Genomic landscape and therapeutic implications in advanced solid cancers: KOSMOS-II (KOrean Precision Medicine Networking Group Study of MOlecular profiling guided therapy based on genomic alterations in advanced Solid tumors, KCSG AL-22-09).

11161 Background: Despite advances in genomic diagnosis and therapeutics, providing precision medicine remains a challenge in the real-world practical application, especially in nations with diverse next-generation sequencing (NGS) cancer panels and a shortage of available targeted agents due to lack of approval or reimbursement. Methods: The KOSMOS-II is an ongoing prospective, nationwide, master observational study with multiple cohorts undergoing molecular-guided treatment (MGT) recommended by a central molecular tumor board based on local NGS cancer panel results for advanced cancer patients (pts) (J Clin Oncol 2023;41:16 suppl TPS 1608). Key genomic features were integrated with clinical data to construct a Clinico-Genomic Database (CGDB). To minimize inter-panel variations, all variant calling format (VCF) data on single nucleotide variants, insertions, and deletions were normalized by lifting over to the same reference genome (hg19). Re-annotation was performed using a consistent pipeline incorporating public databases. This preliminary analysis outlines the genomic landscapes and therapeutic implications in the first half of enrolled patients out of a targeted accrual of 1,000. Results: From September 2022 to December 2023, 489 pts were enrolled with a median age of 60 years (19 - 87), male 56.9%, and a median of 4 lines (1-15) of prior systemic therapy. The most common primary cancer sites were colorectum (n = 96, 19.6%), biliary tract (n = 63, 12.9%), and breast (n=40, 8.2%). CGDB was constructed for 360 pts, employing 20 different panels from 31 participating centers. The most common genomic alterations were TP53 mutation (Mut) (n=183, 50.8%), APC Mut (n=124, 34.4%), and ERBB2 amplification (AMP) (n=90, 25%). MGT was assigned to 59.5% of pts, 45.2% were allocated to investigational drugs targeting genetic alterations approved for other indications (Tier 1), and 12.6 % to clinical trials matching their genetic alterations (Tier 3). The concordance rate between pre-submitted physician choices and the MTB recommendation was 54.6% (Tier 1, 55%; Tier 2 (alternative treatments including palliative care), 46%; Tier 3, 86%). ERBB2 alterations were most frequently led to MGT; trastuzumab emtansine (n = 75, 20.8%; ERBB2 AMP n=36; ERBB2 Mut n=20; ERBB2 AMP NA n=2), or trastuzumab plus pertuzumab (n = 17, 7.5%), followed by EGFR gain to bevacizumab plus erlotinib (n = 18, 5%) and high TMB (≥20 mut/Mb) to atezolizumab (n= 46, 12.8%). Response evaluation of Tier 1 treatment was available in 158 patients, with a 16-week clinical benefit rate of 34.8%. Conclusions: The KOSMOS-II study demonstrated the feasibility of a pragmatic, nationwide precision medicine approach using diverse real-world NGS panels and favorable clinical outcomes.