This study evaluated a methylation‑based circulating tumor DNA assay (Guardant360) that detects tumor‑specific DNA methylation patterns by next‑generation sequencing of plasma cell‑free DNA. The laboratory‑defined threshold for a positive result in this analysis was a methylation signal of 0.05% or greater. Testing was performed during routine post‑treatment surveillance.
The analysis is a retrospective, single‑institution cohort of patients with non‑metastatic, non‑virally associated head and neck squamous cell carcinoma (HNSCC) who completed definitive therapy and had at least one Guardant360 ctDNA test between January 2024 and October 2025. Fifty patients met inclusion criteria.
Primary tumor sites were predominantly the larynx (38%) and oral cavity (36%). Most patients had locally advanced disease: 72% were AJCC stage IVA–IVB, 69% had T3–T4 primary tumors, and 48% had N2–N3 nodal classification. Demographic details, comorbidity profiles, and HPV/EBV testing specifics beyond the non‑viral designation were not reported in the abstract.
Definitive treatment approaches in the cohort were chemoradiation for 54% of patients and surgery for 46%. Median follow‑up reported was 18.0 months. During this follow‑up interval, 13 patients experienced disease recurrence.
Details on timing and frequency of ctDNA sampling relative to treatment completion, imaging schedules, or standardized clinical follow‑up protocols were not provided in the abstract.
Among the 50 patients, 12 had positive methylation‑based ctDNA results during surveillance. Using recurrence as the reference outcome, the assay’s reported diagnostic performance metrics were:
These values indicate that, in this cohort, a positive methylation‑based ctDNA signal identified the majority of patients who recurred, while a negative result was associated with a high probability of no recurrence over the observed follow‑up. The source did not report additional subgroup analyses (for example by primary site, stage, or treatment modality) within the abstract.
The abstract reports that a positive ctDNA result was associated with worse progression‑free survival (PFS). Reported 1‑year PFS estimates were 25.0% for ctDNA‑positive patients versus 96.4% for ctDNA‑negative patients. The text in the provided source truncates at the p‑value and does not present the full statistical result or complete overall survival (OS) data; therefore the exact p‑value and detailed OS comparisons were not available in the source material.
Kaplan‑Meier methods were used to estimate PFS and OS according to the abstract, but the full survival curves, hazard ratios, and multivariable adjustment (if any) were not included in the abstract text supplied.
Key limitations apparent from the source abstract:
Implications:
In this preliminary cohort, Guardant360’s methylation‑based ctDNA assay demonstrated promising diagnostic performance for detecting recurrence in non‑viral HNSCC, with high specificity and NPV and an AUC of 0.858. A positive ctDNA result correlated with substantially poorer 1‑year PFS in the available data. These findings support further prospective validation and more complete reporting of longitudinal outcomes and testing protocols before clinical implementation.
The study provides early evidence that methylation‑based ctDNA testing using Guardant360 may be a useful adjunct for post‑treatment surveillance in non‑viral HNSCC, showing good discrimination for recurrence in this retrospective cohort. However, the preprint status, limited sample size, missing methodological details, and incomplete survival reporting in the abstract warrant cautious interpretation and further validation.