Longitudinal <i>EGFR</i> assessment in plasma and tissue samples in early non–small cell lung cancer (NSCLC).

8038 Background: Osimertinib has become the standard adjuvant treatment for patients (pts) with surgically resected, early-stage, EGFR -mutated non-small cell lung cancer (NSCLC), after the results of the ADAURA trial. However, the temporal distribution of EGFR mutations (mut) and its relation with recurrence patterns in this population have not been established yet. Aim of the study is to describe the prevalence of circulating tumor DNA and EGFR mut at different timepoints in early-stage NSCLC and the correlation between their presence and prognosis. Methods: This is a single center study conducted at the Vall d’Hebron Institute of Oncology, including consecutive pts with surgically resected, stage I-III NSCLC harbouring EGFR mut from 2008 to 2024. Next Generation Sequencing (NGS) with ONCOMINE panel was performed on tissue samples, archival when surgery pre-dated the start of the study. NGS on plasma samples was performed using Guardant360 panel at timepoints: 1, 3 and 6 months after surgery and at recurrence. For pts receiving adjuvant osimertinib, additional plasma samples were collected before drug initiation and during treatment. Pts were referred for genetic consultation if NGS on tissue samples detected a TP53 mut with a variant allele frequency (VAF) &gt;30%, or if NGS on plasma identified TP53 or BRCA mut with a VAF &gt;20%, or a basal T790M mutation. Results: Currently, 70 pts were enrolled, of which 24.3% were male. Median age was 68 years, and 35.7% were former/current smokers. Pts with stage IA were 37.1%, IB 24.3%, IIA 2.9%, IIB 11.4%, IIIA 14.3%, IIIB 5.7%. All pts had an EGFR mut, 54.2% detected by NGS on surgical or pre-surgical specimens. EGFR mut of the other samples were detected using PCR Cobas, and NGS are ongoing. EGFR mut was not detected on plasma after surgery (except for 1). Most pts harboured common mut (47.1% ex19del and 45.7% exon21 L858R on COBAS, 65.6% ex19del and 34.4% exon21 L858R on ONCOMINE). Interestingly, of the 43 pts with post-surgery NGS on plasma, 41.9% had a pathogenic mutation ( TP53 50%, ARID1 and BRCA1/2 11.1% each, 5.6% each SMAD4, MPL, TSC1, NOTCH1, JAK2, APC, FGFR1 and KRAS ) with a median VAF of 0.25% (tissue confirmation is needed to exclude hematopoietic origin). Liquid biopsies were obtained from all pts, with at least one sample collected post-surgery at varying time points. Adjuvant/neoadjuvant chemotherapy was administered to 31.4% of pts, while 17.1% received adjuvant osimertinib (100% after adjuvant osimertinib approbation by local label if indicated) and 11.4% also received radiotherapy. Disease recurrence occurred in 22 patients (31.4%), with 8 being local recurrence. At data lock 83% of pts were alive. Conclusions: Evaluation of EGFR in early-stage NSCLC should be standard procedure. We aim to identify a pattern associated with higher risk of recurrence and worse prognosis. Dynamic monitoring of EGFR could aid in personalising adjuvant treatments and follow-up scheduling.