2767 - A Novel Stratification Score for Oral Cavity Carcinoma Prediction of Locoregional Failure and Overall Survival
Presenter(s)
Z. A. Husain1, C. Khalil2, P. Ataalla2, M. Khoury2, I. Poon3, A. Bayley4, I. Karam5, D. Enepekides6, K. Higgins6, M. Smoragiewicz7,8, K. Chan7, R. Fu6, and A. Eskander6; 1Southern California Permanente Medical Group, Los Angeles, CA, 2University of Toronto School of Medicine, Toronto, ON, Canada, 3Department of Radiation Oncology, Odette Cancer Centre, Sunnybrook Health Sciences Centre, University of Toronto, Toronto, ON, Canada, 4University of Toronto - Sunnybrook Odette Cancer Centre, Toronto, ON, Canada, 5Odette Cancer Centre, Sunnybrook Health Sciences Centre, Toronto, ON, Canada, 6Sunnybrook Health Sciences Centre, Toronto, ON, Canada, 7Division of Medical Oncology, Odette Cancer Centre & University of Toronto, Toronto, ON, Canada, 8Kingston Health Sciences Centre, Kingston, ON, Canada
Purpose/Objective(s): The oral cavity carcinoma predictive score OCCPS (Id Said et al. JNCI 2023) was developed using T stage, N stage, grade, and LVSI to divide oral cavity carcinoma (OCC) patients into standard, intermediate, or high risk of distant failure. We applied this system to a cohort of resected OCC to see if it would also stratify based on the risk of locoregional failure (LRF), disease free survival (DFS) and overall survival (OS).
Materials/Methods: OCC patient charts were reviewed from two tertiary academic centers. Patient and pathologic data were extracted, and the OCCPS was used to stratify patients. Cumulative incidence of LRF and distant metastases (DM) was computed using death as a competing event and compared by risk group using a Gray’s test. DFS and OS were estimated using the Kaplan-Meier (KM) Method and compared by risk group using a log-rank test. The analysis was also repeated in a subset of patients with negative margins and no extranodal extension (ENE).
Results: 665 patients were included, with a mean follow up of 42 months. The mean age was 63 and 62.0% of patients were male. The OCCPS stratified patients as follows: 333 (50.1%) patients fell into the standard-risk group (SR), 200 (30.1%) in the intermediate-risk group (IR), and 132 (19.8%) in the high-risk group (HR). The OCCPS successfully stratified patients based on LRF, DM, DFS, and OS. Details are listed in the table below.
In the subset of patients with negative margins and no ENE (n=410, (61.7%), 269 (65.6%) were SR, 119 (29.0%), were IR, and 22 (5.4%) were HR. Consistent with our primary analysis, we found the cumulative incidence of LRF (P=0.02) and DM (P < 0.01) differed significantly by risk group. Similarly, the calculated risk-group category significantly separated the KM survival function for OS (P=0.02) and DFS (P=0.01).Conclusion: The OCCPS successfully stratifies according to risk of LRF, DM, in addition to DFS and OS. Given the results of RTOG 0920, which suggested a possible OS benefit for selected high risk patients, the OCCPS can be used to identify those at the highest risk of failure, who may benefit from intensification using a concurrent radiosensitizer.
Abstract 2767 - Table 1| Overall cohort n = 665 | Negative margin/no ENE cohort n = 410 | |||||||
| Standard | Intermediate | High | P value | Standard | Intermediate | High | P value | |
| LRF | 26.8% | 34.5 % | 45.8% | <0.01 | 27.0 % | 34.0% | 49.0% | 0.02 |
| DM | 7.1% | 15.9% | 44.9% | <0.01 | 7.1% | 14.1% | 26.6% | <0.01 |
| DFS | 71.2% | 54.3% | 21.1% | <0.01 | 72.0% | 56.5% | 42.0% | 0.01 |
| OS | 90.5% | 75.9% | 38.2% | <0.01 | 92.9% | 80.9% | 61.4% | 0.02 |