Building on previous work with Bristol-Myers Squibb, Celera Diagnostics believes it can help broaden the already huge market for statins by extending their use among patients who carry newly identified alleles associated with myocardial infarction risk.
This strategy, if successful, might sidestep the hallmark risk that drug makers face when deciding how and when to use pharmacogenomics — narrowing prescriptions to responders only.
“The markers that predict risk for a heart attack may also actually be markers for drug efficacy,” Tom White, chief science officer at Celera, told Pharmacogenomics Reporter this week. “For many years now, people have thought about markers for drug efficacy being either based on variation in the target for the drug itself — the statin target is HMG-CoA reductase — and secondly in enzymes that metabolize the drug. That’s been the whole history of the pharmacogenetics of drug-efficacy markers.”
Currently, Celera is researching the link between Bristol-Myers’ statin Pravachol and heart-attack alleles for the Fc receptor of immunoglobulin A. If these studies turn out to work the same way as other genes Celera is studying, the molecular diagnostics company said it will produce a panel test that might find additional patients who should take the drugs, but who otherwise would not have been prescribed the drug.
In 2003, 11 million people in the United States were on cholesterol-lowering drugs, which generated $13.9 billion in sales for a handful of drug makers. This number is the tip of the iceberg: Around 36 million people should have been on the drugs based on then-current guidelines, according to an analysis by Forbes magazine drawn from figures by the US National Heart, Lung, and Blood Institute and IMS Health. In 2003 there were 65 million Americans with elevated cholesterol, and if all of them took statins, total annual sales would have reached about $82 billion, said the magazine.
Moreover, a recent report by the American Heart Association shows there were 865,000 new and recurrent myocardial infarctions and 1.2 million heart attacks of all kinds in the United States each year between 1987 and 2002. This year, the AHA estimates that more than 700,000 million Americans, or about 3.5 percent of the population, will have a heart attack.
“The question is, ‘Do all the people that take [statins] need to take [them], and are they deriving benefit?’” said Eric Topol, provost of the Cleveland Clinic. “It’s very expensive — it’s over $100 a month for any of the statins for life,” he said.
The diagnostic Celera envisions would ideally be recommended by doctors to patients when they undergo traditional testing for cholesterol, at ages 45 and older, said White. Celera hopes that if patients show greater-than-average genetic risks of heart attack, or a combination of genetic and environmental or concomitant health risks, they will be prescribed statins based on the results of its studies. Traditional heart-attack risk factors include diabetes, high-LDL, smoking, high blood pressure, and obesity.
But the possibility remains that using Celera’s heart-attack risk markers to identify patients who need statins might decrease the total number of patients taking the drugs. “[Non-industry] randomized trials suggest [that] out of 100 people taking statins, 12 or so derive benefit in terms of reduction of death, heart attack, or stroke,” said Topol. “If we could guide the therapy, rather than giving it to the entire population, to the people [who] would get the most benefit,” it would be unnecessary to prescribe the drugs outside of populations at risk, he added.
Indeed, this switch to reliance on events (such as heart attack rate) as a gauge of benefit, rather than LDL cholesterol level, would probably decrease the number of people taking statins, particularly if diagnostics become capable of predicting those events, Topol stressed. “It doesn’t look like that’s the link that most people used to think — that if you just lower LDL, you lower events,” he said. “I think that whole blockbuster concept is questionable,” he added.
In a study with Bristol-Myers, Celera found marked differences in response to Pravachol between 4,159 carriers of two FC-receptor variants who had already had one heart attack. “If they do have the risk allele … they have about a 2-fold greater risk of dying. That is, they’re at substantially greater risk compared to not having the risk allele,” White said.
Carriers taking Pravachol showed a drop in risk of about 72 percent in six years, said White. “If they didn’t carry [the] risk allele, and they were treated with [Pravachol], basically [it] didn’t do anything for them,” he said. When the heart-attack risk of carriers and non-carriers on the drug was averaged together, carriers showed approximately three times the benefit of the average patient, he added.
In a similar, second study of people who had not had a first heart attack — in which the placebo arm predicted a 1.6-fold increased risk for heart attack in carriers — a total of 6,595 people were followed. Heart attack due to carrier status and drug treatment was independent of traditional risk factors in both studies, White said. The two studies were named CARE — Cholesterol and Recurrent Events — and WOSCOPS, or West of Scotland Coronary Prevention Study.
It is here that Celera hopes what’s good for the goose is good for the gander. Since the Fc-receptor allele is not related to the mechanism of statin action or its metabolism, it follows that other genes not related to those pathways will also benefit carriers of other genetic risks.
Hoping to discern the effect of statins, Celera is conducting two new studies on the five markers in four genes it has already identified, and which it says are independently predictive of heart attack. “That is a study we currently have in the analysis phase,” said White. Celera has genotyped the 16,000 and 6,000 patients involved in these respective studies, he said. “Sometime this year,” the company will present data on these patients, some of whom have taken statins for 15 years, he added.
The discovery and two replication studies that originally identified Celera's five SNPs-of-interest contained 1,600 cases and controls; the second had 3,000; and the third had 1,500 people. The four genes identified included two kinases, a cytoskeletal gene, an olfactory receptor, and a taste receptor. “Odds ratios [for heart attack risk] for single-variant were in the range of 1.4 to 1.8,” said White. In combinations, the genes produced odds ratios of 1.9 to 2.6, compared to 1.9 for diabetes and 1.6 for people having an LDL level greater than 160.
“There may be a whole new category of drug-efficacy markers, and I think we’re in a good position to find out because we’ve not only been studying markers for myocardial infarction, but also for stroke, rheumatoid arthritis, and Alzheimer’s disease,” he added.