Prospective Study of Plasma Biomarker-Guided Surveillance of HPV-positive Oropharynx Cancer Using TTMV-HPV DNA: The SPHERE Study

Purpose/Objective(s) Previous observational studies suggest that circulating tumor HPV DNA may facilitate early detection of recurrent HPV-positive oropharynx cancer (OPC). This is the first prospective cohort study to investigate whether biomarker-guided surveillance detects recurrent disease sooner than standard of care. Materials/Methods We enrolled adult patients evaluated for HPV-positive OPC at a single center 11/2020-4/2023 and treated with curative intent. Pretreatment plasma and/or tumor tissue were tested for tumor tissue-modified viral (TTMV) HPV DNA from subtypes 16/18/31/33/35 using a commercial assay. Patients were excluded if they had undetectable TTMV DNA in both pretreatment plasma and tumor tissue. Post-treatment plasma TTMV DNA (cTTMV) was assessed every 3 months. Positive or indeterminate tests were repeated in 4 weeks. Two serial positive/indeterminate results prompted imaging. Negative imaging was repeated every 8 weeks while cTTMV was positive, up to 6 months. The primary outcome was the proportion of recurrences first detected by cTTMV testing. NCT04965792 Results The study cohort comprised 150 OPC patients, 80% male with median age 64 years (IQR 58-69). cTTMV was positive pretreatment in 125 (86%), indeterminate in 3 (2%), and negative in 18 (12%); 4 patients enrolled after treatment with unknown pretreatment status. Median follow-up was 18 months (range 5-34 months). Patients had a median of 6 post-treatment tests (range 1-12). Fourteen patients (9%) developed recurrent or persistent disease. Among these, 6 patients (43%, 95%CI 17-69%) had positive surveillance cTTMV as the first sign of disease that prompted the imaging that ultimately detected recurrence. Time from first positive cTTMV test to diagnosis of recurrence was median 132 days or 4 months (range, 47-280 days). Another 4 patients (29%, 95%CI 5-52%) had positive cTTMV corroborate persistent disease detected on standard 3 months post-treatment surveillance imaging. The final 4 patients (29%, 95%CI 5-52%) with recurrence did not have positive surveillance cTTMV (1 had a single transient elevation to indeterminate). Pre-treatment cTTMV was positive for 2 of these, negative for 1, and unknown for 1. Among the 136 patients who remained clinically disease-free, cTTMV was positive or indeterminate for 6 (4%). Two have had low levels of cTTMV for >6 months with no evidence of disease on serial imaging and exams. Four others had temporary ‘spikes’: 3 had a single indeterminate test; 1 had low level positive tests for 4 months that returned to negative. Conclusion Nearly one-half (43%) of patients with recurrence had positive cTTMV results that prompted a search for disease. The remaining patients with recurrence either had positive cTTMV results that confirmed suspicious findings on standard imaging (29%), or had negative cTTMV results despite clinical disease (29%). Biomarker-guided surveillance for HPV-positive OPC likely benefits a subset of patients, and further study of its impact on outcomes is warranted.