–No benefit with evacetrapib despite increase in HDL and decrease in LDL
The fat lady hasn’t sung, the referee hasn’t counted up to 10, but it’s sure not looking good for the once highly promising class of drugs known as the CETP inhibitors.
Bang the Drum Slowly, and Play the Fife Lowly
Following the crash and burn of torcetrapib in the ILLUMINATE trial and dalcetrapib in the dal-OUTCOMES trial, we now have the full obituary details on a third CETP inhibitor, Lilly’s evacetrapib. The one remaining hope for the class lies with Merck’s anacetrapib, which continues to be studied in the ongoing REVEAL study.
Last October Lilly announced that the trial, called ACCELERATE, had been terminated early for clinical futility. Now, at the American College of Cardiology meeting in Chicago, Stephen Nicholls, MBBS, PhD., (University of Adelaide, Australia) presented the full results.
12,000 high risk patients were randomized to evacetrapib or placebo. Evacetrapib, as anticipated, had a potent effect on lipids levels. HDL was dramatically increased by 130%, from 46 mg/dL in the placebo group to a whopping 104 mg/dL in the evacetrapib group. LDL levels also underwent a big change, from 84 mg/dL in the placebo group to 55 mg/dL in the evacetrapib group. One surprising finding was that the hsCRP increased by 4.6% under evacetrapib but decreased by 8% in the placebo group.
But the lipid results did not translate into clinical benefit. The primary endpoint (CV death, MI, stroke, coronary revascularization or hospitalization for unstable angina) was nearly identical in the two groups: 12.8% in the evacetrapib group versus 12.7% in the placebo group. There were no significant differences in any of the secondary endpoints.
There were some additional surprising findings. Investigator-reported hypertension was significantly increased with evacetrapib (11.4% versus 10.1% for placebo, P<0.05). There was a small but statistically significant difference of 1 mmHg in mean systolic blood pressure during the trial (132 with evacetrapib and 131 with placebo).
“Here we’ve got an agent that more than doubles the levels of good cholesterol and lowers bad cholesterol and yet has no effect on clinical events,” said Nicholls, in a press release. “We were disappointed and surprised by the results.”
“The failure of decreases in LDL-C to result in an overall morbidity-mortality benefit emphasizes the limitations of surrogate endpoints,” Nicholls said in his talk. “The findings continue to challenge the hope that CETP inhibition might successfully address residual CV risk.”
The failure of the CETP trials has led many experts to abandon the idea that simply raising HDL levels would prove to be beneficial. Opinion is now divided whether or not HDL has an important independent mechanistic role. Some believe that HDL still plays an important role but the HDL number is less important than measurements of its function.
Elliott Antman, MD, (Brigham and Women’s Hospital) cautioned against any interpretation suggesting that the results in any way seriously threaten the LDL hypothesis. He said that the relatively short followup in the study (less than 3 years) raises concerns that there was not enough time for any benefit from LDL reduction to emerge. The changes in CRP, though statistically significant, were also far too small to raise a clinical concern, he said.
I asked Sekar Kathiresan (Massachusetts General Hospital) whether the ACCELERATE results have any bearing on the LDL hypothesis. Here is his extended response:
I don’t think these results have ANY bearing on the ‘LDL hypothesis’.
The LDL hypothesis is: plasma LDL cholesterol causes coronary atherosclerosis in humans.
Epidemiological, experimental, human genetic, and treatment studies all are consistent with LDL being a causal factor for atherosclerosis. The treatment evidence comes from two LDL lowering mechanisms: statins and ezetimibe. A third LDL lowering mechanism (PCSK9i) has suggestive evidence.
The LDL hypothesis is not: ‘any way of lowering LDL will reduce risk for coronary heart disease’. For example, getting cancer lowers LDL cholesterol. Certainly, wouldn’t expect cancer to improve CV outcomes despite the fact that it lowers LDL.
One can easily envision medicines that are toxic yet lower LDL and such medicines would not be expected to lower CVD risk.
Clearly, CETP inhibition is not beneficial in humans despite lowering LDL.
Potential reasons for this include:
- CETP inhibition leads to other harm that counteracts the LDL lowering.
- RCT considerations (insufficient duration of LDL lowering (e.g., only 2 years; CTT curve expectation based on about 5y treatment)
- off-target toxicity of small molecule inhibitors (both tor and eva increased BP). The degree of BP increase for eva was small but is this an indication of other adverse consequence?
My main takeaways from ACCELERATE are:
- the failure of three CETP inhibitors in large RCTs indicate that this mechanism is unlikely to affect disease risk in humans
- HDL cholesterol in no way can be utilized as a surrogate for CVD benefit in humans