Background Cancer-related fatigue (CRF) is a common side-effect of cancer and its treatments. It is defined as a subjective sense of physical, emotional, and/or cognitive tiredness or exhaustion related to cancer that is distressing, persistent, and not proportional to recent activity. It is generally not relieved by sleep or rest and can have a detrimental impact on functioning, social activities and quality-of-life. Using data from the Head and Neck 5000 cohort, we investigated clinically-important CRF over the year post-diagnosis. Specifically, we: (1) assessed temporal trends; (2) compared CRF across HNC sites and by treatment received; and (3) identified subgroups with higher odds of CRF.

Methods Recruitment of incident HNC patients was undertaken 2011–2014 from 76 UK hospitals. Socio-demographic and clinical data, and patient-reported CRF, measured using the EORTC QLQ-C30 fatigue subscale, was collected at baseline (after diagnosis but pre-treatment) and 4-, and 12- months post-baseline. Clinically important CRF was defined as scoring ≥39 of a possible 100 on the fatigue subscale. Mixed-effects logistic multivariable regression was used to investigate time trends, compare cancer sites and treatment groups, and identify associations between clinical, socio-demographic and lifestyle variables and CRF. Analyses were done in STATA 16. P<0.05 (two-sided) was considered statistically significant.

Results 2,847 patients were included in the analysis. At baseline, 27.8% scored in the range for clinically important CRF. This rose to 44.7% at 4 months (when many patients were still undergoing treatment) and declined to 29.6% at 12 months. When stratified by treatment received, at 4-months more than half of those who had multi-modal treatment had CRF (surgery plus chemoradiotherapy, 57.7%; chemotherapy+radiotherapy, 56.1%); prevalence was much lower among those who had a single treatment (surgery, 31.4%; radiotherapy 37.5%). In the multivariable model, after adjusting for time-point, treatment modality and other variables, those who had depression at baseline had almost 15-fold increased odds of CRF over 12 months (mvOR=14.7, 95%CI 8.81–24.6). Odds of CRF over 12 months were also statistically significantly increased in females and current smokers and those with more advanced cancer and comorbid conditions at diagnosis.

Conclusion Almost one-third of HNC patients report clinically-important CRF at 12-months post-diagnosis. This high prevalence indicates the need for additional focus during cancer follow-up on this debilitating symptom and for interventions and supports to be made available to those affected. These findings identify characteristics of at-risk groups towards whom such interventions could be targeted.