Analysis of chemotherapy effect on the second primary malignancy for head and neck cancer patients by a nomogram based on SEER database

Abstract

Second primary malignancy (SPM) ranks the second leading cause of death in patients with head and neck cancer (HNC), while studies exploring the risk factors for SPM are limited. To clarify this, we investigated the relationship between the chemotherapy and SPM using the Surveillance, Epidemiology, and End Results (SEER) database. 11 345 patients initially diagnosed with HNC between 1998 and 2016 were selected from the SEER database. First, these patients were divided into two groups according to chemotherapy or not. With Fine and Gray model, the subdistribution hazard ratio (sHR) of chemotherapy was calculated based on Propensity Score Matching (PSM). Second, the 11 345 cases were randomized into a training set and a validation set. Based on the training set, the different cumulative incidence of SPMs between the patients with and without chemotherapy was estimated respectively in the high- and low-risk group according to the scores derived from a nomogram. Chemotherapy was negatively correlated to the SPMs (sHR: 0.847, 95% CI: 0.733-0.977, P =.023) by conducting competing risk analysis. With chemotherapy, forest plots showed subgroups of squamous cell carcinoma (SCC, sHR: 0.815, 95% CI: 0.7-0.948, P =.008), 50-64 years old (sHR:0.794, 95% CI: 0.655-0.962, P =.019), male (sHR:0.828, 95% CI: 0.703-0.974, P =.023), and well/moderate histological grade (sHR:0.828, 95% CI: 0.688-0.996, P =.045) were negatively correlated to SPMs; the nomogram showed the high-risk population characterized as SCC, elder age, male, and well/moderate histological grade also tended to have lower incidence of SPMs (sHR: 0.805, 95% CI: 0.669-0.969, P =.022). Despite HNC patients with characteristics of SCC, increased age, male, and well/moderate histological grade had higher risk of a SPM, they were also more likely to be benefitted from chemotherapy to avoid it.