2020 NAMHC Workgroup on Drug Development

Table of Contents

1. Introduction

2. Workgroup Process

3. History of NIMH Support of Drug Development

4. Target Engagement in Experimental Medicine Research

5. Fast-Fail and Rapid Testing of Mechanisms

6. Vulnerable Clinical Trial Populations

7. Confirmatory Efficacy Trial Considerations

8. Changes in Workforce Development

9. Conclusions

10. References

I. Introduction

In 2010, a National Advisory Mental Health Council (NAMHC) Workgroup developed a report on strategies for the development of therapeutic interventions for mental illnesses¹. The report highlighted that while the prevalence of mental illnesses was widespread, the availability of effective therapies was limited. It stated that approximately 1 in 4 U.S. adults lives with a diagnosable mental disorder in a given year², and mental illnesses were the leading cause of disability in those aged 15 to 44 in the United States and Canada³. Nevertheless, available medications to treat these disorders, though widely prescribed, did not sufficiently reduce morbidity and mortality – supported by data from previous National Institute of Mental Health (NIMH) funded pragmatic trials, such as the Clinical Antipsychotic Trials of Interventions Effectiveness (CATIE)⁴ and the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study^{5, 6} – and the mechanisms by which these medications act were not completely understood. Additionally, in the face of increasing development costs and timelines, and high attrition rates due to insufficient efficacy of candidate molecules, particularly in psychiatric drug development, the pharmaceutical industry had been moving resources away from work in this area – and so the pipeline for new medications was lacking⁷.

Over the last decade, NIMH aimed to improve the drug development pipeline by shifting its clinical development strategy from conducting large trials, that promised only incremental benefit in terms of establishing efficacy, to use of an experimental therapeutics approach. In this approach, smaller studies are conducted in the Phase Ib to IIa space that utilize drugs as probes to identify or verify that the drug accessing the intended central nervous system (CNS) target (molecular entity or pathway or process). Specifically, such studies are designed to demonstrate objectively that at doses tested, investigational compounds sufficiently engage CNS targets and thereby have the potential to positively impact a functional outcome. The hope was that this approach would allow for: (1) confirmation of CNS engagement; (2) collection of early pharmacologic evidence of mechanism of action; (3) optimization of drug exposure for efficacy and safety; and (4) efficient utilization of resources to identify a greater number of compounds that might be good candidates for further development, as smaller studies could be conducted more quickly and allow for more rapid results.

In 2020, the NIMH Director, Dr. Joshua Gordon, called for the formation of a Drug Development Workgroup of the NAMHC and tasked the Workgroup with assessing the state of clinical trial research within the Institute in order to ensure that current strategies support effective exploration and expansion of early-stage drug development efforts. The workgroup discussed current progress on efforts around this quantitative pharmacologic trial design approach, whether there should be specific circumstances in which the Institute might expand its focus to include conducting confirmatory efficacy trials for medications, as well as how the Institute should handle the study of vulnerable clinical trial populations and the mentoring of the next generation of clinical trialists in psychopharmacology. This report summarizes the Workgroup’s discussion along with its recommendations for how the Institute can best move forward.

[Top]

II. Workgroup Process

The Workgroup met multiple times throughout 2020 to discuss the background, to hear presentations from NIMH staff on current programs, and to discuss the issues related to the Workgroup charge. Some of the discussion occurred in separate breakout groups to permit more in-depth discussion of:

1. Target Engagement
2. Fast-Fail and Rapid Testing of Mechanisms
3. Vulnerable Clinical Trial Populations
4. Confirmatory Efficacy Trials
5. Workforce

After each of breakout groups met for discussion, participants reconvened to share their recommendations and to discuss with the full Workgroup. The full Workgroup’s conversations and considerations informed the contents of this written report and will be used to guide future directions for drug development efforts in the coming years.

Charge
Dr. Gordon presented the charge to the Workgroup - to address the current state and future directions of drug development for mental disorders at NIMH - through a series of questions:

• What should constitute "target engagement" for medications? Are hypothesis-based readouts sufficiently mechanistic? Are we communicating this priority effectively?
• Should we restart or revamp Fast Fail (contract-based) efforts to explore novel drugs? How else might we encourage rapid testing of mechanisms?
• How do we approach vulnerable clinical trial populations from an experimental therapeutics perspective?
• Should NIMH consider confirmatory efficacy trials for medications? If so under what conditions?
• Do we have the workforce and/or what are the obstacles to developing and maintaining an adequate psychopharmacology research workforce?

III. History of NIMH Drug Development Support

Overview of Funding Mechanisms and Activities
NIMH utilizes a spectrum of funding mechanisms to support investigators in their efforts to advance clinical research. These funding mechanisms roughly fall into four broad categories (contracts, grants, cooperative agreements, and small business research programs), each of which affords investigators a unique balance of support, oversight, and cooperation.

The first of these is the contract mechanism. This is a legal agreement for goods or services for the direct use or benefit of the government. A contract binds parties to defined requirements, specific deliverables and a defined schedule. A contract is used when the government wants a particular scope of work, defined by the statement of work, to be conducted. This funding mechanism also involves measurable milestones to be met during the course of the study, and a very specific set of requirements in order to complete the study. Site monitoring and frequent communications with government staff are other key aspects of research performed as a Contract. Given the degree of government oversight and involvement, contract mechanisms require significant involvement by NIMH staff.

The second category is the grant mechanism (see Table 1). A grant proposes a project that is determined by the investigator and is conducted by the investigator and collaborators, without direction from National Institutes of Health (NIH) staff. As an example, the R01 is a common NIH grant award mechanism made to support a discrete, specified, circumscribed project, and performed by the named investigator(s) in an area representing the investigator’s specific interest and competencies. In 2014, NIMH initiated a series of funding opportunity announcements (FOAs) with set-aside funds as Requests for Applications (RFAs) to encourage the submission and support of clinical trial-based research using an experimental therapeutics approach.

Table 1: NIMH Clinical Trial Grant Mechanisms - Stages of Development*

First in Human	Exploratory Experimental Therapeutics	Confirmatory Efficacy	Effectiveness (Prevention, Treatment, Services)
First in Human and Early Stage Clinical Trials of Novel Investigational Drugs or Devices for Psychiatric Disorders (U01 - Clinical Trial Required)*			Pilot Effectiveness Trials for Treatment, Preventive and Services Interventions (R34 - Clinical Trial Required)
	Early Stage Testing of Pharmacologic or Device-based Interventions for the Treatment of Mental Disorders (R61/R33 - Clinical Trial Required) and (R33 - Clinical Trial Required)*		Clinical Trials to Test the Effectiveness of Treatment, Preventive, and Services Interventions (R01 Clinical Trial Required) or (Collaborative R01 - Clinical Trial Required)
	Development of Psychosocial Therapeutic and Preventive Interventions for Mental Disorders (R61/R33- Clinical Trial Required) and (R33 Clinical Trial Required)	Confirmatory Efficacy Clinical Trials of Non-Pharmacological Interventions for Mental Disorders (R01 Clinical Trial Required)
* The intent of the U01 and R61/R33 for drug development is to support early phase clinical trials to de-risk drug and device therapeutic candidates for further development

These included FOAs seeking applications for early-stage testing of pharmacologic and device-based treatments, and others seeking applications for the development of psychosocial therapeutic and prevention interventions. The early-stage FOAs utilize a phased research award, the R61/R33 grant mechanism, whereby investigators must confirm target engagement within the first two years (R61) of the project before proceeding to the second phase (R33) for up to three more years. In addition, NIMH provides a five-year R01 application for confirmatory efficacy trials of non-pharmacological interventions, as well as a suite of FOAs for effectiveness research, which test the therapeutic value of treatment and prevention interventions for use in community and practice settings. The phased R61/R33 mechanisms, enables investigators to propose their own clinical trials, with a focus on target engagement and use of pharmacodynamic (PD) endpoints to test drugs that they identify and source on their own. The phased approach enables study discontinuation if results are negative; and investigators are required to demonstrate dose dependent target engagement of the drug in order to proceed to the second funding phase. The drug-related R61/R33 awards that have been funded to date primarily studied repurposed drugs, such as oxytocin, ondansetron, L-DOPA, ezogabine, tetrahydrocannabinol (THC), and levetiracetam. However, some projects proposed the testing of investigational compounds that are not approved by the U.S. Food and Drug Administration (FDA) for any indication, including: lanicemine and pomaglumetad methionil.

The third category, the cooperative agreement mechanism (“U” awards), differs from grants in that it provides for significant government staff involvement. Cooperative agreements are assigned an NIH staff member(s) as a Project Scientist(s), who have substantial involvement in the conduct of the study including contributing to the scientific direction of the research. Cooperative agreements provide more oversight than is typical on a grant through the establishment of a steering committee comprised of both government and academic staff. If the agreement is part of a group of funded cooperative grants, then government staff from multiple institutes could be included. Currently, NIMH supports cooperative agreements (which are denoted by a “U” in the award number) for drug development through three different programs. The first of these is the First in Human and Early Stage Clinical Trials of Novel Investigational Drugs or Devices (U01, equivalent in scope to the R01 grant mechanism) which supports early-stage clinical trials of novel mechanisms of action, investigational drugs, or drug candidates for the treatment of psychiatric disorders. The National Cooperative Drug/Device Discovery/Development Groups (NCDDG) is a program to accelerate innovative drug and device discovery, to develop pharmacologic and neuromodulatory tools, and to support early stage clinical trials of investigational drugs. The NCDDG support this research through either the single project U01 or the multi-project (U19) cooperative agreement mechanisms. Lastly, the Novel Assays to Address Translational Gaps in Treatment Development initiative, supported through the UG3/UH3 cooperative agreement activity (analogous to the phased R61/R33 research grant activity), aims to identify neurophysiological measures as potential translational assays (PD measures) for treatment development research. In particular, the pharmaceutical/biotech industry sector has expressed interest in the NIMH cooperative agreement funding mechanism, which enables a higher degree of industry and government involvement than a typical R-grant mechanism, but is not as prescribed as a contract.

Finally, NIMH supports drug development efforts through the Small Business Research Programs. Under the Small Business Act, the Small Business Innovation Research (SBIR) and Small Business Technology Transfer (STTR) programs enable small U.S.-owned and operated businesses to explore federal research and development with an incentive for commercialization. The SBIR and STTR programs aim to encourage and stimulate technological innovation in the private sector. Currently, NIMH supports drug development for mental illnesses by small businesses with Complex Technologies and Therapeutics Development for Mental Health Research and Practice (R43/R44). This supports both biomarker evaluation and the development of pharmacologic agents.

In addition to FOAs, NIMH also utilizes notices to convey our interest in particular areas of research. Focusing on new areas and priority areas of science for which NIMH seeks applications, these notices point investigators to specific topics of research interest and the associated FOAs through which to apply. NIMH has issued specific notices to encourage applications testing these new targets include NOT-MH-19-055  Notice of Special Interest: NIMH’s Interest in Early Stage Clinical and/or Mechanistic Trials in Posttraumatic Psychopathology, Directed at Testing FAAH Inhibitors; NOT-MH-17-044  Notice of NIMH's Interest in Pediatric Pharmacologic Trials in Autism Spectrum Disorders, directed at testing selective GABAergic agents; NOT-MH-17-039  Notice of NIMH's Interest to Highlight High-Priority Pediatric Pharmacologic Trial Research Areas; NOT-MH-20-027  Notice of Special Interest to Highlight High Priority Opportunities for Intervention Development Research in Geriatric Mental Health. These efforts have stimulated the submission of some Cooperative Agreement clinical trial applications to support early stage PD trials through NIMH partnerships with industry, academia, and other collaborators. For example, NIMH recently awarded  Yale University with a Phase IIa clinical trial to test a D1/D5 receptor partial agonist in collaboration with Cerevel Therapeutics.

It is recognized that choosing the most appropriate NIH funding mechanism for a research project can be a complex and daunting process. To address this complexity, NIH provides program staff contacts for every announcement, and investigators are encouraged to contact these program officers to guide them through the application process.

In addition to supporting research projects, NIMH also provides multiple options to support investigators as they progress through their training. These include fellowship awards (through individual "F" awards) or support as a trainee under an institutional T32 training program; career development awards (K awards); as well as some educational opportunities through Center Grants (P50 and P30 mechanisms). The breadth and diversity of these funding mechanisms allows for support of talented investigators as they work to discover and enhance treatments for mental disorders. NIMH is primarily interested in rigorously designed research projects with appropriate NIMH involvement/support based on the project mechanism being employed. This broad toolbox of funding mechanisms and partnerships enables NIMH to collaborate with a diverse group of stakeholders to optimize and leverage NIMH funds and knowledge.

A History of NIMH FAST Fail Program

NIMH initiated the landmark Fast-Fail Trials (FAST) initiative in 2012 in response to industry’s downsizing of psychiatric drug development research programs. The aim of the FAST initiative was to address the critical need for new psychiatric medications by providing an objective approach to test new or repurposed compounds in early stage clinical trials. Moreover, the initiative aimed to identify PD measures in the brain that could also be used in the development of additional candidate compounds. This pharmacologic-based testing approach focused on associating dose with drug exposure and CNS response (PD effect) in order to determine whether drug levels could be achieved that alter brain function. Positive results would provide guidance for dose selection in future larger-scale efficacy trials. Negative results could indicate adequate drug levels cannot be achieved to affect CNS activity in a way that could improve clinical function ("Fast Fail"). Both positive and negative results would be informative and would help to select the most promising compounds for further testing. Additionally, these trial designs could help inform on PD measures that can best capture CNS target activity when designing future trials to assess the mechanism of action of candidate compounds.

Following initiation of the FAST program, three studies were included in the first round. Investigators received guidance from a Council workgroup that was formed to provide input in identifying ideal molecular targets, determining compound availability, and optimizing PD readouts. After targets and compounds were identified, investigators supported by the FAST contracts focused specifically on experimental designs using quantitative PD readouts. The resulting protocols included studies of autism spectrum disorders (FAST-AS), psychotic spectrum disorders (FAST-PS), and mood/anxiety spectrum disorders (FAST-MAS), using AZD7325 (a GABA-A alpha2,3 positive allosteric modulator), pomaglumetad methionil (a mGluR2/3 agonist), and JNJ-67953964 (a kappa opioid receptor antagonist), respectively.

Data publications from the FAST trials became available starting in 2018, with more manuscripts in development. Pharmaceutical and biotech companies began to express excitement about the quality of the data and the collaborative nature of bringing many experts together to identify targets and compounds, and to design trials with informative outcomes. These companies have approached NIMH about testing their compounds using approaches similar to those of the FAST contracts. Additionally, PD measures evaluated in these trials are being incorporated into new industry trials. A recent review discusses the background and goals of each FAST program, and the lessons learned.⁸

[Top]

IV. Target Engagement in Experimental Therapeutics Research

Target Engagement

NIMH requires early stage drug intervention studies to incorporate an experimental therapeutics/proof of mechanism based approach to determine if drug doses can produce a dose- or exposure-dependent PD response that is proximal to the intended/presumed mechanism of action of the compound being studied, and how that action may translate to clinical outcomes. Clinical studies should be designed to avoid interpretive failure, so that either positive or negative results will provide meaningful information, and support the following: 1) the intervention selectively engages/modulates the target (PD effect) in a dose-dependent manner; 2) the doses that demonstrate target engagement are safe and tolerable; and 3) change-in-function or efficacy signals are observed at doses that are safe and produce a PD response. If such results are not obtained, proof of mechanism has not been supported and the program should not be pursued further. 

The Workgroup stated that NIMH should clarify the definition of a "target." The members recommended broadening the definition of target to include a biological process or molecular event that is modified by an intervention and thereby has the potential to produce a therapeutic effect, "proof of mechanism." Target engagement is critical for evaluating therapeutic hypotheses associated with an investigational compound; that is, investigators can reject a hypothesis about a compound when they can conclusively demonstrate that the compound adequately engages the intended target but does not produce the intended effect. The Workgroup discussed the alternative approach of working backwards from an observed clinical signal towards a potential target presumed to underlie disease pathophysiology when there is an absence of prior work validating that clinical assessment directly against a gold standard for target engagement (e.g., receptor occupancy assessed with positron emission tomography (PET). The Workgroup expressed concern that these downstream measures may be too unclear. The Workgroup concluded that while such an approach may be useful for demonstrating the effects of a compound, primary reliance on clinical efficacy signal cannot reliably support or reject hypotheses about mechanism of action.

Biomarkers

The Workgroup stressed the distinction among various utilities of objective, translational biomarkers. In the context of a treatment intervention, an objective biomarker that can specifically demonstrate on-target PD effects is required for rigorous hypothesis-testing. The corollary is that off-target effects—which occur when a compound affects biophysiological parameters other than those mediated by the target of interest—may indirectly suggest target engagement but cannot allow adequate testing of a hypothesis and interpretation of clinical efficacy signal, positive or negative. Other relevant uses of biomarkers include diagnostics, susceptibility risk prediction, therapeutic efficacy assessment, prognostic utility for patient enrichment/stratification, and monitoring disease stage/prognosis. Overall, the heterogeneity of pathophysiology of mental health diseases requires that clinical trials be conducted using: 1) PD biomarkers to demonstrate target engagement/proof of mechanism in early stage clinical trials , and 2) biomarkers to enrich clinical efficacy trials with subjects most likely to respond to intervention due to biological deficits tied to the drug’s potential mechanism.

Related to PD biomarkers, it is critical to establish it in preclinical systems, even if it is based on non-translational, invasive measures. The Workgroup focused additional discussion on the broader definition of biomarkers as measures of target engagement, coverage, and/or desired modulation. The target may be an enzyme, a transporter, receptor, etc. Though molecular imaging using PET generally provide the highest level of confidence in the mechanism being tested, biomarkers of various types of CNS activation (e.g., electroencephalogram (EEG), functional magnetic resonance imaging (fMRI)) can still provide useful information of effects downstream from molecular target engagement. However, in some cases it may actually be more reliable than PET, i.e., high-efficacy GPCR agonists which may express relevant pharmacology at very low occupancy levels. The Workgroup highlighted, however, that the limitations of any biomarker used need to be considered when interpreting the data. These might include technical issues related to the level of validation of the biomarker assay (e.g., range, linearity, reproducibility), feasibility of including the biomarker assessment in the population under study (e.g., adult vs pediatric, healthy vs impaired), and the relationship of the biomarker output to the known mechanism of action/potential off-target activity of the investigational compound, potential confounds of non-CNS (e.g., cardiovascular or hemodynamic) effects of the compound on biomarkers, etc.

Recognizing that the identification of PD measures in the CNS that can indicate the potential for efficacy is a relatively new area of study in CNS clinical research, the Workgroup discussed that a “gray zone” between a molecular target and a potential PD measure may be necessary, particularly in early exploratory studies. To successfully use a biomarker as an index of CNS target engagement in clinical research, the investigator first must understand the intermediate steps between direct target engagement and validation of the downstream effects as a marker of the initial molecular event. The Workgroup noted a fundamental difference between biomarkers that can directly link to target engagement versus those that represent a relevant downstream outcome but cannot be confirmed as a direct casual outcome related to the target engagement. In the absence of clear knowledge about the process from cellular event to system effects, and a lack of PET ligands, investigators may only have putative biomarkers available for study.

The Workgroup stressed the importance of measuring target engagement but recognized that confirming target engagement would have precluded the development of certain psychiatric drugs (such as lithium or gabapentin) in which the target and its mechanisms remain unclear. However, these types of drug "successes" are not a common occurrence and not an efficient way of research and development (R&D). The Workgroup stressed the need to develop rigorous and precise hypotheses that can definitively answer questions related to target engagement/proof of mechanism.

The Workgroup considered the use of peripheral biomarkers, which could be considered proximal to a peripheral target, and whether a peripheral biomarker could be used to predict a central state. However, there would need to be clear validation demonstrating that a peripheral measure accurately represents CNS physiology. To date, these measures are still exploratory.

There was consensus among the group members that investigators can use biomarkers (perhaps phenotype-based or drawn from the Research Domain Criteria [RDoC] framework) to enrich therapeutic assessment trials with more homogeneous subgroups of subjects whose clinical deficit is presumed to be most closely associated with a compound’s mechanism of action. 

The Workgroup then discussed the translation of target engagement from animal models to human systems. The RDoC matrix was developed in response to lack of translation of traditional preclinical behavioral readouts. Instead of these readouts, circuit-level behavioral phenotypes in animal models are believed to better translate into humans. Because mental illnesses are complex and fundamentally human, investigators need to confirm that certain neural pathways in an animal will usefully inform on clinical phenotypes in a human.

While the Workgroup admitted that discussion of preclinical studies was largely beyond the scope of the group’s charge, they felt some brief acknowledgement of translational gaps was important. Investigators frequently refer to animal models to understand the effects of varying levels of target engagement/proof of mechanism on efficacy, often using behavioral readouts as efficacy measures. Unfortunately, there is limited clarity about how the molecular engagement of the primary target produced the chain of events leading to animal behavioral change. For example, if an investigator finds a dose-related effect of a drug on the performance of mice on a non-specific behavioral test (e.g., tail suspension test), they cannot infer the degree of target occupancy; rather, the outcome is used to anticipate efficacy in humans. In this case, struggling on the tail suspension test is not a target engagement biomarker but is used as an efficacy biomarker. Instead, Workgroup members suggested that preclinical studies could focus on the chain of events beginning with molecular engagement, and identify biomarkers associated with proof of mechanism that could be reliably translated to humans, such as EEG. Combining such biomarkers with computational approaches can be very powerful for establishing dosing confidence in humans. For example, researchers currently establish brain penetration of a compound in two species and then extrapolate that quantitatively with a physiologically based pharmacokinetic (PBPK) model to man, to better optimize dosing. This is well precedented in the field of drug-drug interactions and there are now approximately 250 FDA drug labels where in silico predictions were accepted in lieu of actual clinical trials. While that approach has been used for actual marketed drugs, such a paradigm could be feasible in exploratory medicine.

Positron emission tomography (PET)

For more than two decades the NIMH has supported the development of PET and single-photon emission computerized tomography (SPECT) imaging ligands through a series of NIH FOAs, in coordination with other NIH institutes, (recent examples include R01 FOAs: PAR-20-037 , PAR-20-038 ; U-award FOAs: PAR-20-118 , PAR-20-119 ; and the NIMH SBIR/STTR grant mechanisms) as well as in the NIMH Intramural Research Program. The goal has been to prioritize targets of interest for which there are chemical entities with appropriate properties for radiochemistry as well as PK and PD properties.

Molecular imaging with PET is only one methodology for assessing target engagement or demonstrating an effect on a CNS biomarker. It is ideal for use establishing target-proximal PD responses in drug trials and is biased towards G-protein-coupled receptor (GPCR) antagonists which leaves out a number of novel targets and mechanisms (including positive allosteric modulators (PAMs) and negative allosteric modulators (NAMs) which are often difficult to assess with PET). 

The FDA-NIH Biomarker Working Group has established a Biomarkers, EndpointS, and other Tools (BEST) resource⁹: https://www.ncbi.nlm.nih.gov/books/NBK402286/ . The BEST information speaks more to a broader biomarker-based approach beyond receptor occupancy (for which there are relatively few PET/SPECT tracers available).

Recommendations

Recommendations for consideration by NIMH:

• Explore or continue to encourage target engagement/proof of mechanism approaches that are not strictly limited to PET.
• Devote resources to validating analytically qualified biomarkers as PD measures that may predict clinical efficacy response (eventually a surrogate endpoint).
• Measures of target engagement/proof of mechanism ideally should be translatable, which may require computational approaches.
• Measures of target engagement/proof of mechanism should be as proximate to the molecular event as possible.
• Continue to utilize special study sections for review of grant applications focused on target engagement/proof of mechanism trials that include clinical pharmacology expertise.
• Invest in enrichment/stratification biomarker research that can be incorporated into target engagement/proof of mechanism studies.

[Top]

V. Fast-Fail and Rapid Testing of Mechanisms

The FAST Contracts, outlined in Section III, have provided informative and clinically meaningful results aligned with an experimental therapeutics approach. These contracts have also led to progress in biomarker tool development using fMRI, EEG, and magnetic resonance spectroscopy - providing a proof of concept as to whether these techniques can be executed across multiple sites as reliable measures and clear endpoints. Additionally, Workgroup participants agreed that the collaborative platform used by the FAST contracts has stimulated industry interest for early-stage drug development, encouraging partnerships between NIMH and diverse stakeholders.

The Workgroup participants deliberated whether the term “Fast-Fail” accurately captures NIMH goals and priorities for experimental therapeutics research. It was indicated that while Fast-Fail is an industry concept, reflecting drug companies’ goals to conserve resources by quickly moving on from investigational drugs that do not show immediate promise, NIMH aims to engage in rapid and definitive testing of mechanisms in order to determine if follow on studies are appropriate.

The Workgroup members shared some of their experiences with the NIMH FAST application and funding through the contract mechanism process. Several drawbacks of the contract mechanism were shared including complexity in both the contracting process (e.g., technical and business proposal requirements, legally executing contract) and at the level of executive oversight of studies (e.g., regular and extensive NIMH program staff involvement and discussion about target identification, compound selection, alternative drug options, trial designs, review of study progress, use of the NIMH Data and Safety Monitoring Board). Some Workgroup members expressed concern that these issues could cause delays to study initiation. Other members reported a positive, efficient experience and trials that were ultimately successful. These experiences highlighted the differences between a grant and contract and that NIMH should carefully consider the choice of funding mechanism. There was agreement that the cooperative agreement mechanisms may be more efficient and streamlined for some type of NIMH studies than the contract mechanism.

The Workgroup assessed that NIMH should consider working to improve efficiency and operability of contracts by clearly specifying the goals for each particular contract. The project timeline could be accelerated by specifying clear objectives, identifying a compound of interest prior to funding, and incentivizing partnerships between investigators and industry. Additionally, NIMH was encouraged to optimize the oversight process for contracts. It was noted that investigators may work more successfully within a narrowly defined contract mechanism, particularly if background work to select compounds and negotiate partnerships has been completed prior to project initiation.

The Workgroup participants also agreed that investigators would benefit from rigorous, metrics-driven negotiation with sites or specialty contract research organizations (CROs) on a pay-per-patient basis. These agreements typically involve positive and negative incentives for sites or CROs to conduct trials appropriately. Finally, Workgroup members suggested that NIMH take a broad look at other successful projects such as the 2013 StrokeNet  project, in which there were collaborations at international and non-traditional academic sites. However, investigators must provide adequate justification to seek international site partnerships.

 Recommendations

• Modify FAST Contracts to optimize scope and specific objectives.
• Identify Investigational New Drug (IND)-ready drugs of interest (including those shelved by industry) in the Request For Proposals (RFP).
• Consider CRO involvement beyond traditional academic sites.
• Consider whether the "Fast-Fail" term communicates the objectives of the program.

[Top]

VI. Vulnerable Clinical Trial Populations

The Workgroup considered specific issues about drug development research with vulnerable clinical trial populations (geriatric and pediatric populations). Drug developers face unique challenges for target engagement, drug development, and PD studies in pediatric and geriatric populations. For example, an effective therapeutic exposure for adults may not be effective for children, and pediatric-appropriate dosages cannot necessarily be extrapolated from knowledge about adults, though there are some exceptions. Extrapolation failures occur most commonly (40%) in children younger than two years and become less common by adolescence, enabling adolescents to participate in some late-phase adult trials.

Additionally, children’s developing CNS, which may fundamentally differ from developed adult systems, may have an altered PD response. The problem is further complicated by substantial developmental variation across childhood (e.g., premature infants may respond differently than full-term infants, attaining pubescence may modulate response). These concerns have significant implications for safety and efficacy outcomes among children.

On the other end of the spectrum, drug developers must also consider unique challenges for geriatric populations, who experience normal age-related changes in brain structure and function and who tend to suffer from comorbidities that may impact safety and efficacy outcomes. Polypharmacy effects also remain a concern, not just for metabolism or drug-drug interaction, but for changes based on long-term exposure to a therapeutic. The Workgroup also noted that an older adult who has taken an antipsychotic drug for decades might experience varying exposure, efficacy, and target expression over that period of time.

The Workgroup also recognized that drug developers are tasked to overcome significant barriers to treat these populations. The Workgroup brainstormed approaches to support drug development research in this area. There was agreement that NIMH should utilize the full breadth of funding mechanisms available as an effective way to drive and incentivize research in vulnerable clinical trial populations. The phase-based and milestone-based funding mechanism ensures that investigators meet concrete, measurable requirements. Contracts can also effectively support and incentivize investigators to complete PK/PD research that many researchers find tedious or uninteresting. It was noted, however, that some investigators may find small contracts less attractive because the work is not investigator-initiated and the data become property of the government.

There was agreement that investigators benefit from detailed FOA guidance related to fully understanding the criteria under which their applications will be reviewed. NIMH should continue to provide technical assistance prior to a receipt date to remind participants of the criteria. Review panels covering topics involving clinical pharmacology in vulnerable clinical trial populations should include adequate clinical pharmacology expertise.

Additional consensus among Workgroup participants was that NIMH should consider (as outlined in the previous section) the inclusion of specialty CROs on a pay-per-patient basis to help ensure successful recruitment. Incentives may be key for industry involvement in the drug development process, especially given the Pediatric Research Equity Act (PREA ) and the Best Pharmaceuticals for Children Act (BPCA ), the latter of which includes a pediatric exclusivity provision. In general, NIMH must balance productivity expectations (via per-participant enrollment) with the recognition that some investigators may experience infrastructural limitations.

Recommendations

• Offer both phase- and milestone-based mechanisms, such as UG3/UH3.
• Consider the use of the contract funding mechanism for bridging studies to vulnerable clinical trial populations.
• Develop industry involvement.
• Balance site productivity, infrastructure, and leveraging to optimize NIMH investments.
• Provide investigators with specialized resources/expertise to conduct Good Clinical Practice (GCP) trials.
• Provide site compensation on a per-patient basis.

[Top]

VII. Confirmatory Efficacy Trial Considerations

The Workgroup explored the role of confirmatory efficacy trials in the experimental therapeutic medicine approach to drug development. NIMH support of confirmatory efficacy trials (i.e., RFA-MH-18-707 ) is limited to psychosocial and device-based interventions, while specifically excluding drug studies. An overarching theme of the discussion was whether there should be modifications or exceptions to allow for certain types of confirmatory efficacy trials for pharmacotherapies.

The NIMH clinical trial funding announcement is structured for confirmatory efficacy trials to be trials of interventions that have demonstrated target engagement/proof of mechanism and have shown an initial clinical/functional signal. The Workgroup debated if the NIMH should expand the definition of the term confirmatory efficacy trials. For example, because the FDA and the pharmaceutical industry are focused on regulatory approval of drugs, some considered confirmatory efficacy trials to be a second adequate and well-controlled trial to confirm the efficacy results from an initial positive adequate and well-controlled trial. There was consensus that NIMH should consider extending the support of confirmatory efficacy trials to include pharmacotherapies, under certain circumstances. A key determining factor discussed is the presence or absence of industry support for a drug intervention proposed for study. Formal collaborations with pharmaceutical companies may provide a valuable mechanism for NIMH drug development efforts. These partnerships generally include a designated "hand-off" point in the drug development timeline, commonly by the confirmatory efficacy trial stage, and confirmatory drug trials would be conducted by the pharmaceutical industry. On the other hand, pharmaceutical companies have little to no incentive to study drugs without commercial potential, such as repurposed older drugs (e.g., ketamine for treatment-resistant depression). As a result, the commercially focused incentive structure may perpetuate critical knowledge gaps in pharmacotherapy, including considerations for vulnerable clinical trial populations, combinations with psychosocial interventions, and other potential treatment indications.

The Workgroup discussed the potential value of NIMH supporting clinical trials for drugs that lack industry support due to limited commercial incentives even though this does not meet the strict definition of a confirmatory efficacy trial. There are critical gaps in scientific understanding of clinical efficacy and implementation in clinical practice, but a lack of compelling industry priorities has stymied investigation in these areas. To address this urgent public health need, NIMH may wish to assume a role in supporting drug development research which lacks commercial incentives for the pharmaceutical industry. These trials could include off-patent drugs with potential for development for new indications, including those that may already reflect off-label use. Such trials should be designed to investigate optimal dosing, advance scientific knowledge of underlying mechanisms, or change/improve clinical practice as with most current NIMH-supported nonpharmacologic confirmatory efficacy trials being conducted without plans for future regulatory submissions. In special circumstances, it is conceivable that NIMH might wish to partner with an industry sponsor to support confirmatory efficacy trials of significant public health importance. These circumstances might include adding a new indication, changing use practices for a marketed product, or answering important scientific questions during drug development beyond regulatory requirements for supporting approval. Such programs might benefit from consultation with the FDA.

The Workgroup also agreed that confirmatory efficacy trials for drugs could provide a valuable opportunity to validate target engagement. To remain consistent with NIMH’s current research approach, target engagement/proof of mechanism—and subsequent evaluation and confirmation of the pharmacologic relationship between the mechanism and the clinical outcome—would be a mandatory component of confirmatory efficacy trials. Given NIMH’s mission to advance and accelerate scientific discovery, such a requirement would establish objective pharmacology-based dosing strategy, could illuminate underlying mechanisms, and would generate important new knowledge of interest to NIMH. Workgroup members recognized that while investigators should demonstrate target engagement prior to submitting a confirmatory efficacy trial, biomarkers of target engagement may be further validated in a sub-study of participants within the larger confirmatory efficacy study. It was noted that biomarkers often are a critical component of the FDA regulatory process, and in some cases can be considered a primary endpoint or a key safety endpoint with an appropriate level of validation.

The Workgroup also discussed whether investigators may not need to validate target engagement in a confirmatory efficacy trial when target engagement had been previously demonstrated in a prior well-controlled trial. However, in some cases PD biomarkers could enable investigators to show that larger trial populations have similar clinical features to the narrower population in which target engagement has been demonstrated. These studies may also establish the relationship between target engagement and clinical efficacy signal (in other words, the level of target engagement needed to provide meaningful benefit to the patient). This may streamline and accelerate Phase III studies, which tend to include a broader population than more tightly controlled Phase I-II studies. For trials to establish efficacy in pediatric indications, NIMH has worked with the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) BPCA program to support regulatory trials (including pediatric PK/PD bridging studies followed by efficacy trials) with the end goal of label changes through FDA clearances.

Recommendations

• Consider expanding support of confirmatory efficacy trials to include those using drug interventions in very specific and defined situations. These trials should be focused on areas of significant scientific and/or clinical importance that lack commercial incentives for industry support, with clear end goals for informing clinical practice.
• Consider exploring collaborations with the pharmaceutical industry and FDA, as circumstances warrant.
• Balance unmet public health needs with the high cost and complexity of such trials when considering the support of confirmatory efficacy drug trials.
• Establish target engagement/biomarkers prior to performing confirmatory efficacy trials and further validate in at least a subset of participants in confirmatory efficacy trials.

[Top]

VIII. Changes in Workforce Development

Changes in clinical trial research may necessitate changes in workforce training and mentorship. Consequently, the next generation of mental health researchers will face numerous changes in clinical research training. As the science community becomes increasingly interdisciplinary, the definition of "successful" outcomes for trainees may necessitate new supports and offerings in training programs. For example, there is urgent need for new infrastructure to comprehensively train a new generation to conduct experimental therapeutics-based clinical trials. To meet new NIMH research standards and approaches, scientists must develop a skillset beyond the scope of clinical trials, to include expertise in biomarkers, imaging, and other relevant domains. The Workgroup noted that while such an extensive, multi-disciplinary training may not be adequately supported within current NIH funding mechanisms, junior scientists must be sufficiently motivated to choose this pathway early in their careers. Otherwise, workforce expertise in drug development trials tends to bottleneck.

The Workgroup members stressed that NIMH should consider training broadly as it is critical at all stages of the career trajectory, from early-career internships to post-doctoral fellowships and beyond. However, available NIH funding may also be insufficient to cover the resources necessary to implement training programs throughout the entire career pipeline. Lessons can be learned about using novel funding approaches and alternate training methodologies to address training needs. For example, programs such as the GlaxoSmithKline (GSK)/Wellcome Trust co-funded training program in Translational Medicine and Therapeutics affords researchers non-linear career and training opportunities. Additionally, employing cluster training enables institutions to consciously assemble cadres of young, talented trainees with complementary expertise. The result is to build a community of well-connected scientists who comprise a regional academic network and can act as liaisons between academia and industry.

The Workgroup discussed general challenges in clinical research training for the next generation of leaders in mental health research. It was noted that candidates may not take advantage of the available training mechanisms that exists. Trainees may be deterred from involvement in clinical trials because their potential for a first-author publication is limited. To address this concern, some programs allow trainees to conduct their own small clinical trials culminating in two or three portfolio-building pharmaco-epidemiological papers. Along these lines, some investigators consider individual K awards, which are made in the investigator’s name, to be more prestigious than institutional training awards. This may lead the most promising candidates to pursue individual career awards with a single focus over opportunities to engage in integrated, multi-disciplinary institutional training programs. NIMH may want to consider options for balancing career incentives with supporting the broad training required to build the clinical trials research workforce. NIMH engagement may be helpful in ensuring that a strong cadre of trainees and appropriate research projects are supported through training programs in this area. The NIH’s T32 program in pediatric clinical pharmacology was presented as an example of a program supporting this type of cross-disciplinary training.

The Workgroup also discussed the unique challenges the NIMH will face in supporting clinical research training oriented towards industry and in experimental therapeutics. An increased emphasis on partnerships between NIMH and industry is encouraged. For example, there may be a need to develop additional training programs that cultivate a workforce of investigators who are well equipped to move between academia and industry as concerns were noted about the lack of mobility between industry and university sectors at all career stages. To stave off the perception that careers in academia and careers in industry exist in isolated tracks, NIMH/industry partnerships could mitigate the misconception that researchers cannot move fluidly between the sectors. An ideal NIMH/industry collaboration could promote visibility of and expand the capacity for joint academic/industrial pathways in experimental therapeutics for mental disorders.

The Workgroup emphasized that prioritization of workforce diversity in drug development research for mental health, not just among participant pools but among investigators, leaders, faculty, partners, and other stakeholders, is critical. Although some diversity-focused mechanisms exist, drug development research programs often struggle to attract and retain investigators from underrepresented and minority groups. The Workgroup recommended that NIMH expand existing diversity fellowship mechanisms, enhancing early career exposure, and celebrating the value of diversity in science.

Recommendations

• Consider developing NIH/industry co-funded training or career development schemes, using lessons learned from programs such as 1) GSK/Wellcome Trust co-funded program in Translational Medicine and Therapeutics in Cambridge ; and 2) NIH T32 program in pediatric clinical pharmacology .
• Encourage regional networks to support multi-disciplinary training opportunities and foster academic/industrial partnerships.
• Increase visibility and assess feasibility of joint academic/industrial training as a pathway to experimental therapeutics for mental disorders.
• Build capacity for co-mentorship for trainees to build a diverse skillset (e.g., the NIH T32 program in pediatric clinical pharmacology).
• Continue and expand efforts to increase diversity in the clinical trials workforce.

[Top]

IX. Conclusions

NIMH would like to thank the co-chairs and Workgroup members for their input and expertise which informed this report. The NIMH intends to use the report as a guiding framework for drug development initiatives in the years to come. The Workgroup subgroups (Target Engagement, Fast-Fail and Rapid Testing of Mechanisms, Vulnerable Clinical Trial Populations, Confirmatory Efficacy Trials and Workforce) provided recommendations for each topic area. Notably, a theme that was consistent across recommendations was the need to continue building a clinical trials program that focuses on the quantitative assessment of drug pharmacology as it relates to brain function. The Workgroup recommended that NIMH continue to support target engagement/proof of mechanism trials through the use of CNS biomarkers in quantitative pharmacologic-based dosing designs. Training on these biomarker-based trial designs was considered critical for supporting the next generation of researchers. The Workgroup also recommended NIMH support confirmatory efficacy trials for drugs, in specific cases where the drug is used extensively off label, quantitative dosing had been established, and where there is no commercial path for which the studies could be supported by Industry. However, the Workgroup recommended the biomarkers used in proof of mechanism studies should be further assessed in subsets of subjects from the confirmatory efficacy trials to better inform how the pharmacologic response relates to the broader population (and provides an objective readout in the trial). Overall, conducting clinical trials with a neuroscience focus requires multidisciplinary expertise, and therefore applications should be reviewed in special study sections with focused expertise. NIMH should also continue to provide clinical trial resources to investigators to aid them in conducting these types of studies.

[Top]

X. References

¹ From Discovery to Cure: Accelerating the development of new and personalized interventions for mental illness. Report of the National Advisory Mental Health Council’s Workgroup. August 2010. 

² Kessler RC, Chiu WT, Demler O, Walters EE. 2005. Prevalence, severity, and comorbidity of twelve-month DSM-IV disorders in the National Comorbidity Survey Replication (NCS-R). Archives of General Psychiatry, 62(6):617–27. 

³ The World Health Organization. 2004. The World Health Report 2004: Changing History, Annex Table 3: Burden of disease in DALYs by cause, sex, and mortality stratum in WHO regions, estimates for 2002. Geneva: WHO. 

⁴ Lieberman JA, Stroup TS, McEvoy JP et al. for the Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) Investigators 2005. Effectiveness of antipsychotic drugs in patients with chronic schizophrenia. New England Journal of Medicine, 353:1209-23. 

⁵ Trivedi MH, Rush AJ, Wisniewski SR et al., STAR*D Study Team 2006. Evaluation of outcomes with citalopram for depression using measurement-based care in STAR*D: implications for clinical practice. American Journal of Psychiatry, 163(1):28-40. 

⁶ Gaynes BN, Warden D, Trivedi MH, Wisniewski SR, Fava M, Rush AJ. 2009 What did STAR*D teach us? Results from a large-scale, practical, clinical trial for patients with depression. Psychiatric Services, 60(11):1439-45. 

⁷ Jarvis, Lisa M. 2012. Tough Times For Neuroscience R&D. Chemical & Engineering News, 90(12): 22-25. 

⁸ Grabb MC, Hillefors M, Potter WZ. 2020. The NIMH ‘Fast-Fail Trials’ (FAST) initiative: rationale, promise, and progress. Pharmaceutical Medicine, 34(4):233-45.

⁹ FDA-NIH Biomarker Working Group. BEST (Biomarkers, EndpointS, and other Tools) Resource [Internet]. Silver Spring (MD): Food and Drug Administration (US); 2016. Pharmacodynamic/Response Biomarker. 2016 Dec 22. Available from: https://www.ncbi.nlm.nih.gov/books/NBK402286/ Co-published by National Institutes of Health (US), Bethesda (MD).

[Top]

NATIONAL ADVISORY MENTAL HEALTH COUNCIL WORKGROUP ON DRUG DEVELOPMENT ROSTER

Co-Chairs
Kalpana M. Merchant, Ph.D. President and Chief Scientific Officer Trans Thera Consulting Portland, OR Adjunct Professor of Neurology Northwestern University Chicago, IL	John H. Krystal, M.D. Robert L. McNeil, Jr. Professor of Translational Research Professor of Psychiatry, Neuroscience, and Psychology Chair, Department of Psychiatry Chief of Psychiatry and Behavioral Health Yale New Haven Hospital Yale University School of Medicine New Haven, CT
Members
Anissa Abi-Dargham, M.D. Professor of Psychiatry and Radiology Director, Multi Modal Translational Imaging Lab Vice Chair for Research Department of Psychiatry Stony Brook University Stony Brook, NY	Michael Davis, M.D., Ph.D. Clinical Team Leader U.S. Food and Drug Administration Division of Psychiatry Products Office of New Drugs Rockville, MD
Daniel Benjamin, M.D., Ph.D. Kiser-Arena Professor of Pediatrics Professor of Pediatrics Duke University Durham, NC	Michael Ehlers, M.D., Ph.D. Chief Scientific Officer Chief Executive Officer Limelight Bio. Philadelphia, PA
Edward T. Bullmore, FRCP, FRCPsych Professor Head, Department of Psychiatry Behavioral and Clinical Neuroscience Institute Cambridge University Cambridge, UK	Raquel Gur, M.D., Ph.D. Professor of Psychiatry Neurology and Radiology Perelman School of Medicine Vice Chair of Research Development Department of Psychiatry University of Pennsylvania Pittsburgh, PA
Carla Canuso, M.D. Senior Director of Neuroscience Clinical Development Janssen Research & Development, LLC New Brunswick, NJ	John Isaac, Ph.D. Janssen Neuroscience, Johnson & Johnson London, UK
Donald C. Lo, Ph.D. Director Division of Pre-Clinical Innovation Therapeutic Development Branch National Center for Advancing Translational Sciences National Institutes of Health Bethesda, MD	Hartmuth C. Kolb, Ph.D. Vice President Neuroscience Biomarkers Janssen R&D Global Imaging Johnson & Johnson San Diego, CA
Sanjay J. Mathew, M.D. Professor and Vice Chair for Research Menninger Department of Psychiatry & Behavioral Sciences Baylor College of Medicine Baylor, TX	Piet van der Graaf, PharmD, Ph.D. Senior Vice President, Quantitative Systems Pharmacology Professor of Systems Pharmacology Certara Canterbury, UK
Kurt Rasmussen, Ph.D. Director Division of Therapeutics and Medical Consequences National Institute on Drug Abuse Bethesda, MD	Sophia Vinogradov, M.D. Donald W. Hastings Endowed Chair University of Minnesota Medical School Professor and Department Head Department of Psychiatry Minneapolis, MN
National Institute of Mental Health Staff
Jamie Driscoll, B.A. Program Officer Signal Transduction Program Office of Research Training and Career Development Division of Neuroscience and Basic Behavioral Science National Institute of Mental Health Neuroscience Building Rockville, MD	Margaret Grabb, Ph.D. Program Officer, Novel Pharmacologic and Device-based Intervention Program for Pediatrics Coordinator, Small Business Innovation Research (SBIR) and Small Business Technology Transfer (STTR) Programs Division of Neuroscience and Basic Behavioral Science Division of Translational Research National Institute of Mental Health Neuroscience Building Rockville, MD
Jovier Evans, Ph.D. Branch Chief Geriatrics and Aging Processes Research Branch Neuroscience of Mental Disorders and Aging Program Division of Translational Research National Institute of Mental Health Neuroscience Building Rockville, MD	Mi Hillefors, M.D., Ph.D. Deputy Director Division of Translational Research National Institute of Mental Health Neuroscience Building Rockville, MD
Rebecca Wagenaar-Miller, Ph.D. Deputy Director Division of Extramural Activities National Institute of Mental Health Neuroscience Building Rockville, MD	Jean Noronha, Ph.D. Director Division of Extramural Activities National Institute of Mental Health Neuroscience Building Rockville, MD
Matthew Rudorfer, M.D. Chief Adult Psychopharmacology, Somatic, and Integrated Treatment Research Program Child and Adolescent Psychopharmacology and Integrated Treatment Research Program Treatment and Preventive Interventions Research Branch Division of Services and Intervention Research National Institute of Mental Health Neuroscience Building Rockville, MD	Jeffrey Seeburger, Ph.D. Consultant Office of the Director National Institute of Mental Health Neuroscience Building Rockville, MD

[Top]