- Количество слайдов: 42
Presented by Terry Blaschke, M. D. at the meeting of the Clinical Pharmacology Subcommittee of the Advisory Committee for Pharmaceutical Science
Transition of Biomarkers to Surrogate Endpoints Opportunities, Challenges and Some Ways Forward: How can academia-industry-government collaborations facilitate the development of biomarkers and surrogates?
Challenge and Opportunity on the Critical Path to New Medical Products “Adoption of a new biomarker or surrogate endpoint for effectiveness standards can drive rapid clinical development. ”
Recent Examples (From Critical Path Document): • FDA adoption of CD 4 cell counts and, subsequently, measures of viral load as surrogate markers for anti-HIV drug approvals allowed the rapid clinical workup and approval of life-saving antiviral drugs, with time from first human use to market as short as 3. 5 years. FDA convened the data holders, conducted analyses in conjunction with industry and academia, and provided guidance on trial design. • Similarly, FDA adoption of the eradication of H. pylori as a surrogate for duodenal ulcer healing greatly simplified the path of those therapies to the market.
Antiretroviral Drugs for HIV Infection What Allowed Surrogate Endpoints to be Used for Approval?
First, Some History…
(from an HHS Press Release in 1992…) Milestones in dd. C (Zalcitabine) Approval Process: – October 30, 1991 -- NDA filed with FDA by Hoffman-La Roche – November 1991 - April 1992 -- FDA actively working with Hoffman-La Roche to discuss data, work out additional analyses. Additional study and clinical trial data solicited by FDA and the company. – April 21, 1992 -- Antiviral Drug Products Advisory Committee recommends approval of dd. C in combination therapy. – April - June 1992 -- Continuing meetings between Hoffman-La Roche and FDA on labeling. Hoffman La Roche committed to series of studies to further demonstrate efficacy and better define appropriate clinical use. – June 19, 1992 -- dd. C approved by FDA. – June 22, 1992 -- Secretary Sullivan announces approval. – First drug approved since the Vice President announced FDA's accelerated drug approval process. – Process incorporates use of surrogate endpoints to determine efficacy. – Process allows for approval to be withdrawn if further review determines therapy to be ineffective. – Drugs approved in normal process take an average of 10 years in development and 2 years for FDA review of the NDA. dd. C's development took four years and the FDA review, including intensive discussions and soliciting more data, took 8 months.
What factors accelerated the acceptance of CD 4+ cell count and HIV plasma RNA copy number as surrogates? • Urgent need for therapies for a fatal illness • Environment was “risk tolerant” as defined by Ph. RMA • Strong patient advocacy groups • Congressional interest • Subpart E [21 CFR 601. 41] Surrogate - Approval based on a surrogate endpoint or on an effect on a clinical endpoint other than survival or irreversible morbidity. • Willingness of the FDA to take risks by requiring a Phase IV commitment • Collaboration among clinical scientists and statisticians from academia, industry, and Government (FDA, NIH, CDC)
Journal of Acquired Immune Deficiency Syndromes 1990; 3: 1065 -73
The next ARV class: Protease inhibitors Initial approval dates: Saquinavir Ritonavir Indinavir December, 1995 March 1, 1996 March 16, 1996 Comment at the time of approval of saquinavir: Commissioner of Food and Drugs David A. Kessler, M. D. , pointed out that five of the six AIDS therapies approved so far were reviewed in six months or less. "The review of saquinavir is the fastest approval of any AIDS drug so far, and demonstrates FDA's flexibility in situations when saving time can mean saving lives, " Kessler said. "When it comes to AIDS and other life-threatening diseases, we have learned to take greater risks in exchange for greater potential health benefits. "
HIV Protease Inhibitors (Saquinavir, Indinavir, Nelfinavir) v v 4. 9, 3. 2, 2. 6 years in clinical development 26, 31, 25 clinical trials 1283, 1116, 1132 subjects in trials trial features: u u u v 6+, 11 +, 7 randomized, double-blind 3, 11, 6 dose-response 1, 2, 3 confirmatory trials accelerated approval, based on surrogate endpoint & requirement for post-approval clinical confirmation CDDS, Georgetown University Carl Peck, M. D. , Director
The result of using surrogates for ARVs: • The rapid approval of new drugs to treat HIV – Now over 20 ARVs on the market, most of them in record time – Incentives for companies to develop new drugs for HIV – An established pathway to approval of these drugs in the form of an FDA Guidance • In fact, because of the efficacy of these agents, approval without the use of surrogates would not be feasible nor ethical
Let’s look at the process of qualifying the use of HIV plasma RNA and CD 4+ cells as surrogate endpoints in more detail
Surrogate Endpoint Qualification • Begins with a hypothesis about pathogenesis • Ends with establishment of its applicability in clinical trials • The Middle? – Basic and clinical studies of pathogenesis – Discovery of markers of disease progression – Collection of data from both preclinical and early clinical studies – Mechanistic or semi-mechanistic models • Preferable to avoid empiric models alone – Collaboration and sharing of information in order to qualify biomarkers as surrogate endpoints
Hypothesis • Acquired Immunodeficiency Syndrome (AIDS) is caused by an infectious agent that destroys the cellular immune system and results in opportunistic infections that result in the death of the patient – Infectious agent, now named the Human Immunodeficiency Virus or HIV, discovered by Gallo and Montagnier – HIV is the causative agent of AIDS (Koch’s Postulates must be fulfilled) – Suppression or prevention of HIV replication will alter the course of the disease
Pathogenesis • Details of HIV replication and the nature of the interaction between HIV and the immune system were extensively studied in vitro, in animal models and in vivo – Largely an academic endeavor, carried out at the NIH and in academic centers – Led to a detailed understanding of viral structure, replication mechanisms and interaction with CD 4+ cells and involvement of co-receptors – This information was key to the development of antiretroviral drugs, largely carried out by the pharmaceutical industry in collaboration with NIH and academia (e. g. , role of NCI in zidovudine development and in protease inhibitor development)
Discovery of biomarkers of disease progression – Multiple groups, mostly academic, evaluated many possible biomarkers of the progression of HIV to AIDS – Putative biomarkers included: • • • P 24 antigen CD 4+ cell count CD 8+ cell count CD 38+, CD 8+CD 28 - cell count HIV RNA copy number 2 -microglobulin Neopterin Cytokines Immunoglobulins Etc. , etc.
Discovery of biomarkers of disease progression • Cohort studies are essential- Many have been supported – – – – – Women's Interagency HIV Study Multicenter AIDS Cohort Study (MACS) CDC Adult/Adolescent Spectrum of Disease Project HIV Outpatient Study Amsterdam Cohort Studies on HIV/AIDS Swiss HIV Cohort Study UK Collaborative HIV Cohort Study Italian HIV Seroconverter Study the Euro. SIDA cohort Many smaller cohorts, Etc….
Next steps that were necessary • Validation of biomarker assays – For sensitive and ultrasensitive assays for HIV RNA in plasma – For assays of CD 4+ cells by flow cytometry and other techniques • Collection of biomarker data from interventional clinical trials • Creation of mechanistic or semi-mechanistic models incorporating biomarkers • Qualification of biomarkers as surrogate endpoints
Source: Nature, Vol 373, 12 Jan 1995, pp 123 -6
For approval of antiretroviral drugs this process did not occur in the sequential, linear fashion just described
Clinical Infectious Diseases 1997; 24: 764 -74
AIDS Research and Human Retroviruses 2000; 16: 1123 -33
HIV Surrogate Marker Collaborative Group • • • • • • • • A. Babiker, MRC HIV Clinical Trials Unit, University College London Medical School J. Bartlett, Division of Infectious Diseases, Duke University Medical Center A. Breckenridge, Department of Pharmacology and Therapeutics, University of Liverpool G. Collins, the Division of Biostatistics, School of Public Health, University of Minnesota R. Coombs, Virology Division, University of Washington D. Cooper, * National Centre in HIV Epidemiology and Clinical Research, the University of New South Wales T. Creagh, Clinical and Epidemiology Consultants, Atlanta, Georgia A. Cross, Bristol-Myers Squibb, Wallingford, Connecticut M. Daniels, Department of Statistics, Iowa State University J. Darbyshire, MRC HIV Clinical Trials Unit, University College London Medical School D. Dawson, Glaxo Wellcome, Research Triangle Park, North Carolina V. De. Gruttola, Department of Biostatistics, Harvard School of Public Health R. De. Masi, Glaxo Wellcome, Research Triangle Park, North Carolina R. Dolin, Harvard Medical School J. Eron, Division of Infectious Diseases, University of North Carolina at Chapel Hill M. Fischl, Department of Medicine, University of Miami School of Medicine S. Grossberg, Department of Microbiology, Medical College of Wisconsin J. Hamilton, Division of Infectious Diseases, Duke University Medical Center S. Hammer, * Division of Infectious Diseases, Columbia Presbyterian Center P. Hartigan, VA Medical Center, West Haven, Connecticut K. Henry, HIV Program, Regions Hospital, St. Paul, Minnesota A. Hill, Glaxo Wellcome, Greenford, Middlesex, United Kingdom M. Hughes, † Department of Biostatistics, Harvard School of Public Health C. Katlama, Département de Maladies Infectieuses, Hôpital de La Salpêtrière, Paris D. Katzenstein, Division of Infectious Disease, Stanford University Medical Center S. Kim, † Center for Biostatistics in AIDS Research, Harvard School of Public Health D. Mildvan, Beth Israel Medical Center, New York J. Montaner, Canadian HIV Trials Network, Vancouver, British Columbia J. Kahn, San Francisco General Hospital M. Moore, Glaxo Wellcome, Research Triangle Park, North Carolina J. Neaton, Biostatistics Division, University of Minnesota • • • • • • W. O’Brien, Division of Infectious Diseases, University of Texas Medical Branch H. Ribaudo, † Center for Biostatistics in AIDS Research, Harvard School of Public Health D. Richman, Departments of Pathology and Medicine, University of California, San Diego M. Saag, * Division of Infectious Diseases, University of Alabama at Birmingham M. Salgo, Hoffman-La Roche, Inc. , Nutley, New Jersey L. Saravolatz, Division of Infectious Diseases, St. John Hospital, Detroit, Michigan R. Schooley, Infectious Disease Division, University of Colorado Health Sciences Center M. Seligmann, Service d’Immuno- Patholgie et d’Hématologie, Hopital St. Louis, Paris S. Staszewski, Klinikum der J. W. Goethe-Univer sität, Frankfurt, Germany L. Struthers, Roche Products, Ltd. , Welwyn Garden City, Hertfordshire, United Kingdom C. Tierney, Center for Biostatistics in AIDS Research, Harvard School of Public Health A. Tsiatis, * Department of Statistics, North Carolina State University S. Welles, Division of Epidemiology , School of Public Health, University of Minnesota. Group = 55 individuals, international representation from industry & academia
Some general principles of biomarker use and qualification As outlined by Sheiner and colleagues
Establishing Causal Certainty • Establish causality (given empirical association) by • supporting PA as mechanism, not by R/O other causes. • Evidence supporting PA • – Response correlates with (temporally varying) exposure. – Causal path biomarkers change in a mechanistically compatible direction, rate, and temporal sequence (e. g. , viral RNA, CD 4 in HIV). • Learning trials and analyses are well suited to • mechanistic interpretation of time-varying data. • Independent causal evidence is still required: Causal • evidence from (same) single RCT does not rule out transience or interaction.
Causal Path Biomarkers Dosage Pathology Adherence PK Receptor Cp Physiology Biomarker Clinical Effect Biomarker time • Correct temporal sequence Causal certainty. © LB Sheiner, 2002, all rights reserved.
What Are Causal Path Biomarkers? • Biomarkers that serve as indicators of the state or activity of mechanism(s) connecting disease pathophysiology to clinical manifestations – Must be scientifically plausible based on current understanding of disease – As knowledge increases, confidence in the validity of a biomarker of will increase, especially when drugs in the same class and/or with the same indication affect the same biomarker – Increasingly, more biomarkers will be useful in developing models of drug action • Causal Path Biomarkers need not be surrogate markers when used for drug development decisions or as confirmatory evidence of efficacy
Credibility of Causal Path Biomarkers depends on: • State of scientific knowledge of the disease mechanisms • Consistency of association of the clinically approvable endpoint and the biomarker • Proximity of the biomarker to the clinical endpoint on the causal path • Multiple biomarkers changing in correct temporal sequence • Similarity of biomarker exposure and clinical exposure response when both are studied Source: CDDS Workshop Report, Drug Information Journal 2002; 36: 517 -534
Table of Candidate Causal Path Biomarkers (1) Disease Drugs/Drug Class Clinical Endpoint(s) Biomarker Alzheimer’s Disease Any Cognitive testing Entorhinal cortex volume (MRI) Anemia Erythropoeitin Weakness, fatigue, exercise intolerance, etc Hemoglobin concentration, Hematocrit Asthma Lipoxygenase inhibitors, etc Respiratory Distress Pulmonary Function Tests (FEV 1) Bacterial infections Antibacterial agents Clinical pneumonia, pyelonephritis, etc Bacterial counts/sensitivities in biological fluids Cancer (all) Chemotherapeutic Agents Survival, progression Tumor shrinkage Cancer (Colon, in familial polyposis) Celecoxib Prevention of colon cancer Polyp counts (colonoscopy, biopsy) Cancer (Ovarian) Chemotherapeutic Agents Survival, progression CEA-125 Cancer (Prostate, hormone sensitive) Chemical castration agents Survival, progression PSA, Serum testosterone Carotid Artery Disease Anti-atherosclerotic agents TIA, Stroke Carotid artery intima-mediathickness Diabetes Insulin, hypoglycemic agents Coma, Infections, nephropathy, retinopathy, etc Blood glucose (vs. time) Hb. A 1 c Glaucoma Brimonidine, etc Visual acuity Intraocular pressure Human Immunodeficiency Virus Disease Protease inhibitors, reverse transcriptase inhibitors Opportunistic Infections, cancer, survival, etc Viral RNA, CD 4 Hyperlipidemias Cholesterol lowering agents Angina, Myocardial Infarction Serum cholesterol, C-reactive protein Source: Peck, Rubin and Sheiner, Clin Pharmacol Ther 73: 481, 2003
Table of Candidate Causal Path Biomarkers (2) Disease Drugs/Drug Class Clinical Endpoint(s) Biomarker Hypertension Antihypertensive agents Stroke, heart failure, etc Blood pressure (BP) Malaria Antimalarial agents Fever, chills, prostration, coma parasitemia Multiple Sclerosis (MS) Interferon beta 1 b Clinical neurological symptoms MS lesion volume (MRI) Neutropenia Filgrastim ( G-CSF) Clinical infections Blood granulocyte count Osteoporosis Hormone replacement therapy Fractures Bone mineral density (BMD), Alkaline phosphatase, osteocalcin Penile Erectile Dysfunction Sildenafil Completed intercourse Penile plethsymography Peptic Ulcer Disease H 2 blockers Dyspepsia, bleeding, GI-perforation Gastric acid secretion rate Xray, Endoscopy Rheumatoid Arthritis Etanercept, etc Pain, mobility Xray (joint space and erosions), C-reactive protein Sepsis Drotrecogin alfa Shock, organ failure, survival, etc D-dimer, Il 6 Sickle Cell Disease Hydroxyurea Painful clinical crises % hemoglobin modified to have a high oxygen-binding affinity (%MOD or % Hb-F), blood neutrophil counts Thromboembolic disease (arterial) Warfarin Stroke INR Thromboembolic disease (venous) Heparin Pulmonary Embolus Anti-Xa, IIa coagulation factors, APTT Viral Diseases Vaccines Survival Serum antibodies Source: Peck, Rubin and Sheiner, Clin Pharmacol Ther 73: 481, 2003
Establishing Pharmacological Causality • Given an empirical association in preclinical or clinical studies, causality is established by directly supporting pharmacological activity as the mechanism, not by ruling out other causes. • Causal confirmation is more demanding than empirical confirmation • Evidence supporting pharmacological causality: – Identifying and establishing the credibility of Causal Path Biomarkers
Designing Confirmatory (Causal) Evidence Trials Design Feature Phase 2 Randomization Control group Baseline prognostic covariates >2 Dosage Groups X/O dose w/in subjects Clinical Endpoints Serial prognostic covariates Serial biomarkers PK Learning elements Compliance Model-based Analysis Hypothesis testing + + + + + ? Phase 3 + + + ? + + + + © LB Sheiner, 2002, all rights reserved.
Learning While Confirming time patients Confirming Block • • Random assignment Placebo control Clinical Endpoints Baseline Covariates Homogenous patients PK Compliance Serial Biomarkers/Covariates Escalate, or Randomly change to one of multiple other dosage regimens Heterogeneous patients © LB Sheiner, 2002, all rights reserved.
"Learning" in Drug Dev: Where Great Empirical Certainty is Unnecessary • • Development Decisions (all phases) – Development Planning – Study Design (CTS) Labeling (phase 2 B/3) – Dosage Regimens – Dosage Adjustments for Special Populations – Safety Restrictions – Quantifying Benefit • Market Access Testing (phase 3) – Great Potential Benefit – High Prior Presumption of Positive Benefit: Risk – Excessive “Cost” of Objective Evidence – “Confirmatory* evidence” (CE) for SCT approval. * “confirmatory” = “learning”! Lewis B Sheiner 2001
Summary l Conflict: – – l Drug regulation demands certainty & much information. Causal models are inevitably uncertain, but highly informative. Resolution: Use models when – Lesser certainty is permissible: l l – Labeling (“User’s Manual”) Safety/Efficacy: Great potential benefit &/or High prior presumption Modeling can yield high certainty: l l Highly credible models, and Correct performance of tests under null hypothesis. Source: Lewis Sheiner, AAPS 1998
Biomarker Development • What is needed – Data pooling, synthesis, analysis – Identification of what is known – and what are the gaps – Identification of what studies are needed to fill the gaps – Doing the work Source: Janet Woodcock, ACCP meeting, Oct 3, 2004
I believe that: • The Public wants more therapies at reasonable prices • The regulatory issues are no longer a major impediment • The FDA is willing to move forward with new surrogates • Substantial collaboration among academia, industry and regulatory bodies will be necessary – Past history with HIV indicates that such collaboration can occur and benefits all constituencies – Meaningful collaborations are already underway and should be encouraged and supported
Questions and Discussion?