ESC 2017 - Barcelona

A Naturally Randomized Trial Comparing the Effect of Genetic Variants that Mimic CETP inhibitors and Statins on the Risk of Cardiovascular Disease

Presented at the ESC congress 2017 by: Brian A. Ference (Plymouth, MA, USA )

Background

Mendelian randomization studies have consequently demonstrated that LDL-c is causally related with risk of CV disease. In addition, many randomized studies have shown that lowering LDL-c reduces the risk of CV disease with 20% per mmol/L LDL-lowering, irrespective of how LDL levels are reduced. Exceptions to these observations are CETP inhibitors: in the ACCELERATE study, evacetrapib lowered LDL-c by 0.75 mmol/L, but no reduction of CV events was seen. This raises questions about the causal relationship between LDL-c and CV disease and it raises the possibility that clinical benefit of LDL-c lowering depends on how this is achieved.

This study evaluated the causal effect of LDL-c (and other lipoprotein measures) reduction on the risk of CV events, as a consequence of genetic variants that mimic the effect of CETP inhibitors. This was compared with the effect of lower LDL-c as a consequence of genetic variants that mimic the effects of statins (HMG-CoA reductase inhibition), and of ezetimibe (NPC1L1 inhibition) and of PCSK9 inhibition.

In this naturally randomized study, data of 358205 participants from 77 studies (76061 CV events) were analyzed. The primary analysis was conducted in 102837 participants in 14 prospective cohort or case-control studies. External validation analyses were performed in 189539 participants from 48 studies. The primary endpoint was major vascular events (MVE), defined as first occurrence of non-fatal myocardial infarction, stroke, coronary revascularization or coronary death.

Main results

• Results on CETP inhibition alone showed comparable reductions of LDL-c and apoB and a lower risk of CV events, as compared with the absolute reduction of LDL-c (similar effect on MVE per unit LDL-c lowering as with variants that mimic statins, ezetimibe or PCSK9 inhibition).
• A dose-responsecurve was seen with quartiles of the CETP genetic score and risk of MVE.
• In a combined analysis of CETP- en HMG-CoA reductase-inhibition, findings were obtained that were similar to combined trials (all studies tested CETP inhibitors on a background of statins).
• Changes in LDL-c and apoB were discordant; an attenuated change in apoB was seen. A reduction of CV risk in line with the absolute change in apoB, but less than expected per unit LDL-c reduction.

Conclusion

These data show that the causal effect of LDL on CVD is determined by the concentrations of circulating LDL particles (as estimated with apoB), instead of by cholesterol mass carried by these particles (as estimated with LDL-c). The clinical benefit of lowering LDL-c thus depends on the associated reduction of LDL particles, as measured with apoB. The clinical benefit of LDL lowering therefore depends on the method of LDL lowering: the current data suggest that therapies that lower LDL-c by reducing the number of particles (e.g. statins, ezetimibe, PCSK9 inhibitors) should lower the risk of CV events proportional to the absolute reduction of LDL-c or apoB. Therapies that lower LDL-c without proportionally reducing the number of LDL particles (for instance combination therapy of CETP inhibition and statins), should lower CV event risk proportional to the absolute reduction of apoB, which is possibly lower than based on the change in LDL-c.

In the discussion during the press confererence, Ference added that these data might explain why no CV benefit was seen with evacetrapib, and that has been announced that anacetrapib has a beneficial effect on CV events (data of REVEAL were later presented at ESC 2017), because the latter affects apoB levels. The clinical effect may be smaller than expected based on LDL-c reduction, because the effect on apoB is smaller than that on LDL-c concentration. The method of LDL-c lowering therefore affects the attainable clinical effect.

- Our reporting is based on the information provided at the ESC congress -

This study was simultaneously published in JAMA