Background. The impact of treatment for head and neck cancer (HNC) on speech and cognitive functioning is well studied. Moreover, recent research has identified speech and cognitive deficits in HNC patients already prior to treatment, but this baseline cognitive impairment is poorly understood. Importantly, chronic alcohol intake and tobacco use, well-known risk factors for HNC, may also have detrimental effects on cognitive function.
Methods. We analysed objective cognitive scores and patient-reported speech and cognitive complaints of 580 HNC patients from the prospective NET-QUBIC cohort with respect to tumour location, and self-reported alcohol and tobacco use. Objective measures included verbal fluency, delayed recall of a verbal learning task, and psychomotor speed and executive functioning from the Trail Making Test. Patient-reported outcomes comprised the Speech Handicap Index and Cognitive Failures questionnaires.
Results. Moderate to severe cognitive impairment was observed in about 8% of the patients for verbal fluency and delayed recall. Regarding the patient-reported outcomes, the larynx group had worse speech functioning, whereas the rate of cognitive complaints did not differ across groups. A correlation between self-reported speech and cognitive functioning was found. This correlation was strongest in the group with larynx tumours, with a moderate effect size. The self-reported and objective cognitive measures did not correlate with each other. Current alcohol intake and smoking status did not predict any of the self-reported or objective cognitive and speech measures, but tumour site did. The larynx group had worse verbal fluency and self-reported speech scores than the other groups, also when controlled for smoking status and current alcohol intake.
Conclusions. We found a relatively high prevalence of baseline cognitive impairment in HNC. Patients with larynx tumours had the worst verbal fluency performance and the highest degree of speech impairment. Self-reported alcohol and tobacco use remain poor predictors of baseline cognitive impairment.