Prognostic Impact of Lymphovascular Invasion on Distant Metastasis in HPV-Negative Head and Neck Cancer.

Abstract

Lymphovascular invasion (LVI) is a recognized poor prognostic factor in head and neck squamous cell carcinoma (HNSCC), particularly for regional recurrence and overall survival. However, its role in predicting distant metastasis, especially in HPV-negative HNSCC, remains unclear. To evaluate the association between LVI and distant metastasis in adult patients with operable HPV-negative HNSCC who underwent primary surgical resection. A retrospective analysis was conducted on 280 patients with HPV-negative HNSCC treated with primary surgery at two academic institutions from 2017 to 2024. Demographic, clinical, pathological, treatment, and outcome data were extracted. Kaplan-Meier analysis and multivariable Cox proportional hazards models were used to assess the association between LVI and distant metastasis-free survival. Of the 280 patients, 94 (33.6%) were LVI-positive. Kaplan-Meier analysis demonstrated worse distant metastasis-free survival among patients with LVI-positive tumors, with separation persisting at 2 years (69.8% vs. 88.3%) and 3 years (63.1% vs. 87.4%). In multivariable Cox regression adjusting for age, sex, T stage, N stage, PNI, ENE, and margin status, the association between LVI and distant metastasis was attenuated and was not statistically significant (HR 1.56, 95% CI 0.88-2.77, p = 0.131). In a sensitivity model excluding N stage because of overlap with ENE, LVI remained associated with distant metastasis (HR 1.76, 95% CI 1.01-3.06, p = 0.047). LVI is associated with higher rates of distant metastasis and worse distant metastasis-free survival in operable HPV-negative HNSCC. However, this association is attenuated after adjustment for ENE and pN stage, suggesting that LVI may function as part of a broader adverse pathological phenotype rather than as an isolated independent predictor. These findings support the continued evaluation of LVI in postoperative risk stratification models.