Innovative computational approach helps target new cancer diseases

Researchers at Weill Cornell Medicine used artificial intelligence to identify drug targets based on mapping regulatory networks in patients’ tumors. The study, published in Cellular systemsidentified and experimentally validated four drug candidates for neuroendocrine, liver and kidney cancers, which have a poor prognosis with current therapeutic options.

This research provides a much-needed new method for identifying novel therapeutic targets for many cancers. Although targeted therapy for some cancers has improved survival rates, treatment resistance and resulting disease progression are ongoing challenges. Additionally, many types of cancer have no known specific therapeutic targets.

Lead author Dr. Ekta Khurana, associate professor of physiology and biophysics and WorldQuant Foundation Scholar, led the work to map gene regulatory networks for tumor samples from 371 patients, including 22 cancer types, using a novel computational approach. Gene regulatory networks, models that describe the complex relationships between genes in a cell, are often altered in cancer.

Creating precise gene regulatory networks is no easy task. Researchers have incorporated data from tumor cells into messenger RNA, which is translated into proteins and chromatin accessibility, and can help reveal how DNA packaging and other factors affect gene expression.

The researchers developed a novel computational approach, called Cancer Regulatory Networks and Susceptibilities (CaRNetS), to discover key proteins that could serve as therapeutic targets for cancer treatment within gene regulatory networks. They identified known targets, such as BRAF in skin, CTNNB1 (B-Catenin) in colon, and ERBB2 (Her2) in lung cancers.

“With these known positive cases as benchmarks, we sought to validate the best candidates in cancers with limited effective targeted therapies,” the authors said.

The researchers then used their approach to identify key transcription factors and their interacting proteins, which may be vulnerable spots that can be targeted to stop or slow tumor growth. Transcription factors are proteins that bind to specific DNA sequences and regulate gene expression, turning their production on or off.

Using CaRNets on patient tumor samples, the researchers were able to group patients into 22 groups: nine corresponded to a single cancer type and 13 contained patients with multiple cancer types. Importantly, the approach revealed drug targets for all 22 groups. The researchers validated four of these candidate proteins in cells. They found that inhibiting the proteins they identified significantly affected growth in cell lines representing renal, liver, and neuroendocrine cancer types compared to controls.

The researchers believe that with the ease of measuring chromatin accessibility from patient tissues on a large scale, their computational approach will be widely used to find new treatment options for more types and subtypes of cancer.

Dr. Khurana is also a member of the Sandra and Edward Meyer Cancer Center, where she co-directs the Genetics and Epigenetics Program. The paper’s first authors are Dr. Andre Forbes and Dr. Duo Xu, who were working in Khurana’s lab at the time of this research.