Diabetes and cancer: using genomics to probe the biological mechanism

Inga Prokopenko’s research study looks at the biological processes that may link type 2 diabetes and cancer, two of the most significant causes of death globally

  • Topic: Combination of cancers
  • Institution: Imperial College London
  • Country: United Kingdom
  • Status: Completed
Researcher: Inga Prokopenko

Co-applicants

  • Dr Marc Gunter, International Agency for Research on Cancer, France
  • Prof Elio Riboli, Imperial College London, UK
  • Prof Philippe Froguel, Imperial College London, UK

Background

Type 2 Diabetes is a risk factor for several cancers, including breast, colorectal and pancreatic. Individuals with T2D might have decreased risk of developing prostate cancer.

Aims

We aimed to establish the complex relationships and shared genetics between diabetes and four cancers, determined by individual differences in DNA and biological processes, influenced by these DNA variants.

How it was done

We combined data from The European Prospective Investigation into Cancer and Nutrition (EPIC) study, UK Biobank and integrated this information with established diabetes and four cancers risk DNA variants from earlier large-scale genetic studies. Using machine learning approaches, a type of artificial intelligence, we grouped established DNA variants into certain biological tasks such as glucose metabolism, sex hormones, etc. We then took those groups and tested their effects on developing cancer or disease.

Findings

We found that pancreatic, colorectal and breast cancer genetic factors also confer increased risk for diabetes; while prostate cancer genetic variants are inversely associated with diabetes, in line with the Ying-Yang relationship. Diabetes genetic factors decrease the risk for prostate cancer, in line with epidemiological findings. Overall, our research puts forward both, the role of genetic risk definition for co-occurring diseases to identify individuals at higher risk, and the importance of risk classification based on distinct biological mechanisms to implement principles of precision medicine.

Impact

  • We end the debate on the mechanistic link between prostate cancer and diabetes, as we pointed out which biological pathways are shared between the two diseases.
  • We describe causal effects of two biological pathways on colorectal (metabolic syndrome and insulin resistance) and breast cancer (higher adiposity, higher insulin resistance, reduced sex hormones and lower age at menarche) risk estimated using Mendelian Randomisation.
  • We explain that the age dependent link between breast cancer and body mass index is attributed to a certain biological pathway.
  • We re-classify the effects of genetic loci and their association to diabetes and cancer.
  • We show that alkyl-acyl phospholipids with arachidonic acid may have a causal role in colorectal cancer.

Grant publication