Actionable gene alterations in resected non-small cell lung cancer (AGA-R study).

8035 Background: To date, molecular testing indication in resected non-small cell lung cancer (NSCLC) is limited to epidermal growth factor receptor ( EGFR ) mutations (mut) and anaplastic lymphoma kinase ( ALK ) rearrangements, because of demonstrated benefit and approval of adjuvant targeted therapies. The use of next generation sequencing (NGS), routinely adopted in the metastatic setting to guide treatment, is very limited in resected NSCLC. The prevalence of driver gene alterations other than EGFR and ALK , and their impact on adjuvant treatment outcomes, disease recurrence (DR) and survival remain unclear. Methods: We retrospectively analyzed molecular, clinical and survival data from consecutive Caucasian patients (pts) who underwent surgery for stage IA-IIIB NSCLC (AJCC 8 th Edition) and had NGS performed on tumor tissue between January 2020 and December 2023 at our Institute. Primary endpoint was the prevalence of driver gene alterations in the overall cohort. Exploratory analyses were planned to evaluate DR, disease free survival (DFS) and overall survival (OS) and their relationship with the mutational status. Results: Overall, 216 resected NSCLC pts had NGS available. The prevalence of oncogenic driver alterations was 71%, the most common being KRAS (30%; 13% G12C and 17% non-G12C), followed by EGFR (26%; stage I: 29%), with exon 19 deletions (13%), exon 21 L858R substitution (6%), exon 20 insertion (2%), and uncommon mut (5%). Other detected alterations included MET exon 14 skip (6%; stage I: 8%), BRAF (4%; 2% V600), and HER2 exon 20 mut (3%). Only 1% of cases had ALK or RET rearrangements. The overall prevalence of gene alterations was similar in men (69%) and women (70%), however KRAS and MET exon 14 skip were reported more frequently in men, EGFR in women. Among 181 pts with available follow up (median f up 14mo), n=52 DR events were observed. Median time to DR was 13.5 months (95% CI 11-16mo). Of note, among resected stage I pts with EGFR common mut who did not receive adjuvant TKI (n=10), 30% DR occurred. The overall highest DR rates (75%) were observed in the presence of gene fusions, exon 20 ins, and BRAF non-V600, followed by (50%) KRAS non-G12C and HER2 mut. The lowest DR rate was observed in MET ex14 skip (8%) and BRAF V600 (0%). DR rate in other mut subtypes was similar to that observed in wild-type population. Median DFS was 32 months (24-NA), OS data are still not mature at data cut-off. Conclusions: Our study detected driver mutations in 70% of resected NSCLC, including stage I. The prevalence of EGFR and MET exon 14 skip mut in the early stage setting almost doubled the reported prevalence in the metastatic setting. Differential DR rates according to specific mut subtypes, are hypothesis-generating, and suggest the need of NGS testing to inform prognosis and personalize follow up in current clinical practice. Further investigation on tailored adjuvant treatments, also in stage I resected tumors, is supported by our results.