Transcriptomic FHIT^low/pHER2^high signature as a predictive factor of outcome and immunotherapy response in non-small cell lung cancer

Introduction: In recent decades, the development of immunotherapy and targeted therapies has considerably improved the outcome of non-small cell lung cancer (NSCLC) patients. Despite these impressive clinical benefits, new biomarkers are needed for an accurate stratification of NSCLC patients and a more personalized management. We recently showed that the tumor suppressor fragile histidine triad (FHIT), frequently lost in NSCLC, controls HER2 receptor activity in lung tumor cells and that tumor cells from NSCLC patients harboring a FHIT^low/pHER2^high phenotype are sensitive to anti-HER2 drugs. Here, we sought to identify the transcriptomic signature of this phenotype and evaluate its clinical significance.

Materials and methods: We performed RNA sequencing analysis on tumor cells isolated from NSCLC (n=12) according to FHIT/pHER2 status and a functional analysis of differentially regulated genes. We also investigated the FHIT^low/pHER2^high signature in The Cancer Genome Atlas (TCGA) lung adenocarcinoma (LUAD) (n=489) and lung squamous cell carcinoma (LUSC) (n=493) cohorts and used the tumor immune dysfunction and exclusion (TIDE) model to test the ability of this signature to predict response to immune checkpoint inhibitors (ICI).

Results: We showed that up-regulated genes in FHIT^low/pHER2^high tumors were associated with cell proliferation, metabolism and metastasis, whereas down-regulated genes were related to immune response. The FHIT^low/pHER2^high signature was associated with the higher size of tumors, lymph node involvement, and late TNM stages in LUAD and LUSC cohorts. It was identified as an independent predictor of overall survival (OS) in LUAD cohort. FHIT^low/pHER2^high tumors were also predictive of poor response to ICI in both LUAD and LUSC cohorts.

Conclusion: These data suggest that ICI might not be a relevant option for NSCLC patients with FHIT^low/pHER2^high tumors and that anti-HER2 targeted therapy could be a good therapeutic alternative for this molecular subclass with poorer prognosis.