Advancing Education, Research, and Quality of Care for the Head and Neck oncology patient.
Program Number: B313
Session Name: Poster Session
Perineural invasion is associated with decreased T-cell infiltration in salivary gland cancers with facial palsy
Muhammad I Munshi, BA1; Alex E Ladenheim, MD2; Benjamin Schiff, BS1; Zafar Sayed, MD1; Avanti Verma, MD1; Benjamin L Judson, MD, MBA1; Saral Mehra, MD, MBA, FACS1; Ansley Roche, MD, BA1; Sara I Pai, MDPhD1; Suresh Mohan, MD1; 1Division of Otolaryngology, Yale School of Medicine, New Haven, CT; 2Department of Pathology, Yale School of Medicine, New Haven, CT
Introduction: Perineural invasion (PNI) is a key mechanism facilitating tumor progression in head and neck cancers (HNCs) and is associated with a poorer prognosis. The presence of PNI in major salivary gland (MSG) cancers is often associated with facial palsy and a higher risk of local recurrence and distant metastasis. This pilot study aimed to characterize the tumor immune microenvironment in MSG cancers with PNI to elucidate whether direct tumor nerve invasion alone or an associated inflammatory response is responsible for the facial palsy and associated poorer prognosis.
Methods: MSG cancers with PNI were identified from the Yale Head and Neck Biorepository. Tumor infiltrating lymphocyte (TIL) populations in the tumor and perineural zone were quantitated using H&E and immunohistochemistry (CD3, rabbit polyclonal antibody, Biocare Medical, Pacheco, CA) staining. The perineural zone was defined as the area within one high powered field (HPF, 400x) of a tumor-involved nerve. The overall TIL density was characterized on H&E-stained sections as a ratio of the area occupied by TILs to area occupied by the tumor cells and stroma. The absolute number of TILs per high power field were counted in the areas of highest density (‘hotspots’) in both the tumor and in perineural zones. Histologic findings were correlated with clinical signs including facial palsy and survival.
Results: Ten cases with PNI were studied: five primary MSG cancers (adenoid cystic, secretory, and salivary duct carcinomas) and five metastatic squamous cell carcinomas (SCC) to the MSG. Mean patient age was 75 years (range 36-99). Median survival was 1.79 years (range 0.31-8.53). Preoperative facial palsy was present in 60% (6/10) of cases, all of which (100%, 6/6) demonstrated extensive extratumoral PNI. In 90% (9/10) of cases, the CD3+ TIL counts were lower in perineural zones compared to hotspots within the tumor. The mean difference in CD3+ TIL counts was 47.5 cells/HPF (CI: 17.42 to 77.58, p=0.006). Primary MSG cancers as compared to metastatic SCCs had the lowest frequency of TILs in the perineural environment (mean 20/HPF vs 54/HPF, respectively) as compared to the tumor overall (mean 54/HPF vs 120/HPF, respectively).
Conclusion: PNI is associated with a decreased CD3+ TIL population in MSG cancers and metastatic SCCs. Therefore, the associated facial palsy is a result of direct tumor invasion of the nerve rather than a host inflammatory response, as observed in Bell’s palsy. Next steps aim to determine whether PNI induces an immune privileged milieu facilitating tumor growth and invasion. Specifically, immune cell types and cytokines within perineural zones will be characterized using spatial transcriptomics and multispectral immunofluorescence staining.