Data on preoperative chemotherapy in resectable oral cavity cancer are conflicting. We present the long-term results of a randomized trial of induction chemotherapy in resectable oral cavity cancer. A randomized, parallel, multicentre trial evaluated the impact of three cycles of cisplatin 100 mg/m² and fluorouracil 1000 mg/m² (120-h infusion administered every 21 days) in stage T2–T4, N0–N2, previously untreated patients with advanced disease. Control group received upfront surgery. Postoperative radiation was offered to both arms when pathologic risk features were identified. The co-primary end points were the occurrence of locoregional or distant tumour relapse, and death. Among the 198 enrolled patients, with a median follow-up of 11.5 years, there was no difference in the incidence of locoregional relapse between chemotherapy and control group (P = 0.6337), nor in distant metastasis development (P = 0.1527). There was also no difference between groups in overall survival (P = 0.3402). Patients with a pathological complete response (pCR) had higher probability of survival than those without (10-year OS: 76.2% versus 41.3%, P = 0.0004). Late toxicities in patients with a minimum follow-up of 60 months (42 in each group) were similar between arms, except from fibrosis (cumulative incidence 40% versus 22% in chemotherapy arm) and grade 2 dysphagia (14% versus 5%). Long-term follow-up of this randomized trial confirmed the absence of survival benefit with preoperative chemotherapy in oral cavity cancer. Late toxicity was similar in the two arms except for fibrosis and dysphagia, which were less in the chemotherapy arm. The survival benefit for patients achieving a pCR was maintained.

Tumor human papillomavirus (HPV) status strongly affects overall survival (OS) of oropharyngeal cancer (OPC) patients. Recently, three groups with different outcomes were identified based on HPV status, smoking history and tumor stage. Our objective was to validate this model using a single-institutional retrospective database. Patients (n = 120) diagnosed with OPC at our institution, treated with concomitant cisplatin plus radiotherapy (RT) (n = 64), induction chemotherapy followed by concomitant chemoradiation (n = 39) or RT alone (n = 17), were stratified in three groups with respect to the risk of death (low 26, intermediate 46 and high 49 patients) according to tumor p16 expression as surrogate of HPV status, pack-years of tobacco smoking and nodal/tumor stage. Group-stratified Kaplan–Meier OS curves were estimated and compared using the log-rank test. The 2-year OS estimates were 100%, 86% and 70%, respectively. The difference between the survival curves was statistically significant (P = 0.009). The Harrell's concordance index was 0.70. The calibration plot showed a good concordance between our results and those observed in the original study. This study validates the risk grouping previously identified. Risk-driven clinical decision making and trial designs will help in better defining the most appropriate treatment in OPC patients.