40P - MANIFEST: A Multiomic Profiling Platform for Immuno-Oncology Biomarker Discovery

Background

Immune checkpoint inhibitors have become standard of care for many cancer subtypes. Next generation immuno-oncology (IO) agents are in clinical testing, with thousands of combinations in preclinical evaluation. The success of current and future immunotherapies relies on tools, data and technology to rationalize their use and manage toxicity. However, few biomarkers can distinguish responders from non-responders, predict toxicity, or guide treatment choices.

Methods

MANIFEST (https://www.manifest-io.org.uk/), a UKRI/MRC funded platform, leverages scalable methodologies to provide deep profiling of patients receiving immunotherapy; delivering multimodal data integration and modelling. The platform comprises NHS trusts (hospitals), research institutes and universities, and industry partners. The aims of MANIFEST is to harmonise sampling, assays and analyses of IO biomarkers and couple them with large-scale studies in patients with cancer. The platform’s utility will be demonstrated with exemplar projects encompassing multiple tumour types (melanoma, renal cell carcinoma, bladder cancer and triple-negative breast cancer), where predicting treatment outcomes and toxicities remains an unmet need.

Results

We have access to longitudinal samples of >3,000 patients across 10 reported studies. Through partner NHS sites, we aim to prospectively collect and profile samples (blood, stool and tumour) from ∼3,000 patients over 3 years. A tiered approach will implement workflows for high-volume biomarker discovery (Tier 1). In-depth profiling (Tiers 2&3) will further characterize tumours using peripheral immune profiling including high-dimensional flow cytometry, liquid biopsy (cfDNA, immune methylation profiling), spatial tissue image-profiling and molecular profiling (WES, bulk&long-read RNAseq, TCR&BCRseq). For selected patients, we will apply Representative Sequencing (RepSeq), to overcome sampling bias in solid tumours and conduct drug-sensitivity screening through patient-derived tumour fragments.

Conclusions

Machine learning will be applied to derive uni- and multi- modal biomarkers of response and toxicity, and the master databases will be available for ongoing academic and industry research.

Legal entity responsible for the study

The Francis Crick Institute & The MANIFEST Consortium.

Funding

UKRI, Medical Research Council and Office of Life Sciences.

Disclosure

T. Lawley: Financial Interests, Institutional, Leadership Role: Microbiotica. All other authors have declared no conflicts of interest.