Lipid A controls the robustness of intratumoral accumulation of attenuated Salmonella in mice.

Engineered Salmonella have the potential to treat cancers that are not responsive to standard molecular therapies. This potential has not been realized because colonization in human tumors is insufficient and variable as shown in preliminary phase I trials. Recent studies have shown that Salmonella colonization is associated with an inflammatory response mediated by tumor necrosis factor (TNF). An injectable agent, molecular lipid A, could be used to control bacterial accumulation because it induces TNF production and is rapidly cleared. We hypothesized that concurrently administrating lipid A with attenuated Salmonella would increase intratumoral accumulation, improve the robustness of tumor-targeting and be nontoxic. To test this hypothesis, Salmonella and lipid A were injected into mice with 4T1 mammary tumors. Colonization was quantified after 48 hr using anti-Salmonella immunofluorescence. A 2 μg/mouse dose of lipid A increased the area of colonized tissue fourfold, reduced variance 50% and ensured colonization in all mice. Comparatively, Salmonella failed to colonize some control mice, similar to human trials. No toxicity was observed in any treated mice. The fraction of tumor tissue with more than 25% bacterial coverage was eight times greater for treated mice compared to controls. Lipid A treatment also reduced the maximum average distance of tissue to Salmonella colonies from 1348 to 260 μm. A mathematical model of bacterial drug production predicted that 2 μg lipid A would increase tumor cell death by 82%. These results suggest that lipid A could solve the clinical challenges of Salmonella therapy and enable safe and robust treatment of cancer with bacteria.