The 5-year research project involves 14 organizations in seven EU countries and is funded by €15 M from the EU.
The Genome Data Science lab, led by ICREA researcher Fran Supek, will analyze the evolution of ovarian tumors to develop computational models for predicting treatment resistance.
The EU has granted €15 M to fund a 5-year project that seeks to improve personalized treatments for ovarian cancer. The international DECIDER project involves partners from 14 organizations in seven EU countries.
In Europe, over 40000 women die of ovarian cancer every year. In addition to surgery, most patients are treated with platinum-based chemotherapy. Unfortunately, the effect of the chemotherapy often decreases with treatment cycles, and few effective treatments are currently available for patients who develop resistance to platinum-based drugs.
The project kicks off in February and will develop diagnostic tools for earlier and more reliable identification of patients not responding to current treatments. Based on the data collected from the tumors, the project also aims to discover effective combination treatments.
"What makes this project unique, besides the multidisciplinary and extraordinary quality of the partners, is that we will be working with several samples from the same tumors, taken at different time points. We will be studying the evolution of the tumor, and our main goal is to be able to anticipate genetic changes in cancer that lead to treatment resistance," says Fran Supek, head of the Genome Data Science lab at IRB Barcelona.
Cancer is one of the leading causes of mortality in the world and as the population ages, the incidence of cancers will only increase. Any new tools that enable better and more targeted treatment of cancers in the future will not only decrease the amount of human suffering, but also the burden on healthcare."
Sampsa Hautaniemi, Professor and Coordinator of DECIDER Project, University of Helsinki
Artificial intelligence to integrate and visualize patient information for doctors
In the project, a patient's response to treatments is predicted by methods that use, among others, histopathological and genomic data from the patient. Genomic changes and aberrations in gene functions are used to find effective, personalized treatments.
At the Genome Data Science lab, 3-4 researchers will be working on the development of computational models that predict the evolution of every particular tumor and that will aim to make treatment recommendations based on the genetic profile. The group will combine experimental approaches based on gene editing, and computational approaches based on machine learning to understand drug and radiation resistance mechanisms.
"Tumors often have defective mechanisms to repair their DNA, which speeds up their evolution but potentially also results in drug vulnerabilities or drug resistance. If we understand the mechanisms underlying resistance, by sequencing the tumor we would be able to predict whether it would easily be able to gain resistance to a specific drug in the future. This would lead to understanding which other treatments would be more effective on that particular tumor, to slow down its evolution and ultimately improve health outcomes," Supek explains.
An important part of the project will be the collaboration with small-and-medium enterprises (SME) in developing, producing, and registering diagnostic kits, producing a drug-sensitivity test based on the tumor samples, developing image-based diagnostics of digital samples, and developing data pseudonymization and anonymization techniques necessary for the management of sensitive patient data for privacy protection. Patient organizations have an important advisory role in the project.
Legal researchers address ethical and legal concerns
In addition to the medical research, the project also includes a legal work package that addresses the ethical and legal aspects of the project. Furthermore, the legal researchers will also study whether there are inconsistencies between the pharmaceutical regulatory system and other relevant legislation.