Validation of the Brandwein-Gensler Risk Model for Predicting Locoregional Recurrence in Oral Squamous Cell Carcinoma: A Prospective Cohort Study from Western India

Squamous cell carcinoma of the oral cavity (OSCC) is one of the leading malignancies in India and is frequently associated with locoregional recurrence (LRR) despite definitive surgical treatment. Histopathological risk assessment using the Brandwein-Gensler Risk (BGR) model has been proposed as a practical approach for predicting recurrence and supporting postoperative management. The aim of the study is to validate the BGR model for predicting LRR in patients with OSCC undergoing primary surgery at a tertiary care teaching cancer center in Junagadh, Gujarat in western India. A prospective study was conducted among 167 patients with histologically confirmed OSCC who underwent primary surgical treatment between July 2022 and June 2024. Three histopathological variables, namely, lymphocytic host response (LHR), worst pattern of invasion (WPOI), and perineural invasion (PNI) were assessed to calculate the BGR score. Patients were stratified into low-, intermediate-, and high-risk groups, and the predictive performance of the model for LRR was evaluated. Patients' mean age was 48.02 years, and 86.22% were males. Buccal mucosa was the most frequently affected anatomical site (58.68%). A high-risk classification was observed in 50.89% of patients, whereas 26.34% and 22.75% were categorized as low, and intermediate risk, respectively. LRR occurred in 8.38% (14) of patients, of whom 78.57% (11) belonged to the high-risk group. The high-risk category demonstrated a sensitivity of 78.57% for predicting recurrence and showed a statistically significant association (p = 0.0478) with recurrence. The BGR model demonstrated useful predictive ability for identifying patients at increased risk of LRRs following surgery for OSCC. Incorporation of this histopathological risk assessment into routine practice may facilitate risk-based follow-up and individualized postoperative management.