6

Perspectives on Next Steps

The end of the workshop provided an opportunity for session chairs to engage participants in discussing opportunities for improving and accelerating drug discovery as discussed in their respective sessions (see Box 6-1)¹; session chairs’ comments are summarized in this chapter.

Current Therapeutic Development Practices

David Michelson, vice president of clinical neuroscience and ophthalmology at Merck Research Laboratories, observed several key themes throughout the first session and identified next steps to aid in drug discovery. Michelson said several participants indicated that companies do not have confidence in choosing nervous system disorder targets and as a result choose to work in areas where the science is further along. Many times, this leads to a restriction of the focus of research to the disease areas that are best understood, which often are not within the neurosciences. To build confidence in the next stage of research and the investments made, many participants who spoke agreed there is a need to de-risk across steps in the development time line. Some researchers continue to prematurely proceed to clinical trials, without sufficient preclini-

______________

¹The topics highlighted in this chapter are based on the summary remarks made by each session chair during the final workshop session and at the end of his or her respective session. Additional comments by participants related to the closing remarks are also included. As noted in Chapter 1, comments included here should not be construed as reflecting any group consensus or endorsement by the Institute of Medicine or the Forum on Neuroscience and Nervous System Disorders.

cal data, resulting in high clinical failure rates (Leaf, 2013). When companies are at the point of making larger investments and commitments, the probability of success should high, said Michelson.

Several participants noted that the use of multiple models may be beneficial to discover and validate targets; there is no simple “yes” or “no” as to whether a specific animal model will be useful, but little is gained by continuing to pursue a target when data suggest it may be futile. Increasing reagents for novel targets in laboratories and establishing better biomarkers by considering the patient group in which the mechanism of the disease is likely to be homogeneous (patient stratification) could also help de-risk research programs. A few participants reiterated

the need for public–private partnerships to de-risk investments as well as pushing the precompetitive boundaries to Phase IIa by evaluating novel targets in patients. Another concept mentioned by several participants is that large pharmaceutical companies are not the only places where drug discovery and development can occur. Biotechnology companies and federal agencies, for example, might also be resources to consider.

Regarding next steps, Michelson posed the following three questions:

1.   What areas are the ripest for work?

2.   What are the gaps, and how can they be addressed?

3.   What are the steps in the drug discovery process that can be derisked and provide confidence across the development pipeline?

Michelson noted that the focus should not be to accelerate the drug discovery process, but rather to make it better and more efficient.

The Regulatory Pathway

William Potter, senior advisor in the Office of the Director of the National Institute of Mental Health, highlighted how many basic scientists are unfamiliar with Food and Drug Administration approval procedures and resources at NCATS that may potentially resolve these issues. FDA, in its current case-by-case approach, views flexibility as a benefit and helps academic researchers better understand the guidance documents that may alleviate common mistakes and misconceptions. Echoing comments made by several participants, Potter asked, Where are the opportunities for increasing the probability of success? This is an area in which the neuroscience field has been struggling, he noted. Beyond these concerns, Potter stated, the fundamental question is, How can we create a better fit between preclinical assays and experimental human models? Is the field at a point where it can depend on one method, and can this be tested?

New and Emerging Tools and Technologies

There was an overall sense at that workshop that the cell is the smallest common denominator, as opposed to the gene or protein, said

session chair Rajesh Ranganathan, director of the Office of Translational Research at NINDS. Several participants noted that human phenotypes may better inform the translational space than animal models, and, if possible, measuring the same variables in humans and animals might be beneficial. Many participants who spoke agreed that creating a database, such as “www.preclinicaltrials.gov,” as a repository to share data could further the understanding of nervous system disorders and provide a mechanism for researchers to share discoveries.

Ranganathan reflected on the Human Genome Project and asked whether there is a fundamental technology that could change the neuroscience space, and if so, should investment be increased in this specific area? Specifically, for neuroscience, said Ranganathan, target engagement is critical. Ensuring that there is target engagement, taking measurements that are pharmacodynamically related to the target, and testing where it truly matters are important, he added. Developing a better learning curve in clinical trials is needed to eliminate the common responses of “it did not work and we do not know why” or “it’s the patient, not the drug.” Ranganathan concluded by saying that the field should push to race ahead with a broad-based platform of investment to facilitate drug discovery.

Opportunities in Nervous System Disorders

Magali Haas, chief science and technology officer at One Mind for Research, observed that in the course of the workshop, the focus had been on the ability to predict outcomes and determine which steps are or are not informative in the development pipeline, rather than moving toward first-in-human trials from cellular models. During the session, participants discussed an array of topics, including the use of human data for target identification; the willingness of researchers to conduct rigorous experiments both in preclinical models and in an experimental medicine setting; and opportunities to accelerate the pathway to first-in-human trials when the severity of the disease warrants it. However, participants mainly focused their discussions on the investment portfolio in the neuroscience field and where there are opportunities for nonprofits, NIH, and pharmaceutical companies to reconsider their portfolios. There is a large investment in iPSC lines because of the potential to conduct functional pharmacology from a cell derived from a patient or normal individual. Although there was not much discussion about where the next

generation of genetic data, biomarkers, or phenotypes will fit into large-scale human trials, Haas noted that many participants and speakers agreed that this will happen through large-scale public–private partnerships. Several participants advocated for information technology platforms for sharing reliable data, such as a neuroscience community portal (which is currently under development), “neuromine,” or “www.preclinicaltrials.gov” as suggested by a workshop speaker. In addition, Haas noted that there is a strong need to consider the use of new tools such as proteomics, transcriptomics, and computational modeling approaches.

Optimizing Therapeutic Development

Collaboration and the precompetitive space were two key themes in the workshop discussions, noted Daniel Burch, vice president and global therapeutic area head for central nervous system diseases at Pharmaceutical Product Development, Inc. (PPDi). Several participants identified a number of potential areas for collaboration, including target identification and validation, animal assays, and de-risking aspects of research. In addition, many participants highlighted the importance of extending the precompetitive space and establishing small-scale collaborative models. Burch stated that there is inefficiency within the translational space; researchers and groups are reinventing the wheel when it is unnecessary. The development of platforms to explore certain mechanisms might be beneficial, he added.

SUMMARY

Given the challenges surrounding current therapeutic development practices for nervous system disorders, this workshop sought to explore opportunities that could potentially make the pathway from discovery to approval more effective and efficient. Rather than focusing on accelerating the process from cellular models to first-in-human trials, workshop discussion focused on predicting outcomes and determining next steps in the development pipeline. Speakers and participants highlighted several limitations related to de-risking research, target validation and engagement, the regulatory process, inefficient research, and reproducibility. Although clinical outcomes cannot be predicted and additional research is needed, new and emerging tools and technologies, such as iPSCs, are

avenues that could potentially improve the drug development process, according to several participants. Patient stratification, target identification based on human data, systematic evaluations of failed clinical trials, and improved guidance on regulatory issues were just a few potential solutions that participants considered. Finally, several speakers and participants advocated for open data sharing among researchers and a centralized database for all preclinical trials to increase collaborative efforts and decrease replication in the field.