ESTRO 2024 - Abstract Book

S1403

Clinical - Head & neck

ESTRO 2024

has led to a doubling of the survival probability after OPSCC diagnosis for Danish patients. These observations indicate that deintensification of treatment entails the risk of compromising patient outcome, as was also observed in three randomized trials substituting cisplatin with EGFR-inhibition in the treatment of p16-pos OPSCC (1-3). The aim of the present study was to provide real-life data reporting on the efficacy of the use of primary radiotherapy (RT) with radiobiological intensification (hypoxic modification, accelerated fractionation and concurrent chemotherapy) applied according to national DAHANCA guidelines, in a nationwide cohort of consecutive OPSCC patients prospectively collected and treated with curative intent between 2011-2020.

Material/Methods:

The study cohort was identified in the Danish Head and Neck Cancer (DAHANCA) database, which has a 100% coverage of OPSCC in Denmark. For the present study, patients living in Denmark and referred for treatment of biopsy proven stage I-IV OPSCC diagnosed between 2011-2020 were identified. Patients were excluded in case of: distant metastatic disease at presentation, primary surgery and noncurative treatment intent. Outcomes were measured from the first contact with the oncology center to either death, migration (4 patients), end of follow-up (February 17, 2023) or 5 years after diagnosis, whichever came first. Treatment was with primary RT, which in case of node positive disease and/or T3/T4 T-classification was intensified with concurrent weekly cisplatin (40 mg/m2) to fit patients. Most patients received hypoxic modification with nimorazole according to the DAHANCA guidelines. No planned neck dissections were performed. Tumor-HPV-status was determined on pre-treatment tumor tissue according to p16 immunohistochemistry (p16pos>70% tumor cells). Five-year locoregional recurrence rate (LRR) and disease-free (DFS) in the total cohort and stratified by p16-status were evaluated using Aalen-Johansen estimator and Kaplan-Meier estimator, respectively. A Cox proportional-hazards model including sex, age, T- and N- classification, p16, smoking, chemotherapy, hypoxic modification and fractionation schedule was performed to investigate the impact of the various modifications of RT. A total of 3432 patients were identified, 2574 (75%) men and 858 (25%) women, with a median age of 61.7 years. p16-positivity was found in 2288 (67%) of the tumors, whereas 1039 (30%) were p16-negative (105 (3%) unknown p16-status). Sixty-nine% of patients in the p16-positive subgroup were ever smokers (21% current smokers) compared to 98% (67% current smokers) with p16-negative tumors. Median follow-up was 6.7 years. The 5-year risk of loco-regional recurrence was 19.1% (17.8-20.5) in the total cohort, and 12.5% (11.1-13.9) and 33.4% (30.5-36.3) for p16-positive and p16-negative tumors, respectively, Figure 1. The corresponding 5-year DFS probabilities were 65.8% (64.1-67.5) in the total cohort, and 78.2% (76.4-80.0) in p16-positive and 39.7% (36.7-42.9) in p16-negative tumors, respectively, Figure 1. In MVA the prognostic impact of tumour p16-status was confirmed, demonstrating an increased risk of loco-regional recurrence (LRF: 2.46 [2.04-2.96]) and death (DFS: 2.67 [2.32-3.08]) in the p16 negative group of patients, compared to patients with p16-positive disease, Figure 2. In addition, all radiobiological modifications were found to be independently associated with better outcomes: hypoxic modification with nimorazole (LRF: 0.74 [0.59-0.92], DFS: 0.69 [0.58-0.81]), accelerated fractionated RT (LRF: 0.58 [0.40-0.83], DFS: HR: 0.67 [0.50-0.89]) and concurrent chemotherapy (LRF: 0.62 [0.51-0.76], DFS: HR: 0.54 [0.47-0.63]), Figure 2. Results: