Multi-omics data integration for patient stratification in cancer clinical trials

Lead Supervisor
Professor Francesca Ciccarelli
Professor of Cancer Genomics
Cancer & Pharmaceutical Sciences, King’s College London and Francis Crick Institute
francesca.ciccarelli@kcl.ac.uk

Co-supervisor
Professor Chris Yau
Professor of Artificial Intelligence at the University of Manchester and Fellow of the Turing Institute

Project Details

Immunotherapy has introduced a paradigm shift in cancer treatment, improving outcomes in the metastatic setting of several cancer types. However, many patients still fail to benefit because of primary or acquired resistance. Several factors are known to play a role in determining response to immunotherapy. These include extrinsic heterogeneity in the type of immune cells present in the tumour micro-environment and intrinsic heterogeneity in the tumour mutational burden, which is critical to drive neo-antigen formation that stimulates immune responses. The interplay between these factors and how they can be used together with clinical data as biomarkers to predict response and guide therapeutic intervention is still largely unknown.

In this project, we aim to derive maps of the genetic, transcriptomic and immune heterogeneity of the tumours and associated micro-environment through data integration and correlate these molecular maps with clinical data and response to immunotherapy. We have access to gastrointestinal tumours (adenocarcinomas of the oesophagus, gastroesophageal junction, stomach, colon and rectum) and matched clinical information from patients enrolled in clinical trials for the use of immunotherapy agents. These samples have been already subjected to deep whole exome sequencing, shotgun transcriptomics and imaging mass cytometry (solid cyTOF) to profile the immune infiltrates. This project aims to integrate these different types of data to derive correlative maps of the dynamic interplay in space and time between immune and tumour heterogeneity. The resulting molecular maps will then be associated with patient’s clinical data and response to treatment allowing to (1) identify key molecular players of resistance, reveal mechanistic insights and develop biomarkers of response and (2) inform strategies based on these biomarkers to prevent the onset of resistance and convert cold tumours into hot tumours responsive to immunotherapy.

The Ciccarelli group has a long-lasting interest and expertise in the production and analysis of cancer genomics, transcriptomics and immune-phenotypic data (Cereda, Nature Comms 2016; Ciccarelli Nature 2019). The team is highly multi-disciplinary being composed of mathematicians, physicists, computer scientists, biologists and clinicians. In addition to data analysis, we develop methods and databases (Gambardella Bioinformatics 2017; Repana Genome Biology 2019) and collaborate with Professor Yau for the application of machine learning to cancer genomics (Mourikis Nature Comms 2019). The student will be based at the Francis Crick Institute, a world leading biomedical laboratory in central London with access to top-class High Performance Computing facility of the Cancer Research UK City of London Cancer Centre. The collaboration with Professor Yau will expose the student to the scientific environment of the Alan Turing Institute, a research centre devoted to big data analysis.

Datasets

TCGA; clinical data and matched genomic, transcriptomic, immuno-phenotyoic data form patients enrolled in clinical trial. The ethics for the project has been already approved.

Keywords

data integration; patient stratification; clinical trials; cancer immunotherapy