PD.1.06 EGFR Uncommon Mutations Frequency in a 1,688 NSCLC Patients Database in Brazil

Abstract

Background: Lung cancer remains as the leading cause of cancer-related deaths worldwide, with non-small cell carcinoma (NSCLC) being the most frequent histology (85%). Therapies that target activating EGFR mutations are NSCLC first-line treatments that improve patient life expectancy and quality of life. However, despite the current availability of novel TKI drugs in Chile and other Latin-Americans countries, little is known about the molecular epidemiology of EGFR mutations in the region. In particular, Latin American patients are still largely under-represented in genomics databases or clinical trials for this disease. We are presenting the results of a large NGS-based screening of the NSCLC EGFR mutations in Chilean patients. The study is aimed at describing the prevalence of specific somatic actionable mutations, within this population and correlated data with relevant clinical and tumor aspects such as gender, age, specific histology, smoking habits, progression stage and co-occurrencewith KRAS mutations and other mutations. Method: Non-small cell lung cancer FFPE samples from 821 subjects, part of the non-interventional clinical study NIRVANA (NCT03220230), were sequenced using Oncomine Focus Assay (Thermo Fisher Scientific), which covers the most frequent and actionable EGFR mutations through nine amplicons of w100bp, spanning eight exons, to call variant at an expected mean coverage of 1000x. Variants called by the Ion Reporter Server were manually filtered by allele frequency >¼ 0.05 and at least 10 supporting reads, discarding synonymous substitutions and homozygous reference genotypes. Variants were annotated against COSMIC, dbSNP and ExAC databases to classify them in known or novel variants. Functional impact algorithms SIFT and PolyPhen were used. Clinical information was gathered on a GCP-compliant setup and monitored 100% on-site. Prevalence is reported within 95% CI, and associations with clinical characteristics are evaluated by Chi-Squared test or ANOVA, as appropriate (p<0.05). Results: We found that 214 out of 821 subjects, or 26.06% (95% CI: 23.15% to 29.14%) of the studied population, harbor an EGFR mutation. Among known variants, exon 19 E746_A750 deletion (COSM13243) and exon 18 A698T (COSM41905) were the most abundant variations. Non-smokers and under 45 years old subjects were more likely to have a relevant EGFR mutation (p<0.01), in agreement with the literature. Gender did not correlate significantly as a predictor of EGFR mutations (p>0.15) in the cohort analyzed. Conclusion: We expect these results shed light into the Chilean molecular epidemiology and supports initiatives to establish new diagnosis methods and treatment opportunities for Chilean cancer patients.