The landscape of ARID1A variants and correlation with tumor mutational burden in Chinese solid tumor patients.

Abstract

e14624 Background: ARID1A (the AT-rich interaction domain 1A) is one of the most commonly mutated genes in cancer, mostly caused by frameshift variants. Here, we describe the mutational landscape of ARID1A in Chinese cancer patients and its correlation with tumor mutational burden (TMB). Methods: Formalin fixed paraffin embedded (FFPE) samples of 3,828 Chinese solid tumor patients were collected for next‐generation sequencing (NGS) based 450 genes panel assay. Genomic alterations including single base substitution, short and long insertions/deletions, copy number variations, gene fusions and rearrangement were assessed. Microsatellite stable (MSS) status and TMB were also acquired by NGS algorithms. Results: A total of 340 (8.9%) patients were found to have ARID1A variants. The frequency of ARID1A mutations was 4% in lung adenocarcinoma (N = 74/1833), 8% in squamous cell lung carcinoma (N = 23/278), 12% in colorectal adenocarcinoma (N = 61/530), 11% in hepatocellular carcinoma (N = 52/484), 14% in pancreatic adenocarcinoma (N = 25/177), 15% in gallbladder adenocarcinoma (N = 14/94), 22% in intrahepatic cholangiocarcinoma (N = 47/213), 20% in gastric adenocarcinoma (N = 39/200) and 26% in endometrial adenocarcinoma (N = 5/19). ARID1A variants included truncation (78%), missense/indel (non-frameshift substitution/indel) (16%), splicing site (4%) and rearrangement (2%). Truncation was the most common, recurring in a few hot spots (R1276*, D1850Gfs*4, A339Lfs*24, P224Rfs*8 and Q372Afs*28/Sfs*19). In general, ARID1A mutations correlated with higher TMB (median 17.6 vs 7.4 muts/Mb, P < 0.001) in MSS tumors; especially in lung adenocarcinoma (17 vs 7.5 muts/Mb, P < 0.01), squamous cell lung carcinoma (16.8 vs 11.2 muts/Mb, P < 0.01). TP53 (60%), APC (34%) and KRAS (32%) were the most common co-occurred mutated gene in ARID1A (+) tumors. In more than 20% of ARID1A mutations, PI3K/mTOR pathway genes were altered, such as PIK3CA, PTEN, RNF43 and PIK3R1 genes. Conclusions: Since ARID1A mutations correlate with higher TMB, it could be a potential predictor of response to PD-1/PD-L1 immune checkpoint pathway inhibitors. As reported, ARID1A deficiency promotes mutability and potentiates therapeutic antitumor immunity unleashed by immune checkpoint blockade. Patients harboring ARID1A and PI3K/mTOR pathway mutations may provide useful information for drug discovery and biomarker exploration.