SUB-urothelial durvalumab injection: 1 (SUBDUE-1)-Results from the first nine urothelial cancer patients using a dose-escalation schedule.

483 Background: The sub-urothelial administration of checkpoint inhibitors has not been reported. This approach could be safer and more efficacious than systemic delivery for patients with non-muscle-invasive bladder cancer (NMIBC). Methods: This phase 1b study employed a 3+3 dose-escalation design to explore tolerability, safety and immunological efficacy of sub-urothelial durvalumab, a programmed death-ligand 1 (PDL1) monoclonal antibody. Eligible participants had high risk NMIBC or MIBC without prior chemotherapy or immunotherapy (BCG allowed) and were planned for cystectomy. Participants received 25/75/150mg durvalumab diluted in 25mL normal saline injected into the suburothelium at 25 locations (25x1mL injections), at least 2 weeks prior to radical cystectomy. Systematic four quadrant cold cup bladder biopsies were taken immediately prior to durvalumab injection and immediately prior to cystectomy. Tumour, if present, was biopsied before and after injection and bladder maps recorded. International Prostate Symptom Index (IPSS) and O’Leary symptom score at various time points and adverse events (AE) as per CTCAE (Version 4) were recorded. Relative changes in immune cell counts (RCI) on bladder biopsy for CD3, CD8, CD68 and CD168 expressing cells are reported (value >1.0 designating increase). Results: Nine participants were recruited; eight male (89%), 1 female; mean age 72 years (range 56 – 82). No dose-limiting toxicities were observed. No evidence of treatment-related effect on IPSS or
O’Leary Symptom scores was seen. Fourteen AEs were reported by six (67%) patients: 10 were Grade 1, 3 Grade 2, 1 Grade 3. None were considered immune- or treatment-related by investigators. Transient elevation of peri-operative thyroid stimulating hormone was seen in two subjects, which normalised without intervention. No hepatitis was seen. All patients underwent planned cystectomy. RCI of different immune populations was calculated (see Table). Visible tumour was present in only 4 patients limiting interpretation of RCI. RCI varied significantly between cell types (p=0.008*). RCI numerically increased by dose but did not reach statistical significance (p=0.076**). A numeric increase in monocytes was seen at 150mg dose. RCI of different immune populations by dose of sub-urothelial durvalumab. Conclusions: Sub-urothelial injection of durvalumab was safe at all 3 dose levels without any drug-related adverse events. Immunological studies showed differential effects
on immune cells with macrophage population most affected. Further studies investigating the role of 150mg sub-urothelial durvalumab in the management of NMIBC are planned. Clinical trial information: ACTRN12620000063910.