Background Data analysis of specimens from prior clinical trials identified the immune co-simulatory molecule CD137 within the tumor microenvironment(TME) of pancreatic ductal adenocarcinoma(PDAC) that remain to be activated following vaccine induced T cell and PD-1 inhibitory treatments. The requirement of CD137 was subsequently supported by preclinical studies. Therefore, we conducted a clinical trial of combining anti-CD137 agonist antibody urelumab, anti-PD-1 antagonist antibody nivolumab and a GM-CSF-secreting allogeneic tumor cell vaccine(GVAX) as neoadjuvant and adjuvant therapy for resectable PDAC.

Methods Patients of >=18 years old with radiographic evidence of resectable PDACs were eligible for Arm C of this trial(NCT02451982) with an accrual goal of 10 evaluable subjects. The primary objective was to evaluate changes in numbers of tumor infiltrating CD137+CD8+ T cells. Secondary objectives were safety, overall survival, disease free survival, and other immune parameters. Patients who underwent R0/R1 resection were considered evaluable. All subjects received 480mg nivolumab and 8 mg urelumab both intravenously one day prior to receiving GVAX intradermally and two weeks before surgical resection (figure 1). After surgery, eligible patients continued to receive 5 combination immunotherapy cycles in addition to standard of care chemotherapy. Treatment-related toxicity and perioperative complications are monitored.

Results Between February 2019 and August 2020, we completed planned enrollment and treated 10 evaluable patients, who underwent R0 surgical resection of their PDACs. Nausea is the most common adverse event attributed to urelumab (table 1). Other adverse events and perioperative complication were observed in a type, frequency and degree similar to other treatment arms. After repeated dosing, 1 patient demonstrated grade 1 arthritis; 1 patient demonstrated self-limited, transient grade 2 elevated LFTs; 1 patient developed grade 3 rashes, which responded quickly to oral steroid and did not recur after re-dosing. Interestingly, two out of 10 resected patients demonstrated CAP grade 2 pathologic responses in the resected PDACs after a single neoadjuvant treatment; this was not observed with other treatment cohorts(GVAX alone or GVAX+nivolumab) in this neoadjuvant platform trial. Nine out of 10 resected patients remain disease free after a median follow up of 12 months. Immunology endpoints are being analyzed by multiplex immunohistochemistry, DNA sequencing for neoantigen loads, and RNA/TCR sequencing.

• Download figure
• Open in new tab
• Download powerpoint

Abstract 812 Figure 1

Study Schema

Conclusions Previous observations of liver toxicity with urelumab or other T cells agonists and severe immune-related adverse events were not observed in this trial, suggesting urelumab(8 mg) is safe as neoadjuvant/adjuvant therapy in this resectable PDAC patient population. Immune and clinical efficacy of anti-CD137 agonist-based combinations warrant further investigation.

Acknowledgements This is an investigator initiated clinical trial and supported by the funding from the Rare Disease Program at Bristol-Myers Squibb.

Trial Registration NCT02451982

Ethics Approval The study was approved by the Johns Hopkins Medical Institution Institutional Review Board, approved number IRB00050517