Background: There is growing interest in the use of circulating plasma tumor HPV DNA (ctDNA) for diagnosis and surveillance of patients with HPV-associated oropharyngeal squamous cell carcinoma (HPVOPSCC). A recent advance in the technology combining the identification of circulating HPV tumor DNA and tumor DNA fragment analysis (tumor tissue modified viral (TTMV)-HPV DNA) has demonstrated promising reliability of these tests. However, utilization of these newer testing techniques have been limited to small cohort studies and clinical trials. We sought to better understand the test’s clinical utility in the management of patients with HPVOPSCC by evaluating the performance of a commercially available ctDNA assay (NavDx®, Naveris Inc, Natick, MA) in low- and high-prevalence conditions at a tertiary care academic center.
Methods: In this retrospective analysis of a contemporary real-world cohort, we evaluated the performance of the TTMV-HPV DNA assay in detecting HPVOPSCC in the pre-treatment and surveillance stages of care between April 2020 and October 2022. For the pre-treatment phase, patients with HPV-associated and HPV-negative oropharyngeal cancers with at least one TTMV-HPV DNA measurement prior to the initiation primary therapy were included in the analysis. For detection of recurrence, all tests conducted 3 months after completion of primary therapy for patients with pathologically confirmed HPVOPSCC were evaluated. Patients were included in the surveillance cohort regardless of pre-treatment TTMV-HPV DNA testing status. Patients with suspected recurrence, but without pathologic confirmation, were excluded from the final analysis. HPV status was defined utilizing positive p16 staining on immunohistochemistry (92% in Diagnosis cohort, 98.1% in Surveillance cohort) or HPV polymerase chain reaction/in situ hybridization (81.5% in Diagnosis, 90% in Surveillance cohort).
Results: In the pre-treatment diagnosis cohort, 114 patients were included. The diagnosis cohort was 88% male, average age of 63 years old, with 92.1% HPVOPSCC versus 7.9% HPV negative OPSCC. The TTMV-HPV DNA sensitivity in diagnosis was 92.3%; its specificity was 100%.
In the surveillance analysis, 561 tests conducted in 269 patients were evaluated. 20 patients had pathologically confirmed recurrences. The TTMV-HPV DNA test demonstrated 75.6% sensitivity and 100% specificity in detecting recurrence during the surveillance period. Positive predictive value was 100% and negative predictive value was 98.1%. In patients with clinically detectable recurrence, the median lead time from positive TTMV-HPV DNA test to pathologic confirmation was 40 days, with a maximum lead time detection of 507 days.
Conclusions: In this study, we evaluate the performance of TTMV-HPV DNA testing for both diagnosis and surveillance of HPV-associated OPSCC in a contemporary real-world cohort. The TTMV-HPV DNA test demonstrated 100% specificity in both clinical settings and therefore excellent positive predictive value. The lead time detection suggests a window of opportunity for earlier treatment in cases of recurrence during surveillance. However, the sensitivity of the test was higher for diagnosis (92.3%) and lower for surveillance (75.6%). A positive result appears to be confirmatory of the presence of disease; however, a negative result may require further investigation depending on the treating physician’s clinical suspicion for disease.
Table 1. 2x2 Tables for Diagnosis(A) and Surveillance(B) cohorts.