Liquid and tissue profiling of somatic mutational landscape of non-small cell lung cancer from an Indian cohort.

Abstract

e21202 Background: Circulating tumor DNA (ctDNA) encompasses spatial and temporal heterogeneity, tumor evolution, and is ideal for repeat sampling and predictive therapeutic course corrections whereas tissue is the first choice for molecular profiling. We report the mutational landscape of an Indian NSCLC cohort, with focus on liquid biopsy and value addition offered by paired tissue and liquid profiling. Methods: The study comprised 507 tumor tissue and 1271 liquid biopsies from NSCLC patients. Evaluation of single nucleotide alterations, copy number alterations and fusions was conducted using targeted next generation sequencing with paired tissue and liquid profiling in a sub cohort (n = 45). Results: In tissue based analysis (n = 507), EGFR-sensitizing mutations were the most common alterations, detected in 41.4% of samples. Among these, Exon 19 deletion was seen in 46.3%, Exon 21 L858R in 23.9%, Exon 20 insertion in 4.6%, Exon 18 G719X in 3.7% and compound mutations in 5%. TP53, KRAS and PIK3CA mutations were detected in 34.5%,13.4% and 6.3% respectively. MYC was commonly amplified (15.4%) gene followed by PIK3CA (4.1%), CCNE1 (3.3%), EGFR (2.9%), and MET (2.9%) amplifications. Tissue fusion analysis (n = 398) detected ALK in 6.3%, ROS1 in 3.3%, and RET in 1.8%. METEx14 was detected in 4.5%. Liquid biopsy (n = 1271) showed comparable mutation spectrum as tissue, with the EGFR gene being the most altered (32.9%). Mutations in TP53 were observed in 26.4%, KRAS in 7.4%, PIK3CA in 4.4% and BRAF in 3%. In EGFR positive individuals, EGFR T790M was detected at higher frequency in liquid (16.7%) than tissue (8.6%), indicating this acquired resistance alteration being better captured in liquid biopsy than tissue due to tumor heterogeneity and tumor evolution. 2 liquid biopsy samples identified additional KRAS driver mutations not seen in the paired tissue analysis, highlighting the utility of simultaneous tissue and liquid analysis. Concurrent actionable alterations were detected in 11% of tissue and 5% of liquid samples. Conclusions: We report the largest liquid biopsy study reporting the genomic landscape from NSCLC cases of Indian origin along with tissue based molecular landscape. Presence of alterations like KRAS and EGFR T790M in liquid but absence in concurrent tissue biopsy shows utility of paired tissue and liquid biopsy profiling. Higher incidence of targetable and co-existing alterations in the Indian population highlight the utility of broad molecular profiling for successful utilization of targeted therapies. [Table: see text]