Reflecting Back and Looking Forward: Key Themes and Potential Next Steps in Genetics-Enabled Drug Development
Over the course of the workshop, a number of themes emerged related to employing genetics in clinical drug development to advance precision medicine. The realization of precision medicine will likely take a long time, said Robert Plenge, but there may be ways to accelerate the process if genetic research is applied at the very early stages of the drug discovery process. “It is not just about finding those patient subsets where therapies work more effectively,” he said, “but it is also about narrowing the cycle time from therapeutic hypothesis to clinical proof of concept.” Currently drug developers face two major challenges—the rising costs of research and development and the decreased value provided to the overall health care market from new medicines (Yildirim et al., 2016). Overcoming these challenges will require more than just small changes, Plenge said; it will take a comprehensive overhaul of the drug discovery and development trajectory. In the final workshop session, Laura Nisenbaum, Russ Altman, Bray Patrick-Lake, and John Carulli, director of precision medicine at Biogen (at the time of the workshop), summarized the key points from each of the four workshop panel sessions. The highlights from the expert presentations and open discussions, as summarized by the session moderators, are presented in Box 6-1.
This final chapter of this proceedings synthesizes the workshop discussions and provides a forward-looking vision for genetics-enabled clinical drug development.1
1 The rapporteurs’ summary of main topics, recurring themes, and potential next steps is drawn from the presentations, discussions, and summary remarks by the moderators. Items in this chapter should not be construed as reflecting any consensus of the workshop participants or any endorsement by the National Academies of Sciences, Engineering, and Medicine.
Laura Nisenbaum remarked that one major discussion point that came through at the workshop was the need to engage with patients. Patient engagement benefits everyone in the drug development process, she said, because participants may have a better experience in trials, drug developers may generate medicines faster, and, consequently, both patients and developers can access resources that were previously unavailable to them.
Development of Patient-Friendly Resources and Educational Materials
Establishing new and improved educational resources, such as a user-friendly clinical trials database, could empower providers to identify genetic diseases, discuss clinical trials as treatment options, when necessary, and
assist in enrolling patients in trials. Furthermore, Altman asked, given that there could also be advantages to pre-symptomatic treatment (e.g., early treatment for SMA in infants), how could patients who are at risk and who may be pre-symptomatic both be made aware of and recruited to participate in a trial? This and other issues could be addressed collaboratively between trial sponsors and patient advocacy groups; accordingly, advocacy organizations could develop patient-friendly, graphical representations of complex topics or issues, such as genetic variation or benefit–risk assessment, that could be adapted to different diseases.
Solicitation and Incorporation of Patient Input
Patient engagement, David Leventhal said, affects all the main drivers of drug valuation, including cost, revenue, timing, risk, and intangibles.
Soliciting and incorporating patient input throughout the drug development process, including during clinical trial planning, could help ensure that the design, implementation, and follow-up of genetics-enabled clinical research and development is patient-centric. To accomplish this, researchers and trial sponsors could solicit patient input on where preclinical research efforts and clinical trial development could be best focused to address unmet needs. ENPV modeling and similar approaches can support trial sponsor decisions to increase patient engagement throughout the drug development process.
Return of Individual Genetic and Summary Results from Clinical Trials
Many clinical research participants are interested in receiving the results of genetic testing performed during the course of research, Jane Perlmutter said. Patients could benefit from having results returned in a user-friendly and timely manner, she said, and these data should reside with the patient so they can be accessed as patients move across treatment and research settings over the course of their lives. However, Rebecca Blanchard pointed out, researchers who employ genetic testing in their studies often lack the funding to develop data standards or translate results into lay-accessible summaries, and there is a lack of infrastructure to manage the return of genetic data to research participants. Furthermore, in those cases where patients do receive their genetic information, they may not be fully aware of the implications of these results and might benefit from genetic counseling to help understand them. The return of individual genetic and summary-level trial results—including the infrastructure needed to do so—is an issue that requires further discussion from a diverse group of stakeholders.
Representativeness of Genomic Studies
In 2009, 96 percent of individuals included in GWASs were of European descent (Bustamante et al., 2011). A similar analysis in 2016 found that the inclusion of non-European descent individuals in GWASs had increased to 19 percent. In both analyses, however, the vast majority of non-European ancestry individuals were of Asian ancestry (Popejoy and Fullerton, 2016). This can be problematic, because generalizations are made based on European populations and extrapolated to other populations, said Burchard. For example, he said, the current cystic fibrosis screening panel misses about 40 percent of Latinos with cystic fibrosis, and the cur-
rent genetic screening panel to test for hypertrophic cardiomyopathy was developed using data from white Americans (Manrai et al., 2016). This test erroneously identifies African Americans as being at risk when in fact they are not (patients who screened positive were followed and did not develop cardiomyopathy). Incorporating genetics research with data on socioeconomic status and environmental contributors to disease could help move precision medicine forward, Burchard said.
Biomarker Discovery and In Vitro Diagnostic Development
An overarching goal of precision medicine is to identify those patient subsets for which a therapeutic intervention will work more effectively than for the general population, Plenge said. When feasible, the greatest impact of precision medicine will be in leveraging genetics and genomics to identify biomarkers as early as possible in drug development, he continued. While there is more statistical power to detect a pharmacogenomic association in late-stage development or in retrospective analyses, there are also practical or technical difficulties to doing so, and it can be difficult to adjust a sponsor’s development strategy in late-stage clinical research. On the other hand, if a biomarker has been identified early on, research and development programs could be designed to maximize the value of this knowledge by collecting data to support the development and contemporaneous approval of the targeted therapeutic and the codeveloped IVD.
Realizing the goal of identifying biomarkers early in development will likely require a deeper understanding of causal biology and resulting insights into potential therapeutic targets, modalities, and approaches to detecting therapeutic hypotheses, Plenge said. This can be complicated, however, by the genetic complexity of diseases, including patient and locus heterogeneity within even single-gene genetic diseases. Lessons can be learned from gene families that have been associated with safety and efficacy outcomes in pharmacogenomic studies, such as variants implicated in absorption, distribution, metabolism, and excretion; disease mechanism; and drug targets. Researchers could devise innovative approaches to GWASs or trial design that leverage this evidence and target these issues.
Accrual and Retention Strategies
To address the complexity of genetics-enabled clinical trials and the historical difficulties faced in clinical trial accrual and retention, trial sponsors are developing innovative methods for enhancing trial recruitment,
enrollment, and patient engagement. In concert with validated biomarkers and IVD tests, these methods aid in enriching clinical trial populations. Langbaum said that direct outreach to patients and their communities greatly bolstered enrollment and kept patients prepared to participate in Gene Match and in the autosomal-dominant Alzheimer’s disease registry. Strong said that employing many different types of communications and formats to accommodate different learning preferences can help to engage families, and she highlighted the success seen by the Foundation for Prader-Willi Research in using webinars and videos directed toward differing levels education and health literacy. She also encouraged personal contact when possible to more specifically address patients’ questions. Perlmutter noted that the Precision Medicine Initiative’s (PMI’s) All of Us Research Program aims to enroll at least 1 million participants, which will be aided by effective patient engagement and significant upfront investment. Flexibility in trial sites, as exemplified by recruiting globally and setting up “just in time” trial sites, can also aid in improving both patient access and trial accrual, Hornby said. Patient registries can also be leveraged as pre-enrollment, trial-independent recruitment mechanisms so that individuals not eligible for one trial can be re-contacted at a future date for others.
To expedite trial enrollment while also providing sponsors with the ability to more facilely conduct future research, a dual-consenting process can be used for both planned exploratory pharmacogenomics objects in the current trial and also a separate, voluntary consent for future biomedical research (subject to country-specific limitations on the legality of genetics research). Engaging with health systems can also help facilitate the enrollment of diverse populations and allow for patient populations to be followed longitudinally, Williams said.
Novel Clinical Trial Designs to Maximize Evidence Generation
The traditional approach for clinical trials—studying one disease in one population using one treatment—is neither amenable to nor sustainable for genetics-enabled drug development, said Brooks. Accordingly, sponsors have begun implementing novel designs for genetically enabled clinical trials, such as umbrella and basket trials (e.g., Lung-MAP and STARTRK-2, respectively). Grouping patients with rare diseases into clinical trials based on shared molecular etiology (e.g., protein misfolding) is another approach that could make the drug development process more efficient, Brooks said. Innovations such as master protocols, central IRBs, standard informed consent, and “just in time” site activation can be elements that help to deliver economies of scale. N-of-1 precision trials and adaptive designs, such as multiple blinded cross-over studies, may also be warranted for rare genetic diseases in which there is considerable patient heterogeneity,
variable natural histories, wide variation in treatment efficacy, and too few comparable patients to pool for enrollment.
Though umbrella, basket, and SME trial designs have shown promise for more efficient and effective drug development, there can also be challenges to the implementation of each for which further innovative solutions may be needed. For example, as compared to oncology basket trials, which can often rely on a common IVD, disease-specific assays are needed to identify a predictive biomarker in patient cells for SME trials, Brooks noted. To reemphasize a common thread of the workshop, collaboration between stakeholders and the pooling of resources and expertise has been and will continue to be critical to the success of these trials.
The complex challenges associated with genetics-enabled drug development cannot be addressed by a single entity or organization, Nisenbaum said. There is a need for more interaction between biopharmaceutical companies, which often struggle with the economic and business cases for integrating genomics into drug development, and the foundations and patient advocacy groups that are gathering patients, genotyping them, and channeling them into clinical trials, a workshop participant said. There is a natural association to be made between the registry groups and drug development organizations to merge resources in a way that perhaps has not been done before.
Establishment of public–private partnerships and innovative business models that leverage the diverse but complementary contributions—such as expertise, experience, infrastructure, and funding—of these entities could further advance the progress made in discovery and development efforts for genetics-enabled therapies. Furthermore, sponsors could also benefit from interacting and working collaboratively with regulatory agencies as early as possible in drug development.
There is a need to share and collate data more effectively across rare diseases, complex diseases, and other siloed conditions, Patrick-Lake said. Real or perceived hesitation or uncertainty on the part of patients should not be a barrier to data sharing, she said, and researchers should be willing to take this risk if it will benefit the patient. Agreeing, Blanchard said that another challenge is how to operationalize data sharing in the competitive research environment. There are no centralized patient registries and no
single health care system from which to aggregate the data. However, there is interest, including from the biopharmaceutical industry, in participating in an integrated, pre-competitive approach to conducting genetic research in large numbers of people, she said.2 Chandler emphasized the importance of multifaceted partnerships among academia, industry, vendors, patient advocacy groups, and health authorities to enable a common, pre-competitive approach.
Studying why some experimental drugs fail in clinical trials could provide an avenue for better understanding of causal biology, a workshop participant suggested. Some approved drugs that fall out of favor for a particular condition may have value for other conditions, Falk added. The question is how to gain access to these products and information about them. Access to drug libraries of existing molecules, for which there is already some knowledge of the risks and benefits will be very important to facilitating the discovery of treatments for rare diseases in a timely manner, said Falk.
Integrating Genetics-Enabled Research into Existing Drug Development Programs
Integrating genetics-enabled development strategies may be challenging for biopharmaceutical companies with established drug development programs, Falk said. To address this issue, Leventhal said, drug development programs, such as those for rare diseases, could be designed to be nimble, and apparatuses for genetics-enabled drug development could be integrated with legacy operations already in place to more seamlessly achieve this goal. If not, it is possible that small, genetics-enabled studies could become crushed under the infrastructure and organizational complexity inherent to larger development programs. Adaptation of this nature takes time, Leventhal said, and initiatives like TransCelerate,3 a cross-industry collaborative, and CTTI are working to address these types of challenges. New and established innovative trial models might also help to lessen the risk of participation in precision medicine by companies, as the infrastructure for engaging in these models is already in place, Patrick-Lake said.
2 Workshop participants were referred to the proceedings of a prior Roundtable on Genomics and Precision Health workshop that discussed precompetitive collaboration for genomics-based discovery in drug development (IOM, 2011).
Burchard closed the workshop with his perspective on the need for “socially precise” precision medicine. Genetic diversity is an important consideration when recruiting for clinical trials and can both increase the ability to generalize results to the real world and enhance opportunities for new discoveries. Given that there is geographic, ethnic, and racial variability in the prevalence of genetic variants, trials could be designed to better reflect the known distribution of variants based on these factors, he said. Furthermore, to address this issue, patients could be recruited from health systems that serve diverse populations and by launching trial sites that in urban areas. This strategy might encourage participation from a more diverse patient population, and researchers could more extensively leverage real-world information in EHRs. Indeed, the PMI’s All of Us Research Program will employ a multidisciplinary approach, noted Burchard, that will include genetic testing for potential use in genetics-enabled studies or to increase the genetic diversity of clinical trial populations.
One of the most exciting things about this workshop, Plenge concluded, has been that almost every speaker has put the patient at the front and center. This is incredibly important, he said, because at the end of the day the goal is to figure out how to deliver new therapies that will improve the lives of patients.
This page intentionally left blank.