Comparison of Clinical Outcomes, Pathologic Characteristics, and Immune-Related Features of Postradiation vs Sporadic Oral Cavity Squamous Cell Carcinoma

IMPORTANCE: Postradiation oral cavity squamous cell carcinoma (OCSCC) is a common secondary malignant neoplasm affecting survivors of head and neck cancer who underwent radiotherapy. The clinical, pathologic, and immune-related features of postradiation OCSCC are poorly characterized, and treatment options are limited because of surgical difficulty and high morbidity associated with reirradiation. OBJECTIVE: To determine whether postradiation OCSCC has distinctive clinical, pathologic, and immune-related features compared with demographic-matched sporadic OCSCC. DESIGN, SETTING, AND PARTICIPANTS: This retrospective matched cohort study was conducted at a single tertiary oncology center in Hong Kong. Participants included consecutive patients with OCSCC diagnosed between 2000 and 2020. Patients with postradiation OCSCC were matched with patients with sporadic OCSCC using age, year of diagnosis, sex, and anatomic subsites. Data analysis was performed from July to December 2022. EXPOSURE: Head and neck irradiation involving the oral cavity before the diagnosis of OCSCC. MAIN OUTCOMES AND MEASURES: The primary outcomes were relapse pattern, survival, and causes of death. Pathologic features; immunohistochemical staining for programmed death–ligand 1, PD-1, MSH6, PMS2, FOXP3, and Ki67; and mRNA expression of 31 immune-related genes were also analyzed. RESULTS: A total of 173 patients, 60 with postradiation OCSCC (median [IQR] age, 63.8 [53.0-71.7] years; 43 men [71.7%]) and 113 with sporadic OCSCC (median [IQR] age, 64.4 [52.8-70.6] years; 83 men [73.5%]), were included. Patients with postradiation OCSCC had a higher proportion of N0 disease than those with sporadic OCSCC (50 patients [83.3%] vs 56 patients [49.6%]). With a median (IQR) follow-up of 10.2 (1.2-20.5) years, the 10-year relapse-free survival rates were lower in patients with postradiation OCSCC than sporadic OCSCC (29.6% [95% CI, 17.1%-43.2%] vs 52.4% [95% CI, 41.8%-62.0%]; P = .04), and the same was true for overall survival (30.5% [95% CI, 17.6%-44.4%] vs 52.3% [95% CI, 41.4%-62.1%]; P = .03). All relapses in patients with postradiation OCSCC were locoregional, whereas 35.2% of relapses (12 of 34 patients) in patients with sporadic OCSCC were distant. Despite similar 10-year disease-specific survival rates between the 2 groups (68.8% [95% CI, 55.8%-81.0%] vs 67.1% [95% CI, 57.5%-76.5%]; P = .91), patients with postradiation OCSCC had excess mortality due to pneumonia and cerebrovascular events. Postradiation OCSCC exhibited more adverse pathologic features (perineural invasion, worse pattern of invasion, and tumor budding), higher PD-1 expression, and higher gene expression of CD4 and TGF-β compared with sporadic OCSCC. CONCLUSIONS AND RELEVANCE: This retrospective matched cohort study found distinctive pathologic characteristics and relapse patterns of postradiation OCSCC compared with sporadic OCSCC, which may be attributable to the lack of adjuvant radiotherapy, aggressive biologic phenotype, and different host immune response. Further exploration of the role of immune checkpoint therapy may be justified.