Initial efficacy and biomarker analysis of a phase 1b study targeting IL-1β and PD-1 with chemotherapy in metastatic pancreatic cancer (PanCAN-SR1).

Abstract

e16287 Background: Pancreatic ductal adenocarcinoma (PDA) has been refractory to therapeutic targeting of the immune microenvironment. In preclinical work, IL-1β was upregulated in PDA tumors, and in mouse models, IL-1β expression led to activation of pancreatic stellate cells and immunosuppression. We hypothesize that blockade of IL-1β and PD-1 will result in significant alterations in immune and fibroblast subsets within the PDA microenvironment. Methods: We conducted an open-label multicenter Phase Ib study evaluating gemcitabine and nab-paclitaxel with canakinumab (ACZ885), a high-affinity human anti-interleukin-1β (IL-1β) mAb, and spartalizumab (PDR001), a mAb directed against human PD-1. 10 subjects with untreated metastatic PDA and RECIST measurable disease were enrolled. The primary objective was to identify a recommended phase II/III dose by evaluating the incidence of DLTs in the first 56 days. All subjects underwent baseline and on-study tissue and blood collection for extensive exploratory correlative studies. Secondary objectives including safety and tolerability and preliminary assessment of clinical activity. Results: 10 subjects were enrolled between November 2020 and March 2021. 6 out of 10 were evaluable for the dose confirmation. In the dose confirmation analysis there were no dose limiting toxicities (DLTs) and the recommended Phase II/III dose was established as; gemcitabine (1000 mg/m2 IV) and nab-paclitaxel (125 mg/m2 IV) on day 1,8,15; canakinumab (250 mg via subcutaneous injection) and spartalizumab (400 mg IV) on day 1; of each 28 day cycle. Adverse events were consistent with those seen with chemotherapy and were predominately hematologic. A preliminary efficacy analysis confirms 1 PR and 7 pts with SD. Two of the 7 pts have been treated for 11 and 12 cycles, respectively, without progression. Activation of CD8 T cells in peripheral blood and increased serum levels of IFN-induced chemokines CXCL9/10 were observed in patient samples. Using an in vitro suppression assay, we showed that baseline serum, from patients deriving clinical benefit, could induce myeloid derived suppressor cells, but that this effect was abrogated with treatment. Single cell transcriptional profiling also revealed treatment-dependent shifts in T cell activation state and myeloid cells in the tumors of patients experiencing clinical benefit. Conclusions: In this Phase Ib study, we established the Phase II/III dose of canakinumab and spartalizumab with chemotherapy in first line metastatic PDA. Translational studies suggest putative biomarkers that may help identify responding pts. This novel combination is being evaluated in a randomized Phase II/III study through the Precision Promise clinical trial network. Clinical trial information: NCT04581343.