Human cancer is still a widely spread disease. Despite a significant decreased mortality for certain tumor types over the last two decades, it remains among the most common cause of death worldwide. In the last fifteen years, enormous progress in DNA sequencing technology and genome editing has revolutionized cancer genomic research and changed expectations of precision medicine, whereby tumor cells DNA alterations are exploited for patients’ treatment selection. Especially, the use of Next Generation Sequencing (NGS) allowed the scientific community to sequence thousands of tumor samples and to compare their DNA with their healthy counterparts widening the agnostic search for cancer vulnerabilities towards precision medicine in oncology. The SPICE research project has been designed to identify biomarkers of cancer progression and to nominate those vulnerabilities through a combination of computational genomics and experimental efforts. Mapping DNA alterations to the disruption of genes that accelerate disease progression potentially allows for the nomination of new drug targets for patients with advanced disease. Specifically, the search for drug targets was based on the concept of synthetic lethality, where the concomitant disruption of specific pairs (or combinations) of genes is fatal for the cell itself, while the disruption of one single gene is not. In practice, such gene combinations can be exploited by selecting a treatment to intentionally disrupt one gene, in presence of the alteration of a second gene in tumor cells; this can lead to a selective effect on tumor cells only while sparing all healthy cells. This approach could enormously reduce toxic effects typical of chemotherapeutic drugs and improve the quality of life of cancer patients. In order to enhance the chances to nominate these drug targets, in this project we implemented unbiased searches of biomarkers and synthetic lethal gene pairs using computational and mathematical methods to learn from the genomes of thousands of patients’ cancer samples and experimentally validated the most promising findings using state of the art techniques, including the use of genome editing and organoid technologies through collaborative efforts.
The project was designed to address specific clinical questions in the setting of lethal prostate cancers, a genetically and clinically heterogeneous disease with high incidence in the population. While advances in targeted therapy have recently led to more effective management of metastatic disease, prostate cancer still results in the second cause of cancer death in men. Significant effort was also dedicated to chemotherapy-resistant urothelial carcinoma. From a technical perspective, the project aimed at the development of computational approaches to be applicable across cancer, both for the refined analysis of genomic lesions and the inference of tumor evolution and disease progression biomarker identification, and for the nomination of potential synthetic lethal combinations. These approaches together with the results of the large-scale computations performed on thousands of tumors across more than twenty-five tumor types have been made available to the research community.
