Reports

This report is based on medical evidence presented at sanctioned medical congress, from peer reviewed literature or opinion provided by a qualified healthcare practitioner. The consumption of the information contained within this report is intended for qualified Canadian healthcare practitioners only.

MEDICAL FRONTIERS - 81st Scientific Sessions of the American Heart Association

New Orleans, Louisiana / November 8-12, 2008

The strategy of employing high-sensitivity C-reactive protein (hsCRP) as a tool for selecting otherwise healthy patients for lipid-lowering therapy is impressively effective. In JUPITER (Justification of the Use of Statins in Prevention; an Intervention Trial Evaluating Rosuvastatin), there was a 44% (HR 0.56, 95% CI: 0.46-0.69; P<0.00001) reduction in the primary composite end point of cardiovascular (CV) events for a population of patients without hyperlipidemia who were randomized to rosuvastatin relative to those randomized to placebo (Figure 1). In patients with hyperlipidemia, the efficacy of statin therapy for reducing CV risk is well established, but the risk reduction with rosuvastatin in this non-hyperlipidemic population was greater than that seen in any previous statin trial.

If results of JUPITER were translated to the US population, “this simple screening and treatment strategy could conservatively prevent more than 250,000 myocardial infarctions (MIs), strokes, revascularizations, and CV deaths over a five-year period,” reported senior author Dr. Paul. M. Ridker, Center for Cardiovascular Disease Prevention, Harvard Medical School, Boston, Massachusetts. This rate of protection would translate into an anticipated reduction of 25,000 such events in Canada. The number needed to treat (NNT) with rosuvastatin to prevent one event in JUPITER was 25, which is lower than that generated by trials in which statins were employed to treat hyperlipidemia.

In JUPITER, 17,802 men and women were randomized at more than 1300 study sites in 26 countries, including Canada. The major criteria for entry was an hsCRP level >2.0 mg/L and baseline LDL-C level <3.4 mmol/L. All individuals were apparently healthy with no history or evidence of CV disease. Exclusion criteria included any previous or current use of lipid-lowering therapy, current use of hormone replacement therapy, any evidence of significant hepatic or renal dysfunction, uncontrolled hypertension, or diabetes. Once entered into the study, patients were randomized to rosuvastatin 20 mg or placebo once daily.

Follow-up visits were scheduled for five years after randomization, but the data safety monitoring committee (DSMC) recommended termination of the study after only a median of 1.9 years of follow-up because of overwhelming evidence of benefit in the actively treated arm. Dr. Ridker characterized the early termination as “appropriate,” particularly when the complete data set was available for analysis. In less than two years of follow-up, the large benefit on the primary outcome was accompanied by a significant 20% reduction (P=0.02) in all-cause mortality.

Figure 1.

“On the basis of this study, I think the guidelines for risk assessment will now have to be reviewed,” stated Dr. Andrew Tonkin, Head, Cardiovascular Research Unit, Monash University, Melbourne, Australia. Selected as the official discussant of JUPITER at the AHA, Dr. Tonkin cautioned that the study was not designed and cannot confirm that hsCRP is a causal mediator of CV events, but he did agree that hsCRP is a potent marker for identifying individuals at elevated CV risk even in the absence of other major risk factors, including hyperlipidemia.

LDL-C Levels Remain Critical

“We must not forget the lipid hypothesis is alive and well,” stressed Dr. Tonkin, suggesting that he remained circumspect about hsCRP as a specific target of treatment. He emphasized that rosuvastatin “is the most potent of the LDL-lowering agents we have available,” which makes the importance of treating hsCRP independent of LDL an issue that remains unresolved. Rather, the study only shows that hsCRP can be an effective tool for selecting patients for treatment.

Regardless of whether hsCRP is a mediator or a biomarker, Dr. Ruth McPherson, Director, Lipid Clinic, Ottawa Heart Institute, and Professor of Medicine and Biochemistry, University of Ottawa, Ontario, also suggested that JUPITER warrants a change in the guidelines. “These data take us beyond the paradigm of Framingham and show that hsCRP delineates an apparently healthy population who can benefit from lipid lowering,” she stated. “The last Canadian guidelines did stipulate that hsCRP was a useful marker [of CV risk], but these data are likely to change our approach.”

In the JUPITER study, the average 50% reduction in LDL-C levels was accompanied by an average 37% reduction in hsCRP, which is a sensitive marker of inflammation. Triglycerides were reduced by slightly more than 15% relative to placebo, and HDL was relatively unchanged. Despite the important reduction in hsCRP, there was no consensus about its relative contribution to observed benefits, with mixed opinions provided by different experts.

“There was a large reduction in LDL in JUPITER, but the reduction in the primary end point was greater than you would anticipate from the level of LDL reduction based on prior statin trials,” Dr. Ridker noted, adding that the correlation between LDL reductions and CV risk reductions across studies has been relatively consistent in the past. Although Dr. Ridker implied that there might be more to the benefits observed in JUPITER than LDL lowering, he acknowledged that “there is tension created by this study about the inflammation hypothesis. We cannot make any conclusions beyond the data we have.”

These data neither exclude the possibility that the benefit is entirely attributable to LDL-C levels nor the potential for an additional independent benefit from anti-inflammatory activity, whether or not hsCRP is a biomarker alone or a mediator.

JUPITER Patient Populations

Ultimately, this question is less important as an immediate issue for treatment decisions than the evidence that elevated hsCRP identifies a population at risk for CV events that cannot now be detected with other risk markers. In the JUPITER population of men over age 50 and women over age 60, the reduction in the primary end point was consistent across every stratifica
e 2).

Figure 2.

<img2504|center>

“There was no evidence of heterogeneity among any of the subgroups evaluated. This included men and women, race or ethnic group, and Framingham risk score,” Dr. Ridker reported.

Of the nearly 18,000 patients randomized, 38% (n=6801) were women and 29% were non-Caucasian. This is the largest population of women ever studied in a statin trial. The reduction in CV events among women randomized to rosuvastatin vs. placebo was slightly (but not significantly) greater than that observed for men. There were also no significant differences in relative benefit of rosuvastatin vs. placebo for race, geographic region, presence or absence of metabolic syndrome, Framingham risk score of more than or less than 10%, or presence or absence of hypertension.

“There are really no reasons to think that statins would be any less effective in women than in men, but the large number of women randomized in JUPITER closes the door on any concerns,” observed Dr. Beth Abramson, Director of Cardiac Prevention and Rehabilitation, St. Michael’s Hospital, Toronto. “As someone with an interest in CV disease in women, I was pleased that the JUPITER trial took the step of demonstrating an effect in diverse populations.”

The relative benefits of rosuvastatin over placebo were comparable for smokers vs. non-smokers and for patients with a body mass index (BMI) <u>></u>30 vs. those with a BMI <25. Although an anti-inflammatory effect might predict a greater relative benefit in smokers or the obese because of the propensity for these factors to promote an inflammatory state, the consistency of the 40% to 50% reduction in the primary end point for the various groups evaluated was remarkable. The possible exception was the more than 60% reduction in risk of the primary end point among those with a family history of coronary heart disease (CHD). This approached significance (P=0.07) relative to those without a family history, even though the risk reduction among those with no family history was comparable to the study population as a whole.

When individual components of the primary end point were evaluated separately, protection was found to be consistently robust. The risk reductions were 48% (P=0.002) for any stroke, 54% (P=0.0002) for any MI, 65% (P<0.00001) for non-fatal MI, and 46% (P<0.001) for arterial revascularization. Cost-effectiveness analyses have not yet been conducted, but Dr. Ridker told delegates that the reduction in revascularizations over less than two years of follow-up is “particularly likely to be cost-effective.” He also indicated that the relatively low numbers-needed-to-treat for preventing events would predict a favourable cost-effectiveness overall.

Safety

Importantly, the large number of patients provides substantial reassurance about the safety of this strategy. This is true whether the focus is placed on the therapy itself or on achieving very low levels of LDL-C. When the two arms were compared, there was no significant difference in the rate of overall side effects or in any specific side effects. Of the 19 cases of myopathy, for example, nine occurred in the placebo group. Furthermore, there were no significant differences in the rates of muscle weakness or disorders of the hepatic, renal, gastrointestinal, or hematological systems. The rate of newly diagnosed cancer was virtually identical (3.4% vs. 3.5%; P=0.51 for the rosuvastatin and placebo arms, respectively), while the rate of death from cancer was slightly lower on rosuvastatin (0.4 vs. 0.7; P=0.02).

“There is no signal of safety concerns. This is a significant win for the trial,” Dr. Tonkin commented.

The reassuring safety is important because JUPITER produced the lowest LDL-C levels of any large statin trial conducted so far. On rosuvastatin, the median LDL-C levels were 1.4 mmol/L. Moreover, approximately 25% of patients, or more than 2000 individuals, achieved on-treatment LDL-C of <1.0 mmol/L. The safety of these low levels was predicted by previous statin trials, but JUPITER greatly expands these data. Although Dr. Tonkin cautioned that follow-up of more than 1.9 years is needed to confirm long-term safety, Dr. Ridker reported that more than 1000 patients were followed for more than four years by the time JUPITER was terminated, and the safety observed in this study is consistent with other statin trials.

“We have an enormous amount of data from controlled trials with statins. These data have been very consistent for demonstrating high levels of both safety and tolerability,” Dr. Ridker confirmed. He also pointed out that JUPITER associated rosuvastatin with a mortality benefit, the ultimate demonstration of safety and a desirable benefit-to-risk relationship.

Potential New Risk Assessment Tool

The fact that JUPITER was a positive study has been known since the DSMC recommended early termination, but the results have exceeded expectations. Although many guidelines have suggested that hsCRP is a potentially useful risk discriminator, results of JUPITER are expected to produce a reorientation in risk assessment. Currently, such tools as the Framingham risk score are used to select patients for lipid-lowering therapy, but JUPITER demonstrated large risk reductions even in those with Framingham scores that do not now justify therapy. Experts are now challenged with the task of determining how best to incorporate hsCRP into routine management.

“The guidelines are now likely to change. We have to decide as a society what level of risk we are able to accept and what cost we are willing to pay for prevention,” observed Dr. Jacques Genest, Jr., Director of Cardiology, McGill University, Montreal, Quebec. A co-author of the JUPITER study, Dr. Genest cautioned that although hsCRP does appear to be a useful risk stratifier and tool for selecting patients for therapy, it should not be applied indiscriminately. Rather, he expects it be a tool for evaluating patients at intermediate risk who may not be identified with other measures of risk assessment.

“In patients at high risk, we do not need to evaluate hsCRP because we will treat them anyway,” Dr. Genest noted. In young patients at low risk, testing is not economically viable. However, in intermediate-risk patients, this tool may be useful. “We will be meeting with Health Canada to discuss how this new information may be translated into current practice.”

A similar point was made by Dr. Ridker. Although hsCRP proved to be a powerful tool for selecting patients who will benefit from rosuvastatin treatment, he emphasized that the study results are applicable to the study population, which was older men and women. He cautioned against extrapolating results to younger individuals or other types of patients at this time. However, he did suggest that results of JUPITER are likely to draw additional attention to hsCRP as a tool for guiding therapy.

Summary

JUPITER has demonstrated that apparently healthy men over the age of 50 and women over the age of 60 without hyperlipidemia can benefit from lipid-lowering therapy with rosuvastatin when their hsCRP value exceeds 2.0 mg/L. Although JUPITER was stopped after a median follow-up of only 1.9 years, there was a highly significant 44% reduction in the primary composite end point of CV events as well as a 20% reduction in overall mortality. The contribution, if any, of reducing hsCRP levels as well as LDL-C levels to the benefits associated with rosuvastatin in JUPITER is unclear, but the study is expected to alter current guidelines for risk assessment and treatment.