotherapy may have clinically relevant safety differences compared to the chemotherapy-free regimen in MAHOGANY Cohort A.