FIMM researchers excel in prognostic modeling of prostate cancer in an international crowdsourcing challenge

Researchers from FIMM together with their collaborators developed a winning solution for predicting overall survival of patients with metastatic castrate resistant prostate cancer (mCRPC). mCRPC accounts for one third of all prostate cancer patients with metastatic disease. Although a number of options exist for treatment of mCRPC, the impact of these treatments has been modest when it comes to the overall or disease-specific survival of mCRPC patients in the past 20 years. Innovative research approaches are therefore urgently needed for improving the treatment of mCRPC patients.

 headed by Prof. Tero Aittokallio participated in the , a crowd-sourcing competition for highly demanding scientific problems by top research teams around the world. The participating teams were asked to submit predictive models based on over 150 clinical variables from four phase III clinical trials with over 2,000 mCRPC patients treated with docetaxel.

The winning team included researchers with various expertise from FIMM/University of Helsinki, Helsinki University Hospital and University of Turku. The members of this multidisciplinary team put their heads together and brought in their own expertise to improve the collaborative solution.

Image: Prostate Cancer DREAM Challenge

– Personally, I see this kind of experimental-computational modelling as an important part of multidisciplinary cancer research, through which it will be possible to find more precise and effective treatments for advanced and castration resistant prostate cancer, said MD Tuomas Mirtti who supervised the clinical part of the team.

 Through machine learning models, it is possible to make the most of clinical studies and to improve the design of future clinical trials by integrating the knowledge of basic scientists, bioinformaticians and clinicians with the accumulating data from biobanks such as the Helsinki Urological Biobank.

–  The datasets provided by the organizers involved many challenging factors, including integrating data from various sources, controlling potential batch effects, and choosing the most important features which may be partly non-overlapping between the studies", commented Teemu Daniel Laajala, a FIMM-UTU PhD student who coordinated the multi-site team.

– Obviously we are excited to have performed best in the challenge and hope that the mathematical models not only perform well in terms of a statistical metric, but eventually also provide an actual clinical benefit to the patients.

During this autumn the team will be invited to present its predictive model at the Scientific Retreat of Prostate Cancer Foundation in Washington and at the in Philadelphia. The best solution will be published in a top journal, as part of the Challenge overview paper.

The primary benefit of this Challenge will be to establish new quantitative benchmarks for prognostic modeling in mCRPC, with a potential impact for clinical decision making and ultimately understanding the mechanism of the disease progression. The top performing models have therefore the potential to become standards in the field as winning models will be promoted for further vetting by the American Joint Committee on Cancer (AJCC).

More information:

Sage Bionetworks and Project Data Sphere

Tero Aittokallio, PhD, Professor, Institute for Molecular Medicine Finland (FIMM), University of Helsinki, Email: tero.aittokallio at fimm.fi

Tuomas Mirtti, MD, PhD, Adj. Prof., Department of Pathology, HUSLAB, Helsinki University Hospital, Email: tuomas.mirtti at helsinki.fi