Interim analysis of tumor-informed ctDNA detection in head and neck squamous cell carcinoma: The PECAN trial

Background: Head and neck squamous cell carcinomas (HNSCC) are a heterogeneous group of malignancies of the upper aerodigestive tract, encompassing multiple tumor (sub)sites and entities. Current treatment often relies on radiotherapy, with or without concurrent chemotherapy. Clinical surveillance remains the sole indicator of recurrent disease, despite relapse rates of 25-40%. There is a clear unmet clinical need for biomarkers that can predict treatment efficacy prior to and during (chemo)radiotherapy and that can identify residual disease and cancer recurrence early during surveillance. Methods: Seventy-two HNSCC patients were enrolled in the PECAN trial. Plasma and saliva samples were collected at baseline; during (chemo)radiotherapy; and at 2 weeks, 3 months, 6 months, 1 year, and 2 years after treatment. Tumor-informed whole-genome sequencing (WGS) was performed on circulating cell-free DNA (ccfDNA) extracted from plasma for 57 evaluable patients. Molecular findings were correlated with treatment response and clinical outcome. Analysis of the remaining patients, saliva samples, human papilloma virus (HPV) abundance, and imaging are ongoing. Results: At baseline, circulating tumor (ctDNA) was detectable in 92% HNSCC patients based on tumor-informed molecular fingerprinting. All patients with undetectable baseline ctDNA remained negative throughout the monitoring period, and none experienced disease recurrence. Among patients with detectable ctDNA, 84% cleared ctDNA during (chemo)radiotherapy; however, ctDNA clearance during treatment was not predictive of recurrence. In landmark analysis at 3 months after treatment, ctDNA was detected in 6% of patients, all of whom subsequently developed recurrence. During the entire 2-year surveillance period, no ctDNA was detected in patients who remained relapse-free. In case ctDNA was detected recurrence was always diagnosed, resulting in a positive predictive value of 100%. ctDNA detection at end of follow-up (EoF), whether due to recurrence or completion of surveillance, was associated with reduced progression-free survival (hazard ratio [HR] = 5.9) and overall survival (HR = 7.1). Conclusion: This interim analysis of the PECAN trial demonstrates the clinical relevance of ctDNA monitoring following (chemo)radiotherapy, showing substantially reduced survival in patients with detectable ctDNA during surveillance. These data suggest that ctDNA serves as a robust biomarker for recurrence detection and could guide initiation of follow-up treatment. Sensitivity of ctDNA detection in the heterogeneous disease spectrum of HNSCC may be improved by complementary analysis of saliva samples and alternative biomarkers (e.g., HPV). For this cohort, the added value of incorporating additional time points, saliva-based detection, HPV abundance, and imaging correlations for all evaluable cases will be presented at AACR.