Genomic Characterization of Unresectable/recurrent Early-Onset Gastric Cancer by Comprehensive Genomic Profiling Tests.

393 Background: The incidence of early-onset gastric cancer (EOGC: age ≤45 years) is showing an increasing trend. Although EOGC has unique clinicopathological features compared with late-onset gastric cancer (LOGC: age>45 years), disparities in genetic alterations (GAs) between EOGC and LOGC are not fully understood in patients with unresectable/recurrent GC. We aim to investigate to elucidate unique GAs in unresectable/recurrent EOGC. Methods: We retrospectively reviewed data on patients with unresectable/recurrent GC who underwent comprehensive genomic profiling (CGP) tests and were registered as “Stomach adenocarcinoma (STAD)” cases in the Center for Cancer Genomics and Advanced Therapeutics (C-CAT) in Japan from June 2019 to April 2023. CGP tests were conducted by NCC Oncopanel System (NOP) or FoundationOne CDx (F1). We examined somatic GAs, microsatellite instability (MSI) status, and tumor mutation burden (TMB) of GC. TMB-high was defined as ≥10 mutations/Mb. Actionable GAs were counted if they were classified as “pathogenic/oncogenic” or “likely pathogenic/likely oncogenic variants”. Druggable GAs are actionable GAs that can be treated with drugs approved for GC or other types of cancer or have shown efficacy in clinical trials for either GC or different types of cancer. Results: In total,112 patients with EOGC and 810 patients with LOGC were identified. In EOGC group, most common actionable GAs included TP53 (59.8%), ARID1A (17.0%), KRAS (15.2%), CDKN2A (11.6%), FGFR2 (11.6%), and CDH1 (9.8%). Compared to the LOGC group, the EOGC group exhibited a significantly higher frequency of actionable GAs in FGFR2(11.6% vs. 4.6%, P=0.0059), CDH1 (9.8% vs. 4.1%, P=0.015) and BAP1 (3.6% vs. 0.6%, P=0.016). On the other hand, the EOGC group showed a significantly lower frequency of actionable GAs in ERBB2(7.1% vs. 20.1%, P=0.0004), CCNE1 (4.5% vs. 19.0%, P=0.0001), and AURKA (0% vs. 7.9%, P=0.0003) compared with the LOGC group. In the EOGC group compared to the LOGC group, the proportions of MSI-high were 0.9% vs. 2.6% (P=0.345), and those of TMB-high were 4.5% vs. 11.6% (P=0.011). Excluding ERBB2 amplification, druggable GAs in the ECOG group included FGFR2 amplification (9.8%), MET amplification (4.5%), EGFR amplification (1.8%), KRAS G12C mutation (1.8%), and MDM2 amplification (0.9%). Conclusions: Among unresectable/recurrent GC population, EOCG has displayed distinct genetic profiles of GAs compared to LOGC. CGP tests may be useful in expanding treatment opportunities for patients with unresectable/recurrent EOGC.