Advancing Education, Research, and Quality of Care for the Head and Neck oncology patient.
Objective/Background: Circulating tumor (ct) DNA is a biomarker of disease status in patients with Human Papillomavirus (HPV) + oropharyngeal squamous cell carcinomas (OPSCC). Observational studies suggest that HPV ctDNA clearance after chemoradiation and surgery may predict disease control and elevated levels during surveillance may facilitate early detection of recurrence. Existing literature in surgical populations is limited by low sample sizes and variability in test performance characteristics. The objective of this study was to assess clinicopathologic variables associated with pre- and postoperative HPV ctDNA using a commercially available assay.
Methods: A retrospective review of patients with HPV + OPSCC treated with upfront surgery at a tertiary academic medical center between September 2021 and July 2024 was performed. Patients had pre- and postoperative ctDNA measured by tumor tissue modified (TTMV) HPV DNA (NavDx). Covariates included age, sex, race, TNM stage (AJCC 8th edition), pathologic tumor size, nodal size, margin status, lymphovascular invasion, perineural invasion, extranodal extension, primary disease subsite (tonsil versus base of tongue), smoking status, adjuvant treatment, and recurrence. Multivariate linear and logistic regression determined factors associated with pre- and postoperative ctDNA. ROC analysis with Youden Index determined the preoperative ctDNA level which predicts detectable postoperative ctDNA.
Results: This study identified 80 patients (mean age: 58) with HPV+ OPSCC treated with upfront surgery. Mean preoperative HPV ctDNA level was 2820 copies/mL. On multivariate linear regression including covariates with significant univariate comparisons, factors associated with higher preoperative HPV ctDNA included larger involved lymph node size (β 1225 (95% CI 40-2412), p=0.043) and active smoking status (β 3501 (95% CI 12-6991), p=0.049). There was no association between the magnitude of preoperative ctDNA level with other pathologic features, receipt of adjuvant treatment, or recurrence.
Postoperative HPV ctDNA levels were obtained prior to initiation of adjuvant treatment (average 13 (SD 10) days after surgery) and were available in 50 patients. Of these patients, 21.1% (11/50) had detectable postoperative HPV ctDNA. On multivariate binary logistic regression, the only factor associated with detectable postoperative ctDNA level was higher preoperative ctDNA level (OR 5.9 (95% CI 1.1-33.5, p=0.043). Detectable postoperative ctDNA was not associated with pathologic features, receipt of adjuvant treatment, or recurrence. On ROC analysis, preoperative HPV ctDNA level predicted detectable postoperative ctDNA (AUC = 0.71 (95% CI 0.53-0.88), p=0.022), with an optimal cutoff point of 214 counts/mL. In patients with preoperative HPV ctDNA < 214 copies/mL, 92.6% (25/27) had undetectable and 7.4% (2/27) had detectable postoperative ctDNA. When preoperative HPV ctDNA ≥ 214 copies/mL, 60.9% (14/23) had undetectable and 39.1% (9/23) had detectable postoperative ctDNA.
Conclusion: In surgically treated patients with HPV mediated OPSCC, higher preoperative HPV ctDNA levels are associated with larger nodal disease burden and active smoking. Detectable postoperative ctDNA is more likely to be observed in patients with preoperative ctDNA levels ≥ 214 copies/mL but does not seem to correlate strongly with adverse pathologic features in this cohort. The sensitivity of this assay to detect small volumes of disease after surgery may be limited in patients with lower starting HPV ctDNA.