A trove of freely accessible cancer data helps researchers find new combination treatments.
Targeted cancer therapies work by blocking a specific cancer-driving signal, thus limiting harm to healthy cells and tissues. The problem is, tumors often develop resistance by switching to alternative signaling pathways to resume their life-threatening drive to multiply and spread.
Figure: A drug-protein connectivity map generated from the new dataset.To leap ahead in this arms race, researchers now have mapped out in rich detail how cancer cells of many types adapt when challenged with an array of targeted drugs. And they’ve made all of the data freely available to other scientists via a user-friendly web portal.
“This is the largest data set of its type in existence,” said Gordon Mills, M.D., Ph.D., a senior author of the paper in the journal Cancer Cell describing the work. Mills is a professor of cell, developmental and cancer biology in the OHSU School of Medicine and director of precision oncology for the Knight Cancer Institute in Portland, Oregon.
The data portal is designed, he said, to help cancer researchers to discover new biomarkers to track response to treatment, to unravel the mechanisms tumors use to resist treatment, and to identify new and more effective drug combinations to overcome resistance. “This is an outstanding example of team science and the importance and power of sharing data with the research community for improving patient outcomes,” he said.
To build the dataset, the team used large-scale perturbation experiments in which they expose tumor cells grown in a lab dish to a large variety of anti-cancer drugs and then track how each exposure changes the output of key proteins in the tumor cells over time. The changes in protein expression – up, down or neutral – reflect how cancer cells are rewiring their signaling pathways to survive and adapt to the stress of specific drug treatments.
“Through this dataset, one can immediately see the consequences of a given drug, including perturbed pathways and adaptive responses, which can help to identify optimal drug combinations,” senior co-author Han Liang, Ph.D., said in a news release. Liang is a professor of bioinformatics and computational biology at The University of Texas MD Anderson Cancer Center.
The experiments included more than 300 cancer cell lines, including cell lines grown from breast, ovarian, uterus, skin, blood, and prostate tumors. The researchers tracked expression changes in more than 200 proteins after treatment with 168 drugs, used alone or in combinations.
The researchers used the dataset to predict potentially useful drug combinations. The protein expression data made it possible to infer which signaling pathways were upregulated and downregulated in response to a given drug. Further analysis showed whether the pathways correlated with drug resistance. Then they identified which drugs were affected by the resistance pathways. Using this strategy, they identified 150 drug combinations. A search of published studies and clinical trials showed that more than half of the predicted combinations had supporting evidence.
“The data set has already provided support for many of the drug combinations that are being tested in SMMART clinical trials,” Mills said. SMMART (Serial Measurements of Molecular and Architectural Responses to Therapy) is a clinical trials program at the Knight Cancer Institute that’s made it possible to study each person’s tumor in great detail, track how cancer cells evolve in response to treatment, and use the information to select combinations of drugs tailored for the individual.
Large-scale Characterization of Drug Responses of Clinically Relevant Proteins in Cancer Cell Lines by Wei Zhao, Jun Li, Gordon B. Mills, Han Liang and others, Cancer Cell (Nov. 5, 2020)
Making cancer clinical trials SMMART by Joe Rojas-Burke, Cancer Translated (Nov. 16, 2018)