Killing cancer cells, saving normal ones
A new experimental-computational approach developed by researchers from Denmark and Finland shows promise for treating individual patients with blood cancers or solid tumours. The research team used large-scale public pharmacogenomics data to pre-train a machine learning model called scTherapy that accurately predicts clone-specific treatment options solely based on single-cell RNA sequencing (scRNA-seq). The research team’s novel approach was developed with financial support from the EU-funded projects REMEDI4ALL, CROC, RESIST3D and DISCOVER. Financial support was also provided by another EU-funded project, ERA PerMed, through the PARIS, JAKSTAT-TARGET and CLL-CLUE projects funded through its 2020 and 2022 Joint Transnational Calls. A single tumour can consist of distinct tumour cell populations, or clones, with different molecular and phenotypical profiles. Treating advanced malignancies therefore requires therapies that can act on different targets at the same time. However, systematically identifying patient-specific treatments that selectively co-inhibit cancer clones poses a challenge since there are many more possible drug-dose combinations than what could be tested in patient cells. The study published in ‘Nature Communications’ shows how the new approach makes it possible to identify personalised multi-targeting treatment options in different patient and cell populations. The case studies the researchers focus on are blood cancers (relapsed and refractory acute myeloid leukaemia) and solid tumours (metastatic ovarian carcinoma). Ovarian cancer material and data were provided in part by the EU-funded DECIDER project.
Low toxicity for healthy cells
“We demonstrate that the predicted combinations do not only show synergistic effect in overall cancer cell killing, but also result in minimal toxic side effects in non-cancerous cells, thereby increasing the likelihood for clinical translation,” states study joint first author Aleksandr Ianevski in a news item posted on the website of DECIDER project coordinator and REMEDI4ALL project partner University of Helsinki, Finland. Ianevski is a doctoral researcher at the Institute for Molecular Medicine Finland (FIMM), a translational research institute at the university that focuses on human genomics and precision medicine. The results show that over 95 % of the predicted treatment combinations have synergy, meaning they have the potential to improve treatment efficacy. Also, more than 80 % of combinations resulted in low toxicity to normal cells. “Since the approach uses only a limited number of patient primary cells, it is widely applicable to any patient samples that are amenable to scRNA-seq profiling. Selective combinations among approved drugs also provide straightforward repurposing opportunities for cancer treatment,” remarks joint first author Kristen Nader, also a doctoral researcher at FIMM. The ERA PerMed (ERA-Net Cofund in Personalised Medicine) and RESIST3D (Targeting drug resistance in ovarian cancer through large-scale drug-response profiling in physiologically relevant cancer organoids) projects have ended. The CROC (Unravelling ChemoResistance mechanisms and improving first-line therapeutic strategies in high-grade serous Ovarian Carcinoma using multi-culture patient-derived organoids) project ends in December 2024 and DECIDER (Improved clinical decisions via integrating multiple data levels to overcome chemotherapy resistance in high-grade serous ovarian cancer) in 2026. REMEDI4ALL (BUILDING A SUSTAINABLE EUROPEAN INNOVATION PLATFORM TO ENHANCE THE REPURPOSING OF MEDICINES FOR ALL) and DISCOVER (DISCOVERing treatment from biomedical research) both end in 2027. For more information, please see: DECIDER project website ERA PerMed project website REMEDI4ALL project website CROC project RESIST3D project DISCOVER project website
Keywords
DECIDER, ERA PerMed, REMEDI4ALL, CROC, RESIST3D, DISCOVER, cancer, cancer clone, RNA sequencing, blood cancer, solid tumour, tumour, scTherapy