How big data is reshaping the drug industry

By Gail Dutton, Contributor

Big data is transforming the way drugs are being developed and prescribed. Among the innovations: drugs that already have been approved are reformulated or repurposed for other ailments.

Elysium Therapeutics, for instance, is using this method to combat “prescription drug abuse in areas scientists either overlooked or considered impossible,” says CEO Greg Sturmer.

The company is experimenting with redesigning opioids, which lose about 1 percent of their potency each year, to make them inactive soon after the prescription expires.

Historically, drug development has been relatively isolated, with little information sharing among researchers. Now, however, scientists have the ability to tap information from public and controlled-access databases, mine real-world data from insurance claims and crowdsource additional information.

Consequently, much of the research work shifts from data generation to data analysis.

“Typically, companies developing new chemical entities need 10 to 12 years to bring a drug to market,” says Sturmer. “We expect to get multiple drugs to the market within five years, and at much lower costs.”

That’s possible because Elysium leverages the wealth of human studies already performed to determine the ideal chemical composition, as well as safety and dosage.

“We’ll have to prove our formulation gets the same bioequivalence as the other drugs,” Sturmer says.

New uses for old drugs

NuMedii is mining disparate databases to identify existing drugs that can be repurposed or even refined to be safer and more effective.

NuMedii’s big data approach already has predicted the efficacy of drug repurposing for treating colitis a form of inflammatory bowel disease small-cell lung cancer and other conditions, according to Scott Saywell, vice president, corporate development.

One recent finding, for instance, suggests the epilepsy drug topiramate also may be effective in treating inflammatory bowel disease. The company is discussing trials with Stanford University to test that hypothesis.

If those trials are promising, NuMedii will then assess the drug’s commercial viability based upon patents, safety, drug behavior in the body (pharmacokinetics), the potential for reformulation, and other factors.

Finding the causes of disease

Big pharma is embracing big data, too. GlaxoSmithKline teamed with analytics powerhouse SAS last summer to develop a secure website where drug developers can share not just pre-competitive data but patient-level clinical trial data on a controlled-access basis.

Consequently, companies are gaining access to previously proprietary data. This enables drugs and diseases to be matched based upon their gene signatures.

“Big data is getting individual information about patients, the general population and trends that are influencing health and well-being on an unprecedented scale,” notes Andy Last, COO of Affymetrix. His company recently won a tender from the UK Biobank to genotype the biological samples from 500,000 individuals. The genotypes will be correlated with de-identified medical records and other data.

“This means researchers will have a phenomenally large set of highly detailed data that includes whole genome analysis and the health conditions of a population historically and going forward,” Last says.

Previous studies typically have involved only a few hundred to a few thousand people, which delivered limited insights. “Genotyping such a large data set, such as in a biobank, holds the promise of identifying the variables that actually influence disease and health conditions,” Last says.

More effective drugs result

For patients, the use of big data analytics in drug development results in less trial and error when physicians prescribe drugs. This tighter targeting of drugs to disease also results in fewer side effects.

For drug developers, big data has multiple benefits. Early in the discovery cycle, it helps link diseases to specific causes for specific populations of people. It also helps leverage existing data from multiple databases to make development more efficient and faster.

Under traditional drug development, a company can screen 5,000 to 10,000 compounds, with 250 advancing to preclinical testing, five advancing to clinical trials and only one being possibly approved, according to the Pharmaceutical Research and Manufacturers of America, or PhRMA.

Last year, American companies had 3,400 drugs in development, with theFDA approving only 27 new molecular entities, PhRMA says.

“Traditional drug discovery and development isn’t sustainable,” Saywell says. “That’s why we’re integrating disparate data, technologies and knowledge bases into every aspect of drug discovery and development.”

About the Author: Power More