Association of a clinical model based on skeletal muscle radiation attenuation with efficacy for patients with gastric cancer treated with chemotherapy plus PD-1 antibodies: A single-center retrospective study.

Abstract

e16029 Background: Biomarkers to select gastric cancer (GC) patients who can benefit from PD-1 antibody-based treatment are under investigation. We aimed to assess the association between body composition parameters and efficacy of chemotherapy plus PD-1 antibodies in GC patients. Methods: Clinical information of GC patients receiving perioperative chemotherapy with or without PD-1 antibodies was reviewed. Tumor regression grade (TRG) of gastric cancer was recorded after resection. Radiation attenuation and index of skeletal muscle, visceral adipose tissue and subcutaneous adipose tissue were calculated based on computed tomography (CT) images at the third lumbar (L3) level using SliceOmatic software. Associations between clinical parameters and patients’ outcomes including TRG, progression-free survival (PFS), and overall survival (OS) were analyzed. Results: A total of 120 patients treated with chemotherapy plus PD-1 antibodies (IO cohort) and 82 patients treated with chemotherapy alone (CTx cohort) were enrolled. TRG0/1 rates in the IO cohort and CTx cohort were 45.8% and 32.9%. High skeletal muscle radiation attenuation (SMRA), neutrophil to lymphocyte ratio (NLR) < 2.65, and weight loss < 5% were associated with TRG0/1 in the IO cohort but not in the CTx cohort. High SMRA was an independent prognostic factor of PFS in the IO cohort (HR = 0.333, 95%CI 0.115-0.970, P= 0.044). A clinical model based on SMRA, NLR and weight loss was established to stratify patients into 3 groups, including group 1: High SMRA with NLR < 2.65 and weight loss < 5%, group 2: High SMRA with NLR≥2.65 and/or weight loss≥5%, and group 3: Low SMRA. In the IO cohort, patients in group 1 showed higher TRG0/1 rate (n = 27/40, 67.5%, P< 0.001) than group 2 (n = 26/63, 41.3%) and group 3 (n = 2/17, 11.8%). The clinical model was also associated with patients’ survival. Conclusions: A clinical model combining baseline SMRA, NLR and weight loss could identify well-responding GC patients treated with chemotherapy plus PD-1 antibodies.