Abstract 2948: The impact of drinking water pH on anti-PD-1 efficacy in syngeneic mouse models

Introduction: Mounting evidence has revealed the vital role of the gut microbiome (GM) in response to cancer immunotherapy in both animal models and patients. In preclinical in vivo oncology pharmacology studies, factors such as diet, housing, bedding and the pH of drinking water can influence the GM. Water acidification by adding hydrochloric acid is commonly used in many rodent facilities to inhibit bacterial growth. However, this may result in a different microbiome composition in animals compared to those receiving pH neutral water. The impact on response of tumor-bearing mice to immunotherapy is not yet clear. We therefore investigated the effects of drinking water pH on anti-PD-1 efficacy in MC38 and CT26.WT murine colon cancer models. Methods: C57BL/6 and BALB/c mice were housed with bedding exchange for 2 weeks upon arrival to standardize the gut microbiota. Each stain of mouse was then assigned into 3 cohorts and provided with pH neutral water, hydrochloric acid water or sodium bicarbonate water 7 days prior to tumor inoculation. MC38 and CT26.WT tumors were inoculated subcutaneously into C57BL/6 and BALB/c mice, respectively, and then randomized at average tumor volumes of 70-80 mm3 into two groups, receiving i.p. injections of 10 mg/kg of anti-PD-1 (CrownVivo™) or PBS twice weekly. The mice were sacrificed one day post last dose with tumors collected for immunophenotyping by FACS. Fecal samples were subsequently sequenced for 16S ribosome RNA. Results: Both models were modestly responsive to anti-PD-1 treatment at all 3 pH levels. In the MC38 model, anti-PD-1 treatment showed the largest tumor growth inhibition (TGI) with alkaline water (TGI=74%), followed by pH neutral water (TGI=66%) and then acidified water (TGI=58%). Interestingly, unlike MC38, CT26.WT tumors showed little difference in response to anti-PD-1 treatment with different water pH levels. Nevertheless, significantly increased CD3+ and CD8+ T cells, while decreased myeloid cells (CD11b+) and macrophages in tumor infiltrating lymphocytes (TIL) were shown in mice provided with alkaline water compared to mice with neutral water, which implied a more immune active tumor microenvironment (TME) in alkaline conditions for the CT26.WT model. The TIL phenotyping of the MC38 model was consistent with the in vivo efficacy data, with the mice receiving acidified water having the lowest total lymphocyte infiltration (CD45+), T cells (CD3+,CD4+ and CD8+) and highest macrophages, thus prone to an