Sep 30

QP 08 - Head and Neck 3: Quick Pitch: Prognosis in Focus: Predicting Outcomes in Head and Neck Cancer

1046 - The Role of TP53 in Nodal Metastasis: A Molecular Guide for Neck Management in Oral Cavity Cancer

08:25am - 08:30am PT

Room 160

Presenter(s)

Victor Lee, MD, BS - Yale New Haven Hospital, Middletown, CT

V. Lee¹, A. Rybkin¹, P. Oh¹, S. Izreig², H. S. M. Park³, L. E. Morris⁴, T. J. Hayman¹, M. R. Young³, C. Pickering⁵, S. Mehra⁶, N. Y. Lee⁷, and J. J. Kang⁸; ¹Department of Therapeutic Radiology, Yale School of Medicine, New Haven, CT, ²Department of Surgery, Section of Otolaryngology, New Haven, CT, ³Department of Therapeutic Radiology, Yale University School of Medicine, New Haven, CT, ⁴Yale New Haven Hospital, New Haven, CT, ⁵Yale School of Medicine, New Haven, CT, ⁶Department of Surgery, Section of Otolaryngology, Yale School of Medicine, New Haven, CT, ⁷Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, NY, ⁸Yale University, New Haven, CT

Purpose/Objective(s):

Lymph node metastasis (LNM) is a critical prognostic factor influencing both treatment paradigm and survival in oral cavity carcinoma (OCC). Level 1 evidence supports upfront elective neck dissection (END) for improving survival in patients at risk for LNM. However, current clinical trials are evaluating the feasibility of sentinel lymph node biopsy (SLNB) in unselected patients as a less invasive alternative. Abnormal p53 immunohistochemistry has been recently identified as a potential biomarker for LNM in OCC. Given these findings, there is rising potential for molecular and genomic data to guide neck management decisions such as END, SLNB, or postoperative radiation target volumes. This study aims to refine risk stratification in OCC by characterizing genetic mutations associated with LNM.

Materials/Methods:

The TCGA head and neck cancer dataset was analyzed to extract OCC primary tumor mutational data. 321 genes included in the Foundation One CDx panel were individually assessed for their association with LNM on univariate logistic regression (UVA), where the presence of mutations in each gene was treated as an independent predictor of LNM. Genes significant on UVA were then evaluated in a multivariate logistic regression model (MVA), adjusting for clinical T (cT) stage, age, and gender. The EAp53 system was used to evaluate specific p53 mutation risk scores. Statistical analyses were performed using R and Python 3.7 with scipy.stats, statsmodels, and sklearn libraries.

Results:

The final cohort included 219 patients, 123 (56.2%) of whom were pathologically node-positive. Mean age was 60.5 years. TP53 was the only gene found to be associated with LNM on UVA. TP53 mutations were present in 160 patients (73.1%), with a higher prevalence among node-positive (79.7%, 98/123) versus node-negative (64.6%, 62/96) patients. On UVA, mutations in TP53 were associated with double the risk of LNM: (Odds Ratio [OR]: 2.15, 95% Confidence Interval [CI]: 1.17–3.94, p = 0.013). TP53 mutations remained predictive of LNM on MVA (OR: 2.05, 95% CI: 1.08-3.95, p = 0.029). cT stage did not predict LNM on MVA. On subset analysis of the cN0 cohort, TP53 remained a robust independent predictor of LNM on MVA (OR: 3.26, 95% CI: 1.10-11.75, p = 0.047). TP53 deemed high risk by the EAp53 system remained an independent predictor of LNM on MVA (OR: 2.01, 95% CI: 1.13-3.60, p = 0.018).

Conclusion:

Our study highlights the role of tumor genomic alterations, particularly TP53 mutations, in predicting LNM in OCC. TP53 was an independent predictor of LNM, conferring greater risk than cT stage. These findings provide a foundation for considering the integration of molecular biomarkers into neck management decisions such as those surrounding SLNB, END, or postoperative radiation therapy volumes. Prospective trials are needed to validate these findings and to establish precision medicine-driven treatment algorithms for OCC.