ERA PerMed 2021 - Multidisciplinary Research Projects on Personalised Medicine – Development of Clinical Support Tools for Personalised Medicine Implementation

Epigenomic and machine learning models to predict pancreatic cancer: development of a new algorithm to integrate clinical, omics, DNA methylation biomarkers and environmental data for early detection of pancreatic cancer in high-risk individuals. – IMAGene

Submission summary

Pancreatic cancer (PC) has the lowest survival rate of all cancers in Europe, with no early detection strategies available. The IMAGene project will develop, implement and test a comprehensive Cancer Risk Prediction Algorithm (CRPA) to predict PC in high-risk (HR) asymptomatic subjects; it will investigate the potential for DNA methylation biomarkers to improve currently available risk indexes, and validate the feasibility of using liquid biopsies for early detection of cancer in such HR individuals. A sample of 200 healthy first-degree relatives of PC patients will be recruited, and their epidemiological factors related to PC risks assessed through initial interviews. Subjects will receive medical and psychological visits, and will undergo screening for germline mutations, DNA methylation profiling plus Whole-body MRI at baseline. Biostatistical analysis of data will be performed to develop algorithms able to extract risk profiles from biological and imaging data. All subjects will undergo epigenetic follow-ups plus radiological exam at 1 year after baseline. Participants’ lifestyle, epidemiological and psycho-decisional assessment will be performed through a monitoring process over the 3 years of the project. HR subjects’ lifestyle data will be correlated with DNA methylation profiles. All data collected will feed the supervised machine learning CRPA. Assuming the risk assessment through CRPA and DNA methylation profiling allows a two or three-fold enrichment in early detection of suspicious cancer in HR individuals (compared to the detection rate of pancreatic cysts with malignant potential observed in non-stratified asymptomatic population, 9.3%), we expect a detection rate of suspicious lesions of 20-25% in the selected population. A cost-utility and a detailed ethical analysis will be conducted. The IMAGene project will adopt a transnational, multi-level, multidisciplinary and multi-methodological approach to achieve its aims.

Project coordination

Serena OLIVERI ()

The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.


CHUT Centre Hospitalier Universitaire de Toulouse

Help of the ANR 1,248,175 euros
Beginning and duration of the scientific project: March 2022 - 36 Months

Useful links

Explorez notre base de projets financés



ANR makes available its datasets on funded projects, click here to find more.

Sign up for the latest news:
Subscribe to our newsletter