Age and Mutations as Predictors of the Response to Immunotherapy in Head and Neck Squamous Cell Cancer

Abstract

The immunosuppressive tumor microenvironment plays an essential role in the treatment of head and neck squamous cell carcinoma (HNSC). Compared to traditional chemoradiotherapy, immune checkpoint inhibitors (ICIs) have become increasingly important in HNSC therapy. Prior studies linked the efficacy of ICIs to PD-L1, microsatellite instability (MSI), HPV infection, tumor mutation burden (TMB), and tumor lymphocyte infiltration in patients with HNSC, but further verification is needed. Additional predictors are needed to recognize HNSC patients with a good response to ICIs. We collected the clinical information and mutation data of HNSC patients from Memorial Sloan Kettering Cancer Center (MSKCC) and The Cancer Genome Atlas (TCGA) databases to generate two clinical cohorts. The MSKCC cohort was used to recognize predictors related to the efficacy of ICIs, and the TCGA cohort was used to further examine the immune microenvironment features and signaling pathways that are significantly enriched in the subgroups of predictors. Multivariate Cox regression analysis indicated that age (HR = 0.50,p= 0.014) and ARID1A (HR = 0.13,p= 0.048), PIK3CA (HR = 0.45,p= 0.021), and TP53 (HR = 1.82,p= 0.035) mutations were potential predictors for ICI efficacy in HNSC patients. Age > 65 years and ARID1A or PIK3CA mutations correlated with good overall survival (OS). TP53 mutant-type (MT) patients experienced a worse prognosis than TP53 wild-type (WT) patients. The subgroups associated with a good prognosis (age > 65 years, ARID1A-MT, and PIK3CA-MT) universally had a high TMB and increased expression of immune checkpoint molecules. Although TP53-MT was associated with a high TMB, the expression of most immune checkpoint molecules and immune-related genes was lower in TP53-MT patients than TP53-WT patients, which may reflect low immunogenicity. Pathways related to the immunosuppressive tumor microenvironment were mostly enriched in the subgroups associated with a poor prognosis (age ≤ 65 years, low TMB, ARID1A-WT, PIK3CA-WT, and TP53-MT). In conclusion, the factors age > 65 years, PIK3CA-MT, and ARID1A-MT predicted favorable efficacy for ICI treatment in HNSC patients, and TP53 mutation was a negative predictor.