1. 1 Study Design Issues and Considerations in HUS Trials Yan Wang, Ph.D. Statistical Reviewer Division of Biometrics IV OB/OTS/CDER/FDA April 12, 2007

2. 2 Outline Adequate and well-controlled studies Study design issues and considerations in HUS trials –Choice of primary efficacy endpoint –Efficacy evaluation and sample size –Safety evaluation and sample size Conclusion

3. 3 Adequate and Well Controlled Studies 21 CFR 314.126 (b) 1.Clear statement of objectives 2.Study design permits valid comparison with appropriate control to provide quantitative assessment of drug effect 3.Select patients with disease (treatment) or at risk of disease (prevention) 4.Baseline comparability (randomization) 5.Minimize bias (blinding, randomization, etc.) 6.Appropriate methods of assessment of outcome 7.Appropriate methods of analysis

4. 4 Ideal/Preferred Study Design in HUS Prevention Trials Randomized, double blind, placebo-controlled Target population may include –Patients with STEC infection and at risk of developing HUS Primary efficacy endpoint –Incidence of HUS Adequate power to detect treatment effect Adequate data to demonstrate safety

5. 5 Challenges for Designing HUS Prevention Trials Rare and sporadic nature of STEC infection Unpredictable nature of HUS development from STEC infection Therapeutic window may be narrow (possibly within 48 hours after infection) Low incidence rate of HUS in patients with STEC infection (5%-15%)

6. 6 Sample Sizes Required for Efficacy in HUS Prevention Trials Under Various Scenarios PlaceboIncidence Rate of HUS (STEC Infection)

7. 7 Use of a composite endpoint can reduce sample size because the number of events increases Can we use a composite endpoint in a HUS trial (e.g. “HUS + other clinically relevant events) ? PlaceboIncidenceRate

8. 8 Difficulty in interpretation when treatment effects on components are not homogenous Hypothetical example of a HUS trial using a composite endpoint Composite Endpoint Component #1: HUS Treatment Effect Component #2 Component #3

9. 9 Considerations for Using a Composite Endpoint Are the individual components clinically relevant and of similar importance to patients? Do the more and less important endpoints occur with similar frequency? Is the underlying pathophysiology of the components similar? Are the components likely to have similar relative risk reductions?

10. 10 Issues/considerations for efficacy evaluation in HUS trials –Sample size can be prohibitive if “prevention of HUS” is used for efficacy measurement –Use of a composite endpoint (if possible) can reduce sample size for testing treatment effect –Potential difficulty in interpreting results of the composite endpoint when the treatment effect on composite endpoint cannot be translated to an effect on HUS prevention Safety evaluation – –Need to have an adequate number of treated patients for safety evaluation

11. 11 Chance of Observing No Serious Adverse Events What does it mean when no serious event (SAE) was observed or no safety issues were identified in a clinical trial of a new product? It doesn’t necessarily mean that the new product is safe because the chance of observing no events can be high when trial size is small. Chance of Observing No Events = (1-p) N Incidence of SAE (p) Number of Patients Treated in a Trial (N) 5010020030060010002000 0.5% 78%61%37%17%4.9%0.67%0.00% 1%61%37%13%4.9%0.24%0.00%  

12. 12 Rule of Three (3/N) Provide an answer to “What is the worst possible scenario for the risk of a serious adverse event when no events occur in a trial?” If no events occur in N treated patients, the upper bound of the 95% confidence interval for the risk (p) can be estimated as 3/N.

13. 13 Majority of Patients With STEC Infection May Not Benefit From Prophylactic Therapy for HUS # of Patients Needed to be Treated to Prevent one HUS Case Assumed Treatment Effect Placebo HUS Incidence Rate 5%10%15% 33% Reduction603020 50% Reduction402014 75% Reduction27149    98% 95% 89% 85% to 95% of patients with STEC infection will not develop HUS Number of patients needed to be treated to prevent one HUS case could be large

14. 14 Conclusion Need to have adequate and well controlled clinical trials to evaluate efficacy and safety of new products for HUS prevention Challenges for designing HUS prevention trials with –Adequate statistical power to evaluate a clinically meaningful efficacy endpoint –Sufficient data to demonstrate safety Return to Main Menu.