Evaluation of the genomic and immune profiles of patients with lung adenosquamous carcinoma (LUAS) and association of response to treatment.

8567 Background: LUAS is a rare subset of non-small cell lung cancer (NSCLC) comprising up to 4% of cases, characterized by at least 10% of both adenocarcinoma and squamous components. Limited studies exist on LUAS regarding the genomic/immune landscape and response to immunotherapy (IO). Methods: NSCLC samples (n = 35404) underwent NextGen Sequencing of DNA (592 genes or whole exome)/RNA (whole transcriptome) at Caris Life Sciences. PD-L1 expression was assessed using IHC (22c3; positive: TPS >1%). High tumor mutational burden (TMB-H) was defined as ≥10 mut/MB. Cell infiltration in the tumor microenvironment was estimated by quanTIseq. Gene expression profiles were analyzed for transcriptomic signatures (IFN-γ) predictive of IO response. Real-world overall survival (rwOS) was obtained from insurance claims data, calculated from time of biopsy to last contact. Time on treatment (TOT) was calculated from the first pembrolizumab (pembro) treatment to the last. Mann-Whitney U and X 2 /Fisher-Exact tests were applied where appropriate, with p-values adjusted ( p < .05). Results: Among NSCLC samples, 1.0% (n=374) were LUAS, 26.5% (n = 9365) were squamous (SCC) and 72.5% (n = 25665) were adenocarcinoma (AC). LUAS exhibited actionable alterations in 29% (n = 110/374) of cases, with MET exon 14 skipping alteration ( METex14) significantly higher in LUAS compared to SCC or AC (9.02% vs 1.21% vs 2.61%, p < 0.01). BRAF V600E (0.54% vs 0.13%, p = 0.041) and ALK-Fusion (1.13% vs 0.12%, p = 0.0145) were higher in LUAS compared to SCC, while PIK3CA mutation was higher in LUAS compared to AC (10.8% vs 4.1%, p <0.01). Notably, LUAS exhibited higher prevalence of KRAS, targetable EGFRand STK11 mutations compared to SCC but lower compared to AC ( KRAS G12C: 8.63% vs 1.37% vs 13.24%, p <0.01; EGFR: 13.1% vs 1.12% vs 17.53%, p<0.01; STK11: 6.97% vs 1.83% vs 17.22%, p<0.01). LUAS exhibited significantly higher PD-L1+ (65.5% vs 59.8% vs 53.7%, p <0.01), along with elevated expression of immune checkpoint genes and the IFN-γ signature compared to AC and SCC. Moreover, LUAS exhibited significantly higher neutrophils, CD8 T cells, M2 macrophages and Tregs compared to SCC. LUAS showed improved rwOS compared to SCC (HR 0.792, p<0.01), with no difference noted in the rwOS between LUAS and AC (HR = 0.98, p = 0.81). While PD-L1+ was not associated with TOT on pembro in LUAS, high TMB showed improved TOT on pembro in LUAS. Conclusions: This study provides a comprehensive genomic landscape of LUAS, highlighting actionable alterations in ~ 29% of cases. LUAS exhibits distinct molecular and clinical features compared to AC and SCC, with enrichment in actionable METex14, emphasizing the need for NGS testing. Although PD-L1 status may not significantly impact TOT in LUAS, it remains relevant in other NSCLC subtypes on IO and high TMB emerges as a potential biomarker for favorable TOT on pembro in LUAS, warranting further investigation.