Technologies for Measuring Cognition in Clinical Trials Critical - PowerPoint PPT Presentation

Technologies for Measuring Cognition in Clinical Trials Critical Path for Alzheimer’s Disease Consortium Stephen P. Arnerić , PhD, Executive Director ISCTM Conference October 15, 2018

WHY IS DEVELOPING BETTER WAYS TO MEASURE COGNITIVE FUNCTION CRITICAL? https://www.gbhi.org/about/ CRITICAL REASONS • Unmet need for effective treatments • Economic burden to Healthcare Systems Worldwide • A new therapeutic for AD has not been approved in over a decade! • Any trial in AD is dependent upon the outcome measures used, and must require the necessary sensitivity and specificity - especially true for the “ presymptomatic ” stage of the disease. • Need for early detection assessments to detect those “at risk” for cognitive decline. www.c-path.org/cpad 2

CRITICAL PATH INSTITUTE Fifteen global consortia collaborating with 1,450+ scientists and 84 organizations FOCUS: Data standards; clinical trial simulation tools from actionable data, disease progression models; biomarkers; clinical outcome assessment instruments www.c-path.org/cpad 3

CPAD DATABASE UTILIZATION (as of August 2018) 562 Total Applicants 29 AD Clinical Trials 6,995 Patients from 129 300 338 Distinct Institutions 82 USE BY SECTORS USE BY REGION North America: 57% Academia: 254 Europe: 24% Pharmaceutical: 165 Asia: 15% 4 7 Other: 70 Australia: 2% Non-profit: 32 South America: 1% Government: 11 Africa: 1% 8 Industry Academia & Foundations Abbvie; Allergan; AstraZeneca; Biogen; Amherst College; Arizona State Univ.; Bill & Biomarkable; CoreLab; Melinda Gates Foundation; CHDI Foundation; Daewong; Eisai; GE Duke Univ.; Fraunhofer Institute; Goethe Government & Healthcare; IBM; Johnson & Univ.; Harvard Univ.; Karolinska Institute; Other Johnson; Lundbeck; Merck; King’s College London; Michael J Fox NeuroCog; Novartis; Pentara; Foundation; Rockefeller Univ.; Seoul National NIH; Neurology Pfizer; Siemens; SAS Univ.; Univ. of Oxford; Yale Univ. Today; Gigatrust www.c-path.org/cpad 4

VALUE, HISTORY, & FUTURE OF CDISC STANDARDS FOR AD Value Proposition: FDA REQUIREMENT (as of December 2016) • A reproducible research framework with controlled terminology which All clinical data from accelerates our understanding of AD across trials trials using a uniform registration trials must format. be in CDISC format • Improves our ability to detect signals in new compounds; maximizes learnings from successes and failures. Historical Perspective (CPAD/CDISC partnership): Version 1.0 (2011) – • First user guide for AD CDISC standards (did not include biomarkers) focused on key demographic, genetic and clinical outcome assessments (COAs). Version 2.0 (2016) – • Added global consensus data standards for key CSF AD biomarkers, vMRI imaging and PET ligands. Future: Version 3.0 (~2019) – • Focus on promising exploratory biomarkers and biometric assessments . www.c-path.org/cpad 5

DEMONSTRATED UTILITY OF THE CLINICAL TRIAL SIMULATION TOOL IN MILD-to-MODERATE AD Balancing power, sample size, and duration, given varying effect magnitudes Crossover Parallel 91 weeks 78 weeks • Better power • ~50% cost savings • 13 weeks less time www.c-path.org/cpad 6

HIPPOCAMPAL VOLUME IMAGING BIOMARKER QUALIFICATION WITH FDA: ICV-HV enrichment yields trial size reduction for MCI ~29%, ~37.5%, and ~66% reduction of sample size by enrolling only subjects with baseline ICV-HV <97.7 th , <84.1 th and <50 th percentile, respectively. The sample size savings estimated by the two models with either LEAP™ or FreeSurfer ™ ICV -HV were approximately within 6 to 8% of each other. Under these assumptions: ▪ 24-month placebo-controlled parallel group trial. ▪ Drug effect of 50% reduction in the progression rate. ▪ Power was calculated as the proportion of trials for which the effect of treatment on progression rate was beneficial with a two-tailed P -value < 0.05 www.c-path.org/cpad 7

DEFINING DISEASE Requires a Composite Assessment = Signs Symptoms Observer / Performance Outcomes Patient & Physician Reported Outcomes Genetics Examination Temperature • Cognition (MMSE, CDR-SB, etc.) • Real World Data Behavior (sleep/mood scales – QOL-AD, GDS) • Motor function (UDPRS) • Sensation (NRS, etc.) • Balance & Coordination Vision Forgetfulness Infection Mobility • Autonomic GI/Lung/ Kidney EKG EEG/ Glucose tests function HR/BP Sleep/ Fatigue Imaging Modalities www.c-path.org/cpad 8

WHY CONTINOUS MEASUREMENT IS RELEVANT AND CRITICAL! Which patient is rapidly declining? • These data highlight the challenge of infrequent cross-sectional assessments • Understanding vector trends (the relevant 90%) in individual continuous performance would be more reflective of true long-term trends in performance/health maintenance, i.e., Aligned with Precision Medicine Objectives ! • GOAL: Validate Digital Assessments as DDTs to identify the “right patients”, enhance Clinical Trial efficiencies, and enable tailored treatment approaches 24 months baseline 12 months Courtesy of Dr. Jeff Kaye www.c-path.org/cpad 9

WE CAN ONLY STUDY AND UNDERSTAND WHAT WE CAN MEASURE & OBSERVE! PATIENTS & CLINICAL TRIAL SUBJECTS AS ICEBERGS www.c-path.org/cpad 10

DIGITAL DRUG DEVELOPMENT TOOLS Biometric Monitoring Devices (BMDs) as Regulatory Accepted Clinical Trial Assessments for Specific Contexts-of-Use WHY HOW WHAT • Improve our understanding of real- • • Continuous physiological Data (signal output) collected time changes in FUNCTION during monitoring with devices from a biosensor that the progression of life in health and (wearables/smart phones, measures a biological disease clothing, implants/ response • Improve the efficiency of AD clinical ingestibles, remote trials to accelerate the delivery of biosensors) novel treatments • Deliver precision care KEY COUs • Understand disease progression • Measure treatment responses • Deliver Precision Care www.c-path.org/cpad 11

THE VISION: DEVELOP AN END-TO-END ALZHEIMER DISEASE MODEL www.c-path.org/cpad 12

COLLECTING REAL WORLD DATA The Rise of Consumer Health Wearables: Promises and Barriers. Piwek L, Ellis DA, Andrews S, Joinson A PLoS Med 13(2): e1001953. pmed.1001953, Feb 2016 RWD = RWE: Careful data standardization, aggregation, and quantitative modeling will be required to transform RWD to RWE www.c-path.org/cpad 13

DOMAINS OF FUNCTION: WHERE DO WE FOCUS? CHALLENGES: • Data availability • CDISC Standards Development Drug • Understanding clinically meaningful Concepts-of-Interest Adherence relationships for healthcare decision making ; speech www.c-path.org/cpad 14

MEDICATION ADHERENCE AS A KEY IADL www.c-path.org/cpad 15

NEUROCOG [VeraSci] www.c-path.org/cpad 16

COGSTATE www.c-path.org/cpad 17

eVOX www.c-path.org/cpad 18

AURAL ANALYTICS: www.c-path.org/cpad 19

OTHER TECHNOLOGIES FOR COGNITIVE ASSESSMENTS www.c-path.org/cpad 20

EVIDENTIARY CONSIDERATIONS FOR TECHNOLOGIES www.c-path.org/cpad 21

CONSIDERATIONS FOR POSSIBLE ENDPOINTS WHERE TO FOCUS DIGITAL COGNITIVE ENDPOINT ASSESSMENT DEVELOPMENT! www.c-path.org/cpad 22

EMA GUIDELINES FOR AD 8.2.3. Efficacy endpoints in Preclinical AD For the time being there is no "gold standard" for assessment of treatment effect in patients with preclinical AD (see section 9). Cognitive endpoints used in primary and secondary prevention trials have been the diagnosis of dementia (based on cut-off scores), significant Section 6 (page 12): “So far, one specific biomarker cannot cognitive decline and change in cognitive function based on longitudinal performance on be endorsed over other alternatives for the purpose of certain tests. Novel outcome tools sensitive to small neuropsychological changes in this population are being developed, however they are not yet validated and cannot be endorsed identifying those patients who may progress more rapidly. solely as primary endpoints in this population. A time to event analysis could be a The trajectory of cognitive decline may further be modified complementary measure in order to support the relevance of any chosen outcome, although by cognitive reserve, medical comorbidities, lifestyle factors feasibility issues including length of the trial and number of drop-outs are recognized. The event must be of clear clinical importance such as onset of cognitive impairment (see section and cognitive training (see section 9). Hence increasing 9). Until a biomarker will be qualified as a reliable surrogate measure of treatment effect in clinical trial efficiency and qualification opinion procedures absence of a clinically observable change, patients should be followed up for a sufficient time are encouraged.” to capture relevant cognitive changes. www.c-path.org/cpad 23

TYPES OF BIOMARKERS & DRUG DEVELOPMENT TOOLS NCBI NLM NIH. BEST (Biomarkers, EndpointS, and other Tools) Resource. NCBI Bookshelf, 2016 Available from : https://www.ncbi.nlm .nih.gov/books/NBK3 38448 www.c-path.org/cpad 24