Brief Overview Of Biomarkers: Value, Limitations, And The Biomarker .

Brief overview of biomarkers: value,limitations, and the BiomarkerQualification Program (BQP)Peter Stein, MDDeputy DirectorOffice of New Drugs / CDER / FDA

Disclaimers Views expressed in this presentation arethose of the speaker and do not necessarilyrepresent an official FDA position I do not have any financial disclosuresregarding pharmaceutical drug products 2

Overview Types of biomarkers – a brief review Surrogate endpoints – value and limitations Challenges of biomarker development3

BEST Resource: Biomarkers, EndpointS, andOther Tools A glossary of terminology and uses of biomarkersand endpoints in basic biomedical research,medical product development, and clinical care Created by the NIH-FDA Biomarker Working Group Publicly available athttp://www.ncbi.nlm.nih.gov/books/NBK326791/ BEST harmonizes terms and definitions andaddresses nuances of usage and interpretationamong various stakeholders, including: Biomedicalscientists BiomedicalscientistsTranslationaland Translationalandclinicalclinical researchersresearchers MedicalproductdevelopersdevelopersMedicalproduct Patient/disease advocacy groupsPatient/diseaseadvocacy groups Government officialsGovernmentofficials CliniciansClinicians4

Biomarker: definition“A defined characteristic that is measured as anindicator of normal biological processes,pathogenic processes, or responses to anexposure or intervention, including therapeuticinterventions. Molecular, histologic, radiographic,or physiologic characteristics are types ofbiomarkers.”5

Why develop biomarkers? High quality biomarkers can markedly accelerate and enabledrug development in areas of unmet need Biomarkers can improve trial efficiency and feasibility– Improve ease and accuracy of identifying patient population– Enrich the study population Study population with more events – so detecting change in outcomewith treatment becomes feasible Study a more responsive study population to detect a drug effect– Improved monitoring Enhanced patient safety – earlier detection of drug toxicity Detect changes in patient status– Improve assessment of treatment response; predict clinical benefit Address unmet medical needs, where progress is halted ordelayed by lack of adequate drug development tools,including biomarkers6

BEST (Biomarkers, EndpointS, and other Tools)Classification: Range of Biomarker Types Susceptibility / risk biomarker Diagnostic biomarker Prognostic biomarkerMeasures of diseasepresence and status Monitoring biomarker Predictive biomarker Pharmacodynamic/Responsebiomarker – including surrogateendpointsMeasure aspects of responseto treatment Safety biomarker7

Disease-Focused BiomarkersNon-disease populationSusceptibility or risk predictor biomarkersIndividuals at high risk of disease orpre-clinical disease populationDiagnostic biomarkerDiagnostic biomarkerPatients with diseaseDiagnosticbiomarkerDiseaseSubtype 1DiseaseSubtype 2

Disease-Focused BiomarkersNon-disease populationSusceptibility or risk predictor biomarkersIndividuals at high risk of disease orpre-clinical disease populationDiagnostic biomarkerDiagnostic biomarkerPatients with diseaseDiagnosticbiomarkerDiseaseSubtype 1DiseaseSubtype 2Prognostic biomarkerPrognosticbiomarkerPatients with disease at higher risk ofdisease-related outcome(s)

Disease-Focused BiomarkersNon-disease populationSusceptibility or risk predictor biomarkersIndividuals at high risk of disease orpre-clinical disease populationDiagnostic biomarkerDiagnostic biomarkerPatients with diseaseDiagnosticbiomarkerDiseaseSubtype 1DiseaseSubtype 2Prognostic biomarkerPrognosticbiomarkerPatients with disease at higher risk ofdisease-related outcome(s)Diagnostic biomarkerDisease-related outcome(s)

Disease-Focused BiomarkersNon-disease populationSusceptibility or risk predictor biomarkersMonitoring biomarker: assess disease statusIndividuals at high risk of disease orpre-clinical disease populationDiagnostic biomarkerDiagnostic biomarkerPatients with diseaseDiagnosticbiomarkerDiseaseSubtype 1DiseaseSubtype 2Prognostic biomarkerPrognosticbiomarkerPatients with disease at higher risk ofdisease-related outcome(s)Diagnostic biomarkerDisease-related outcome(s)

BEST (Biomarkers, EndpointS, and other Tools)Classification: Disease Focused Biomarkers Susceptibility / riskbiomarker: Diagnostic biomarker: Monitoring biomarker: Prognostic biomarker:Examples: BMI or 2 hr post-meal glucose for diabetes risk Apo E genotype risk for Alzheimer’s diseaseKey uses: Define population for more efficient preventiontrialsExamples: Blood pressure in hypertension FEV1 for COPDKey uses: Define disease population for studyExamples: HCV-RNA PSA in prostate cancerKey uses: Monitor patient status in trialsExamples: Gleason score in prostate cancer Total kidney volume in AD-PCKDKey uses: Define higher risk disease population, enhancing12trial efficiency

Treatment-Related BiomarkersPatients with diseasePredictive biomarkerLower ornon-responderpopulationResponder or higher responder population

Treatment-Related BiomarkersPatients with diseasePredictive biomarkerLower ornon-responderpopulationResponder or higher responder populationDrugTreatmentPhase 1, 2a (early stage biomarkers)Target engagementbiomarkerDrug is hitting targetPharmacodynamic/Response BiomarkersPharmacodynamicbiomarkerRelevant pathway is modulatedProof of PharmacologyProof of Concept

Treatment-Related BiomarkersPatients with diseasePredictive biomarkerLower ornon-responderpopulationResponder or higher responder populationDrug TreatmentPhase 1, 2a (early stage biomarkers)Target engagementbiomarkerDrug is hitting targetPharmacodynamic/Response BiomarkersPharmacodynamicbiomarkerRelevant pathway is modulatedProof of PharmacologyProof of ConceptPhase 2b, 3 (later stage biomarkers)Reasonably likelysurrogate endpointEndpoint likely to predictoutcome or clinical benefit (AA)Validatedsurrogate endpointEndpoint predicting outcome orclinical benefit (Full Approval)Intermediateclinical endpointEarlier change in definitiveendpoint (Accel Approval, AA) Disease morbidity / mortality (“survives”) Clinical benefit (“feels or functions”)Surrogate EndpointsClinicalOutcomes

Treatment-Related BiomarkersPredictive biomarkerLower ornon-responderpopulationResponder or higher responder populationPhase 1, 2a (early stage biomarkers)Drug TreatmentSafety biomarkers: assess treatment-related toxicityPatients with diseaseTarget engagementbiomarkerDrug is hitting targetPharmacodynamic/Response BiomarkersPharmacodynamicbiomarkerRelevant pathway is modulatedProof of PharmacologyProof of ConceptPhase 2b, 3 (later stage biomarkers)Reasonably likelysurrogate endpointEndpoint likely to predictoutcome or clinical benefit (AA)Validatedsurrogate endpointEndpoint predicting outcome orclinical benefit (Full Approval)Intermediateclinical endpointEarlier change in definitiveendpoint (accelerated approval) Disease morbidity / mortality (“survives”) Clinical benefit (“feels or functions”)Surrogate EndpointsClinicalOutcomes

BEST (Biomarkers, EndpointS, and other Tools)Classification: Treatment-focused biomarkers Predictivebiomarker:Examples: Cystic fibrosis genotypes response to ivacaftor Microsatellite-high predicts response to pembrolizumabKey uses: Trial enrichment – improves efficiency, reduces samplesize, increases response to treatment Pharmacodynamic/Responsebiomarker:Examples: Blood pressure in hypertension FEV1 or 6 minute walk test LDL-CKey uses: Demonstrating drug-target engagement, dose-ranging Surrogate endpoints (validated or reasonably-likely) Safety biomarker:Examples: ALT, creatinine / eGFR Urinary kidney injury biomarkers (KIM-1, etc.)Key uses: Detecting / assessing drug toxicity17

BEST (Biomarkers, EndpointS, and other Tools)Classification: Pharmacodynamic / Response BMsTo support approval, FDA expects substantial evidence ofeffectiveness – that shows that a drug improves meaningful clinicaloutcomes: how a patient feels, functions, or survives A validated surrogate endpoint: accepted by FDA that the effecton the biomarker predicts a specific clinical outcome. Validatedendpoints have strong and diverse evidence supporting therelationship of the BM and the outcome A “reasonably likely” surrogate endpoint: an endpointsupported by strong mechanistic and/or epidemiologic rationalesuch that an effect on the surrogate endpoint is expected to becorrelated with a clinical benefit, but not yet reaching thestandard for validation.18

Types of Surrogate EndpointsCausal BiomarkerPathway or mediatorbiomarkerReflecting causal factorIn pathway of disease:mediator of damage Genetic or geneticrelated defectivefunction (e.g., genevariant, decreasedenzyme level orfunction) Environmentalexposure (e.g., bloodlead level, etc.) Microbiologic (e.g.,HIV, HCV, bacterialculture, AFB smear)BADCEOrgan injurybiomarkerSites of InjuryReflecting organinjuryOrgan 1ClinicalfunctionEvent orfunctional lossClinicalFunctionalMeasuresOrgan 2Organ 3Tissueinjury BMOrganFunctionBMClinicalEvents19

The limitations of surrogate endpoints Not a direct measure of how a patient feels, functions orsurvives Intended to reflect and predict clinical benefit not measure theoutcome With a surrogate endpoint, the benefit / risk assessmenttherefore must be based upon assumptions / predictions ofbenefit– Translating the extent of clinical benefit from an indirect measure, andalso using a limited dataset on risk to assess harms– Challenging when a drug shows clear effects on a surrogate endpoint –but also has safety issues And biomarkers may fail to predict clinical benefit20

The limitations of surrogate endpointsSurrogate on causal pathwaymodulated by drugSurrogate not on causal pathway bywhich drug leads to benefit, ormultiple pathways of leading toclinical outcome, BM may or maynot reflect key pathwaysDrug may induce adverse effects ondesired clinical outcome through apathway not reflected by BM, or maylead to other toxicities BM doesnot reflect benefit (or cityClinicalOutcomeAfter Fleming Statistics in Medicine 201221

Biomarker integration into drugdevelopment: 3 pathways IND pathway: based upon agreement with the division, inthe context of a specific drug development program Scientific community consensus: broadly/widely usedbiomarker, appropriate scientific support, generallyaccepted by experts in the field Biomarker qualification program: review and acceptancebased upon appropriate submission qualification package;available for use in any development program withinapproved context of use22

The challenges of biomarker development Many disease areas with unmet needs have insufficient drug developmenttools to maximize trial efficiency (or even feasibility) Biomarker development is a long and resource-intensive task– Biomarker discovery: biased or unbiased screening in animal, clinical, epidemiological(include RWE)– Early animal translational models– Clinical or epidemiology observational studies– Analytic validation efforts: assure accuracy / reproducibility of measure– Interventional studies with “gold standard” endpoints compared to candidate – withmultiple different treatments (different MOAs) to show that BM works across drugclasses Many stakeholders in the mix:– Academic investigators at multiple institutions, US and ex-US– Often several academic societies in disease area with different viewpoints andmembership– Different companies – both drug and device-focused may be working in the area– May be different patient stakeholder organizations The challenge: how to prioritize biomarker needs, focus resources, andintegrate efforts across stakeholders23