Drug Repurposing and Repositioning: Workshop Summary (2014)

6

Increasing the Efficiency and Success of Repurposing

Repurposing and repositioning have gained considerable attention from a range of stakeholders, including NIH, academic institutions, pharmaceutical companies, and even some technology companies, for its potential to improve human health, particularly in rare diseases, said Aidan Power, vice president and head of PharmaTx Precision Medicine at Pfizer. When these stakeholders have joined together in collaborations, they have achieved notable successes in finding new and sometimes unexpected uses for existing drugs, he said. (The suggested best practices and opportunities that individual speakers offered concerning drug repurposing are found in Box 6-1.) Indeed, given the degree of success experienced with repositioning for rare diseases, global strategies could be instituted to pursue this approach, Power said.

Initial drug discovery and repurposing are not very different from each other in the sense that they share some of the same challenges encountered during the drug development process, Cardon said. When thinking about drug development, the initial approach may be to focus on the drug itself, but with drug repurposing in rare diseases the thought process is different and the approach typically begins with a focus on the disease pathology, and clinical insights remain critical, Cardon said.

THE ROLE OF GENOMICS

An integrated approach to drug repurposing that is not focused specifically on genomics or other omics-based research will be the most successful approach for finding new indications, Cardon said. But genomics and other new technologies add a dimension to repurposing that could greatly speed progress. With diseases such as cancer, for example, immense quantities of data are being generated that can foster innovation, both in the kinds of compounds that are being developed and in the specific treatments used for molecularly profiled diseases.

Genomics can be used throughout the repurposing process to identify therapeutic approaches and disease pathways, said Ronald Bartek, president of Friedreich’s Ataxia Research Alliance. The technology is a tool for developing an understanding of mechanisms and for gaining clues as to why some patients respond to drugs during clinical trials while others do not, said Allen Roses, the Jefferson–Pilot Professor of Neurobiology and Genetics at Duke University. Dietz agreed, adding that in the case of failed trials, genomics may enable the identification of the patient populations in which a drug is effective.

Collecting Other Types of Data

Geoffrey Ginsburg, director of the Center for Genomic Medicine at Duke University’s Institute for Genomic Sciences, asked whether it makes sense to monitor the entirety of the transcriptome, metabolome, proteome, immunome, and anything else that could be informative about the diversity of actions of a drug. The answer to that question involves looking at relative costs and benefits, Frail said. Even analyzing just the metabolome is quite expensive, and the returns from such an investment are uncertain. Genetic studies, whether of the genome or of the transcriptome, are more cost-effective, but the cost of analyzing the data still can be prohibitive. The difficulty arises from the need to screen large numbers of compounds in individual patients, although such screens may be possible with a more limited number of compounds, McKew said.

While biological understanding can drive drug development and the choice of treatments, clinical data also can inform research activities, observed Ringel. The most promising approach, individual participants observed, is an integrated one that combines genomics, clinical observations, basic research, animal models, and clinical trials. However, Power reminded the group, all approaches need to culminate in human experiments if drugs are to be used in patients.

USE OF ELECTRONIC MEDICAL RECORDS

Early experiences have shown that EMRs can be used for research, Ginsburg said, even though they have not been designed for that purpose. EMRs could be especially valuable if the information they contain were coordinated so that signals from the data more clearly emerged from the noise, Ringel added.

However, the use of EMRs is still in its infancy, Cardon said. Much depends on the quality and quantity of the clinical information available. As Kaufmann pointed out, EMRs have been constructed more for compliance and billing purposes than for research, and the information they contain often is proprietary. Where EMRs could be helpful in working with rare diseases would be identifying patients or linking clinicians with trial opportunities, she said. However, with so few patients the data would probably not be good enough to generate useful results.

The current weaknesses of EMRs provide an opportunity to think about which stakeholders should be at the table in deciding about their

future evolution, Kaufmann continued. If more research capabilities were built into these systems, they could prove to be valuable. Ginsburg also pointed out that patient-reported datasets are becoming more common, and it may be possible to derive information from such datasets that is not available from EMRs.

EXISTING BARRIERS

Repurposing and repositioning are still in their early stages, Ginsburg said. Many strategies are being pursued, none of which has been described as ideal. Furthermore, the stakeholders involved in repurposing and repositioning have different interests, and no obvious way yet exists to harmonize these interests.

A number of barriers to repurposing exist. One is that companies may decline to release information out of a fear that a blockbuster drug will be tarnished if a repurposing program uncovers safety or efficacy issues, said Michael Pacanowski, acting associate director for genomics of the Center for Drug Evaluation and Research at FDA. For rare or severe diseases, in contrast, the bar may not be set as high with regard to safety; thus, in this regard repurposing is less risky for these indications, although such drugs also produce lower returns on investment because of their limited patient populations. However, Taylor noted that companies do not measure returns on investment just in terms of dollars. An organization may be willing to accept financial risks if the patients it represents have a personal stake in the undertaking.

Pharmaceutical companies continue to have problems with intellectual property protections and with the costs of transactions for licensing, Rai said. Other industries share information without the formality seen in drug development, but this is because clinical trials are so expensive. An alternative would be to publicly fund trials, but this is not likely to happen, she said. Ringel pointed out that there have been success stories of companies transferring rights among themselves, but these success stories are rare. Sometimes, when another company expresses interest in a compound, the company owning the rights to the drug will renew its own internal effort. In addition, each company has its own disease area strategy, and it can be hard to get traction within the company if an idea is outside that area.

Difficulties can occur in gaining timely IRB approval for fast-turnaround studies using social media, Kaufmann said. People who self-

enlist in such studies are information altruists who want their data to be used, while the medical system has the goal of protecting their data. Centralized IRBs that see more and larger projects may be more forthcoming with approvals.

POSSIBLE NEXT STEPS

Collaborations, such as consortia, represent an opportunity to create a clearinghouse or other means of information exchange and dissemination designed to expedite drug repurposing programs, Bartek said. A venue where ideas, resources, and knowledge can be freely shared without excessive procedural and legal formality could maximize the number of drugs being tested for different drug targets, Dietz said.

Technology transfer officers have different levels of expertise in negotiating complex arrangements with private companies and government, and they are often overburdened at their jobs, Rai said. An alternative model that has been discussed is to centralize the technology transfer process so that a smaller number of technology transfer officers with expertise in particular areas represent a number of universities. In such a case, technology transfer officers could be experts not only in negotiating contracts but in the technology covered by the contract.

Frail emphasized how much can be learned from negative, or less interesting, findings. Journals are more receptive to publishing such studies now than they have been in the past, and clinical trial results are now more transparent. Colvis observed that negative data are really just data, especially in the context of repurposing, where information about what does not work and why can be extremely valuable. Bartek noted that it would be helpful to know whether the drug itself or the clinical trial failed. Colvis pointed out, though, that data of value can sometimes be difficult to share.

Though NCATS and the MRC have begun demonstrating what is possible, many other government agencies could do more to promote repurposing and repositioning, Bartek said. First, NCATS is not the only supporter of translational science at NIH. All the other institutes have their own translational and clinical programs that could become engaged with drug repurposing, and these programs could be especially productive if they were coordinated within NIH to accelerate translational science. NIH could potentially play a role as a clearinghouse with academia for identifying points of contact for learning more about how drug repur-

posing technologies are being used for specific diseases, Bartek said.

Pacanowski observed that FDA has large quantities of data that could be extremely useful if they were integrated and synthesized. In addition, the further development of regulatory science will allow science to inform the regulatory process, observed Tong. Government agencies can coordinate efforts to address issues related to emerging technologies, as with quality control of microarray applications. CMS and other public and private healthcare payers also could advance repositioning, particularly through collaborations with other stakeholders in the drug development ecosystem, Tong suggested.

Patient groups have an important role to play, said a workshop participant. Although patient groups do not receive a funding stream in the current drug development system, they can quickly put together cohorts and provide samples that are matched to genotypes, phenotypes, and available compounds. Kaufmann added that patient data will be subject to some selection bias in terms of who decides to volunteer information, but the opportunity is sufficiently great that this limitation needs to be accommodated. Researchers will need to work with patient organizations to make sure that patients feel comfortable and that privacy concerns are adequately understood. Data quality is also a consideration, but simple measures, such as an app that asks patients to report on their pain level, can offer good ways of collecting information, even if not all data will be useful in, for example, securing a label change from FDA. Kaufmann pointed out that common data elements and outcome analyses would facilitate future meta-analyses. Even better would be coordinating studies from the beginning rather than doing studies separately.

As an example of what patient groups can do, Frail cited the Poly-cystic Kidney Disease Foundation, which set up a small scientific advisory board that could provide guidance on compounds and animal models to test. Similarly, the Michael J. Fox Foundation has supported drug repositioning for Parkinson’s disease. In that case, the foundation brought together industry and academic partners to pursue promising leads. However, such collaborations need to harmonize with the business model of a company.

FUTURE REPURPOSING EFFORTS

Drug developers and their partners in government and in the patient community should continue to pursue low-hanging fruit, especially with

rare diseases, Bartek said. New technologies can lead to a better understanding of disease mechanisms and molecular pathways and can continue to create new opportunities to repurpose and reposition drugs for alternative targets. Even after a high-throughput screen has detected hits for diseases such as Friedreich’s ataxia, new assays or better cell models can create new possibilities, he said.

An emphasis should be placed on unmet medical needs as opposed to searching for convenient discoveries, Pacanowski said, and investments should be made where they will have the biggest impact on public health instead. Colvis urged that repositioning programs be broadened to include active as well as deprioritized compounds. Preventive disease strategies and therapeutics should also be included when thinking about drug repurposing, a workshop participant added.

Finally, individual workshop participants who spoke stated that the ultimate measure of success is improved health. As Ringel put it, by creating value for patients, repositioning can change people’s lives.