Cause-specific mortality in patients with head and neck cancer:long-term follow-up of a population-based cohort from 1986 to 2012 accounting for competing risks

Abstract Objectives: Recent recommendations for treating head and neck cancer (HNC) patients favor an individualized approach. Expected long-term survival — together with short-term survival — after diagnosis is the primary focus in assessing the treatment modality and follow-up scheme. “Disease-specific” survival up to five years is often used for measuring the prognosis and for assessing treatment methods. However, especially long-term survival is strongly affected by competing causes of death among HNC patients. Materials and Methods: The long-term prognosis of patients with HNC in terms of mortality from both cancer and competing causes was analyzed according to recent methodological guidelines by examining cumulative incidence functions and models for cause-specific hazards and sub-distribution hazards in a population based cohort of 220 patients treated in a tertiary care center in Northern Finland. Results: In addition to well-known tumor-related factors, mortality from HNC was associated with older age. The mortality from other causes of death was strongly dependent on age and Charlson’s Comorbidity Index, but less on gender. When demonstrating the importance of individualized approach in simulated patients, the mortality was highly variable across patients with similar cancer status, but with different comorbidities or age. Conclusion: The overall survival pattern of HNC patients depends not only on their cancer characteristics, but also varies greatly according to their age and comorbidities. Our findings support the need for individualized treatment and follow-up protocols, and active management of comorbid diseases. Appropriate methods for analyzing competing risks should be used when presenting survival estimates of cancer patients