e15654 Background: Myeloid-derived suppressor cells (MDSCs) in the tumor microenvironment are potential therapeutic targets in immune checkpoint cancer therapy, particularly for cancers that are unresponsive to anti-PD-1 therapy. It has previously been demonstrated that trefoil factor family 2 (TFF2), a secreted anti-inflammatory peptide, can partially suppress MDSC expansion and partially activate tumor immunity through agonism of the CXCR4 receptor. We investigated whether a novel fusion protein, TFF2-murine serum albumin (TFF2-MSA), has single agent activity and can improve on the therapeutic effects of anti-PD-1 in the MC38 and CT26.wt syngeneic mouse models of advanced colorectal cancer (CRC). Methods: Two syngeneic colon carcinoma mouse models were developed using the MC38 and CT26.wt CRC cell lines grafted subcutaneously into C57BL/6 and BALB/C mice, respectively. We generated a recombinant fusion protein, designated TFF2-MSA, which contains murine TFF2 fused to murine serum albumin (MSA), for the purpose of increasing half-life and reducing the frequency of dosing. Mice subsequently received TFF2-MSA, anti-PD-1 antibody (clone 29F.1A12) or combination TFF2-MSA/anti-PD-1. Tumor volume, and survival were measured. At the endpoint, flow cytometry was performed on the draining and axillary lymph nodes, to examine treatment-induced effects on cellular immune profiles. Results: In the MC38 model, on Day 49 of treatment, tumor growth was suppressed (TGI) by TFF2-MSA alone, anti-PD-1 alone, and by combination TFF2-MSA/anti-PD-1 by 50%, 82% and 87% TGI, respectively. The survival rates at 50 days of mice treated with vehicle, TFF2-MSA, anti-PD-1, or combination TFF2-MSA/anti-PD-1 were 20%, 50% and 80%, and 90%, respectively. In the combination TFF2-MSA/anti-PD-1 treated mice, 40% demonstrated a complete response. In the CT26.wt model, tumor growth was suppressed by TFF2-MSA, anti-PD-1 and by combination TFF2-MSA/anti-PD-1 by 17%, 43% and 67%, respectively. Survival in the CT26.wt model at Day 30 treated with vehicle, TFF2-MSA, anti-PD1 and the combination of TFF2-MSA/anti-PD-1 was 0%, 40%, 60% and 60%, respectively. The percentage of exhausted CD8+ T cells was markedly reduced in the draining lymph node by the combination TFF2-MSA/anti-PD-1 treatment, as measured by flow cytometry using antibodies against LAG3, TIM3, and PD-1. Conclusions: TFF2-MSA has single agent activity and is additive to anti-PD-1 antibody checkpoint inhibition in treating syngeneic mouse models of advanced CRC. In a separate abstract, additive effects between TFF2-MSA and anti-PD-1 antibody were also demonstrated in a separate ACKP (Atp4b-Cre; Cdh1-/-; LSL-KrasG12D; Trp53-/-) gastric cancer model, suggesting combination therapy may also be applicable to gastric cancer.