Multi-genomic analysis of compound and uncommon EGFR mutations in patients with non-small cell lung cancer.

8534 Background: There are similar and different properties between the epidermal growth factor receptor mutations (mEGFR), which are classified into common, uncommon, and compound mutation subtypes depending on their location and pattern on the EGFR gene. We investigated the RWD and RNA expression profile of uncommon and compound mutations in the mEGFR+ NSCLC patients. Methods: We collected the medical information for stage I-IIIA mEGFR+ NSCLC patients receiving curative surgical resection in Seoul St. Mary’s Hospital. We explored the RNA expression profile using multiplex IF and High-Plex Digital Spatial Profiling. A total of 18,676 DEG libraries were constructed, and 151 significance genes with > 2-fold changes, 4-log 2 normalized read counts, and p-value = 0.05 (up and down) were identified. Results: In hospital cohort, 941 mEGFR+ cases of total 6,609 cases consist of common, 860 (91.4%), uncommon, 60 (7.4%), and compound, 21 (2.2%). Median relapse-free survival in uncommon and compound mEGFRs (31.4M and 33.7M) was shorter than common mEGFRs (40.1M, p = 0.877), but median overall survival did not differ between the groups. CNS recurrence rates in the uncommon and compound mEGFRs (40.0% and 35.3%) tended to be higher than that of common mEGFRs (27.4%, p = 0.361). Spatial RNA-seq data of the ratio of deconvolution individual cells showed that genes involved in inflammatory and immune response were relatively lower in mEGFR+ groups. According to the difference in DEG library through clustering heat map, uncommon or compound mEGFR in the immune, inflammatory response category, extracellular matrix, RNA splicing gene category, JAK-STAT, p53, PI3K-AKT, and notch signaling pathway related genes were down regulated compared to common mEGFR+ group. Gene ontology string analysis revealed RNA expression of genes in compound mEGFR was functionally associated with chemokine production (23.4%), electron transfer activity (14.9%) and T cell selection (8.5%), and, in uncommon mEGFRs, was associated with regulation of oligopeptide transport (26.1%), hypermethylation of CpG island (21.7%), and CD4+ CD25+ αβ regulatory T cell lineage commitment (13.0%), compared to common mEGFR+ groups. Based on the GSEA, relatively highly clustered genes in compound mEGFR were the protein modification process, G-protein coupled receptor, regulation of cytosolic calcium ion concentration, and IL-5 production pathways, and lower ES levels in oxidative phosphorylation and RNA processing pathways, when compared to common mEGFR+ groups. Conclusions: In conclusion, our results suggest that uncommon and compound mEGFR+ subtypes exhibiting distinguished RNA expression profile of functional signaling pathways and protein network be not same disease entity as common mEGFR+ subtype.