Skip to main content
Premium Trial:

Request an Annual Quote

NCI Taps GNS Healthcare to Build In Silico Models of Non-Small Cell Lung Cancer

Premium

By Uduak Grace Thomas

GNS Healthcare said this week that the National Cancer Institute will use its reverse engineering and forward simulation, or REFS, platform to build hypothesis-free computer models of non-small cell lung cancer based on datasets from imaging, genetics, pathology, and other sources.

Specifically, GNS will analyze data from the laboratory of Terry van Dyke, director of NCI's Center for Advanced Preclinical Research. The firm will use the REFS platform to create in silico network models that will be used in simulations that connect drug doses to transcriptional and imaging measurement networks.

The data were generated from genetically modified mouse models of non-small cell lung cancer and include transcriptomic and MRI data relating to NSCLC induction, regression, and combination drug treatments.

The results from these simulations are expected to provide insights into the mechanisms of NSCLC and its response to drug treatments. The ultimate goal is to match patients and treatments better as well as to develop more effective therapies for the disease, GNS said.

Initially, the partners will work on developing data-exchange standards that will allow future collaboration in other mouse model systems. The partners then plan to create combined experimental and computational workflows aimed at enabling hypothesis generation, testing hypotheses both in silico and in vivo, generating confirmatory data, and then performing additional modeling, GNS said.

Iya Khalil, executive vice president and co-founder of GNS, told BioInform that the firm will work with the NCI team to discover key markers of drug response in transcriptomic and imaging data with plans to expand the approach to other datasets and eventually other projects.

“We are going to predict how [a] gene or combinations of genes can tell you if a patient will or will not respond” to a particular treatment with the ultimate aim of using the findings to tailor treatments to patients genetic code in clinical settings, she said.

Although she could not provide specific details about the treatments that will be tested in the models, Khalil did say that the partners are looking at drugs that are currently on the market as well as some that are still in development.

In addition, she said the REFS platform will be used to generate hypotheses that explore the impact of these drugs on cancer individually as well as in combination.

REFS "can handle large-scale data sets as they are and at the scale they’re generated,” Khalil said, noting that these datasets are comprised of “tens of thousands of molecular transcripts and changes that are being measured.”

She added that the GNS platform improves on standard statistical approaches that are currently employed in most research settings.

The REFS platform creates network modes that explain the “cause and effect relationship in a system” as well as answer questions like “what nodes are the drugs hitting first and how does that affect the downstream pathways and networks,” she said.

The agreement marks GNS’ second partnership with the NCI this year. In January, the firm said that it had signed a subcontract with NCI contractor SAIC Frederick to create in silico simulations of drug interaction data generated by applying several well-known cancer treatments to the institute’s NCI-60 cancer cell line panel (BI 1/28/2011).

While there isn’t an established timeframe for the NSCLC project, Khalil said the initial phase is expected to last for a few months.


Have topics you'd like to see covered in BioInform? Contact the editor at uthomas [at] genomeweb [.] com.