Substantial variation of time to benefit between cholesterol-lowering agents
Variations in time to benefit among clinical trials of cholesterol-lowering drugsBarter PJ, Waters DD. - J Clin Lipidol. 2018 Jul - Aug;12(4):857-862.
Introduction and methods
When analyzing clinical trial results, the course of the outcome curves receives great attention. The point where they diverge is often the center of interest, here considered to be equivalent to time to benefit (TTB).
TTB is interesting as it may provide insight into the mechanism of action of the treatment being evaluated. Moreover, TTB is a relevant consideration in deciding whether or not, and when to stop a clinical trial because of futility; examples of trials exist in which the curves did not separate appreciably until after 2-3 years, but in which a statistically significant benefit was observed. It has been proposed that TTB should be taken into account in decisions regarding whether or not to treat elderly patients with multiple disease and reduced life expectancy .
This study aimed to compare TTB among clinical trials of cholesterol-lowering drugs. 24 Trials with positive outcomes were included in the analysis. TTB was determined by visual inspection of the primary endpoint outcome curves. This is a method that lacks statistical validity, as in most cases, the point where curves separate, the difference does not approach statistical significance. It should be noted that this analysis is an approximation, as it is not derived from patient data directly.
- Across all 24 trials, TTB varied from 1 to 36 months, with an average of 13.1 months.
- In statin vs. placebo trials (n=14), mean TTB was 11.1 months.
- In all statin trials (n=17), mean TTB was 10.3 months.
- In non-statin trials (n=7), mean TTB was 20.0 months.
- In two trials evaluating PCSK9 inhibitors, TTB was 12 months.
- Among statins, atorvastatin (n=6, 4.74 months) appeared associated with a shorter TTB than other statins (n=11, 13.4 months).
The authors list a number of factors that could potentially influence TTB in cholesterol-lowering trials, and these should be considered when interpreting the results.
- Number of endpoint events, which depends on sample size and event rate
- Baseline LDL-c levels
- LDL-c reduction
- Clinical substrate/patient population: primary prevention, stable coronary disease, post-ACS
- Characteristics of the specific drug (statin/non-statin, differences between statins).
These factors do not account for the difference seen in TTB between atorvastatin and other statin trials. Some evidence suggests that atorvastatin has an active metabolite that functions as an antioxidant, and that it favorably affects lipoproteins [2-4], but it remains speculative whether this is related to the shorter TTB.
These insights may inform those who design or evaluate the design of clinical trials of cholesterol-lowering drugs. For most of these agents, early benefit should not be expected. Indeed, stopping a trial early for futility (event curves not separating), may cause a later benefit of the active drug to be missed. False-negative trials negatively affect the whole field, not only for financial reasons, but it can also affect development of other drugs in the same class, and the understanding of pathophysiology and targeted pathways.
TTB showed a wide variation (1-36 months) between different trials of lipid-lowering therapies, with statins showing shorter TTB’s than other lipid-lowering drugs. In trials evaluating novel lipid-lowering therapies, patience is indicated before declaring futility, as benefit may develop only after two years.