Pathologic and radiographic responses in a window of opportunity for durvalumab plus metformin trial for squamous cell carcinoma of the head and neck (HNSCC).

Abstract

6068 Background: Durvalumab is a human monoclonal IgG1 antibody directed against programmed death-ligand 1 (PD-L1). PD-1/PD-L1 immune checkpoint inhibition (ICI) shows promise in HNSCC, but durable responses have been seen in only a fraction of patients. Metformin, a biguanide oral anti-hyperglycemic, has shown promise in altering immunity within the tumor microenvironment (TME) towards a stronger anti-tumor distribution of immune cells. We aimed to investigate the combined effect of metformin and durvalumab in patients with HNSCC. Methods: This was a single-center prospective phase 1, window of opportunity clinical trial in which previously untreated patients with any stage resectable HNSCC were randomized 3:1 to durvalumab + metformin (Arm A) or durvalumab alone (Arm B) during a four-week period between diagnosis and surgical resection. Six patients were included in a safety lead-in of durvalumab and metformin and an additional 32 patients were randomized. The primary endpoint was immune cell polarization. Here we report pathologic and radiographic effect. Pathologic effect was graded independently by two pathologists. Radiographic effect was evaluated using the immune-related Response Criteria (irRC). Results: Thirty-eight patients were enrolled (29 Arm A, 9 Arm B). Three patients withdrew consent prior to intervention (2 Arm A, 1 Arm B) and were excluded from analysis. AJCC 8th edition staging was as follows: Stage I (n = 21), Stage II (n = 2), Stage III (n = 3), Stage IVa (n = 6), Stage IVb (n = 3). Primary tumor sites included the oropharynx (n = 20, all p16+), oral cavity (n = 11), larynx (n = 2), maxillary sinus (n = 1), and unknown (n = 1). Pathologic effect was observed in 55% (18/33) of evaluable patients: 60% in Arm A vs 37.5% in Arm B (p = 0.418). 40% of patients with involved lymph nodes had discordance of pathologic effect at the primary site versus lymph node. Radiographic response based on irRC among 30 evaluable patients included 1 CR, 1 PR, 24 SD, and 4 PD. There was a significant correlation between pathologic effect and radiographic disease control, defined as CR, PR, and SD (p = 0.021), but no correlation when looking only at radiographic responders (p = 0.925). No patients experienced Grade 3–4 treatment or immune-related adverse events or a delay in surgery due to trial participation. All patients remained resectable. Conclusions: Our data demonstrate that the study intervention was well-tolerated in HNSCC patients. There was a trend towards an increased proportion of pathologic responders in the group receiving metformin. Additional studies targeting the TME are needed to further elucidate whether synergistic effects between metformin and durvalumab were seen in this patient cohort. Clinical trial information: NCT03618654.