The idea that lowering cholesterol is the key to preventing heart attacks and cardiovascular disease has taken a couple of big hits recently. The first came on Mar. 30, when a panel of cardiologists recommended that Zetia and Vytorin, cholesterol-lowering drugs marketed by a joint venture of Schering-Plough (SGP) and Merck (MRK), be used only as a last resort. The reason: A clinical trial adding Zetia to other cholesterol-reducing drugs had failed to show a benefit (BusinessWeek.com, 3/31/08).
But in the furor over Zetia and Vytorin, an equally dramatic but largely unnoticed development occurred the next day. On Mar. 31, AstraZeneca (AZN) announced that it was halting early a 15,000-patient trial of its cholesterol-lowering drug, Crestor—because the drug was working better than expected. The surprising twist: When the patients started taking the drug, their "bad" cholesterol levels were already very low—so low, in fact, that drugs normally would not have been recommended or used. Yet patients on the drug had fewer heart attacks than those untreated in the trial, which was dubbed Jupiter, and the benefit showed up much earlier than expected. "It was stunning to have Jupiter stopped so early," says Dr. James Liao, a researcher in the vascular medicine unit of Brigham & Women's Hospital in Cambridge, Mass., the lead research center for the trial. "It suggests a new paradigm. These drugs may be working in ways other than lowering cholesterol."
That's a heretical notion, given the overwhelming message from doctors, companies, and the media that high levels of bad cholesterol can lead to an early grave and must be reduced. According to national treatment guidelines, everyone's LDL (or bad cholesterol) levels should be brought under 130 mg/dL, and in many cases, lowered as close to 100 as possible.
Yet there have always been doubters about the almighty importance of cholesterol levels, and there is evidence that LDL may be only a part—and a small part—of the story. Half of all heart attacks and cases of cardiovascular disease occur in people with normal or even low levels of LDL (BusinessWeek, 1/17/08), for instance. And the Zetia trial showed that different types of cholesterol-lowering drugs don't bring the same benefit. In that trial, the additional LDL reduction from adding a second drug, Zetia, to the standard statin-type drug, which works differently in the body, seemed not to help patients.
Dr. Paul Ridker, director of the Center for Cardiovascular Disease Prevention at Brigham & Women's Hospital and a professor of medicine at Harvard University, was one of those who thought something else must be going on. The evidence, he believed, pointed to a major role for inflammation in causing cardiovascular disease. He became a proponent of testing blood for a biological marker of inflammation, called C-reactive protein (CRP). Could inflammation be a better indicator of risk than cholesterol levels, he wondered? Maybe statins such as Lipitor work, in part, by reducing inflammation.
Ridker convinced AstraZeneca that it was worthwhile for the company to fund a major trial to test the idea. After all, there was little risk and lots of potential gain for the Anglo-Swedish drugmaker. Its drug, Crestor, had been late to the cholesterol-lowering game, and it lagged behind other drugs in the same "statin" class, such as Pfizer's (PFE) Lipitor. But if the trial showed that Crestor worked in patients with low cholesterol, then it could reach a wider market than the other statins. Instead of just selling it to people with high cholesterol, it might also be used in those with high CRP or other indications of inflammation even when they had normal or low cholesterol. That's a potential expansion of the market by some 25 million to 30 million Americans. "The trial was originally conceived to see how important a factor inflammation is in cardiovascular disease," explains AstraZeneca Chief Executive David Brennan.
The trial Ridker launched was huge: 15,000 patients with high CRP levels. Their average cholesterol level: 108, which is very low. Half the volunteers got Crestor; half got a placebo. Ridker designed the study so that, if the drug reduced events like heart attacks by 25%, the benefit in those getting the drug would be noticeable in 3½ years. "Our expectation was that the trial would take until 2010 or 2011," says Brennan.
Instead, the benefit was so obvious that the trial was stopped in March, more than two years early, so that patients getting the dummy pill could also benefit. Continuing the trial would have been unethical since it would have denied the benefit to those still on the placebo. The company says it will make the actual data public this fall at a scientific meeting. But the effect is surprisingly large. For the reduction in heart attacks to have been seen so early, the benefit in these patients is as high, or higher, than the benefits seen in patients who start with high bad-cholesterol levels. The implication is remarkable: The main reason why blockbuster statins work may not be because they lower cholesterol, but because they reduce the inflammation that leads to heart attacks. "I think statins do work, but maybe not because they lower LDL," says Liao.
Liao's own research has proved that statins have other biochemical effects than lowering cholesterol. Most important, they reduce the amount of an enzyme called Rho-kinase. That, in turn, dials back damaging inflammation in arteries. When Liao knocks down the level of Rho-kinase in rats, they don't get heart disease. And in new, still unpublished work, he has showed that simply reducing Rho-kinase in certain immune system cells is enough to reduce heart disease in rats.
Depending on details of the data from the Jupiter trial, it may be possible for AstraZeneca to convince the Food & Drug Administration that Crestor should be approved for people with low cholesterol but high levels of inflammation in their arteries. That could turn the drug, now a $2.7 billion-a-year best-seller, into a mega-blockbuster. Without FDA approval, the company can't yet market the drug for this new use, "but we think it will take the science in a certain direction, towards inflammation," Brennan says. A year's supply of the 20mg Crestor dose used in the Jupiter trial costs just under $1,200 when purchased online. If just 5 million more people went on the drug, that would represent an additional $5.2 billion for AstraZeneca—a nearly 200% boost from current sales. And if 10 million in the potential market expansion of 25 million to 30 million people were to take Crestor, the company would add some $11.8 billion in additional annual sales.
One hurdle, however, is exactly how that inflammation should be measured. CRP is seen as a bit of a blunt instrument, since it varies considerably among people and requires multiple tests to establish a true reading. The body's immune system kicks into high gear at the first sign of injury or disease, be it a head cold or broken bone, and inflammation increases. In addition, Ridker has been criticized for having patented the test for CRP at the same time that he is pushing for more widespread testing—a conflict of interest. Others, such as Liao, believe that other biological substances, such as the Rho-kinase enzyme, might end up being a better marker.
Either way, the Jupiter trial adds to the growing evidence that the American obsession with "bad" cholesterol levels may be misplaced. In the coming years, doctors and patients may become far more concerned with how inflamed our arteries are.