Risk-adapted therapy guided by human papillomavirus (HPV) circulating tumor DNA in patients with HPV-positive oropharyngeal cancer (ReACT 1.0).

Abstract

6009 Background: Human papillomavirus-positive oropharyngeal cancer (HPV+ OPC) has favorable outcomes with platinum-based concurrent chemoradiation (CRT), but long-term toxicity can be significant. Various treatment de-intensification strategies have been explored to maintain survival and mitigate treatment-related morbidity. We present the first study using tumor tissue modified viral (TTMV)-HPV DNA in real-time to stratify non-surgical patients (pts) to receive de-intensified, curative-intent CRT. Methods: This phase 2 two-cohort, clinical trial (NCT04900623) enrolled pts with AJCC 2017 8 th ed. stage I-III (no fixed nodes) HPV+ OPC treated with CRT if they had detectable TTMV-HPV DNA (type 16) pre-treatment (pre-tx). Pts were assigned to the low-risk (LR) arm with T0-3 N0-2 disease, a ≤10 pack-year smoking history, and any detectable HPV DNA result pre-tx. Pts were assigned to the intermediate-risk (IR) arm with T4 disease or a >10 pack-year smoking history if they had a pre-tx HPV DNA score of >200. LR arm pts received de-intensified CRT (54-66 Gy with reduced dose platinum or RT alone). IR arm pts who cleared their pre-tx HPV DNA by >95% were also de-intensified; those ≤200 pre-tx or who failed to clear received standard 70 Gy CRT with up to 300 mg/m 2 cisplatin. Primary endpoint was 2-year progression-free survival (PFS) among all de-intensified LR and IR pts. Group-sequential testing for non-inferiority (NI) was performed Q4 months after the first 40 pts completed >6 months of follow-up. Interim analyses (IAs) were declared NI if the 2-year PFS lower CI bound was >80% among de-escalated LR/IR pts (Bootstrap approach; alpha 0.05). Secondary endpoints: safety, overall survival, distant metastasis-free survival, quality of life (QoL) metrics, and exploratory radiomic data. Results: From 7/2021 to 5/2024, 138 pts screened, 71 accrued (cohort 1). Most were male (62, 87%) and white (70, 99%) with a median age of 63 (range: 46-80). Forty-four (62%) were assigned to the LR arm (23% had N2 disease) with a median pre-tx HPV DNA of 439 (range: 6-221995). Twenty-seven (38%) pts were assigned to the IR arm (33% T4, 78% smokers) with a median pre-tx HPV DNA of 2346 (range: 207-84971). Among the 27 IR pts, 18 (67%) had >95% DNA clearance and were de-escalated. At a median follow-up of 14 months, the 2-year PFS estimate at IA2 was 92% (95%CI, 85-100) among 62 de-escalated LR/IR pts including 18 (29%) T4/smokers and 15 (24%) with N2 disease. Two distant and 2 locoregional plus distant failures occurred; with no isolated locoregional failures. QoL data is forthcoming; 1 patient died from disease. Cohort 2 explores further de-escalation and is ongoing. Conclusions: Using HPV DNA-guided CRT de-intensification we achieved our primary endpoint and report a favorable 2-year PFS >90%, which included T4 pts and smokers. HPV DNA as a biomarker to guide de-intensification warrants further study. Clinical trial information: NCT04900623 .