3664 - Evaluation of a Novel Prognostic System for Overall Survival in Surgically Resected Oral Cavity Carcinoma

Z. A. Husain¹, A. Rybkin², V. Lee², M. R. Young³, A. Eskander⁴, B. Burtness⁵, and H. S. M. Park²; ¹Southern California Permanente Medical Group, Los Angeles, CA, ²Department of Therapeutic Radiology, Yale School of Medicine, New Haven, CT, ³Department of Therapeutic Radiology, Yale University School of Medicine, New Haven, CT, ⁴Sunnybrook Health Sciences Centre, Toronto, ON, Canada, ⁵Yale University School of Medicine and Yale Cancer Center, New Haven, CT

Purpose/Objective(s): Oral cavity carcinomas are aggressive neoplasms associated with a high mortality rate and minimal therapeutic progress. Recently, an oral cavity carcinoma predictive score (OCCPS) model for distant metastases using grade, lymphovascular space invasion (LVSI), T3/4 disease, and N stage was created based on data from high-volume hospitals in Brazil and Canada (Id Said J Natl Cancer Inst 2023). We hypothesized that this model is prognostic for overall survival (OS) in a large United States hospital-based cohort. Materials/

Methods: The National Cancer Database was used to select patients with oral cavity cancers diagnosed in 2010-2020 who underwent surgical resection to the primary site and neck. Points were assigned based on pathologic risk factors per the published model as follows: T3/4 disease: 0.4, LVSI: 0.4, Grade 3: 0.7, N1: 0.8, N2: 1.0, N3: 1.5. A total score of <0.7, 0.7-1.7, and >1.7 differentiated patients into standard-risk (SR), intermediate-risk (IR), and high-risk (HR) groups, respectively. Overall survival was calculated using the Kaplan-Meier method, with differences between groups evaluated using the log-rank test.
Results: With a median follow-up of 38.4 months, 62,035 patients were identified. Mean age was 64, and 60.6% of patients were male. There were 60.9% of patients with T3/4 tumors, 21.4% with LVSI, and 19.6% with grade 3 tumors. Distribution by N stage was as follows: N0: 57.8%, N1: 12.0%, N2: 14.0%, N3: 16.2%. The OCCPS stratified patients as follows: SR: 50.2%, IR: 26.5%, HR: 23.4%. With a median follow up of 38 months, 5-year OS was 58.5%, 47.5%, and 34.3% for the SR, IR, and HR groups, respectively (log-rank p<0.01). Additionally, the model showed discriminative ability for 5-year OS in subsets of patients who received surgery alone, surgery followed by radiation, and surgery followed by chemoradiation (all log-rank p<0.01).
Conclusion: While initially designed for the prediction of distant metastases, the OCCPS also successfully discriminates between patients at standard, intermediate, and high risk for death in a national hospital-based dataset. This suggests a potential broader utility for the OCCPS, for individualized treatment selection as well as clinical trial stratification and design. Abstract 3664 – Table 1

	Current Study	Id Said et al.
pT1-2	39.1%	47%
pT3-4	60.9%	53%
pN0	56.5%	58%
pN1	12.4%	9%
pN2	14.4%	15%
pN3	16.7%	17%
Surgery alone	46.4%	34%
Surgery + RT	30.1%	53%
Surgery + CRT	23.5%	13%