Circulating tumour DNA to augment PET-CT in determining clinical outcome after head and neck cancer treatment.

AIM: PET-CT and MRI are used to assess disease response after head and neck cancer treatment. Equivocal findings can delay the potential for salvage curative treatment or result in over- treatment with further surgery. The aims of this study were to establish if liquid biopsy (LB) of circulating tumour DNA, could be used to aid decision-making after treatment. METHOD: Patients (n = 41) undergoing chemoradiotherapy (CRT) and surgical treatment had serial blood testing; pre-treatment and 10-12 weeks post-treatment, alongside imaging. PET-CT/MRIs were categorised as Cancer, Equivocal or Complete Response and true disease status was determined retrospectively. A bespoke 17-gene panel and probes targeting 5 HPV subtypes were used for next-generation sequencing in the ctDNA assay. RESULTS: PET-CT was equivocal in 13/27 CRT patients; sensitivity and specificity for determining true disease status was 67 % and 42 % respectively. LB performed better at determining true disease status than PET-CT post-treatment in CRT patients; sensitivity and specificity of 83 % and 95 % respectively. A combined LB/PET/MRI test performed better than imaging alone; sensitivity and specificity of 100 % and 90 % respectively. The improved specificity was statistically significant (Fisher's exact test OR=7.1; 95 % CI:1.6-45.8, p = 0.0005), as was the balanced accuracy (Welch's t-test, p