Elderly patients at higher risk of laryngeal carcinoma recurrence could be identified by a panel of two biomarkers (nm23-H1 and CD105) and pN+ status.

Laryngeal squamous cell carcinoma (LSCC) recurrences are very difficult to manage in elderly patients (age ≥65 years), because treatment carries significant morbidity and mortality. The aim of this study was to develop a panel of parameters (clinicopathological variables or biomarkers) to improve our ability to detect elderly patients at higher risk of LSCC recurrence. Maspin, nm23-H1, and CD105 were investigated using immunohistochemistry on surgical specimens from 46 elderly patients treated for LSCC. After univariate analysis identified parameters associated with LSCC recurrence, a multivariate prognostic model was constructed. At univariate analysis, a higher recurrence rate was significantly associated with nm23-H1 nuclear expression in carcinoma cells ≤2.0% (p = 0.01), CD105 expression in intratumoral vascular endothelial cells ≥5.28% (p = 0.04), and pN+ status (p = 0.04). Multivariate modeling confirmed that nuclear nm23-H1 ≤2.0% (p = 0.009) and CD105 ≥5.28% (p = 0.013) had a negative prognostic significance in terms of disease recurrence, while pN+ status showed a trend toward significance (p = 0.05). We thus obtained a panel comprising two biomarkers and neck lymph node status that revealed an excellent discriminatory power [AUC (ROC) of 0.81] in terms of the risk of LSCC recurrence. The panel achieved a specificity of 96% and a positive predictive value of 93%. We identified a panel with an excellent discriminatory power in identifying elderly patients at higher risk of recurrence after treatment for LSCC. These patients would benefit from a more aggressive primary treatment.