Multibiomarker analysis homing in on EGFR-TKIs therapy in non-small cell lung cancer.

Abstract

e18025 Background: EGFR mutation has been identified to be a strong predictive biomarker to EGFR-TKIs in non-small cell lung cancer (NSCLC), but it isn't exclusive molecular mechanism to TKI therapy. So developing platforms of serial of biomarkers detection is important for personalized TKI therapy. This study investigated the relationships between genetic factors and clinical outcome in Chinese NSCLC patients treated by gefitinib. Methods: A total of 156 NSCLC were enrolled for genetic screening and 101 of which were treated by gefitinib. EGFR AKRAS mutation (mEGFR, mKRAS) were identified by denaturing high performance liquid chromatography (DHPLC) and EGFR copy number AEML4-ALK rearrangement were detected using fluorescent in situ hybridization (FISH). A polymorphic dinucleotide repeat (CA simple sequence repeat1 [CA-SSR1]) was determined by DNA sequencing. Results: mEGFR, mKRAS, EGFR FISH positive and EML4-ALK rearrangement were detected in 62 (39.7%),13 (8.3%), 49 (20.2%), 6 (3.8%) pts, respectively. Pts with shorter (CA)n repeat genotype (< 18) account for 45.5%. The mKRAS was found to be significantly higher in the shorter (CA)n than in the longer group (≥ 18) (p = 0.018), but EGFR FISH positive was higher in longer (CA)n group (p = 0.004) and the associations between (CA)n length and EGFR mutations and EML4-ALK rearrangement were observed. Among 102 pts with gefitinib therapy, EGFR mutant had significantly higher ORR than wild-type group (28/47, 60% versus 9/54, 17% p = 0.000). Further analyzing 54 patients with EGFR wild-type, 32 was in longer (CA)n and 22 in shorter (CA)n. Of 9 pts carrying wild-type EGFR who had response to gefitinib, 8 showed longer (CA)n polymorphism, and only 1 was in shorter (CA)n (p = 0.048). All of 6 mKRAS pts were present in the 22 shorter (CA)n group. The survival analysis showed the pts with mEGFR (9.4months ver 6.8months, p = 0.055), KRAS wild-type (8.4 months ver 3.1months, p = 0.019) and longer (CA)n (9.4months ver 6.2months, p = 0.073) had longer PFS. Conclusions: Multibiomarkers analysis may be more accurate for predicting response to TKI than single biomarker. The valid combination of molecular testing should be validated by large- samples study. No significant financial relationships to disclose.