Combination therapy is a standard treatment for multiple diseases, such as HIV, Tuberculosis and different cancers. High throughput drug combination screening can be used to identify novel drug combination against a disease by profiling phenotypic effects of thousands of drug combinations in patient-derived cell lines or other pre-clinical model systems.
However, interpreting the clinical significance of the massive datasets that the combination screening produces is not straightforward. Currently, the identification of most promising drug combinations often relies merely on the observed synergy (excess beneficial response observed due to the use of multiple drugs at a time) between drugs. This approach neglects the other key determinants for the therapeutic success of drugs in the clinics, i.e. the actual efficacy and potential toxic effects.
The FIMM research team wanted to overcome this challenge by developing a systematic, computational-experimental tool that can incorporate efficacy and toxicity with the synergy data of a drug pair. The tool, called SynToxProfiler, was described in a recent PLOS Computational Biology article and is freely available online.
The input needed for the SynToxProfiler is a drug combination dose-response data from both diseased cells (e.g. cancer cells, virus-infected cell models) and healthy cells (e.g. blood cells). SynToxProfiler calculates the synergy and efficacy of drug pairs using the measurement from diseased cells and estimates toxicity from healthy cells. This enables users to select effective drug combinations with less toxicity for a single cell line or individual, therefore facilitating real-time personalized medicine applications.
In the publication, the research team also demonstrated the applicability of SynToxProfiler as a systematic tool for the prioritization of top combinatorial hits both in T-cell prolymphocytic leukemia (T-PLL) and anti-Ebola drug combination screening.
Using a compendium of 20 drug combination matrices tested in T-cell prolymphocytic leukemia cells and 77 combinations in anti-Ebola infection models, integrative analysis of synergy, toxicity and efficacy prioritized clinically established drug pairs as the top hits. The results also demonstrated that the top hit selected based on the proposed method has a better chance of success in clinics.
“Notably, around 20% of drugs fail in the early development phase because of safety concerns, and over 50% fail due to a lack of sufficient efficacy. We hope that our method will decrease the failure rate of drug combinations during drug development”, says Doctoral Student Aleksandr Ianevski, who implemented the tool.
The method makes use of a novel scoring algorithm (STE score) developed in Tero Aittokallio’s lab to rank the drug combinations. The method development was carried out in collaboration with the FIMM High Throughput Biomedicine (HTB) unit to support future combinatorial screens.
The tool was developed as part of the European Union's Horizon 2020 Project, ERA PerMed JAKSTAT-TARGET.
Original publication:
Aleksandr Ianevski, Sanna Timonen, Alexander Kononov, Tero Aittokallio, Anil K. Giri. SynToxProfiler: An interactive analysis of drug combination synergy, toxicity, and efficacy. https://doi.org/10.1371/journal.pcbi.1007604
Further information:
Aleksandr Ianevski
Doctoral Student
Institute for Molecular Medicine Finland FIMM, HiLIFE, University of Helsinki
e-mail: aleksandr.ianevski@helsinki.fi
Anil Kumar
Postdoctoral Researcher
Institute for Molecular Medicine Finland FIMM, HiLIFE, University of Helsinki
e-mail: anil.kumar@helsinki.fi