Symptom-based remote assessment in post-treatment head and neck cancer surveillance: A prospective national study

Objectives To report the incidence of locoregional recurrence in head and neck cancer (HNC) patients under surveillance following treatment undergoing symptom-based remote assessment. Design A 16-week multicentre prospective cohort study. Setting UK ENT departments. Participants HNC patients under surveillance following treatment undergoing symptom-based telephone assessment. Main outcome measures Incidence of locoregional recurrent HNC after minimum 6-month follow-up. Results Data for 1078 cases were submitted by 16 centres, with follow-up data completed in 98.9% (n = 1066). Following telephone consultation, 83.7% of referrals had their face-to-face appointments deferred (n = 897/1072). New symptoms were reported by 11.6% (n = 124/1072) at telephone assessment; 72.6% (n = 90/124) of this group were called for urgent assessments, of whom 48.9% (n = 44/90) came directly for imaging without preceding clinical review. The sensitivity and specificity for new symptoms as an indicator of cancer recurrence were 35.3% and 89.4%, respectively, with a negative predictive value of 99.7% (p = .002). Locoregional cancer identification rates after a minimum of 6 months of further monitoring, when correlated with time since treatment, were 6.0% (n = 14/233) <1 year; 2.1% (n = 16/747) between 1 and 5 years; and 4.3% (n = 4/92) for those >5 years since treatment. Conclusions Telephone assessment, using patient-reported symptoms, to identify recurrent locoregional HNC was widely adopted during the initial peak of the COVID-19 pandemic in the United Kingdom. The majority of patients had no face-to-face reviews or investigations. New symptoms were significantly associated with the identification of locoregional recurrent cancers with a high specificity, but a low sensitivity may limit symptom assessment being used as the sole surveillance method.