Molecular residual disease (MRD) interception in locoregionally advanced head and neck squamous cell carcinoma (LA-HNSCC): MERIDIAN study.

TPS6122 Background: The detection of circulating tumor DNA (ctDNA) after definitive treatment or MRD is associated with recurrence across different tumor types. We previously showed that ctDNA detection using a bespoke tumor-informed assay (RaDaR), in LA-HNSCC patients (pts) after surgery, definitive radiation (RT) or chemoRT (CRT), is associated with recurrence within a year (Sanz-Garcia et al, ASCO 2023). To date, the impact of intercepting MRD prior to progression in HNSCC has not been explored. Anti-PD-1 blockade has shown benefit in recurrent/metastatic HNSCC but the efficacy in LA-HNSCC is still uncertain. TIGIT is overexpressed in HNSCC and could be a potential target to improve immunotherapy (IO) outcomes. Rilvegostomig (AZD2936) is a monovalent bispecific TIGIT/PD-1 antibody that has shown acceptable safety in phase I studies, currently in phase III studies. We hypothesize that IO (AZD2936) could induce ctDNA clearance post-definitive treatment and avoid, or delay recurrence. Methods: This study will recruit 200 pts with high-risk LA-HNSCC treated with curative intent: surgery ± adjuvant therapy, definitive RT or CRT; stage III Human Papilloma virus (HPV) positive or III-IVB HPV negative. Archival tissue must be available for whole exome sequencing (WES). Pts will be enrolled in part A (definitive therapy) and part B (follow up – FU - post-definitive therapy). Plasma samples will be collected in part A (pre-treatment and post-surgery prior to adjuvant therapy) and part B (at 4-6 weeks: FU1 and 8-12 weeks: FU2). These FU samples will be analyzed in real time for ctDNA using RaDaR, an assay that targets patient specific somatic variants identified by WES of matched tumor tissue. MRD+ pts (defined as having ctDNA detected at FU2) will be enrolled to part C (interception) and randomized 3:1 to receive AZD2936 750mg IV q3w for 6 cycles or observation. 30% of pts (N=60) will be enrolled in part C. Tumor must be positive for PD-L1 (CPS≥1) and there should not be residual primary tumor or distant metastases at FU2; residual lymph nodes are allowed if neck dissection is performed. Part D will consist of long term FU of MRD+ pts; RaDaR will be performed at week 2 (W2) and week 10 (W10), and radiological assessments will be performed at W2. MRD- pts will be enrolled in part E (long term FU). The primary endpoint of MERIDIAN is ctDNA clearance in MRD+ pts, defined as no detection of ctDNA at Part D W2 and W10. We will have 87% power to identify a significant improvement in ctDNA clearance rate from 10% (observation) to 40% (AZD2936) given a significance level of 0.1. Secondary objectives include survival and safety. Exploratory analyses include: ctDNA detection using other assays (HPV DNA, methylated ctDNA), ctDNA detection in part E and beyond W10 in part D, quality of life assessments (FACT-ICM, EORTC-HN43), health economics and radiomics. As of February 2024, 19 patients are recruited. Clinical trial information: 05414032.