Regulatory Experience in Reviewing CV Safety for Diabetes Drugs Kristina Dunder MD, PhD, Alternate CHMP member MPA, Sweden Disclaimer The views and opinions expressed in the following PowerPoint slides are those of the individual presenter and should not be attributed to Drug Information Association, Inc. ( DIA ), its directors, officers, employees, volunteers, members, chapters, councils, Special Interest Area Communities or affiliates, or any organization with which the presenter is employed or affiliated. These PowerPoint slides are the intellectual property of the individual presenter and are protected under the copyright laws of the United States of America and other countries. Used by permission. All rights reserved. Drug Information Association, DIA and DIA logo are registered trademarks or trademarks of Drug Information Association Inc. All other trademarks are the property of their respective owners. Drug Information Association www.diahome.org 2 1
Disposition Do diabetes drugs increase CV risk? CHMP and FDA guidance Interpretation of CHMP guideline Recent Experience Would we have asked for an outcome study pre approval for rosiglitazone today? Drug Information Association www.diahome.org 3 Do Diabetes Drugs Increase CV risk?? Diabetes drugs have different mechanism of action Relevant with outcome studies for all drugs? Only those with a suggestion of "detrimental t effect"? Other potential serious events may be of more relevance Pioglitazone Urinary bladder cancer? GLP-1 analogues Medullary thyroid cancer? Insulin analogues Tumor promoting effect Current concern based on findings for one specific drug class with potential negative cardiac effect due to mechanism of action Drug Information Association www.diahome.org 4 2
Do Diabetes Drugs Increase CV risk?? Do Diabetes Drugs Increase CV risk?? All diabetes drugs do not increase CV risk Diabetes drugs should not increase, but preferably reduce the risk of CV events Type 2 diabetes = CV risk factor HbA1c not surrogate for CV risk How to assess CV safety included in CHMP and FDA guidance Lipid lowering and antihypertensive guidelines will be updated Drug Information Association www.diahome.org 6 3
CV Safety; FDA requirements 7 Current status of the CHMP Diabetes Guideline Current Guideline 2002 Concept Paper for update adopted 2008 Second consultation period (dead line 18/11) 4
CHMP Guideline More flexible approach; totality of the data Type of Studies Study Population Cardiovascular Safety Outcomes Evaluation of the Results Drug Information Association www.diahome.org 9 Type of studies for assessment of CV safety Non-clinical studies important Athero-thrombotic findings, fluid retention, blood pressure, renal function, electrolytes homeostasis, cardiac functionality, repolarisation and conduction abnormalities Clinical studies; 2 approaches are possible; Metaanalytic approach. The size of database, as well as the mean duration of the studies is expected to be adequate to detect signals for serious and uncommon events As an alternative or when there is suspicion of an adverse CV signal specific long-term controlled clinical safety study with at least 18 24 months follow-up would be expected at the time of submission of the MAA. 5
Interpretation Does every drug in the same class need an outcomes study? When there is suspicion of adverse CV signal, otherwise metaanalytic approach How do you define a signal? Safety concern is intrinsic in the molecule/ mechanism of action or has emerged from pre-clinical/ clinical registration studies. Increase in body weight, oedema/ fluid retention, occurrence of hypertension or arrhythmia How reliable are the meta-analysis results in studies not intended to assess CV outcomes? Events adjudicated, study population, duration of studies Other support needed Drug Information Association www.diahome.org 11 Study Population Drug Information Association www.diahome.org 12 6
Interpretation Patients with high risk for CV events representing a relevant proportion of the diabetic population (according to validated cardiovascular risk scoring systems), are strongly recommended to be included in phase III Majority of patients at high risk? More events, shorter duration Would not mirror the target population Extrapolation to low risk population? p Drug Information Association www.diahome.org 13 Safety outcomes Emphasis on major CV events (MACE) (CV death, non fatal myocardial infarction and stroke) Hospitalisation for unstable angina could be included in a composite endpoint if the main objective is to exclude a safety signal Other safety outcomes, based on safety profile of the product class, the mechanism of action and/or the non-clinical findings. Prospective definition of CV safety outcomes, common for all phase II-III studies, facilitating pooled analysis strategies. Applicants should foresee a consistent central adjudication system for all predefined CV events during the phase II-III program. 7
Interpretation Inclusion of unstable angina High percentage of the weaker part of the endpoint (i.e. unstable angina), difficult to draw appropriate conclusions regarding the hard MACE endpoint Drug Information Association www.diahome.org 15 Evaluation of results Drugs belonging to a well-known class (and mechanism of action) Careful evaluation of the available medical literature together with the absence of pre-clinical and clinical signals of increased CV risk may lend some support to a meta-analytic approach Whenever a safety concern is intrinsic in the molecule/ mechanism of action or has emerged from pre-clinical/ clinical registration studies. Fully powered CV safety assessment, e.g. based on a dedicated CV outcome study, should be submitted before marketing authorization 8
Evaluation of results Acceptability of the data presented will be decided based on its overall quality, the point and interval estimates obtained for the calculation of specific risk and the reliability of these estimations.. Indications of increased risk of certain adverse events or unacceptable lack of precision are an important concern and may trigger the request for additional specific long-term outcome trials to exclude an unacceptable increase in CV or other identified risks associated with the new agent. Drug Information Association www.diahome.org 17 Interpretation In contrast to FDA guidance, CHMP draft guideline does not request specific limits to be met for the upper bound of the 95% CI of the hazard ratio. Evaluate the overall data package including non-clinical and clinical data for a conclusion on CV safety. Knowledge on the whole substance class and how similar the mechanism of action and receptor specificity are compared to available data for other drugs in the class Pay attention not only to the upper bound of the 95% CI of the hazard ratio for CV events but also to the point estimate Drug Information Association www.diahome.org 18 9
Cardiovascular Outcome Studies; Reflections Pros and cons with " the flexible approach" Possibility for regulators to interpret the data based on Mechanism of action Findings in preclinical/clinical studies First in class or "me too" Any cut-off arbitrary Difficulties giving advice concerning design of studies Drug Information Association www.diahome.org 19 Recent Experience (from EPAR): Exenatide QW No preclinical or mechanistic rationale Meta-analysis of exenatide BID studies versus placebo HR CV events 0.70 (95% confidence interval 0.38-1.31). Available study data for the QW formulation do not indicate any adverse cardiac effects Duration of studies short, patients with CV disease and risk factors were to a large degree excluded Long term cardiovascular outcome study planned including patients at a higher CV risk No mechanistic rationale for an increased CV risk,,no observed increased risk associated with exenatide BID CV outcome study not needed pre approval Drug Information Association www.diahome.org 20 10
Recent Experience (from EPAR): linagliptin 5239 patients with type 2 diabetes included in a CV meta-analysis (MACE and hospitalisation due to unstable angina) versus combined active and placebo comparators [Hazard ratio 0.34 (95% confidence interval 0.17;0.70)]. In total there were 11 primary events on linagliptin and 23 on comparators Low risk population Results reliable? Other DPP-4 inhibitors, no suspicion of a detrimental CV effect. No mechanistic rationale for an increased CV risk CV outcome study not needed pre approval Drug Information Association www.diahome.org 21 Rosiglitazone; Would FDA guidance lead to requirement of CV outcome study before approval? SAEs relating to the cardiovascular system (angina pectoris, MI, chest pain, CAD) were reported with a similar or lower frequency on RSG monotherapy (0.2-0.3%) than on placebo, SU or metformin. Reports under the body system Cardiovascular General were the same for RSG and placebo (5.0 and 5.3% respectively) There were isolated cases of worsening of pre-existing CHF during RSG treatment, but no increased incidence of new-onset CHF, compared with placebo. RSG was not associated with aggravated hypertension or specific ECG changes WOULD UPPER BOUND OF CI 1.8?? Drug Information Association www.diahome.org 22 11
Rosiglitazone; Would CHMP guidance lead to requirement of CV outcome study before approval? DATA FROM MAA 3.7 % oedema 2.7 % hypercholesterolemia( LDL och HDL) 0.5 % CHF (2.5% in combination with insulin) Increase in body weight of 4-6% PRODUCT INFORMATION WARNING Fluid retention, possible heart failure. Monitoring of signs of fluid retention, such as weight increase CONTRAINDICATIONS CHF (NYHA stadium I - IV), Combination with insulin Post approval outcome study requested (RECORD) Drug Information Association www.diahome.org 23 WOULD WE HAVE REQUIRED A PREAPPROVAL STUDY? YES! Drug Information Association www.diahome.org 24 12
CONCLUSION Diabetes drugs should not increase CV risk Requirements to show lack of harm will depend on Mechanism of action, Preclinical and clinical data, Previous experience within the class CV outcomes studies not needed for all new drugs Other safety issues may be more relevant Thank you for your attention! Drug Information Association www.diahome.org 25 13