Cholesterol-reducing drugs are big earners for the pharmaceutical industry, despite the fact that cholesterol reduction is of dubious benefit and these drugs don’t work too well in terms of savings lives. In addition, some evidence of late has called into question the mantra that ‘lower is better’ when it comes to levels of supposedly unhealthy LDL (low density lipoprotein) cholesterol.
By way of example, consider the results of the so-called ENHANCE trial, which pitted two cholesterol-reducing drugs (ezetimibe and simvastatin against just simvastatin. In this study, researchers tracked the thickness of the main artery supply blood to the brain (technically known as carotid artery intima media thickness or ‘CIMT’) in the study participants. The results (which had to be forced out of the manufacturers) showed that those taking the two drugs actually saw a significant drop in LDL cholesterol levels of 58 percent compared to 41 percent in the single-drug group. Yet, the amount of plaque progression was about twice in the two-drug group than the mono-drug takers (although this was not statistically significant). More about this study can be read here.
Of course the manufacturer of ezetimibe (Schering-Plough) was not going to leave it there. This week saw the publication of a study which looks, on the face of it, as an attempt to re-instil our faith in the ezetimibe [1]. The study is in fact a ‘post hoc analysis’ of a previously published study [2]. Post-hoc analysis essentially means reviewing study data for findings that the study was not actually designed to elicit. In scientific circles, the results of post-hoc analyses are generally viewed less reliable as those that come from analysis of data in accordance with the original study design.
The previously published study involved treating about 500 diabetics with drugs to reduce blood pressure and cholesterol. Some of these were designated to received ‘aggressive treatment’ to reduce LDL levels to up to 70 mg/dl, while the rest were left on ‘standard’ treatment with a target LDL of up to 100 mg/dl. The aggressively treated group were treated with statins. But if that did not do the job, ezetimibe was added. In total, just 69 individuals took ezetimibe.
This week’s study was published on-line in the Journal of the American College of Cardiology (JACC) [1] and found that the more aggressive treatment was associated with improvements in carotid artery thickness compared with less aggressive treatment. This study is being used to vindicate the use of ezetimibe for driving LDL levels to very low levels.
I have a few problems with the JACC study and its conclusions. The first thing, as pointed out above, is that it’s a post-hoc analysis. Also, the way individuals ended up on ezetimibe was not ‘randomised’. Normally, whether a study participant gets a treatment or not is down to chance. This helps to ensure that the characteristics of the group being treated and those not receiving treatment are the same. That didn’t happen here: individuals got ezetimibe if their LDL was deemed not to have fallen low enough on statins alone.
I suppose it’s also worth pointing out. lest we forget, that in the far larger ENHANCE study that was properly randomised, ezetimibe plus statin brought no benefits in terms of carotid artery thickness compared to statin alone.
More disquiet for me comes from comments from the lead author of this study, Jerome Fleg. Here you can see that he has stated: “The importance is getting down to the target and it doesn’t matter how you get there.” He also suggests that an LDL target of 70 mg/dl should be considered for patients at high risk of heart disease, such as diabetics. The reason for my disquiet is that there is evidence linking ezetimibe with an increased risk of cancer and even death from cancer. See here and here for more details about this.
The link between ezetimibe and cancer is also underpinned by quite a body of evidence linking low cholesterol levels with enhanced cancer risk. Not so long ago I wrote about the results of a study which analysed the relationship between LDL cholesterol levels and cancer risk in diabetics [3] (remember, it’s diabetics who were the subject of this week’s ezetimibe-trumpeting study).
The researchers involved in this study looked at the relative risk of cancer in 3 groups of individuals divided according to their LDL levels. The three ‘bands’ were:
LDL level of less than 108 mg/dL (2.8 mmol/L)
LDL level of between 108 – 147 mg/dL (2.8 up to 3.8 mmol/L)
LDL level of 147 mg/dL (3.8 mmol/L) or above
Compared to those in the middle band, those individuals with lower cholesterol levels were found to be, overall, at a 74 per cent increased risk of cancer. Remember, this lower LDL cholesterol level is higher than the target levels now being recommended for diabetics.
The authors also analysed not just the risk of cancer, but the overall risk of death, in each of the three LDL-cholesterol bands. Compared to those in the middle band, those individuals with lower cholesterol levels were found to be, overall, at a 50 per cent increased risk of death.
So, in short, what this study showed is that in a group of diabetics, an LDL level of less than 108 mg/dL (2.8 mmol/L) is associated with an increased risk of cancer and overall risk of death. With this in mind, I’m not sure I’d be as enthusiastic as Jerome Fleg (and many others) appear to be about driving LDL levels to ever-lower levels. Let’s make sure this isn’t killing people first.
References:
1. Fleg JL, et al. Effect of Statins Alone Versus Statins Plus Ezetimibe on Carotid Atherosclerosis in Type 2 Diabetes. The SANDS (Stop Atherosclerosis in Native Diabetics Study) Trial. J Am Coll Cardiol (Published online 3 December 2008)
2. Howard BV, et al. Effect of lower targets for blood pressure and LDL cholesterol on atherosclerosis in diabetes: the SANDS randomized trial.
JAMA. 2008 Apr 9;299(14):1678-89.
3. Yang X, et al. Independent associations between low-density lipoprotein cholesterol and cancer among patients with type 2 diabetes mellitus. Canadian Medical Association Journal. 2008;179(5):427-437
