Research

Within the research focus of IARC, GEM has shaped a 5-year portfolio of 10 projects that harness its innovative science, multidisciplinary and unique expertise to lead international consortia and develop cancer research projects using advanced technologies, and a unique scientific partnership with international institutions. Situated within IARC’s Pillar II: Understanding the Causes, GEM’s overarching aim is to elucidate and refine the risk factors and exposures that drive cancer initiation and progression, and to translate these discoveries into effective cancer prevention strategies. The unifying theme of GEM is the continued deployment and development of genomic epidemiology as a transformative framework for cancer prevention and control: drawing from genomic studies and precise exposure assessment to illuminate causal pathways, refining strategies for early detection, and generating evidence that can be operationalized within cancer control frameworks. By combining the convening power and reach of IARC/WHO, GEM’s multidisciplinary expertise, and state-of-the-art infrastructure, GEM will continue to deliver high-impact science that reduces the incidence and burden of cancer worldwide – advancing knowledge from molecular mechanisms to population health and ensuring that genomic epidemiology directly informs global strategies to prevent cancer.

Axis 1: Discovering exposures

1. Mutographs – early-onset colorectal cancer (eoCRC): Focuses on the rising incidence of early-onset colorectal cancer using whole-genome mutational signatures to attribute causal exposures and resolve age-specific etiological heterogeneity.
2. Mutographs-ENV (various cancer types): Maps exposure to common and novel mutagens through large-scale whole-genome sequencing of tumour and matched normal tissues across diverse populations, building an exposure–signature atlas for causal inference.
3. DISCERN (various cancer types): Applies a broad exposomics framework to discover and validate novel risk factors for renal, pancreatic, and colorectal cancers, linking exposure profiles to mechanistic pathways of carcinogenesis.

Axis 2: Understanding carcinogenesis

4. PROMINENT (various cancer types): Dissects how environmental and lifestyle promoters – including non-mutagenic factors – drive the transition from normal cells to cancer. Integrates longitudinal exposure assessment with somatic trajectories and clonal dynamics to pinpoint actionable promoting processes.
5. GENESIS (small cell lung cancer, pancreatic cancer): Uses high-resolution sequencing technologies and innovative analyses such as “slow-motion” modelling of cell-state transitions (temporal multi-omics, functional genomics) to expose the earliest steps of small cell lung cancer and pancreatic carcinogenesis and their modifiable levers.
6. Ecosystems (various cancer types): Develops mathematical and evolutionary models of tumour ecosystems across cancer types to identify constraints, tipping points, and optimal windows for preventive or interceptive strategies.

Axis 3: Precision early detection

7. Genetic susceptibility and international consortia: As polygenic and rare-variant profiles mature – particularly for heritable lymphomas – evaluates the clinical and surveillance utility for clearly defined high-risk subsets (with attention to equity, transportability, and ethical oversight).
8. WSI-AI (various cancer types): Discovers histopathological patterns of aggressive phenotypes via deep-learning embeddings of whole-slide images, molecularly guided and linked to -omics, to deliver clinically interpretable digital pathology tools.
9. UbioBca (bladder cancer): Develops and rigorously validates urine-based biomarkers for early detection and disease monitoring. The programme encompasses analytical validation, clinical performance against current standards (e.g. cytology), evaluation in haematuria and follow-up clinics, and readiness for implementation and cost–effectiveness assessment.
10. IMPACT–HNC (head and neck cancer). Identifies robust prognostic profiles for micrometastatic risk and disease progression at diagnosis by integrating multi-omic data, histopathology, and tumour microenvironment features.