CNS Summit 2021 Abstracts of Poster Presentations

A message from the editor

Dear Colleagues:

Welcome to the annual CNS Summit Abstracts of Poster Presentations supplement to Innovations in Clinical Neuroscience. We are pleased to provide you with this reference guide to some of the innovative research presented during the CNS Summit 2021 event. The supplement is also available online by visiting www.innovationscns.com.

This supplement is just a small representation of the cutting-edge research, innovative ideas, and collaborative efforts shared via the CNS Summit platform. Though the COVID-19 pandemic continues to globally impact us in many ways, the strong sense of community and collaborative support the Summit continues to foster among its participants has enabled us to face the pandemic and its ensuing challenges confidently, purposefully, and as a united force of individuals committed to developing better, safer, and more accessible therapies for patients around the world. We believe this mission is our ethical duty to the patients we serve. The annual CNS Summit and its year-round programming are designed to encourage and facilitate open channels of communication and data sharing across all disciplines of medical research—with the ultimate goal of achieving optimal patient outcomes. 

In this abstract supplement, we’ve organized CNS Summit 2021 poster abstracts into the following groups for your convenience and easy reference:

• Biomarkers
• Digital Tools and Technology
• Investigative Drug Compounds and Therapies
• Patient Assessment
• Social Issues/Global Challenges in Healthcare Research and Practice
• Trial Methodology

You will also find an alphabetical index by author and poster title on pages 15 and 16 of this publication.

We hope you find the CNS Summit 2021 Abstracts of Poster Presentations supplement to Innovations in Clinical Neuroscience informative and that it provides you with a useful snapshot of the research being presented at the annual CNS Summit event. Visit www.cnssummit.org for more information.

As always, we welcome your feedback and participation.

Sincerely,

Amir Kalali, MD

Editor, Innovations in Clinical Neuroscience

---

Table of Contents

Biomarkers 

• EEG biomarkers as clinical outcome measures in Lewy body dementia and Alzheimer’s disease

Digital Tools and Technology

• Application of E-Value analysis to gauge unmeasured confounding in real-world data (RWD) studies
• How to use Natural Language Processing to detect, categorize and further support documentation of clinical trial issues
• JINKO—The first clinical trial simulation platform to refine and augment the traditional pharmaceutical research and development process 
• RNA editing tools to improve drug development in mental health
• Using wearables and artificial intelligence to improve diagnostic decisions and treatment in youth with attention-deficit hyperactivity disorder

 Investigative Drug Compounds/Therapies

• Adjunctive troriluzole—a novel glutamate modulator—in patients with obsessive compulsive disorder: impact of baseline disease severity on treatment outcomes
• A Phase II clinical trial evaluating the efficacy, safety, tolerability, and pharmacokinetics of PRAX-944 in adults with essential tremor
• Modulation of stress responses in acute and chronic variable immobilization models following treatment with Teneurin C terminal associated peptide
• PRAX-114 produces robust increases in EEG beta frequency power, a translational biomarker of GABA_A receptor modulation, without dose-limiting sedation
• The reduction of opioid withdrawal symptoms and modulation of behavioral response in mice following treatment with Teneurin C terminal associated peptide
• TNX-601 CR: a once-daily formulation of tianeptine in development for the treatment of major depressive disorder in the United States
• Translational pharmacology of PRAX-944, a novel T-type calcium channel blocker for the treatment of essential tremor 

Patient Assessment

• Challenging PANSS items for raters in adolescent schizophrenia trials: expansion of earlier findings
• Investigating the adaptation of sleep scoring algorithms for pediatric data

Social Issues/Global Challenges in Healthcare Research and Practice

• Applying best practices in teaching, learning, and assessment to the University of Oxford’s Post Graduate Diploma in Global Health Research 
• Assessing comorbid behavioral health conditions, healthcare resource utilization, and medication use in patients seeking mental healthcare within a large, representative, real-world multiple sclerosis registry
• Diversity, equity, and inclusion: Alzheimer’s disease and dementia research-related global challenges, solutions, and opportunities
• Impact of COVID-19 pandemic on seriously mentally ill and nonpsychiatric control subjects in clinical trials
• Societal costs of schizophrenia in 2020

Trial Methodology

• Best practices in clinical trials of antidepressants: overcoming challenges to optimize success
• Best practices for online clinical trial recruitment in psychiatry: time to first contact  
• Furthering our understanding of participant retention: an analysis of research sites’ study data and methodological correlates
• Operationalizing the enhancement of diversity in clinical trials
• Optimizing assay sensitivity by combining exclusion of highly variable pain subjects with adjusted analysis
• Pioneering a decentralized clinical trial in depression
• Placebo response rates in randomized, controlled psychiatric studies over the past 10 years
• The financial burden of screening and enrolling ineligible subjects in clinical trials

---

Biomarkers

EEG biomarkers as clinical outcome measures in Lewy body dementia and Alzheimer’s disease

Authors: Chris Berka,¹ Amir Meghdadi,¹ Joanne Hamilton,^2,3 Bradley F. Boeve,⁴ Erik K. St. Louis^4,5

Affiliations: ¹Advanced Brain Monitoring, Carlsbad, CA, USA; ²Advanced Neurobehavioral Health of Southern California, San Diego, CA, USA; ³Scripps Health, La Jolla, CA, USA; ⁴Center for Sleep Medicine, Departments of Neurology and Medicine, Division of Pulmonary and Critical Care Medicine, Mayo Clinic College of Medicine and Science, Rochester, MN, USA; ⁵Department of Clinical and Translational Research, Mayo Clinic Health System Southwest Wisconsin, La Crosse, Wisconsin, USA

Objective: To demonstrate the specificity of resting-state electroencephalogram (EEG) biomarkers for Alzheimer’s disease (AD) and Lewy body dementia (LBD).

Methods: Participants with AD (n=26), LBD (n=16), and age-matched controls (HC, n=56) were recruited. Diagnosis criteria for AD included a presence of objective cognitive impairment in the memory domain and at least one other cognitive domain and decline in activities of daily living. Diagnosis of LBD were based on the recommendations of the DLB Consortium (McKeith, 2017). EEG power spectral density (PSD) during five-minute resting-state, eyes closed, were computed. PSDs were characterized by measuring the power at low frequencies (Delta and Theta bands, 1–7Hz), the frequency and prominence of the dominant rhythm, and its posterior dominance. AD and LBD groups were compared against HC and EEG measures, and significant differences were identified. The EEG biomarkers that were specific to either AD or LBD were identified.

Results: Compared to the control group, both AD and LBD groups exhibited EEG slowing (increased power in Delta band [1–3Hz]). This effect was largest in the LBD group. Delta power was positively correlated with cognitive fluctuation (CAF) in patients with LBD and negatively correlated with MMSE in patients with AD. Slowing of the dominant rhythm (DR) was observed in all patient groups, more so in patients with LBD. A significant reduction in power of the DR (prominence of the peak in frequency domain) was observed only in the AD group. While the AD group evidenced normal spatial distribution of the dominant rhythm (higher at posterior sites), this posterior dominance was lost in the LBD group.

Conclusion: Resting-state EEG measures can differentiate LBD from AD. Converging evidence including the present study suggests that these measures might be associated with disease severity, providing an inexpensive and sensitive outcome measure in clinical trials. 

Funding/financial disclosures: Chris Berka: CEO, co-founder, and share holder of Advanced Brain Monitoring; Amir Meghdadi: employee of Advanced Brain Monitoring

Digital Tools and Technology

Application of e-value analysis to gauge unmeasured confounding in real-world data (RWD) studies

Authors: Jessica Paulus, ScD; Cristi Cavanaugh, MHS; Ia Topuria, MPH; Zhaohui Su, PhD

Affiliations: All with OM1 Inc., Boston, MA, USA

Objective: Sensitivity analysis is an essential practice in assessing how robust real-world (RWD) results are to potential unmeasured confounding. The E-value was recently introduced as a measure of the evidence for causality in observational studies that are subject to confounding, it but has been used relatively infrequently in research practice to date. 

Methods: A set of 10 statistically significant effect estimates with 95-percent confidence interval (CI) were abstracted from published RWD studies across an array of therapeutic areas. E-values for each point estimate and CI were calculated using software available at www.evalue-calculator.com. 

Results: The selected studies described treatment-outcome associations ranging from a hazard ratio of 0.62 to 7.04 and sample sizes from approximately 5,000 to 5,131,00 patients. E-values for these point estimates ranged from 1.71 to 13.56, with 90 percent indicating “moderate” uncontrolled confounding effects (E-value ≥2) that would be required to fully explain the observed association. The E-values for the CIs ranged from 1.43 to 6.08, with only 30 percent considered “moderate” confounding effects. 

Conclusion: E-values are computationally simple sensitivity analyses for RWD associations evaluating causality but require significant subject matter expertise to judiciously interpret them. E-values might be useful tools in evaluating RWD-based studies used to support regulatory or clinical decision making. 

Funding/financial disclosures: None to report 

How to use Natural Language Processing to detect, categorize, and further support documentation of clinical trial issues

Authors: Nicolas Huet,¹ Laura Trotta,¹ Steve Young,² Patrick Hugues²

Affiliations: ¹CluePoints S.A., Louvain-la-Neuve, Belgium; ²CluePoints Inc., King of Prussia, PA, USA

Objectives: In recent years, there have been major advances in deep learning, with successful applications in computer vision, natural language processing (NLP), and medical image analysis. With larger volumes of data continually being generated by clinical research activities, there is a growing interest in applying those techniques to further inform ongoing and future clinical investigations. This research work highlights how NLP can be used to improve the detection, categorization, and documentation of issues within clinical trials. 

Methods: We leveraged recent developments in NLP to train an algorithm in analyzing the text entered by study teams to document risk signals generated by central monitoring activities. The algorithm was applied to a database of more than 70,000 risk signals from 444 studies to automatically detect and categorize issues. 

Results: The algorithm detected more than 13,000 issues with various root causes, including data entry mistakes, misunderstanding of study protocol, noncompliance to Good Clinical Practice protocols, and miscalibration of equipment. 

Conclusion: These results suggest that NLP can effectively be applied to detect and document clinical trial issues to further inform ongoing and future trials and improve the efficiency and quality of clinical investigations.

Funding/financial disclosures: NH, LT, SY, and PH are employees of CluePoints. LT, SY, and PH hold stock in CluePoints.

JINKO—The first clinical trial simulation platform to refine and augment the traditional pharmaceutical research and development process 

Authors: François-Henri Boissel, Frédéric Cogny, Jean-Pierre Boissel 

Affiliations: All authors with Novadiscovery

Objective: Novadiscovery (NOVA) has developed the clinical trial simulation platform— JINKO­—to generate disease models and predict clinical outcomes prior to studies in humans. JINKO is based on computational modeling that combines biomedical knowledge with existing data. With JINKO, biotech and pharma companies can enhance their drug research and development (R&D) programs and run virtual clinical trials in virtual populations to test treatment scenarios, improve study design and trial execution, and optimize resource allocation. 

Methods: JINKO’s modeling simulations represent virtual patients, disease pathophysiology, and drugs in thousands of biological entities, which allows the design, execution, and refinement of clinical trials and other study-related factors. Therapeutic strategies, outcomes, and hypotheses can be tested in a multitude of scenarios corresponding to real-life clinical trials (including parallel, cross-over, and external control arms [ECAs]). Each modeling and simulation step is documented and seamlessly traceable to the primary source of information to ensure full auditability. 

Results: The JINKO platform has successfully been used for dose optimization, combination therapies, biomarker development, target analysis, and responder stratification, accelerating research by leveraging virtual patients. Specific applications of JINKO are the generation of ECAs and guidance of adaptive trial designs. 

Conclusion: The JINKO clinical trial simulation platform helps biotech and pharma companies generate real-world-like evidence to make informed decisions, mitigate the R&D risk, accelerate clinical research programs, and reduce overall development cost.

Funding/financial disclosures: All authors employed by Novadiscovery, the developer of the clinical trial platform.

RNA editing tools to improve drug development in mental health

Authors: Jean-Daniel Abraham, Christopher Cayzac, Nicolas Salvetat, Francisco Jesus Checa-Robles, Benjamin Dubuc, Jacques Dainat, Charline Trento, Diana Vetter, Dinah Weissmann

Affiliations: All with ALCEDIAG/Sys2Diag, Montpellier, France

Objective: ALCEDIAG is a personalized medicine company that develops ribonucleic acid (RNA) editing biomarkers and tools for drug development and diagnosis. RNA editing is an epigenetic mechanism with a key role in mental health and central nervous system (CNS) disorders. With more than 10 years of expertise and research and development (R&D) in this field, ALCEDIAG is at the cutting edge of world scientific research. Our ambition is to dramatically change mental illness diagnosis and management by introducing precision medicine, as well as accelerate biomarker-driven clinical trials in CNS to improve patient care.

Methods: ALCEDIAG takes advantage of its unique integrated RNA editing and artificial intelligence (AI) platforms—EDITECH®—and its expertise to identify novel targets for different indications. After an editome, biostatistics and network, and functional analyses, this process allows selection of innovative targets for drug development and/or diagnostic biomarkers.

Results: The company has identified a panel of RNA editing blood biomarkers to diagnose depression and evaluate suicide risk. For its proprietary tests, ALCEDIAG ran from biomarkers discovery to clinical validation and regulatory approval. For example, EDIT-B® is a blood test differentiating unipolar and bipolar depression, clinically validated on a 255-patient cohort, showing sensitivity and specificity of 85 percent. ALCEDIAG also collaborates with leading pharmaceutical companies to apply its technology platform across a broad range of therapeutic areas for clinical and preclinical development. 

Conclusion: Based on its unique know-how, ALCEDIAG makes it possible to identify innovative RNA editing targets for use in brain, blood, and cell lines as biomarkers, to design patient stratification tools and companion tests. 

Disclosures: All authors employed by ALCEDIAG/Sus2Diag

Using wearables and artificial intelligence to improve diagnostic decisions and treatment in youth with attention-deficit hyperactivity disorder 

Authors: Rich Brancaccio, Joe Koziak, Lindsay E. Ayearst

Affiliations: All authors with Revibe Technologies

Objectives: The rise of wearable devices enables constant data collection and allows wearable sensors to leverage artificial intelligence (AI) tools to provide novel metrics for clinical decision making. This research explores the ability of a new wearable sensor to provide meaningful predictive analytics that could improve the diagnosis and treatment of attention-deficit hyperactivity disorder (ADHD) in youth. 

Methods: Data from 1,646 users were explored, including focus rate (FR; % of time spent on-task), attention span (AS; length of potential workflow), and activity level (motion sensor data). Six-axis accelerometer data were collected and analyzed using deep learning to identify and track behaviors specifically related to ADHD. 

Results: Average AS increased from 10.3 minutes at baseline to 26 minutes after three weeks (p<0.001, d= 0.57). Average FR increased from 58.4 percent at baseline to 68.7 percent after three weeks (p<0.001, d= 0.50). Interesting correlations were found between activity patterns and on-task behavior (e.g., greater activity level in the morning led to increased FR later in the day). 

Conclusion: Having a wealth of real-world data points on a given individual makes it possible, using deep learning methods, to identify novel trends and correlations in large, aggregated cohorts, as well as in individuals on a personalized basis, that were previously undetectable. These data have the potential to increment the accuracy of diagnostic decisions and inform personalized interventions by predicting what specific daily behaviors and actions might benefit focus and attention for a given youth that can be tracked for efficacy and modified through continuous real-time monitoring.

Funding/financial disclosures: Rich Brancaccio: Founder and Chief Innovation Officer of Revibe Technologies, owns stock, and is a patent holder (US10624590B2/US20190254522A1); Joe Koziak: CEO of Revibe Technologies, owns stock options; Lindsay Ayearst: Chief Scientific Officer at Revibe Technologies, owns stock options.

Investigative Drug Compounds/Therapies

Adjunctive troriluzole—a novel glutamate modulator—in patients with obsessive compulsive disorder: impact of baseline disease severity on treatment outcomes

Authors: Loren Aguiar, MD;¹ Carolyn Rodriguez, MD;² Alexander Bystritsky, MD;³ Azim Munivar, MD;¹ Chris Pittenger⁴

Affiliations: ¹Biohaven Pharmaceuticals; ²Department of Psychiatry, Stanford University School of Medicine,;³CalNeuro Research Group; ⁴Department of Psychiatry, Yale School of Medicine

Objective: Glutamatergic dysfunction has been implicated in the pathophysiology of obsessive-compulsive disorder (OCD), and troriluzole is a novel glutmate modulator currently in development that might normalize this dysfunction. This proof-of-concept study evaluated the efficacy of troriluzole as adjunctive therapy added to standard-of-care (SOC) medication. 

Methods: The 12-week, double-blind, placebo-controlled trial tested the effect of troriluzole 200mg versus placebo. Subjects were required to have a Yale–Brown Obsessive Compulsive Scale (Y-BOCS) score of greater than 18 while on treatment with a stable dose of a SOC medication. The primary endpoint was the change in the Y-BOCS total score from baseline to the Week 12 timepoint. Subanalysis for illness severity included only the subset of patients with baseline Y-BOCS of 26 or greater. 

Results: Two-hundred and forty-four patients were enrolled in the study. Troriluzole treatment resulted in a numerically greater improvement versus placebo in the change from baseline in the Y-­BOCS during all efficacy assessment visits. At Week 8, the mean Y­BOCS change from baseline was -5.1 points for the troriluzole group (n=96) versus ­3.6 points for the placebo group (n=108), (nominal p value=0.041). At Week 12, the mean Y-­BOCS change from baseline was ­5.9 points for the troriluzole group (n=99) versus ­4.9 points for the placebo group (n=102) (p=0.22). 

In post-hoc analyses, the troriluzole treatment difference compared to placebo was greater both at Week 8 and Week 12 in subjects with more severe OCD symptoms at baseline (Y­-BOCS total score ≥26). At Week 8, the Y-BOCS mean change from baseline in this subset was -5.7 points for the troriluzole group (n=66) versus ­3.8 for the placebo group (n=76) (nominal p­ value=0.051). At Week 12, the mean Y-­BOCS change from baseline was ­6.7 points for the troriluzole group (n=69) versus ­5.0 for the placebo group (n=73) (nominal ­­p value=0.105).

Conclusion: This proof-of-concept study in adult patients with OCD having an inadequate response to SOC treatment revealed a consistent treatment benefit of adjunctive troriluzole over time. Subjects with more severe illness at study entry demonstrated larger effect sizes. Phase III studies with adjunctive troriluzole for the treatment of OCD are currently ongoing (NCT04641143 and NCT04693351), with modifications to trial design including a larger sample size, a higher Y-BOCS severity score cut-off, and higher drug dose. 

Funding/financial disclosures: Funded by Biohaven Pharmaceuticals, Inc.

A Phase II clinical trial evaluating the efficacy, safety, tolerability, and pharmacokinetics of PRAX-944 in adults with essential tremor

Authors: Gabriel Belfort, Shane Raines, Ted Snyder, Bernard Ravina

Affiliations: All with Praxis Precision Medicines, Boston, MA, USA 

Objectives: Essential tremor (ET) is the most common movement disorder, with high unmet patient needs. Treatment options are limited and discontinuation rates are high due to poor tolerability and modest efficacy. We present preliminary findings from a Phase II clinical trial evaluating efficacy, safety, tolerability, and pharmacokinetics in adults with ET.

Methods: This is an ongoing, open-label, two-part trial in adults (18–75 years of age) with ET (NCT05021978). In Part A, participants received PRAX-944 20mg orally once-daily for seven days, followed by 40mg orally once-daily for seven days. Eligible participants were receiving either no medications or one stable dose tremor medication. The primary outcome was upper limb tremor assessed by the Essential Tremor Rating Assessment Scale (TETRAS-UL). Secondary outcomes included other measures of tremor severity, as well as safety and tolerability.

Results: Seven participants received PRAX-944 (Part A) with six completing all study visits. Mean baseline TETRAS-UL score was 12.4 (range 10–15; corresponding to moderate disease). Five (83%) participants included in efficacy analysis were taking concomitant propranolol during the intervention period. Preliminary results revealed a mean reduction in TETRAS-UL score of 2.83 points on Day 14, corresponding to a 42-percent reduction in upper limb tremor amplitude. PRAX-944 20- and 40mg doses were generally well-tolerated.

Conclusion: Our findings provide preliminary evidence that PRAX-944 can reduce tremor symptoms in ET at well-tolerated doses. Continuing investigations, including an expanded dose range up to 120mg daily and a placebo arm (Part B), and an eight-week, dose-range finding, placebo-controlled trial (NCT05021991), will determine optimal doses for evaluation in later phase studies. 

Funding/financial disclosures: Funded by Praxis Precision Medicines; all authors employed by and/or stockholders of Praxis Precision Medicine

Modulation of stress responses in acute and chronic variable immobilization models following treatment with Teneurin C terminal associated peptide

Authors: Andrew Slee,¹ Robert Stein,¹ Jennifer Buell,¹ Garo Armen,¹ David Lovejoy,² Alexa Buffa³

Affiliations: ¹Protagenic Therapeutics, New York, NY; ²University of Toronto, Toronto, Ontario, Canada; ³Agenus, Lexington, MA

Objective: Teneurin C terminal associated peptide (TCAP) is a natural peptide that plays a critical role in stress. Our objective was to determine whether TCAP modulates stress responses, as expressed by a selected biomarker, and performance in a rat model.

Methods: In an acute tube-restraint model, rats were immobilized for 40 minutes. TCAP was administered at selected doses in preventative or treatment mode. Corticosterone levels were determined before stress and at five and 30 minutes upon release. Rat activity was monitored when in the open-field apparatus. In a chronic paradigm, rats were restrained daily for variable durations over 12 days, subjected to noxious stimuli, overcrowding, or tilt positioning. TCAP was administered subcutaneously or intranasally. A corticotropin releasing factor (CRF) antagonist was also studied and administered daily for 10 days.

Results: Subcutaneous TCAP administration reduced corticosterone levels, and dose response was linear. TCAP given intranasally reduced corticosterone levels at doses significantly lower than TCAP given subcutaneously. TCAP was more efficacious than CRF antagonist. Under variable chronic stress conditions, TCAP reduced corticosterone levels by 45 percent, and animals moved freely in the open field. In this chronic model, the CRF antagonist failed to reduce corticosterone levels or stress responses in the open field.

Conclusion: TCAP demonstrated control of negative stress responses in a rat model, suggesting its potential utility in modulating stress responses in humans. 

Funding/financial disclosures: Andrew Slee and Robert Stein employed by Protagenic Therapeutics; Jennifer Buell and Garo Armen directors of Protagenic Therapeutics; David Lovejoy scientific advisor at Protagenic Therapeutics; Alexa Buffa employed by Agenus, Inc; no conflicts of interest to report. 

PRAX-114 produces robust increases in EEG beta frequency power, a translational biomarker of GABA_A receptor modulation, without dose-limiting sedation

Authors: Zoë Hughes, Shane Raines, Liam Scott, Nicholas DeMartinis, Gabriel Belfort, Marion Wittmann

Affiliations: All authors with Praxis Precision Medicines, Boston, MA, USA 

Objective: PRAX-114 is a gamma-aminobutyric acid (GABA) A receptor-positive allosteric modulator (GABA_AR-PAM) neuroactive steroid with a preference for extrasynaptic versus synaptic GABA_AR. GABA_AR-PAMs cause increases in EEG beta-frequency band power; therefore, PRAX-114 effects on quantitative EEG (qEEG) were measured in rodents and healthy human participants as a translational biomarker of GABA_AR activation to inform dose selection for future major depressive disorder (MDD) trials. 

Methods: The effects of PRAX-114 (0.3–10mg/kg i.p.) on rat qEEG were measured in male rats in a within-subject, crossover study (n=7–8/group). In healthy human participants (n=9/group), qEEG effects of PRAX-114 (15, 30 or 60mg, qPM) were measured in a parallel group, 14-day, multiple ascending dose trial investigating safety and tolerability of PRAX-114 (ACTRN12618000650291). For rats and humans, spectral analysis of qEEG recorded post-dose was compared to pre-dose baseline using mixed model repeated measures.

Results: In rats, PRAX-114 produced dose-dependent increases in qEEG beta-power (14–32Hz) with peak effects 1 to 2 hours after administration, reaching a maximum of approximately 2.5-fold greater power versus baseline. In humans, PRAX-114 (15–60mg) also elicited dose-dependent increases in beta-frequency (12.5–30Hz) power, reaching a maximum of approximately 2.7-fold greater power versus baseline. All doses in humans were well tolerated, with only mild somnolence reported. 

Conclusion: In rats and humans, PRAX-114 produced robust increases in beta-power. In rats, the dose that increased beta-power approximately 1.6-fold was more than 11 times lower than that which previously demonstrated reduced locomotor activity, while also showing antidepressant-like effects. In humans, the 60mg dose achieved large beta-power effects without dose-limiting sedation, suggesting extrasynaptic GABA_AR preference is associated with substantial pharmacodynamic effects and favorable tolerability. 

Funding/financial disclosures: All work was funded by Praxis Precision Medicines. All authors are employees of Praxis Precision Medicines and may be stockholders.

The reduction of opioid withdrawal symptoms and modulation of behavioral response in mice following treatment with Teneurin C terminal associated peptide

Authors: Andrew Slee,¹ Robert Stein,¹ Jennifer Buell,¹ Garo Armen,¹ David Lovejoy,² Alexa Buffa³

Affiliations: ¹Protagenic Therapeutics, New York, NY; ²University of Toronto, Toronto, Ontario, Canada; ³Agenus, Lexington, MA, USA

Objective: Teneurin C terminal associated peptide (TCAP) is a natural peptide that plays a critical role in the negative effects of stress. Our goal was to evaluate TCAP’s effect on behavioral responses to naloxone administration and assess corticosterone levels, as a biomarker of stress response, in a modified mouse opioid withdrawal model. 

Methods: In the modified Saelens test, mice were administered seven morphine injections over two days. TCAP was administered subcutaneously prior to study initiation or just before the naloxone injection. Responses to delivery of naloxone were observed directly, along with camera recording. The duration of observation after naloxone ranged from 20 to 30 minutes. A small molecule corticotropin releasing factor (CRF) antagonist was also evaluated in the same model. 

Results: TCAP demonstrated a linear dose response curve and reduced the stereotactic jumping following precipitated withdrawal from an opiate. Reduction in the stereotypic response was seen in either a preventive mode or a treatment-like paradigm. TCAP also reduced corticosterone levels in the mice, as a consequence to stress. The mobility of the test animals in any of the studies was not impacted by TCAP, compared to control agents. TCAP was significantly more potent that the CRF antagonist that was evaluated.

Conclusion: TCAP demonstrated prophylactic potential in a mice model, suggesting its potential utility in addiction responses in humans. 

Funding/financial disclosures: Andrew Slee and Robert Stein employed by Protagenic Therapeutics; Jennifer Buell and Garo Armen directors of Protagenic Therapeutics; David LoveJoy scientific advisor at Protagenic Therapeutics; Alexa Buffa employed by Agenus, Inc; no conflicts of interest to report. 

TNX-601 CR*: a once-daily formulation of tianeptine in development for the treatment of major depressive disorder in the United States

Authors: Gregory M. Sullivan,¹ David T. Hsu,¹ Ashild Peters,¹ Perry Peters,¹ Siobhan Fogarty,¹ Regina Kiu,¹ Bernd Meibohm,² Seth Lederman¹

Affiliations: ¹Tonix Pharmaceuticals, Inc. (Tonix); ²University of Tennessee Health Science Center

Objective: Tianeptine sodium 12.5mg (Stablon®), administered three times daily (TID), is an atypical antidepressant approved in Europe, Asia, and Latin America for clinical depression. The efficacy of tianeptine sodium has been shown to be comparable to tricyclics and selective serotonin reuptake inhibitors (SSRIs), with superior tolerability. We previously identified a new crystalline salt, tianeptine oxalate (TNX-601), which has improved physiochemical properties relative to amorphous tianeptine sodium. In this study, we developed a precursor once-daily formulation with pharmacokinetics (PK) comparable to tianeptine sodium, dosed TID during daytime. 

Methods: Single-center, open-label, sequential period Phase I study of 12 healthy participants. PK of tianeptine and active metabolite MC5 were determined, and safety was assessed. 

Results: PKs were similar for tianeptine sodium 12.5mg and TNX-601 13.1mg (which contained equimolar tianeptine base). A modified-release (MR) prototype of tianeptine oxalate (TNX-601 MR 39.4mg) was selected from several tested compounds due to similar PK to simulated tianeptine sodium TID dosing and minimized food effect. Multidose steady-state simulations of selected prototype support the concept that a final once-daily formulation of TNX-601 controlled-release (CR) tablets, will be pharmacodynamicly similar to TID tianeptine sodium. TNX-601 and prototypes were well-tolerated, and adverse events were consistent with the established safety profile of tianeptine sodium.

Conclusion: The selected precursor prototype demonstrated PKs appropriate for once-daily dosing, with minimal food effect, which is a potential adherence advantage over TID tianeptine sodium. These Phase I findings in the selected precursor prototype support upcoming Phase II testing of the once-daily TNX-601 CR (controlled-release tianeptine oxalate 39.4mg and naloxone 1mg) tablet in MDD in early 2022.

Funding/financial disclosures: Developmental studies funded in total by Tonix; Bernd Meibohm PK consultant to Tonix; remaining authors employed by Tonix and/or hold stock and/or stock options in the company. *TNX-601 CR is an investigational new drug and has not been approved for any indication. 

Translational pharmacology of PRAX-944: a novel T-type calcium channel blocker for the treatment of essential tremor 

Authors: Corey Puryear, Liam Scott,
Gabriel Belfort, Shane Raines, Zoë Hughes, Bernard Ravina, Marion Wittmann

Affiliations: All with Praxis Precision Medicines, Boston, MA, USA 

Objective: Essential tremor (ET) is the most common movement disorder, with clear need for new therapeutic options. Mounting evidence suggests that tremor is caused by increased neuronal burst firing and oscillations in cerebello-thalamo-cortical circuitry, thought to be driven by T-type calcium channel (TTCC) activity. TTCCs regulate sigma encephalographic power during nonrapid eye movement sleep. To assist in informing dose selection for future efficacy trials, we used sigma-power as a translational biomarker to assess whether pharmacodynamically active doses of PRAX-944 would be well-tolerated.

Methods: Rodent harmaline-induced tremor and spontaneous locomotor activity were used to assess efficacy and tolerability of PRAX-944 (0.1–30mg/kg, taken by mouth), respectively. Sigma-power was used as a translational biomarker of TTCC blockade following PRAX-944 treatment in rats and in a Phase I dose-escalating (5–120mg, administered every morning) clinical trial in healthy participants (ACTRN12620000675921).

Results: In rats, PRAX-944 dose-dependently reduced tremor by 50 percent and 72 percent at 1mg/kg and 3mg/kg doses, respectively, without locomotor side-effects; these doses reduced sigma-power by 30 to 50 percent. In healthy participants given repeated doses of PRAX-944, sigma power was similarly reduced by 33 to 47 percent at 10 to 100mg, with no further reduction at 120mg. All doses were well-tolerated. 

Conclusion: Administration of PRAX-944 in rats and humans produced strong and consistent effects on sigma-power, which might represent a robust and translatable biomarker of TTCC blockade. In rats, PRAX-944 reduced sigma-power at concentrations that reduced tremor without locomotor side effects. In healthy participants, comparable sigma-power reductions indicate that TTCC blockade was achieved at well-tolerated doses that could hold therapeutic promise for tremor reduction in patients with ET.

Funding/financial disclosures: All work was funded by Praxis Precision Medicines. All authors are employees of Praxis Precision Medicines and may be stockholders.

Patient Assessment

Challenging PANSS items for raters in adolescent schizophrenia trials: expansion of earlier findings

Authors: Joan Busner,^­1,2 David G. Daniel,¹ Robert L. Findling,²

Affiliations: ¹Signant Health; ²Virginia Commonwealth University School of Medicine, Department of Psychiatry

Objective: Measures designed for adults, such as the Positive and Negative Syndrome Scale (PANSS), are frequently used in adolescent schizophrenia trials. To identify PANSS items for which raters in pediatric trials might have particular difficulty, we previously reported PANSS item scoring variability from two standardized videos in three adolescent schizophrenia industry trials. We have since secured data from two additional trials and videos, allowing for combined analysis and expansion of our findings. 

Methods: Standard deviations (SDs) were calculated for each of the 30 PANSS items scored by 408 investigators/raters from 23 countries who had viewed one of four standardized adolescent patient videos as part of the qualification process for their respective clinical trial. PANSS item SDs were rank ordered by variability, per video, and compared for cross-video similarity using Kendall W. 

Results: Variability rankings of the 30 PANSS items for the four videos were statistically similar: W=0.57, p<0.0001. Three PANSS items were ranked among the 10 most variable in all four videos: N4, P7, and P4; and two were ranked among the 10 least variable in all four videos: P3 and G14. 

Conclusion: Variability rankings across four adolescent videos were statistically similar, suggesting that scoring ease or difficulty of individual PANSS items is independent of the specifics of the patients rated. Identification of challenging PANSS items for pediatric raters allows targeted training and in-study intervention. Findings differed from those noted in adult videos, making even more clear the need for focused attention and perhaps modification for PANSS items when applied to the pediatric age range. 

Funding/financial disclosures: Presented in part at the 2021 Congress of the Schizophrenia International Research Society (SIRS) virtual meeting, April 17-21, 2021; financial support provided by Signant Health.

Investigating the adaptation of sleep scoring algorithms for pediatric data

Authors: Georg Dorffner,^1,2 Georg Gruber,² Marco Hirner,² Peter Sorantin,² Lukas Pinka,¹ Patrick Erber,¹ Tawab Shoja,¹ Ruth Luigart,¹ Barbara Schneider,³ Peter Anderer,²

Affiliations: ¹Medical University of Vienna, Section for Artificial Intelligence, Vienna, Austria; ²The Siesta Group, Vienna, Austria; ³Children’s Hospital St. Marien, Landshut, Germany

Objectives: We investigated the applicability of previously developed sleep scoring algorithms to pediatric data. Our hypothesis was that the main quantitative characteristic of infant electrophysiology is reflected by two important values: the average amplitude of the EEG and the cutoff frequency between the alpha and the theta band.

Methods: We used a data set of approximately 150 recordings from a children’s sleep laboratory, manually scored by two independent experts, covering several age groups from 1 to 14 years of age. Based on the above hypothesis, we applied our standard sleep scoring algorithm in nine different settings, by varying the average amplitude (original, scaled by a factor of 75% or 50%) and the cutoff between alpha and theta (7Hz, as in the original algorithm, 6Hz, or 5Hz). We validated the results by calculating a set of standard sleep endpoints (e.g., sleep efficiency, percentage of time in each sleep stage) and comparing the values of these endpoints to those derived by a visual expert scorer in terms of bias (e.g., average difference) and consistency (e.g., correlation coefficient between the two methods). 

Results: Results demonstrated that, indeed, varying the two main parameters led to much improvement in automated scoring, especially for younger age groups, compared to the visual baseline. For all three age groups—1 to 5 years, 5 to 9 years, and 9 to 12 years—the optimal configuration turned out to be a 50-percent amplitude scaling and a 5Hz cutoff value for alpha/theta. 

Conclusion: These results constitute an important step toward full validation of computer-supported sleep scoring of infant and child data. 

Financial/funding disclosures: Georg Dorffner, Georg Gruber, Marco Hirner, and Peter Anderer: employees and (except for Marco Hirner) shareholders of The Siesta Group, a service provider for measuring electrophysiological signals including sleep in clinical trials; all other authors had no conflicts to disclose.

Social Issues/Global Challenges in Healthcare Research and Practice

Applying best practices in teaching, learning, and assessment to the University of Oxford’s Post Graduate Diploma in Global Health Research 

Authors: Jeremy Whitty,^1,2 Sinéad Whitty,^1,2 Trudie Lang,¹ James Whalen,^2,3

Affiliations: ¹The Global Health Network, Centre for Tropical Medicine and Global Health, University of Oxford; ²The Faculty of Capacity Development; ³ERG Clinical

Objective: The new Post Graduate Diploma in Global Health Research will bring training in global health research to participants who, up to now, could not access high-quality university education due to geographical, social, or financial constraints. 

Methods: Oxford’s first academic award designed to be taught 100-percent online is for those who fund, design, deliver, regulate, or implement research, whether in Massachusetts or Mozambique, giving participants a thorough understanding of the processes required to conduct effective research within the contextual realities in which they operate. We highlight the key steps in the design of the diploma, including a study with over 7,000 participants from 153 countries. We also demonstrate how accessibility was designed into the program, including scholarships and nontraditional pathways to entry. 

Results: Results to date include a successful pilot program, a curriculum approved by Oxford University’s rigorous review process, and permission by the universty to advertise the course. Future results will include in-depth monitoring and evaluation of the program after it starts in October 2022. 

Conclusion: We live in an era when diseases rapidly cross borders, yet the skills to implement global health research are scarce, surveillance is patchy, the experience to analyze data is inadequate, and much of the systemic support required to turn research into practice does not exist. A critical mass of scientific, management, and leadership skills to detect and stop the next pandemic are not located where they are needed most. Oxford University’s Diploma in Global Health Research was designed to help solve these challenges. 

Funding/financial disclosures: None to report

Assessing comorbid behavioral health conditions, healthcare resource utilization, and medication use in patients seeking mental health care within a large, representative, real-world multiple sclerosis registry

Authors: Haley S. Friedler, MPH; Kyra Mulder, BSPH; Shannon Cerf, MHSA, MBA, PharmD; Sarah Marks, BSN, RN; Ia Topuria, MS; Zhaohui Su, PhD; Gary Curhan, ScD, MD; Richard Gliklich, MD

Affiliations: All authors with 1OM1 Inc., Boston, MA, USA 

Objective: To describe comorbid behavioral health conditions, healthcare resource utilization (HCRU), and medication use among patients with multiple sclerosis (MS) who were seeking mental health care (SMHC).

Methods: Data were derived from the OM1 MS Registry (OM1, Boston, MA), a multisource real-world registry with linked healthcare claims and electronic medical records data on US patients with MS (2013–2021). All patients who were SMHC were included. Patients were considered to be SMHC if they had any encounter with a behavioral health specialist, a behavioral health-related procedure code, or a record for a behavioral health-related medication on or after the first observed MS diagnosis date (index). Patient characteristics were assessed pre-index; comorbidities, HCRU, and medication use included data after index.

Results: Of 19,133 patients with MS, 15,384 patients (80.4%) were SMHC. Patients who were SMHC were on average 49.3 years of age (standard deviation: 12.5), 78.4 percent female, 62.2 percent White, 3.9 percent Hispanic or Latino, 43.9 percent had commercial insurance, and 7.3 percent were current smokers. Most patients (71.3%) had a behavioral health comorbidity. Anxiety disorders (42.4%), depression (40.9%), and substance use disorders (37.3%) were most frequent. Patients most commonly saw psychiatrists (28.6%) and had non-substance use-related psychotherapy (25.8%). Most patients (87.8%) used behavioral medications. Highest use was anti-anxiety medications (63.3%), antidepressants (58.1%), and mood stabilizers (46.9%).

Conclusion: Most patients with MS were SMHC. Anxiety and depression were the most common behavioral health comorbidities, and patients were frequently treated with medications.

Funding/financial disclosures: All authors are employees of OM1 Inc.

Diversity, equity and inclusion: Alzheimer’s disease and dementia research-related global challenges, solutions, and opportunities

Author: Charles S. Wilcox, PhD, MPA, MBA

Affiliations: Praxis Research Consulting; Alzheimer’s Association Orange County; Alzheimer’s Impact Movement (AIM)

Objectives: To identify diversity, equity and inclusion challenges with recommended solutions within the global Alzheimer’s disease community that will provide guidance for advancing these goals in the United States (US). Our “ask” on the Equity in Neuroscience and Alzheimer’s Clinical Trials (ENACT) ACT, on this likely-to-be-approved legislation, is $60 million per year for five years. The need to maximize the $300 million cannot be overstated and, domestically, we can and should leverage applicable lessons learned from more than 150 other countries. 

Results: In the US, 28 percent of Hispanic, 30 percent of Asian, 31 percent of Native American, and 42 percent of African American caregivers reported feeling “Not listened to.” Sixty-three percent of patients with dementia and/or their caregivers in Southeast Asia, and 67 percent of patients with dementia and/or their caregivers in Africa reported that “their dementia symptoms were joked about.” Additionally, 40 percent of the public throughout foreign countries felt that “healthcare practitioners ignore people with dementia.” The elderly Pakistani community in Norway, like the elderly Moroccan community in Belgium, have remarkably similar barriers to ethnically diverse and economically challenged elderly communities throughout the US. Living below the poverty line, in poor housing conditions, with difficulties reading and writing, in addition to transportation challenges, exacerbates health disparities and markedly reduces opportunities for diverse, equitable, much-needed inclusion in dementia trials globally.

Conclusion: As countries such as The Netherlands have had success providing minority groups “personal budgets” to compensate family members as caregivers, outreach efforts to diagnose minority ethnic groups with more culturally appropriate assessment and diagnostic tools is a global challenge and opportunity.

Funding/financial disclosures: None to report

Impact of COVID-19 pandemic on seriously mentally ill and nonpsychiatric control subjects in clinical trials

Authors: Robert E. Litman, MD; Maria Fe Garcia-Rada, MPS

Affiliations: Both authors with CBH Health, CenExel Clinical Research

Objective: This study aimed to examine the impact of the COVID-19 pandemic on seriously mentally ill (SMI) patients, specifically relating to psychiatric morbidity, pandemic-induced stress, and ability to cope with pandemic-related precautionary measures, restrictions, and disruptions to daily life. 

Method: A cross-sectional survey study of 287 clinical trial patients was conducted. This sample included SMI patients (n=139) with a diagnosis of bipolar disorder (n=23), major depressive disorder (MDD, n=46), or schizophrenia (n=69) and nonpsychiatric controls (n=149), located at five clinical trial sites across the United States. A univariate analysis was performed to obtain general frequencies of the sample. Unpaired t-tests were used in comparing the groups on numerical variables, and an analysis of variance (ANOVA) was performed to identify differences when comparing three or more categories. 

Results: SMI patients were more likely to report wearing face masks, avoid large gatherings, and endorse the use of precautionary measures despite receiving a COVID-19 vaccine (p<0.001). In SMI patients, 70.3 percent (n=97) reported experiencing at least one episode of worsening mental health, 48 percent reported experiencing suicidal ideation, and 66 percent reported a need for increased mental healthcare due to COVID-19 distress. SMI patients reported higher levels of stress, compared to the controls, with MDD patients reporting the highest levels of stress (p<0.001). 

Conclusion: These findings demonstrate an increased vulnerability to symptom worsening in SMI patients during a pandemic and suggest the need to account for pandemic-induced psychological stress in clinical trials design, subject selection, and symptoms ratings.

Financial/funding disclosures: None to report

Societal costs of schizophrenia in 2020

Authors: Holly Krasa,^1,2 James Baumgardner,³ Kelly Birch,³ Jacki Chou,³ Kenneth Ching,³ Theresa Frangiosa,¹
Gordon Lavigne,² Andrew Palsgrove,¹ Kyi-Sin Than,³ Timothy Murphy²

Affiliations: ¹Blue Persimmon Group, Washington, DC, USA; ²Schizophrenia & Psychosis Action Alliance, Alexandria, VA; ³PRECISIONheor, Los Angeles, CA, USA

Objective: To measure the societal cost of schizophrenia and related disorders in the United States (US), including direct medical, direct nonmedical, and indirect costs to provide evidence necessary to support changes in population-specific healthcare and nonhealthcare services.

Methods: A prevalence-based total excess cost model was developed combining inputs from a targeted literature review and an analysis of Medical Expenditure Panel Survey (MEPS) data, adjusted to 2020 US dollars. Costs were calculated by subcategory and subpopulation and combined to estimate costs of schizophrenia and related disorders in the US in 2020. 

Results: The total costs of schizophrenia in 2020 were $281.6 billion based on a one-year prevalence estimate of 0.8 percent. Direct costs accounted for $62.2 billion ($27.2 billion, healthcare; $24.7 billion, supportive housing/homelessness; $16.8 billion, criminal justice system; $5.1 billion, Social Security, with $11.6 billion cost of living offsets). Indirect costs accounted for $219.6 billion ($36.1 billion, reduced quality of life; $41.6 billion, unemployment/reduced wages; $37.4 billion, shortened life expectancy; $20.8 billion, caregiver burden; $83.7 billion, caregiver unpaid labor). 

Conclusion: Despite the relatively low prevalence of schizophrenia and related disorders, excess costs attributable to the disease are substantial. Significant investments into data collection, research, and evidence-based system implementation are needed to ultimately impact health policy and improve outcomes for individuals with schizophrenia in a comprehensive system of care.

Funding/financial disclosures: HK, JB, KB, JC, KC, TF, AP, and KT are or were employed or contracted by organizations that received funding for conduct and oversight of the study. 

Trial Methodology

Best practices in clinical trials of antidepressants: overcoming challenges to optimize success

Authors: Paula Jacobsen, Nicholas DeMartinis

Affiliations: Both authors with Praxis Precision Medicines, Boston, MA, USA 

Objective: Despite considerable advances in pharmacological options for the treatment of major depressive disorder (MDD), conducting successful MDD clinical trials is challenging, primarily due to the subjective nature of assessments and a substantial placebo response. We aim to provide best practice recommendations for optimizing clinical trial design and execution to provide novel antidepressants the greatest chance of success for approval.

Methods: Factors that negatively impact signal detection in clinical trials, such as failure to enroll participants with the correct diagnosis, inclusion of participants with insufficient symptom severity, and issues with high placebo response, were reviewed. Approaches to addressing these factors, thereby increasing the probability of a successful trial, were outlined, and best practice recommendations were summarized.

Results: The key components of clinical trial best practices that can both mitigate the placebo effect and improve data reliability include using a straightforward, uncomplicated, randomized, controlled trial design with a substantial placebo cohort; standardizing participant selection by providing external oversight; standardized participant education on expectations regarding treatment response or nonresponse; directly monitoring participant adherence; rater training alignment with interactive monitoring of rating assessment quality; and establishing specific intervention plans should challenges arise for these approaches. Obstacles can be effectively addressed when relationships based on mutual commitment to scientific collaboration are maintained between the sponsor, sites, and clinical research organizations.

Conclusion: Clinical trials for antidepressants can be optimized for success when best practices are followed. A well-designed and well-executed clinical trial can produce robust results despite the inherent complexities of psychiatric conditions such as MDD.

Funding/financial disclosures: All work was funded by Praxis Precision Medicines. All authors are employees of Praxis Precision Medicines and may be stockholders.

Best practices for online clinical trial recruitment in psychiatry: time to first contact 

Authors: Jenicka Engler, Daniel Thorpe, Miriam Evans, Dennis Fagundo, Sarah Starling, Grace Koo, Hasiba Zandi, Hannah Tucker, Dyanna Domilici 

Affiliations: All with Adams Clinical 

Objective: Clinical trial recruitment issues are the single largest cause of delays. One solution has been recruiting via online advertising campaigns, yet this presents novel engagement difficulties compared to drawing from an established patient population. We examined the impact of time to first phone contact among a cohort from the Massachusetts Major Depressive Disorder (MDD) clinical trial seeking population on response to calls, likelihood of completing phone screening, and enrollment. 

Methods: Between August 2019 and July 2021, 12,277 potential study participants responded to online trial advertisements on Facebook, Instagram, and Google by submitting their contact information. Site recruiters then attempted to call participants to complete a phone screening interview. Initial calls were completed as soon as possible, and if not answered, several additional contacting attempts were made via call or text. 

Results: Time to first contact ranged from 2 minutes to 6.5 days (mean [M]=21.78 hours, standard deviation [SD]=23.5). Overall call response rate was 25 percent and phone screening completion rate was 48 percent. Call speed was a significant predictor of call response, with participants who were called sooner being more likely to answer the phone (β= –0.008, p<0.001), and to ultimately complete a phone screen interview (β= –0.002, p<0.05). 

Conclusion: Best practices for online trial recruitment should emphasize the initial phone contact being completed as soon as possible, ideally within the first hour after application. Given the increasing popularity of online recruitment of trial participants, time to contacting potential participants should be carefully considered by sites as a means of speeding potential trial enrollment. 

Funding/financial disclosures: All presenters are employees of Adams Clinical, an independent CNS research site that conducts industry-sponsored pharmaceutical trials. 

Furthering our understanding of participant retention: an analysis of research sites’ study data and methodological correlates

Authors: Elan A. Cohen,¹ Howard A. Hassman,¹ David P. Walling,² Vera M. Grindell,² John G. Sonnenberg,^3,4 Katarzyna Wyka,⁵ Brett A. English,^1,2 Jaclyn M. Lobb,¹ Djouher Hough,¹ Cassie L. Blanchard,¹ Mia Robb Stahler,¹ Larry Ereshefsky,^1,2,6

Affiliations: ¹Hassman Research Institute, Science Division; ²Collaborative Neuroscience Network; ³Uptown Research Institute; ⁴Northwestern University Feinberg School of Medicine; ⁵The City University of New York, Graduate School of Public Health and Health Policy; ⁶Retired Professor, University of Texas

Objective: Clinical trials in psychiatry are beset by participant dropouts post-randomization, leading to substantially reduced power, internal validity, and generalizability, as well as expanded trial duration with heightened expenses or the opposite, which is potential early study termination. This investigation adds further insight to dropout correlates previously empirically explored (e.g., study length) as well as new variables formerly not assessed (e.g., number of scales administered) to enhance attrition understanding and recommend retention strategies. 

Methods: Across six United States (US) research sites in the East, Midwest, and West Coast regions, 176 outpatient psychiatric clinical trials (e.g., schizophrenia, depression, bipolar depression, post-traumatic stress disorder [PTSD], attention deficit hyperactivity disorder [ADHD]) were evaluated. Analyses were conducted on completed placebo-control (PC), open-label (OL), and extension studies. Dropouts post-baseline were evaluated directly due to participant decisions or other factors (e.g., lost to follow-up and withdrew consent) rather than matters out of their control (e.g., investigator discretion due to labs or adverse events). 

Results: Pearson correlation analyses revealed, for PC studies across all indications and regardless of site location, retention was significantly correlated with higher visit frequencies (r=0.33; p=0.001), shorter study duration (r= –0.36; p<0.001), less scales (r= –0.29; p<0.01), and higher compensation (r=0.25; p<0.05). For extension studies, regardless of whether they were PC, premature discontinuation was significantly correlated with higher study visits (r=0.60; p<0.05) and study completion significantly correlated with higher compensation (r=0.59; p<0.05). 

Conclusion: The current study results compliment as well as add to previous dropout research and recommendations for protocols to increase retention while simultaneously enhancing data integrity. 

Financial/funding disclosures: None to report 

Operationalizing the enhancement of diversity in clinical trials

Authors: Stacey Versavel, Alicia Subasinghe, Kenasha Johnson, Nicole Golonski, Janna Muhlhausen, Pamela Perry, Raymond Sanchez

Affiliations: Cerevel Therapeutics, Cambridge, MA, USA

Objective: Adequate representation of diverse participants remains a notable challenge in the design and execution of clinical trials. Investment in representing affected populations and encouraging clinical trial participation across socioeconomic status and demographics should not be rate limiting. Early engagement in the clinical trial lifecycle to incorporate diversity, equity and inclusion (DEI) assists with gaining a precise understanding of a therapeutic intervention’s effects prior to commercial availability. 

Methods: A DEI guidebook was developed for clinical trial teams to address potential barriers that may impact clinical trial participation. The guidebook contains suggested strategies to inform feasibility while considering trial-specific objectives, operational parameters, phase of development, indications, and therapeutic target population(s). This guidebook acts as a primary resource focused on enrolling clinically relevant populations to provide sufficient information pertaining to the safety and efficacy of the therapeutic intervention. 

Results: The DEI guidebook is focused on the clinical trial lifecycle. Planning concentrates on eligibility and early engagement with participant and patient advocacy organizations. Start-up supports the informed selection of investigators and sites to ensure representation of geographically diverse populations and under-represented businesses. Conduct strategies reflect consideration of cultural differences, provision of financial support, and leveraging a visualization platform to track demographic representation. Closeout emphasizes continued communication to share learnings and cultivate partnership with clinical trial participants and patient advocacy organizations. 

Conclusion: Application of strategies in the DEI guidebook should enable more effective, evidence-based recommendations. This tool was designed to champion improvements to the representation of population(s) who will ultimately utilize, if approved, the therapeutic intervention under study. 

Funding/financial disclosures: Stacey Versavel, Alicia Subasinghe, Kenasha Johnson, Nicole Golonski, Janna Muhlhausen, Pamela Perry, and Raymond Sanchez employed by Cerevel Therapeutics.

Optimizing assay sensitivity by combining exclusion of highly variable pain subjects with adjusted analysis

Authors: Arthur Ooghe, Samuel Branders, Dominique Demolle, Alvaro Pereira

Affiliations: All with Tools4patient

Objective: The baseline pain variability (BPV) has often been presented as positively correlating with the placebo response (PR) and associated with a lack of consistency in the subjects’ pain evaluation. Excluding subjects with high BPV should then improve the precision of the treatment response. Another common method to increase the assay sensitivity is to adjust the analysis for covariates associated with the response. We aimed here at optimizing this assay sensitivity by combining both the exclusion of high BPV subjects with adjusted analysis.

Methods: We used data from a randomized, placebo-controlled study, which included 171 subjects with osteoarthritis. Up to 25 percent of the subjects with the highest BPV were excluded from our analysis. In parallel, an adjustment for the Placebell placebo covariate was tested, and the improvement in assay sensitivity was estimated.

Results: Subject exclusion did not significantly change the endpoint variance in the study, despite significant correlations between BPV and endpoints. On a positive note, this exclusion did not attenuate the performance of the placebo covariate adjustment, which improved the assay sensitivity by up to 37 percent.

Conclusion: The exclusion of subjects with high BPV can be used in combination with covariate adjustment to improve the assay sensitivity. Surprisingly, however, this exclusion had no impact on endpoint variance in this study. Overall, the adjustment seemed to bring larger gains in precision and power than the exclusion of subjects with high BPV.

Funding/financial disclosures: None to report 

Pioneering a decentralized clinical trial in depression 

Author: Christopher Reist, MD

Affiliation: Science 37

Objective: For this Phase II study, Science 37 conducted a first-of-its-kind decentralized clinical trial (DCT) in depression, enabling universal access to patients.

Methods: Implementing DCT is a complex undertaking, requiring a unified technology platform that orchestrates workflow, generates evidence, harmonizes data, and seamlessly connects the networks of patients, investigators, nurses, coordinators, and devices. The success of this approach was demonstrated in a randomized, controlled study evaluating the safety and efficacy of an adjunctive compound in patients with MDD with inadequate response to antidepressant monotherapy. All elements, including patient recruitment, telemedicine investigators, remote coordinators, mobile nurses, and a remote assessment team, were activated. The patient recruitment campaign included both digital direct-to-patient and physician-referral strategies, utilizing online content and behavioral targeting, chart review, and digital look-alike audiences, to identify eligible participants.

Results: Although the study is ongoing, the many complex elements described above are fully operational. Data quality is excellent, and subject experience is positive, with no dropouts or early terminations to date. Almost all participants report high satisfaction.

Conclusion: Each DCT execution brings a unique set of challenges and, therefore, demands a unique set of solutions. A successful and agile clinical trial relies on an operating system that combines sophisticated technology and specialized networks. This is a first-of-its-kind study in depression and, so far, DCT appears to be a great fit.

Funding/financial disclosures: Study funded by Boehringer Ingelheim; Dr. Reist employed by Science 37

Placebo response rates in randomized, controlled psychiatric studies over the past 10 years

Authors: Andreas Schreiner,¹ Cherilyn Boller,¹ Shari Medendorp,¹ Martin Strassnig²

Affiliations: ¹Premier Research, Research Triangle Park, NC, USA; ²Department of Psychiatry, Miller School of Medicine, University of Miami, Delray Beach, FL, USA

Objective: Double-blind, randomized, placebo-controlled clinical trials (RCTs) in patients with psychiatric disorders increasingly fail to separate active treatment from placebo. Contributing factors include diagnosis, geographical setting, patient expectations, clinician ratings, and the trial design itself. Here, we aim to analyze treatment responses over the past 10 years, focusing on changes in response to investigational compounds versus placebo and factors contributing to these changes, while suggesting potential mitigation strategies.

Methods: Retrospective review of interventional RCTs (Phase II and III) in psychiatric disorders, supported by Premier Research, between the years 2011 and 2020.

Results: Fifteen studies in ADHD were evaluated, with a total of 2,390 subjects. Placebo rates varied between nine percent and 33 percent. Trial design-related factors contributing to placebo response included high variability in measurements among placebo subjects, including baseline, and entry criteria violations. Most trials reviewed had washout periods before treatment start. There was a trend toward higher placebo-response rates in psychiatric RCTs over the past 10 years. Mitigation strategies include avoiding unequal randomization ratios (active vs. placebo), providing placebo response mitigation training, central versus local ratings, blinding raters for key inclusion criteria, or a single-blind placebo run-in to eliminate placebo responders. However, not all measures demonstrated a consistent significant effect on placebo response. 

Conclusion: Overall, high placebo response represents a substantial risk for psychiatric studies and should be addressed in Phase II and III studies.

Funding/financial disclosures: Research conducted by Premier Research; AS, CB, and SM employed by Premier Research; Consultancy services for Premier Research provided by MS 

The financial burden of screening and enrolling ineligible subjects in clinical trials

Authors: Andre M. Pinho; Kerri K. Weingard, ANP; Mitchell D. Efros, MD, FACS 

Affiliations: All authors with Verified Clinical Trials (VCT)

Objectives: The inability to verify a subject’s research history can result in inclusion/exclusion violations (IEPVs). The financial impact of IEPVs by research subjects has not been historically documented. This study will fill the gap in the literature by creating a prospective measure of IEPVs utilizing the VCT global research subject database to explore the relationship between IEPVs and the resultant financial burden.

Methods: The VCT database was utilized to prospectively detect and prevent IEPVs in a series of clinical trials conducted in the United States since 2017. Subjects (N= 9,529) were verified utilizing the VCT database. Following execution of an institutional review board (IRB)-approved consent, subjects were verified using the VCT database. A near-instantaneous response will identify any IEPVs revealed after authentication and comparison of each subject’s research history to the protocol criteria via a proprietary algorithm. Data was analyzed using Excel and R v. 3.5.0.

Results: Of these subjects, 746 (7.8%) were flagged for potential protocol violations and disqualified from the clinical trial. Using estimated costs, the loss was approximately $29,680,000.

Conclusion: Our research indicates that use of a global research subject database can detect and prevent IEPVs, thus improving subject safety and data quality. Additionally, our research indicates that such a tool can reduce study costs and potentially prevent study or investigational product failure. 

Funding/financial disclosures: All authors employed by Verified Clinical Trials