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 - 34th International Stroke Conference

San Diego, California / February 18-20, 2009

In Canada, there are approximately 70,000 myocardial infarctions (MIs) and 50,000 strokes each year. By itself, stroke is the third leading cause of death. Only about 10% of patients fully recover from stroke and the risk of a second stroke within two years of the index event is about 20%. While stroke prevention in survivors of previous vascular disease events, including MI or stroke, should be aggressive, primary prevention provides the best opportunity to reduce morbidity and mortality. Hyperlipidemia is one of the most important treatable cardiovascular (CV) risk factors, but about half of all MIs and strokes occur in individuals with LDL-C levels below the current treatment threshold.

A Measure of Stroke Risk

In a study called JUPITER (Justification for the Use of statins in Prevention: an Intervention Trial Evaluating Rosuvastatin), which tested elevated high-sensitivity C-reactive protein (hsCRP) as a tool for identifying patients with “normal” LDL-C who might benefit from lipid lowering, the reduction in first strokes with rosuvastatin vs. placebo was almost as robust as the reduction in first MIs. Despite evaluating a population with lipid levels widely considered to be optimal in almost all current prevention algorithms, the relative benefit observed on stroke in JUPITER was greater than that observed in almost all prior statin trials.

“Findings were consistent in all subgroups evaluated with notable benefits in high-risk groups and no increased risk of hemorrhagic stroke,” reported Dr. Robert J. Glynn, Center for Cardiovascular Disease Prevention, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts.

The primary results of the JUPITER study were published several months ago (Ridker et al. N Engl J Med 2008;359:2195-207), but stroke, a predefined secondary outcome of JUPITER, deserves a separate analysis for several reasons, according to Dr. Glynn. One is that elevated hsCRP, the major criterion for selecting patients in JUPITER, has been identified as a good predictor of stroke in previous studies.

“Reduction in stroke risk might be particularly notable in individuals with elevated levels of hsCRP. There is a large and consistent body of evidence from observational prospective studies demonstrating that individuals with elevated levels of hsCRP are at an increased risk of stroke. The data are consistent across age groups, men and women, and people from varying geographic locations,” Dr. Glynn remarked.

Reviewing JUPITER

In JUPITER, 17,802 individuals at 1315 sites in 26 countries including Canada were randomized to a relatively modest dose of rosuvastatin 20 mg q.d. or placebo. All participants were apparently healthy without previous history of CV disease and without hyperlipidemia (LDL-C <3.4 mmol/L). Men were required to be <u>></u>50 years of age and women <u>></u>60. The median age was 66. Although many participants had CV risk factors, such as family history (approximately 12%), metabolic syndrome (approximately 41%) or smoking (approximately 16%), diabetes and uncontrolled hypertension were exclusion criteria. The median blood pressure (BP) at entry was 134/80 mm Hg. The median LDL-C at entry was 2.79 mmol/L.

Although participants had a level of LDL-C that is not currently considered elevated in otherwise healthy individuals, they were required to have an elevated hsCRP, defined as <u>></u>2.0 mg/L. The median entry hsCRP was 4.2 mg/L. A marker of inflammation, upregulated hsCRP has been shown to be an independent risk factor for CV events in previous studies. The most recent Canadian treatment guidelines stipulate that hsCRP can be employed for risk assessment, but JUPITER was specifically designed to test the hypothesis that individuals with elevated hsCRP but without hyperlipidemia would benefit from intensive LDL-C lowering.

The degree of benefit was unanticipated. Although the trial design allowed for follow-up visits to be conducted for five years after randomization, the data safety monitoring committee recommended termination of the study after a median of only 1.9 years of follow-up because of overwhelming evidence of benefit in the actively treated arm. In addition to the 44% (HR 0.56; 95% CI, 0.46–0.69; P<0.00001) reduction in the primary composite end point, there were large reductions in all secondary end points, including stroke (Figure 1). Also unanticipated over such a relatively short period of follow-up, there was a 20% reduction (P=0.02) in all-cause mortality.

Figure 1.

Statins and Stroke

The primary protection against stroke was exceptional even in the context of previous statin trials. According to Dr. Glynn, LDL-C has not been considered to be a major risk factor for stroke and this outcome has not always been included as an end point even in secondary prevention. For example, statin therapy reduced stroke by 29% relative to placebo in the 4S (Scandinavian Simvastatin Survival Study), but it was a post-hoc analysis. Although this is consistent with the 30% relative reduction in stroke produced by a meta-analysis of secondary prevention trials, several of these studies also failed to include stroke as a prospectively defined outcome measure.

In primary prevention, the evidence has been even weaker. The reduction in stroke in WOSCOPS (West of Scotland Coronary Prevention Study) and AFCAPS/TexCAPS (Air Force and Texas Coronary Atherosclerosis Prevention Studies) was not only smaller but non-significant, according to Dr. Glynn. Although Dr. Glynn cautioned that the outcomes are difficult to compare because of different trial enrolments and designs, the difference in statins (pravastatin in WOSCOPS, lovastatin in AFCAPS/TexCAPS, rosuvastatin in JUPITER) cannot be discounted as a factor. In one meta-analysis of primary prevention trials, the reduction in the risk of stroke on a statin vs. placebo was only 11%. Again, this reduction did not reach significance.

The most striking difference in results may be between JUPITER and SPARCL (Stroke Prevention by Aggressive Reduction in Cholesterol Levels). In SPARCL, a secondary prevention trial, 4731 patients with a stroke or transient ischemic attack (TIA) within six months of enrolment were randomized to atorvastatin 80 mg q.d. or placebo. Patients with previous coronary heart disease were excluded. The primary end point was a fatal or nonfatal stroke. The median LDL-C at entry, 3.5 mmol/L, was higher at entry in SPARCL relative to JUPITER but still modest. Entry LDL-C levels >4.9 mmol/L were an exclusion criterion.

After a median follow-up of 4.9 years in SPARCL, the 16% reduction in stroke (HR 0.84; 95% CI, 0.71–0.99; P=0.03) on atorvastatin relative to placebo was far lower than the 49% reduction (HR 0.51; 95% CI, 0.35–0.74; P<0.001) in non-fatal MI, which, although not the primary end point, was also measured. The relative reduction in all major CV events was 20% (HR 0.80; 95% CI, 0.69–0.92; P=0.002). In addition, the reduction in ischemic strokes was slightly and non-significantly offset by an increased number of hemorrhagic strokes (55 in 2365 patients randomized to atorvastatin vs. 33 in 2366 patients randomized to placebo).

In JUPITER, there was no signal of an increased risk of hemorrhagic strokes. Serious adverse events on the relatively modest rosuvastatin dose of 20 mg in JUPITER were slightly and non-significantly lower in the active treatment arm than in the placebo arm (15.2% vs. 15.5%). There was no significant increase in reports of muscle weakness, myopathy or rhabdomyolysis relative to placebo. The slightly lower rate of death from cancer on rosuvastatin reached borderline significance (0.4 vs. 0.7; P=0.02). Although patients randomized to rosuvastatin had a significant elevation in hemoglobin A1c at 24 months relative to placebo, Dr. Glynn also reported that they had a significant improvement at 12 months in glomerular filtration rate (Table1).

In JUPITER, the protection against stroke was observed almost immediately. According to Dr. Glynn, the event curves diverged within the first month in favour of rosuvastatin and continued to separate over the full course of follow-up. If stroke outcome is confined to ischemic strokes, the relative risk reduction increased to 51% (P=0.004), “which is nearly spot on for the reduct
legates.

Table 1.

<img2856|center>

When subgroups were evaluated, the relative benefit was consistent by age, gender, body mass index, hypertension, Framingham score stratification, and presence or absence of metabolic syndrome or smoking. The relative benefit within subgroups did not always reach statistical significance but there was no divergence in the direction of benefit. For example, even though JUPITER had the largest female enrolment of any statin trial yet conducted, a wide CI prohibited the stroke reduction from reaching significance. However, Dr. Glynn, who was specifically asked about relative protection in women, reiterated the consistency of the data.

“This was the largest number of women ever randomized. It is important to say that if one considers the total composite end point, women had a significant benefit so the trial did prove that rosuvastatin is effective in women,” Dr. Glynn stated. “The difference [in significance for stroke] is most likely due to sample size, but I think it is safe to say that the results are applicable to women. There was certainly no significant heterogeneity.”

LDL-C or hsCRP Reduction

The large protection against stroke with rosuvastatin in JUPITER relative to other statins in previous studies is attributed by many experts to the 50% reduction in LDL-C, which brought median levels to 1.42 mmol/L over the treatment period. This is the lowest median LDL-C level ever achieved in a large randomized trial despite the relatively modest statin dose. Although those enrolled did not have hyperlipidemia by conventional standards, elevated hsCRP appears to identify a group of patients at high risk. However, rosuvastatin also reduced hsCRP levels by 37%, and the contribution of the anti-inflammatory effect is being debated.

Asked to comment on the JUPITER stroke results, Dr. J. David Spence, Director, Stroke Prevention & Atherosclerosis Research Centre, University of Western Ontario, London, emphasized the importance of LDL-C and said he considers hsCRP a non-specific marker. He does not consider it a treatable risk factor, and he attributes the benefits observed in rosuvastatin to the lipid reduction.

Taking the opposite position, Dr. Ashfaq Shuaib, Director, Division of Neurology, University of Alberta, Edmonton, considers inflammation to be a catalyst to the atherosclerosis which ultimately results in stroke. Although he agrees that LDL-C is a critical target of therapy, he believes that hsCRP is an independent risk marker that can be useful for identifying patients without hyperlipidemia who can still benefit from aggressive lipid lowering that also produces an anti-inflammatory effect.

“For sure, you treat LDL-C in patients with hyperlipidemia regardless of their hsCRP, but on the other hand, we now have evidence from JUPITER that we can identify another population who can benefit from aggressive lipid lowering even when we do not see very high lipid levels,” confirmed Dr. Shuaib, who remarked that he is now measuring hsCRP routinely as a tool for evaluating patients for treatment.

Unlike LDL-C, which actively leads to atherosclerotic plaque growth and contributes to plaque instability and rupture, hsCRP is a marker of inflammation with an uncertain direct pathogenic role in the risk of thrombus formation. Ultimately, however, the debate about the contribution, if any, of lowering hsCRP does not change the application of the JUPITER findings. Whether or not the anti-inflammatory effect of this treatment contributes to benefits, it does not alter the strong and direct correlation between a reduction in LDL-C and a reduction in CV events. The level at which a further reduction in LDL-C provides no further reduction in CV risk has yet to be identified, but JUPITER redefines the level at which LDL-C can be reduced safely for the largest risk reductions yet achieved. Furthermore, it reveals that this risk reduction is available to individuals with elevated hsCRP.

Summary

The JUPITER study has demonstrated that apparently healthy men over the age of 50 and women over the age of 60 without hyperlipidemia can achieve reductions in stroke on rosuvastatin that approach the reductions in MI if their hsCRP value exceeds 2.0 mg/dL. Despite early termination by the JUPITER data safety monitoring committee due to the large risk reductions achieved within the first two years of study, findings associated rosuvastatin with a 48% reduction in all stroke, a 51% reduction in ischemic stroke, a 54% reduction in MI, and a 44% reduction in a primary composite end point of CV events that included stroke and MI. There was also 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 for stroke as well as other CV events.