Peripheral blood values as predictors of autoimmune status in oral cavity squamous cell carcinoma

• Main Authors: Anjali Pillai, Cristina Valero, Kathleen Navas, Quaid Morris, Snehal G. Patel
• Format: Article
• Language: English
• Published: Elsevier 2021-12-01
• Series: Translational Oncology
• Subjects: Autoimmune disease; Oral squamous cell carcinoma; Inflammatory response; Head and neck cancer; Immune dysregulation
• Online Access: http://www.sciencedirect.com/science/article/pii/S1936523321002126

Background: Recent literature has highlighted the role of the host in prognosis in oral squamous cell carcinoma (OSCC). Autoimmune (AI) disease represents a macroscopic depiction of host status. The goal of this study was to predict an AI “status” and to analyze the utility of this “status” as a prognostic indicator in OSCC. Methods: From a departmental database of OSCC patients (n = 1377), 125 patients with an AI disorder were identified. PBL values were obtained and standardized for analysis. A LASSO regression model was used to determine the best predictors of AI status and an AI score was developed. The score was then analyzed across various survival endpoints. Results: When AI score was divided into a binary variable, patients in the highest quartile had a significantly worse overall survival (OS), local recurrence-free (LRFP) and distant recurrence-free probability (DRFP). Survival curves showed significant differences for OS, DSS, LRFP, and DRFP. Conclusions: AI diseases are immune dysregulations that could play a role in prognosis. Therefore, development of an AI score is necessary to depict host status in a ubiquitous manner. AI score as a binary variable may be more utilitarian in a clinical setting, compared to the continuous score. This novel tool needs validation and integration into more tumor and host characteristics. This investigation showed utility of such a score, similar to PBL data in OSCC prognosis. Future studies should incorporate other relevant variables known to affect outcome and implement a more comprehensive predictive model.