Using a computational tool, DeepTarget, physicians were able to predict both primary and secondary targets of small-molecule agents for cancer treatment, according to findings from a study published in npj Precision Oncology. The study authors suggest that this represents a potentially significant advancement that could be used to accelerate drug development in oncology.
“Sometimes the field looks at these [small-molecule] drugs with tunnel vision in terms of them having a single target along with some side effects labeled as ‘off-target effects,'” said Sanju Sinha, PhD, Assistant Professor, Cancer Metabolism and Microenvironment Program, Sanford Burnham Prebys Medical Discovery Institute, La Jolla, California. “Taking a more holistic view reveals that small molecules can have different targets and effects depending on the disease and cell type, and we can use this knowledge to repurpose more drugs to treat more patients.”
Study Methods
Researchers wanted to explore the malleability of small-molecule drugs and identify both on-target and off-target effects of cancer drugs. Dr. Sinha and his colleagues developed DeepTarget as an open-source tool to integrate large-scale drug and genetic knockdown viability screens plus omics data to determine cancer drugs' mechanisms of action.
The computational tool was tested on eight datasets of high-confidence drug-target pairs for cancer drugs. The test was used to benchmark DeepTarget. Then, the researchers experimentally validated the tool's predictive ability on two case studies: the antiparasitic agent pyrimethamine, and ibrutinib in the setting of solid tumors with EGFR T790 mutations.
Key Study Findings
Benchmark testing for DeepTarget revealed that the tool outperformed currently used tools such as RoseTTAFold All-Atom and Chai-1 in seven out of eight drug-target test pairs for predicting drug targets and their mutation specificity. DeepTarget showed a strong predictive ability across diverse datasets for determining both primary and secondary targets.
In the validation case studies, DeepTarget demonstrated that pyrimethamine affects cellular viability by modulating mitochondrial function in the oxidative phosphorylation pathway. In the second case study, DeepTarget showed that EGFR T790 mutations influence response to ibrutinib in BTK-negative solid tumors.
DeepTarget also showed that kinase inhibitors expected to have higher target specificity led to increased progression in clinical trials.
Overall, the computational model predicted target profiles for 1,500 cancer-related drugs as well as 33,000 natural product extracts that have previously been unpublished.
“We believe that the tool’s superior performance in real-world scenarios is due to it more closely mirroring real-world drug mechanisms, where cellular context and pathway-level effects often play crucial roles beyond direct binding interactions,” Dr. Sinha said. “It also underscores DeepTarget’s potential to accelerate drug development and repurposing efforts as a complementary approach alongside structural methods focused on chemical binding.”
“The potential pool of chemicals is much larger than what we are able to screen for even with modern, high-throughput drug screening methods,” concluded Dr. Sinha. “Improving treatment options for cancer and for related and even more complex conditions like aging will depend on us improving both our ways to understand the biology as well as ways to modulate it with therapies.”
Disclosure: The study was supported by the National Institutes of Health, National Cancer Institute, and Sanford Burnham Prebys. Drs. Ruppin and Sinha are the inventors on a pending provisional patent application related to the DeepTarget algorithm. Dr. Ruppin is a co-founder of Medaware, Metabomed, and Pangea Therapeutics (divested from the latter); he is also a nonpaid scientific consultant to Pangea Therapeutics. The rest of the study authors declared no conflict of interest.
