Quantitative lymph node burden as a ‘very-high-risk’ factor identifying head and neck cancer patients benefiting from postoperative chemoradiation

Adjuvant chemoradiation (CRT) is standard for head and neck squamous cell carcinoma (HNSCC) patients with positive margins or extranodal extension (ENE) following surgery. However, emerging evidence suggests the number of positive lymph nodes (LNs) is the dominant determinant of survival in non-oropharyngeal HNSCC and thus may better identify those benefiting from treatment intensification. Patients from the National Cancer Database diagnosed with non-oropharyngeal HNSCC (oral cavity, larynx, hypopharynx) between 2004 and 2014 and undergoing surgical resection, neck dissection, and postoperative radiotherapy (RT) were included. Multivariable regression with first-order interaction terms was used to model the interaction between postoperative CRT and continuous number of positive LNs with respect to overall survival. In total, 7144 patients met inclusion criteria. In multivariable analysis, increasing number of positive LNs was associated with both increasing mortality (P < 0.001) and increasing benefit from postoperative CRT versus RT alone (interaction P < 0.001). While there was no benefit from postoperative CRT in patients with 0–2 LN+ [hazard ratio (HR) 0.96, 95% confidence interval (CI) 0.86–1.07, P = 0.47], increased benefit was seen in those with 3–5 LN+ (HR 0.84, 95% CI 0.70–1.00, P = 0.05) and those with ≥6 LN+ (HR 0.65, 95% CI 0.51–0.82, P < 0.001) in multivariable models. By contrast, margin status and ENE did not reliably identify patients benefitting from postoperative CRT based on statistical tests of interaction. Even in patients with ENE, positive margins, or both, only those with ≥6 LN+ had improved survival with postoperative CRT. Increasing metastatic nodal burden was associated with increased benefit from CRT compared with RT alone, surpassing conventional high-risk factors in identifying patients benefiting from CRT. Stratification by metastatic LN number may characterize a very-high-risk patient cohort best suited for treatment intensification.