Summary of the context and overall objectives of the project

Prostate cancer (PC) is the third leading cause of cancer death amongst men in the developed world. Most aging men will develop PC, yet the life-time risk of a PC-caused death is only 3%. For this reason, care givers aim to distinguish between clinically significant and insignificant disease. However, risk evaluation remains a challenge due in part to PC’s both inter- and intra-tumour heterogeneity. Despite recent findings that highlight the wide range of tumour variability, standard PC diagnostics does not include genomic profiling resulting in a lack of personalized therapies for patients with castration-resistant PC. The PrECISE consortium’s partners share the aim of developing algorithms and technologies to improve risk evaluation of PC patients. This improvement is much needed to avoid unnecessary treatments that heavily deteriorate the patient’s quality of life, reduce the financial burden associated with over-treatment, and focus available treatment capacity on those patients that actually benefit from it. Moreover, few treatment options are available for aggressive drug-resistant PCs. Diagnosis and treatment of these tumours will benefit from identification of predictive biomarkers and the master regulators of drug-resistance, but these markers can only be found if appropriate techniques that take into account tumour heterogeneity are developed. The consortium has tackled the problem by profiling DNA from multiple biopsies per tumour and by integrating different type of omics data in order to characterise the molecular state of each patient at different malignancy stages. Furthermore, PrECISE also aims to develop new approaches to suggest chemotherapy drugs and targeted therapies for each patient. We have investigated molecular mechanisms and identified suitable intervention points for therapy that helps designing personalized therapies based on each patient’s molecular signatures. Finally, we want to make our results easily accessible to scientists, clinicians and patients. With this aim in mind, we focused our efforts in developing PrECISE algorithms as open-source software or open access web services, mostly available on a common platform known as the SmartBiobank.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

The consortium developed a set of computational methodologies to integrate publicly available multi-omics datasets, well-characterized multiple-biopsies cohorts, and literature-driven knowledge that have been available as open access services on IBM Cloud. These methodologies unravel molecular mechanisms that drive prostate cancer and focus on identifying new complex biomarkers that can distinguish benign from malignant tumours. Additionally, by characterizing intra-tumour heterogeneity, which is known to play a fundamental role in the diverse clinical evolution of prostate cancer patients, the partners have been able analyse disease evolution at patient-specific level. Specifically, the consortium developed a novel computational approach to unravel tumour heterogeneity using multiple biopsies from a patient and understand its influence in clinical phenotypes. The algorithm has been validated in castration resistant and metastatic biopsies produced by data provider institutions part of the consortium. In PrECISE we also developed novel approaches to suggest chemotherapy drugs and targeted therapies for each patient. By developing computational methodologies to analyse molecular mechanisms, we have been able to identify suitable intervention points for therapy and to suggest personalized therapies based on each patient’s molecular signatures. These algorithms have been validated on data produced inside the consortium. In addition, the consortium developed machine learning approaches that leveraged and combined various sources of prior knowledge such as the available biomolecular interaction databases in order to analyze different omics data and identify biomarkers for patient stratification or drug sensitivity predication.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

PrECISE produced multiple outcomes in terms of scientific publications, reusable computational tools and different omic datasets. All these elements constitute an extremely valuable set of resources for the European society, since they will allow considerable advances in systems biology research for complex diseases beyond cancer research. Another crucial aspect for the impact of the project is the accessibility of the solutions developed. All partners devoted efforts to make the developed algorithms and methods publicly available in form of open-source software or web services, We have implemented a cross-platform to share data and gain access to the majority of the developed approaches. This user-friendly framework will enable the access, analysis and visualisation of molecular data by clinicians and future stakeholders. To summarise, PrECISE brings together high profile researchers from the biological and clinical domains, computational and technology experts and innovative companies. The consortium has not only created jobs for its research employees, but is also training the next generation of leading scientists in this highly interdisciplinary project. But the PrECISE contribution to society is much larger than the impact on the partners’ research activities. PrECISE delivered enhanced innovation capacity and the translation of new knowledge into clinical settings, therefore strengthening the competitiveness and growth of participants, project stakeholders and the overall society. For this very important task, the coordinator was supported by the technical lead IBM, which has over 21 years of patent leadership and has demonstrated to translate basic research and ideas into innovation and market impact.