Photo of scientists courtesy of the National Institutes of Health; photographer Rhoda Baer.
Compared with trying to discover and develop a novel compound from scratch, repurposing an approved drug saves time and money. It might be for a different indication or to rescue a candidate drug that may have failed in early trials and developing it to treat a different condition. Repurposing also carries less risk to companies and to patients, according to SLAS Drug Repurposing SIG Co-chairs Roger Bosse, Ph.D., senior global product leader for PerkinElmer, Life Sciences & Technology and Mathieu Arcand, Ph.D., co-founder of Biopra.
A recent Institute of Medicine (IOM) report confirms that pharmaceutical companies have become increasingly interested in finding new uses for existing drugs to cut the escalating costs of taking a new drug to market. According to the report, if a drug has already been approved or even if it has simply moved through Phase I or Phase II trials, data on safety, efficacy and toxicity have already been gathered, reducing the time (typically 10 years) and money (typically $1 billion or more) of a de novo investment. The report adds that today about 30 percent of repurposed drugs are approved, versus about 10 percent of new drug applications. This gives companies "a significant market-driven incentive for finding ways to repurpose existing drugs," according to the report.
The fact that an active drug "almost never has just a single activity" means a "back to the future" approach makes sense, says pharmaceutical consultant and SLAS member Christopher Lipinski, Ph.D. Most readers are familiar with the repurposing of the antihypertensive sildenafil as a treatment for erectile dysfunction and pulmonary hypertension, and the re-emergence of thalidomide, removed from the market after it was linked to fetal limb defects, as a treatment for leprosy and for multiple myeloma. In a presentation on repurposing several years ago, Lipinski pointed to other drugs under investigation for new indications, including the anti-cancer agent tamoxifen for bipolar disorder; the neuroleptic thioridazine for tuberculosis; and the antihistamine astemizole for malaria.
These and other successes have prompted companies to add repurposing projects to their research portfolios and the U.S. National Institutes of Health to add drug repurposing as a key function of its National Center for Advancing Translational Sciences (NCATS), according to NCATS director of Drug Development Partnership Programs Christine M. Colvis, Ph.D., "Never before has there been such a sense of urgency to rethink the way in which therapeutics are discovered and developed," write SLAS members Colvis and NCATS Director Christopher Austin, M.D., in the journal Neuropsychopharmacology. "Drug development has grown less efficient with every passing year, at the same time as fundamental understanding of human biology in health and disease has exploded...."
"Once you've gone even part way down the path of development," Colvis says, "you've invested so much that you really want to make sure no stone has been left unturned."
The discovery process for repurposing drugs has accelerated in recent years due to the emergence of new tools and technologies and new stores of publicly available data, according to the IOM. Authors of a recent review article on the topic summarize the main repurposing approaches and outline situations in which one may be more appropriate than others. They highlight the following key methods:
Blinded search or screening: Primarily utilize Food and Drug Administration off-label use and phenotypic screening to identify potential applications of an existing drug to other diseases.
Target-based: Use high-throughput screening/high-content screening to identify a protein or biomarker of interest, as well as in silico screening of drugs from compound libraries, to find new drug indications.
Knowledge-based: Apply bioinformatics/cheminformatics analyses of information from drug-target networks, chemical structures of drugs and targets, clinical trials and other input to drug-repurposing studies.
Signature-based: Use gene signatures derived from disease data to discover unknown targets or disease mechanisms.
Pathway- or network-based: Use disease "omics" data analyses to reconstruct disease-specific pathways for targeting by repositioned drugs.
Targeted mechanism-based: Integrate treatment "omics" data analyses with other sources of information to delineate unknown mechanisms of action of existing drugs.
Deciding which strategy to use depends on multiple factors, such as available data on the disease or drug and technologies available to mine that data effectively. However, regardless of the approach, it makes sense for companies to "use the knowledge they gained from their initial investment and try to repurpose it to a slightly different angle or disease area—particularly the compounds that failed in clinical trials. It's one of the easiest ways to recoup some of the money they've already spent," Bosse says. However, for repurposing efforts to succeed, "everyone in the organization must be committed, from the top down," adds Arcand. "If it's just something researchers do on the side, it will most likely fail."
The ability to form collaborative partnerships is vital to a successful repurposing program, according to the IOM, and NCAT's drug repurposing effort is an example of how such a partnership can bear fruit. For its 2012 pilot program, Colvis and her colleagues recruited Abbott, AstraZeneca, Bristol-Myers Squibb, Eli Lilly and Company, GlaxoSmithKline, Janssen Pharmaceutical Research & Development, Pfizer, and Sanofi, all of whom collectively agreed to make 58 compounds available to academic researchers.
"Each company had to offer up at least three assets—small molecules or biologics—that were essentially Phase II-ready. The idea was that the investigators could then propose a new indication and perhaps even jump right into a Phase II trial," Colvis explains. "The companies also gave us some information—mainly safety, tolerability and mechanism of action—about each asset so the investigators could make an intelligent guess as to whether or not the asset could be useful for the disease they were interested in treating." Companies also had to provide a supply of any product that might be needed for new clinical trials.
The researchers came back "with an incredible diversity of potential indications," Colvis says. "Sixteen of the 58 agents received at least five applications, and only one of those had the same indication five times. In every other instance, if there were five or more applications, three or more were for very different indications. Because we at NCATS didn't have a really good way of mining all the data on all the entities to come up with potential new indications ourselves, crowdsourcing in this way worked very well."
Could the collaboration be viewed as favorable to companies, since the goal was to identify new uses for their drugs or compounds? "It's really a win-win," Colvis says. "In the end, if the effort is successful, yes, a new indication might be something the company would be interested in. But then they would have to pursue it with the investigator. If it were a genuinely new use that the company hadn't thought of, then the investigator or his/her institution would have the use patent on it, while the company would have the composition of matter IP. So it really takes the two to tango.
"The flipside of the collaboration is that getting to test one of these assets puts an investigator in an advantageous position," Colvis continues. "For a university or the NIH to have helped a company get to the stage where a compound is being tested for a new indication is very significant. So the process genuinely is of benefit to both parties and ultimately, hopefully, to patients."
Pre-applications for NCATS' most recent round of drug-repurposing funding were due in July, and companies were notified earlier of the need for compounds. However, Colvis says there is an ongoing need for compounds for future initiatives, and her door is open to new ideas as well. For example, investigators have begun to request structures for virtual screening in addition to small molecules and biologics. Those interested in working with NCATS on a future initiative and/or offering new ideas can contact Colvis directly by calling 301.451.3903 or e-mailing email@example.com.
Although drug repurposing today is driven largely by the need for companies to maximize the return on investment in new clinical entities (NCEs), historically repurposing has served the more altruistic goal of identifying drugs to treat neglected diseases, Bosse says. "In the past, neglected and orphan diseases had low target value because of the relatively small numbers of people who could be treated with a new drug. However, now that university and industrial laboratories have joined forces, they are in a better position to take up the challenge of neglected diseases. This is a topic that was discussed in the SLAS2014 Drug Repurposing Special Interest Group meeting in San Diego, CA, and is likely to come up again at SLAS2015 in Washington, DC."
A study published recently in The Lancet Global Health affirms the importance of repurposing in this arena. Of the 850 new therapeutic products approved from 2000-2011, 37 (4%) were indicated for neglected diseases. Of those, 25 were products with a new indication or formulation, meaning they were repurposed. By contrast, only four NCEs were approved for neglected diseases, accounting for just 1 percent of the 336 NCEs approved during the period.
Bioinformatics-based approaches can be particularly helpful in identifying new indications for existing drugs that might be used for rare and neglected diseases, as well as identifying less expensive and safer drugs, according to Weida Tong, Ph.D., director of the Division of Bioinformatics and Biostatistics at the U.S. Food and Drug Administration's National Center for Toxicological Research, who did a presentation on the topic. Such approaches also could facilitate and support neglected diseases collaborations, according to a recent study. The FDA has established a Rare Disease Repurposing Database, and NCATS has a Therapeutics for Rare and Neglected Diseases program to help spur the development of new treatments, either through repurposing or by the discovery of new entities.
Similar initiatives also exist in Europe. For example, researchers from the Liverpool School of Tropical Medicine report on the challenges of drug discovery, including repurposing, for neglected diseases in the March 2014 issue of the Journal of Biomolecular Screening. The authors describe their experience with the A-WOL Consortium, a collaboration of academic and industrial partners funded by the Bill & Melinda Gates Foundation, whose aim is to develop new drugs to treat onchocerciasis and lymphatic filariasis.
In addition, a recent report by Minna Allarakhia, Ph.D., director of the nonprofit BioEndeavor, provides a roundup of open-source approaches for repurposing drugs for neglected as well as common diseases, and includes a number of collaborative efforts around the globe (also see Sidelines, at right).
Despite the success of some current repurposing efforts both for common and orphan/neglected diseases, challenges remain. "We need more access to compounds," says Bosse. "Although some companies recently have made a very impressive number available for repurposing activities, others need to get on board."
Intellectual property and other legal considerations also can pose barriers to repurposing, Bosse observes. "When your efforts are limited by IP restrictions, you're not putting the patient first; you're putting business first." Arcand agrees, noting that "the IP space is becoming more and more restrictive in terms of both chemical IP and disease IP. Tackling one disease with a specific family of chemicals might actually prove very difficult from both ends because of those IP restrictions." A recent publication from the Marquette Intellectual Property Law Review provides details on the current state of IP issues in repurposing.
As noted earlier, which repurposing strategy to use depends on multiple factors, including available data on a disease or drug. Yet even within a specific strategy—e.g., mining human genetics data for repurposing opportunities—disagreements can surface, Bosse says. He cites the example of an article on the use of genome-wide association studies that utilized data from the U.S. National Human Genome Research catalog as the source. That article "sparked a subsequent debate," he says. Other investigators suggested that a different database be used as a starting point; the authors of the original study countered by defending their choice while welcoming "continued discussion."
"Personally, I see a positive outcome from such scientific debates," Bosse says. "There may be more than one way to do things right, and discussions represent an opportunity to challenge some views and improve methods."
July 21, 2014