Research & Innovation

Stevens Researchers Harness Computing Power in Hunt for the Next Wonder Drugs

The human cost of disease is staggering: cancer, in aggregate, kills more than 8 million people worldwide each year (half a million in the United States alone). Anxiety has been estimated to trouble perhaps 550 million. And depression has been estimated to afflict more than 350 million globally.

In an effort to treat or ameliorate diseases and disorders such as these, the pharmaceutical industry has grown into a global powerhouse, producing more than $300 billion in revenues annually in the U.S. and nearly $1 trillion in revenues worldwide in a typical year.

Yet cures or proper management of many conditions remain surprisingly elusive or incomplete.

Now Stevens has entered this conversation with an innovative new technology that can save years — and millions of dollars — in the search for new cures, medications and therapies.

"We're actually playing on the same field as Big Pharma," points out Peter Tolias, director of Stevens' Center for Healthcare Innovation (CHI). "Only we're playing at a fraction of the cost. We can do almost exactly what they can do, in terms of drug discovery, but much more quickly.”

"When we find an opportunity that looks especially promising, we can move on it in days and obtain results within weeks. This is at least six to 24 months faster than large pharmaceutical companies can move on promising new medications, due to their size, internal approval paths, and endless discussions on strategy and marketing that slow the initial discovery research."

And that agility could lead to a major medical breakthrough one day on Castle Point.

Narrowing millions of chemicals to hundreds

To do it, Stevens has enlisted one of the leading thinkers in the field. In May of 2014, Stevens brought Dr. Sid Topiol and his longtime colleague Michael Sabio onboard to CHI's Biotechnology and Drug Discovery Laboratory.

Topiol was an early expert in the developing field of computational chemistry, one of the disciplines that helped spawn modern-day drug discovery; Sabio has been his collaborator for decades, heading breakthroughs at Sandoz and elsewhere. Working on room-sized, 250-kilobyte mainframe computers in the late 1970s, Topiol helped write and fine-tune some of the software packages that first analyzed chemical and biological reactions and interactions both for U.S. government laboratories and key university chemistry laboratories, including that of Nobel Prize winner John Pople at Carnegie Mellon University.

After subsequent stops at Mount Sinai School of Medicine and pharmaceutical heavyweights including Berlex (a subsidiary of Schering AG), Sandoz (now part of Novartis), where they unearthed potent new compounds to attack a variety of disease areas, and most recently at Lundbeck, where they focused on central nervous system diseases, Topiol and Sabio have brought their expertise and technical firepower to Stevens.

Deploying the same computational approach, and equipped with far more advanced hardware and software, the two are probing the structural biology of cancer, depression, anxiety and other afflictions for potentially effective therapies.

"It is as much an art as a science," explains Topiol. "The software is a tool to rapidly crunch the numbers and automate the process of matching compounds with target proteins. But with the recent advances in protein X-ray crystallography, we are able to take a new, more forward-thinking approach.

"Rather than simply designing small molecules to a given target site from known small molecules, which was the previous way of doing discovery, we can now look at the compounds in the target protein and their interactions."

Any and all disease areas are potential areas for inquiry at Stevens, says Topiol, although his team will focus on certain high-profile targets at first.

"We will be extremely strategic in the areas we choose to investigate," says Topiol. "We constantly stay connected to and current on the medical literature, and our priorities will also reflect those areas that seem most promising both for society and for potential commercial application and benefit. Our current oncology project is one promising example."

The screening process works like this. Utilizing leased high-power computational resources, Topiol and Sabio select disease 'targets' and fine-tune highly specialized (and extremely expensive) software to analyze millions of chemical compounds that might interact with those targets. The first rounds of investigation typically take two to three weeks of continuous computation; by the time the final results of those operations are ranked, however, Topiol and Sabio have screened out millions of candidate compounds and narrowed them down to the most promising 50,000 or fewer.

"This is before a single new compound has even been synthesized," Topiol points out. "We have calculated that this saves years in manpower and millions of dollars in research time and expense for essentially the same results."

Topiol and Sabio then tap into additional software packages, using long experience in the field to further refine their compound list based on the disease protein targets' unique geometrical features. The software can also suggest new lines of inquiry, such as analogous compounds and chemical classes.

"This simply wasn't possible, at all, in the past," notes Topiol. "The computations would have been too complex and time-consuming. This capability to search for analogues is new, and we are running with it."

Within another week, Topiol and Sabio may have narrowed their search down from 50,000 to just a few hundred which are tested in the lab and often yield five to ten chemical classes — at which time they can turn over the findings of their research to industry and academic partners with the laboratory scale and firepower to synthesize new compounds and apply them to disease sites. That's a first step toward possible animal and eventual human clinical trials.

In many cases, that step will also mean closely involving Stevens' team of a dozen drug discovery biologists and pharmaceutical industry veterans.

"They're our close collaborators, and we rely heavily on their partnership," says Topiol. "They do outstanding work."

A true breakthrough discovery isn't out of the realm of possibility.

"You begin with millions of candidates, but you get it down to hundreds of 'hits' relatively quickly. And Stevens is in the fairly unique position to being able to 'jump on' something that looks really promising right away," says Topiol. "If you're fortunate, one of these compounds then proves potent and effective."

If that should happen, it could mean eventually patented compounds, methods, and medications owned or licensed by the university — bringing both enormous societal benefit and huge commercial potential.

"That certainly could happen," says Tolias. "There is a long road from basic medical research to clinical trials and approval of a new medication, but it is absolutely not unrealistic to suggest that our team might one day discover compounds that are first to treat a given disease target."