Circulating Tumour DNA as a Complementary Tool for Treatment Evaluation in HPV-Associated Head and Neck Squamous Cell Carcinoma: An Observational Cohort Study

Abstract

Objectives: HPV-positive oropharyngeal squamous cell carcinoma (OPSCC) and head and neck carcinoma of unknown primary (HNCUP) are increasing. Despite good prognosis, recurrence rates range from 10% to 25%. Surveillance with clinical controls and imaging is not always reliable. Circulating tumour human papillomavirus DNA (ctHPV-DNA) has emerged as a potential biomarker for treatment evaluation and detection of recurrence. We aimed to investigate the correlation between ctHPV-DNA in HPV+ OPSCC/HNCUP and radiologic tumour burden. Additionally, we sought to assess whether ctHPV-DNA could serve as a tool in treatment evaluation.

Design: A prospective observational cohort study.

Setting: This multicenter study involved three otolaryngology units located in central Sweden. We utilised HPV genotype-specific assays for droplet digital PCR (ddPCR) to detect ctHPV-DNA in plasma at diagnosis and follow-up. ctHPV-DNA levels were correlated to radiological tumour burden and radiological response using the Kendall Rank correlation coefficient and the Kruskal-Wallis test.

Participants: Patients with HPV+ OPSCC/HNCUP undergoing definitive (chemo)radiotherapy and enrolled in the CIRCOS study.

Results: Out of 54 patients, 51 were eligible for analyses. At baseline, ctHPV-DNA was detectable in 88%. A majority of patients with a favourable radiological evaluation according to RECIST had a corresponding undetectable ctHPV-DNA at follow-up. The levels of ctHPV-DNA at baseline correlated with total tumour volume and nodal volume (rτ = 0.39, p < 0.01, respectively rτ = 0.26, p < 0.01).

Conclusion: ctHPV-DNA shows correlation with tumour burden. This study strengthens the role of ctHPV-DNA as a promising biomarker for treatment evaluation in HPV-related OPC/HNCUP. With further research on serial plasma sampling, ctHPV-DNA could complement radiological treatment evaluation in HPV+ OPSCC/HNCUP.

Trial registration: NCT05904327 [ClinicalTrials.gov].

Keywords: RECIST; biomarker; cancer of unknown primary; ctHPV‐DNA; head and neck squamous cell carcinoma; human papilloma virus; oropharyngeal cancer.

FIGURE 1

Presentation of available data at the different time‐points. The cohort consists of the patients enrolled between December 2020 and June 2022.

FIGURE 2

Diagram showing the correlation between ctHPV‐DNA (copies/mL) at baseline and radiological volumetric measurements on images from diagnostic CT. (A) The correlation between ctHPV‐DNA and diameter of target lesions (mm). (B) The correlation between ctHPV‐DNA and total tumour volume (cm³). Kendall rank correlation coefficient was applied for calculation of correlations. The short axis of the lymph nodes was used for diameter estimations in nodal lesions (mm), while the longest diameter (mm) was applied for the primary tumour lesions (n = 51).

FIGURE 3

Treatment evaluation. (A) Comparison of baseline values of ctHPV‐DNA depending on RECIST evaluation. (B) Comparison between molecular response measured in percentage change in ctHPV‐DNA from baseline to the follow‐up sample taken at close proximity to radiological evaluation, and RECIST results. The Y‐axis has been modified to enhance interpretability. Kruskal–Wallis. n = 47. CR = complete response, PD = progressive response, PR = partial response.