Expression Of Immune Response Biomarkers (Pd-L1, P16, Cd3(+) And Cd8(+) Tils) In Recurrent Head And Neck Squamous Cell Carcinoma Within Previously Irradiated Areas

The immune landscape of head and neck squamous cell carcinoma in pretreated areas remains poorly documented. We aimed to assess the tumor microenvironment for biomarkers of antitumor immune responses in tumors in previously irradiated areas compared with de novo tumors. This retrospective monocentric study analyzed 100 paraffin-embedded surgical samples of invasive head and neck squamous cell carcinoma (oral cavity, oropharynx, larynx, hypopharynx) from patients who underwent surgery between January 2010 and November 2017. We compared the immune microenvironment in 50 de novo tumors and 50 tumors recurring within irradiated areas. We used immunohistochemistry to assess p16 status, CD3(+)/CD8(+) tumor-infiltrating lymphocytes (TILs), and programmed death-ligand 1 (PD-L1) expression on tumor and immune cells in stromal and intratumoral components. CD3(+) TIL counts were significantly lower in intratumoral and stromal components (P=0.003 and P=0.020, respectively) in the irradiated area cohort; there was no significant difference between CD8(+) TIL counts in the two cohorts. The percentage of tumors with PD-L1(+) tumor cells (tumor proportion score >= 1%) was significantly lower within the irradiated area cohort than the de novo cohort (56.0% vs. 86.0%, P<0.001). There were also significantly fewer tumors with PD-L1(+) immune cells in the irradiated area cohort. Predominantly, tumors from the irradiated area cohort had microenvironments classified as 'adaptive immune resistance'. There was persistence of cytotoxic cells in tumors in the irradiated areas but lower PD-L1 expression and CD3(+) TIL counts than in the de novo tumors. This offers an initial hypothesis to explain why these lesions are less responsive to immunotherapy, even though they may still have antitumor capacities. Assessment of immune response biomarkers in patients treated with immunotherapy in randomized trials is required.