Overall survival (OS) and pathologic response rate from a phase II clinical trial of neoadjuvant GVAX pancreas vaccine (with cyclophosphamide) in combination with nivolumab and stereotactic body radiation therapy (SBRT) followed by definitive resection for patients with borderline resectable pancreatic adenocarcinoma (BR-PDAC).

e16309 Background: The optimal management of BR-PDAC remains undefined due to the high rate of local progression and/or metastasis with upfront surgery alone. Given the limited activity of single agent PD-1 blockade in PDAC, combination strategies aimed at increasing high-avidity T-cells in the tumor microenvironment (TME) are under investigation. GVAX is an allogenic, whole cell, GM-CSF-secreting vaccine that activates T-cell immunity against tumor-associated antigens. Prior work has shown that GVAX in combination with low-dose cyclophosphamide (Cy) can inhibit T-regulatory cells and induce both PD-L1 expression and the formation of tertiary lymphoid aggregates. This multi-center phase II clinical trial evaluates the safety and clinical efficacy of the addition of GVAX/Cy/nivolumab and SBRT to neoadjuvant chemotherapy in patients with BR-PDAC. Methods: Patients received neoadjuvant mFOLFIRINOX or gemcitabine/abraxane if intolerant to mFOLFIRINOX. Two to six weeks after chemotherapy, the first dose combination immunotherapy with Cy (200 mg/m2), nivolumab (240mg) and GVAX (5 x 10 8 vaccine cells) was given. Three weeks later, patients received the second dose of combined immunotherapy concurrently with SBRT (6.6 Gy x 5 days) prior to surgical resection. The primary endpoint is CD8 + T-cell density (CD8) in surgical specimens compared with CD8 in historical control samples treated with neoadjuvant mFOLFIRINOX and SBRT only. Additional endpoints include pathologic response rates, adverse events (AEs) as graded by NCI CTCAE version 4.03, OS, and evaluation of immune changes to the TME. Results: 31 patients were enrolled from 2/2/2018 to 7/27/2021. Of these, 18 patients received at least one dose of combination immunotherapy. Following study treatment, one patient had progressive disease, three had occult intra-operative metastatic disease and 14 underwent definitive surgical resection (13 R0 resections and one R1 resection). The major pathologic response rate was 35% (defined as <10% residual viable tumor), including one pathologic complete response (pCR). At a median follow-up of 20.5 months, median OS was 20.4 months (95% CI 18.2, NA). There were two grade 3 or higher AEs (one case of nivolumab-related autoimmune hepatitis and one case of SBRT-related pancytopenia). Conclusions: The addition of combined immunotherapy and SBRT to chemotherapy was safe and feasible in BR-PDAC. Observed median OS, pCR and R0 resection rates were comparable to contemporary studies administering neoadjuvant mFOLFIRINOX and SBRT. Additional immune correlative data will be available and reported at the time of the conference, guiding future applications of novel vaccine approaches. Clinical trial information: NCT03161379 .