James Schofield, Founder & Chief Executive Officer of TopMD Precision Medicine, and his team are working on a new project which builds on findings from the U-BIOPRED project, a complex, in-depth research project undertaken to understand severe asthma. Top MD is a spin-out company from Southampton University that uses enormous volumes of multi-omics data to stratify patient populations. Schofield explained that standard statistical approaches to multi-omics data are very limited, so he wanted to invent a new way to tackle the data.
Biology is often noisy and some markers do not accurately represent a biological system. The team developed a biology-focused topological analysis tool that measures pathway activation rather than individual gene abundance. This enables more robust identification of disease mechanisms and biomarkers. Differential abundance of individual molecules is not enough to determine the relationships between certain molecules.
Schofield applied his approach to a large IBD dataset. They obtained a large dataset from the Crohn’s Colitis Foundation, including ileal biopsy gene expression data from 629 patients (217 with Crohn’s disease). The data types that were integrated included lipidomics, proteomics, metabolomics and genomics.
Schofield’s approach stratified patients with Crohn's disease and ulcerative colitis into molecular phenotypes based on pathway activation, which can guide targeted therapy selection. Using the Top MD tool, the team mapped each patient’s molecular phenotype based on pathway activation in their ileal samples. Some patients were TNF-high, others IL-6-high, and some showed high activation of other pathways like TLR9 or interferon gamma.
Moving onto predicting patient response, the group used pathway activation signatures, especially for TNF to identify gene sets that predict which patients will respond to anti-TNF treatments. This could potentially improve outcomes and reducing unnecessary treatments. The method was validated on independent datasets, which demonstrates its utility for high-resolution patient stratification and personalised treatment strategies in IBD.
