Clinical Trials Homework

Question 1 - You are studying the role of an IV agent (the new drug) compared to standard oral therapy in-hospital loading to prevent AF recurrence. Please describe how you would introduce a pharmacoeconomic sub-study. It is also known that the IV drug needs to be activated by the CYP enzyme system. How could one introduce a genetic study to determine drug responders? Comment on generalizability. (2 page limit)

Question 2 - In analyzing a clinical trial, a number of factors can make analysis problematic. Discuss each concern below and give an example of the problem and how it can be addressed.

a) Factors that disproportionally affect one "arm" of the study - covariate adjustments.

b) Results that are outliers, how does one handle?

c) Multiple group comparisons and the issue of sub-group analysis.

d) The use of arbitrary "cut-points".

e) The handling of competing events. (2 page limit)

Question 3 - The DSMB has important functions in a multi-center trial.

1) Discuss the primary role of the DSMB and how that at times can cause tensions regarding maintaining "study integrity".

2) How can a DSMB decision reduce the acceptance of a study and leave lingering questions unanswered? Give example(s). (2 page limit)

Question 4 - Draw a diagram to describe the difference between a superiority trial and a non-inferiority trial using an effect size point estimate, 95% lower confidence interval and a defined delta. Do this for a superiority observation, non-inferiority observation and an inconclusive observation, as well as an inferior observation. Explain the different observations shown in your diagram for each category.

In a non-inferiority trial, which type of analysis will tend to favor a non-inferiority outcome: intention to treat, or on-treatment analysis? Explain? (2 page limit)

Question 5 - A new IV version of Nebivolol, a B1 B2 B3 blocker has been studied in a trial to reduce BP acutely in patients presenting with an "hypertensive emergency". It's a difficult trial being conducted in the Emergency Department and a lot of data is missing, in fact over 25% of BP measurements after therapy is administered are missing. The primary endpoint was the mean reduction in BP in the first 2 hours of the study. The trial reported a significant difference and handled the missing data by censoring all missing data. As an FDA reviewer, you are suspicious of the conclusions of the trial and want to undertake a data analysis using imputation techniques, employing sensitivity testing for a "tipping point". Please explain how this is done and what are the possible outcomes using this analysis and how it effects the "robustness" of the study conclusions. (2 page limit).