Dallas, TX - Intriguing data presented this past week at the American Heart Association Scientific Sessions 2005 may help explain why whites benefit more from statin therapy than African Americans. Researchers have identified two haplotypes of the gene that codes for hydroxymethylglutaryl coenzyme-A reductase (HMG-CoA), and these haplotypes appear to influence LDL-cholesterol levels and the LDL response to statin therapy in African Americans.

"Statin drugs are widely and appropriately used for heart-disease risk reduction, primarily for their ability to lower LDL cholesterol," lead investigator Dr Ronald Krauss (Children's Hospital Oakland Research Institute, CA) told heartwire. "But we do know that there is quite a range of variation in how people respond to statin drugs—in particular, the primary target, LDL cholesterol. Some patients show a very modest reduction, and some show a very large reduction. This could have a very big impact on outcomes. We were interested in determining whether there was a genetic component to this variability of response."

Krauss noted that other research groups have shown that there are gene markers that have been shown to be predictive of response. One obvious candidate believed to have an effect on LDL-cholesterol levels and response to therapy is the HMGCR gene, the gene that encodes for the targeted enzyme of statin therapy. The purpose of this study, Krauss explained, was to determine whether there were single nucleotide polymorphisms (SNPs) in the HMGCR gene associated with differences in LDL response to statin therapy and whether such effects contributed to differences in response between African Americans and whites.

In this analysis, Krauss and colleagues sequenced the entire HMGCR gene and identified 79 SNPs. Of these, the researchers keyed in on 12 individual SNPs, clustered in nine haplotypes. At the same time, the group carried out a clinical study, enrolling 302 African Americans and 579 whites and treated these subjects with simvastatin 40 mg/day. Compliance as assessed by pill count at intervals during the study was well over 95%, noted Krauss.

Investigators identified three haplotypes related to baseline LDL-cholesterol levels—two associated with a lower LDL level and one associated with a higher cholesterol level—providing evidence that variants of the HMGCR gene contribute to variations in LDL-cholesterol levels. Regarding reductions in LDL cholesterol with statin therapy, one haplotype (HAP7) was strongly associated with changes in LDL cholesterol, but only in African Americans. This haplotype, Krauss explained, is found in about 5% of the population and is slightly more prevalent in African Americans than whites.

"Double-whammy" combination

Investigators then sought to determine whether HAP7 interacted with any other of the HMGCR haplotypes in influencing response to statin therapy. What they discovered was that when haplotype 2 (HAP2) and HAP7 occurred together, there was a profound effect on LDL levels and the response to statin therapy, specifically in African Americans. The combination of HAP2 and HAP7 was essentially protective in terms of LDL-cholesterol levels, resulting in a 38-mg/dL lower LDL-cholesterol level at baseline. On the other hand, the percent reduction achieved by statins was blunted by nearly 50%. In subjects with both HAP2 and HAP7, treatment with simvastatin reduced LDL cholesterol levels by only 22%, compared with the average 40% reduction in the rest of the population.

Krauss told heartwire that HAP2 adds something very significant to HAP7, and in combination, the haplotypes cause a "bit of a double whammy." He pointed out that HAP2 is markedly enriched in African Americans, occurring in 31% of the population, while it is rather rare in whites, occurring in just 2% of the population. Overall, the prevalence of the combined haplotypes is rare, about 3% in African Americans and even less prevalent in whites.

"One of the things that is intriguing about this, stepping back and looking at it, is that people tend to think of a smaller response to the statin, or any drug, as bad," said Krauss. "But here, we found the baseline risk in the same genetic variant was lowered substantially. These people are likely to be at a much lower risk of heart disease because their baseline LDL is lower. If people just focus on the response, you might miss the fact that it might not be nearly as important because they have lower baseline risk."

Krauss said that before such information is clinically relevant and useful, researchers will need to open up the gene field to the entire genome, because there is no way of knowing in advance which genes contribute to differential LDL levels and responses to therapy. Krauss said his group plans to scale up their efforts by looking at more SNPs that might be responsible for both LDL levels and responses to statin therapy.

"The goal of the whole project is to identify a subset of SNPs to be packaged into a clinical profile to assess statin efficacy and safety and to identify those individuals most likely to benefit and those not likely to benefit from drug therapy," said Krauss. "Because statins are so widely used and will continue to be widely used, I think it will be important to be able to sort out where it really makes sense to use big doses and where it doesn't, using a variety of clinical and genetic criteria," he said.