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.

JOURNAL CLUB 2009 - Cardiology

March 2009

Editorial Overview:

Jacques Genest, Jr., MD, FRCPC

Professor of Medicine, McGill University Novartis Chair in Medicine, Montreal, Quebec

Otherwise healthy older individuals without hyperlipidemia benefit from intensive reductions of LDL-C if they have elevated high-sensitivity C-reactive protein (hsCRP), according to the recently published JUPITER (Justification of the Use of statins in Prevention; an Intervention Trial Evaluating Rosuvastatin) results. In the multinational JUPITER trial, rosuvastatin 20 mg brought median LDL-C levels to 1.4 mmol/L with a substantial proportion of individuals achieving LDL-C levels <1.0 mmol/L. Compared to a lack of change in LDL-C among those who received placebo, these levels of LDL-C were associated with very large relative reductions in the primary composite end point of CV events. The results of JUPITER prove that hsCRP can identify a group of individuals without hyperlipidemia who can benefit from intensive LDL-C reductions. The results also confirm intensive LDL-C reductions are as safe in primary prevention as they have been in secondary prevention.

The lower-is-better approach to LDL-C has assumed an added dimension as a result of JUPITER, which tested the hypothesis that intensive lipid lowering can reduce cardiovascular (CV) events in apparently healthy people without hyperlipidemia if they have elevated hsCRP. In this primary prevention trial, the protection against CV events from intensive lipid lowering was robust. A once-daily, 20-mg dose of rosuvastatin was associated with a 44% reduction (HR 0.56, 95% CI:0.46–0.69; P<0.00001) in the primary end point of myocardial infarction (MI), stroke, arterial revascularization, hospitalization for unstable angina, or death from CV causes.

JUPITER establishes that hsCRP is a CV risk marker and that a strategy based on its measurement helps in clinical decision making. In JUPITER, hsCRP was shown to be a marker of risk on par with established CV disease and high-risk diabetes. In a group of patients solely selected for treatment by the presence of elevated hsCRP, the benefits were of such magnitude that JUPITER was terminated prematurely by the Data Safety and Monitoring Committee 1.9 years into a planned five-year trial period.

In JUPITER, 17,802 individuals (men = 50 yearsold or women = 60 years old) were enrolled in 26 countries including Canada. The primary inclusion criterion was an hsCRP =2 mg/L.The primary exclusion criteria were LDL-C =3.4 mmol/L and any history or evidence of CV or cerebrovascular disease. Individuals were also excluded for previous or current use of lipid-lowering therapy, current use of hormone replacement therapy, uncontrolled hypertension, diabetes, or any evidence of significant hepatic or renal dysfunction. The study results were recently published (Ridker et al. NEJM 2008;359:2195-207).

Table 1.

In addition to the reduction in the composite primary end point, 20 mg once daily of rosuvastatin also produced significant reduction in the major components, including a 54% reduction (P=0.0002) in MI, a 48% reduction (P=0.002) in stroke, and a 56% reduction (P<0.0001) in arterial revascularization. All-cause mortality, which was not part of the composite primary end point but which is perhaps the ultimate documentation of clinical benefit, was reduced by 20% (HR 0.80, 95% CI: 0.67– 0.97; P=0.02) despite early termination of the trial.

The reduction in allcause mortality is also indicative of the overall benefit-to-risk profile of a strategy that was not associated with any major safety issues from intensive lipid lowering in an otherwise healthy population. Although there is an enormous databank supporting the safety of statins, the results of JUPITER specifically reinforce the safety of rosuvastatin, which achieved a median 50% reduction in LDL-C on the relatively modest dose of 20 mg once daily, and the safety of very low LDL-C levels whether or not patients have a history of CV disease. Despite initial reports of rhabdomyolysis and muscle weakness when statins were first introduced, there is now a large body of evidence that such side effects are largely confined to high doses. In lower doses, such as 20 mg of rosuvastatin, these events appear to be rare. Indeed, there was no association between rosuvastatin and muscle weakness, myopathy, rhabdomyolysis, or any serious adverse event in JUPITER in a study with a size large enough to detec
events.

Figure 1.

<img2825|center>

In addition, there was no signal of increased risk of cancer, another persistent concern even though the preponderance of data has never supported such an association. In JUPITER, there was a significant reduction in death from cancer among those randomized to rosuvastatin (0.4 vs. 0.7; P=0.02), which is almost certainly a statistical anomaly, much like previous reports of a cancer association with other statins. Both are highly unlikely given the slow progression of cancer. In JUPITER, there was a small but statistically significant increase in the hemoglobin A1c (5.9% vs. 5.8%; P=0.001) and in physicianreported new cases of diabetes (3.0% vs. 2.4%; P=0.01) among those taking the statin, but this finding is consistent with numerous other statin trials. The physician-reported new cases of diabetes is similar to what has been observed in other statin trials (PROVE-IT, HPS, ASCOT-LLA and PROSPER).

The consistency of benefit from intensive lipid lowering on the basis of elevated hsCRP alone was one of the most remarkable results of JUPITER. When patients were stratified by a broad array of characteristics and risk factors, the advantage of rosuvastatin over placebo remained consistently large. There was no statistical interaction for hsCRP with these characteristics. The possible exception was family history of coronary heart disease (CHD). In those with a family history, rosuvastatin provided a greater than 60% risk reduction as opposed to a 40% risk reduction in the absence of family history (P=0.07). Compensating for the poor representation of women in many previous statin trials, JUPITER demonstrated a slight and non-significantly greater benefit in the 6,801 women (38%) when compared to the 11,001 men. When non-Caucasians, who were also well represented in this study
ucasians, the differences were also non-significant.

Table 2.

<img2826|center>

In addition, the benefit of intensive lipid lowering was of similar magnitude for those older than age 65 when compared to those younger, those with a history of hypertension and those without and those with or without the risk factors that might be specifically associated with a pro-inflammatory state, such as smoking, metabolic syndrome, or elevated body mass index. The benefit of rosuvastatin was also independent of Framingham risk score. The relative event reduction was also exactly the same when patients with a greater than 10% risk of an event over 10 years were compared to those with less.

One question unanswered by JUPITER is whether the reductions in hsCRP contributed to the outcome. Although rosuvastatin has previously demonstrated a high degree of potency against both LDL-C and hsCRP in modest doses, it is unclear whether hsCRP is a marker or mediator of inflammation that can be targeted independently. Previous studies demonstrate that LDL-C and hsCRP are not well correlated. While further studies are needed to address any independent contribution for achieving a reduction in hsCRP, JUPITER does strongly reinforce the premise of “lower the better” for LDL-C with no plateau yet identified at which there is no further benefit. The median level of LDL-C in JUPITER is the lowest yet achieved in a major statin trial. Approximately 25% of patients achieved a LDL-C <1.0 mmol/L with no signs of adverse effects.

In Canada, as elsewhere, a large number of CV events occur in patients without hyperlipidemia. In JUPITER, the relative risk reduction from 20 mg rosuvastatin was of similar magnitude in those with no risk factor other than hsCRP when compared to those with multiple risk factors, even though the absolute number of events was lower in those with fewer risks. The results of JUPITER suggest that it is now possible to protect a proportion of those patients at elevated CV risk who had not been well identified by other means.

New guidelines influenced by JUPITER are likely. One important consideration will be how to use this marker cost-effectively. It is important to recognize that JUPITER was conducted in individuals, who are already at elevated risk for events by their age alone. Screening hsCRP is not likely to be useful in high-risk patients for whom intensive lipid lowering can already be recommended on existing risk factors. It is not appropriate to extrapolate JUPITER data to younger age groups, but given the strength of hsCRP as a marker of risk, some physicians may consider using this tool in younger patients with a borderline risk profile for whom a high hsCRP might provide an impetus to move to a statin therapy.

Summary

Elevated plasma hsCRP level identifies a group of otherwise healthy individuals who can reduce their CV risk through intensive lipid lowering with rosuvastatin. The benefits of a 20-mg dose of rosuvastatin, which reduced LDL-C by 50%, were observed in the absence of hyperlipidemia and CV disease. The large risk reductions with this primary prevention strategy, which was safe and well tolerated, are expected to generate new treatment guidelines. Such guidelines are likely to recommend hsCRP screening in the types of individuals recruited for JUPITER. The role of hsCRP for identifying risk in other individuals is yet to be determined.

questions and answers

Panel

Jean Bergeron, MD, Université Laval

Robert Dufour, MD, Université de Montréal

Steven A. Grover, MD, McGill University

George Honos, MD, Université de Montréal

Serge Lepage, MD Université de Sherbrooke

Does JUPITER provide new information about evaluating patients for CV risk?

Dr. Bergeron: JUPITER has confirmed the utility of hsCRP as a clinical tool. Now we see that it is not just individuals with elevated lipids who can benefit from reductions in LDL-C. I think this provides a new way of assessing CV risk and identifying some of those patients who are dying from heart attacks without elevated lipids.

Dr. Dufour: JUPITER tells us that there is a substantial proportion of men over the age of 50 and women over the age of 60 at intermediate risk, who do not have diabetes or elevated lipids but can still achieve a large reduction in CV risk with intensive lipid lowering. The impressive aspect of this study was that it was a primary prevention trial. Currently, we have no other way of identifying this group of individuals other than hsCRP.

Dr. Honos: The INTERHEART study showed that nine easily assessed clinical parameters predicted 90% of MIs in all populations studied worldwide. Measuring hsCRP may assist in further refining the CVD risk of a subset of individuals who are at intermediate risk by conventional risk predictors such as Framingham. However, hsCRP is currently expensive to measure (about 200 times the cost of a serum creatinine or glucose measurement) and requires the availability of specialized laboratory equipment not yet widely available in Canada. Serially repeated hsCRP measurements vary substantially and result in as many as 30 to 40% of patients having to be re-classified to a different hsCRP risk category. Moreover, many coexisting conditions, including smoking, chronic obstructive pulmonary disease, HRT, exercise and weight loss, as well as commonly prescribed medications such as glitazones and statins, will affect hsCRP values. In view of this, I predict that it will be challenging for physicians to integrate hsCRP into CVD risk assessment algorithms in routine clinical practice.

Dr. Grover: This was an extraordinary study that clearly demonstrates a benefit associated with rosuvastatin use in primary prevention among individuals with elevated hsCRP. However, the role of hsCRP in routine clinical practice is more complicated. To date, there is conflicting evidence regarding whether the addition of hsCRP testing improves our ability to risk-stratify patients beyond the traditional risk factors included in the Framingham equations. A recent study of the Reynolds Risk Score for men developed by Dr. Ridker and colleagues suggested a small but significant improvement in risk prediction if hsCRP and parental history of premature coronary artery disease (CAD) were added to traditional risk factors. However, the contribution of hsCRP independent of family history was not reported. On the other hand, the Framingham group has published studies demonstrating little incremental benefit to adding hsCRP to traditional risk factors while family history does appear to improve risk pred
CRP, a family history is relatively cheap and not affected by other causes of acute inflammation.

Figure 2.

<img2827|center>

Accordingly, it remains to be determined if elevated hsCRP, given its non-specificity for CAD, will prove to be a useful additional risk factor beyond taking a family history.

Dr. Lepage: Current Canadian guidelines already recommend hsCRP as a tool for gaining more information about CV risk. The contribution of JUPITER is that it has proven that hsCRP is an effective tool by itself for selecting patients who can benefit from intensive lipid lowering.

Do you expect JUPITER results to alter cholesterol treatment guidelines?

Dr. Lepage: I fully expect changes in the guidelines based on the results of JUPITER. We have long known that a substantial proportion of CV events occur in patients with few or no risk factors, and JUPITER has identified a new population we can target. The relative risk reductions in those enrolled in JUPITER were quite significant.

Dr. Dufour: I do expect the guidelines to be changed. One change I think we can expect is to advocate routine measurement of hsCRP in subjects at intermediate risk, particularly in age ranges included in JUPITER. This has been shown to be useful information for guiding therapy.

Dr. Grover: I would not be surprised if the guidelines evolve to include recognition of the JUPITER study. Patients will undoubtedly ask for it to be measured. I also think more clinicians will be interested in measuring hsCRP when they cannot decide whether to treat a specific patient. Nonetheless, I don’t think we know yet how to incorporate these results into current treatment guidelines. Secondary analyses of the JUPITER study may provide additional clues.

Dr. Bergeron: Based on JUPITER, hsCRP appears to be most useful for guiding treatment in patients with intermediate CV risk. I think the guidelines will be changed and the changes will primarily address how these patients should be managed.

Dr. Honos: The JUPITER results may lead to an increase in the “orbit” of individuals who may benefit from long-term statin therapy in the primary prevention setting. The LDL-C target will likely be lowered for selected high-risk primary prevention patients, given that the magnitude of relative CVD benefit in primary prevention statin trials appears to correlate with the extent of LDL-C reduction without any apparent hazard to high-dose statin therapy.

With the addition of 18,000 patients to the statin safety database, are you reassured that aggressive lipid lowering is well tolerated and safe?

Dr. Dufour: I was not particularly concerned about very low LDL-C levels before JUPITER, but more data are always reassuring. There have been several concerns expressed about the risks of very low LDL-C, such as the potential for an increased risk of hemorrhagic stroke. We have other data that refute these concerns, but a study of this size provides very compelling evidence of safety.

Dr. Grover: Statins do appear to be very safe drugs particularly when used among the relatively healthy individuals who were enrolled in JUPITER. In these patients, very low LDL-C levels were also very well tolerated. However, the study was stopped after a median follow-up of only two years, which is relatively short for evaluating a therapy that must be continued indefinitely. I am still reluctant to prescribe any medication for individuals whose overall risk of CVD is low as the absolute risk can’t be reduced much lower. For instance, even a 50% reduction is small if one’s pre-treatment risk over 10 years is 2%.

Dr. Bergeron: JUPITER was a very large study that provided an opportunity to detect relatively infrequent side effects. I found results reassuring because there were no safety issues. This included the safety issues that have been previously attributed to statins. For example, there was no increase in muscle complaints among those on active therapy, and there was only one case of myopathy, which may not have been treatment-related. If there were important risks from rosuvastatin and very low LDL-C levels, JUPITER was of sufficient size to have revealed them, at least for a short period of follow-up.

Dr. Honos: Prior to JUPITER, over 18,000 patients had already been studied in ”high-dose” (80 mg) atorvastatin trials that demonstrated the safety and efficacy of this aggressive approach. I was therefore already reassured! JUPITER, however, did confirm that a median LDL-C of 1.4 mmol/L is well tolerated out to a two-year follow-up and should reassure physicians who are uncomfortable with such low LDL-C values.

Dr. Lepage: We saw very low LDL-C levels in a substantial proportion of patients who were randomized in the TNT [Treatment to New Targets] trial, so I think we already had the evidence that low LDL-C is safe. JUPITER was stopped after two years, so I think you still have to be cautious about predicting safety over a lifetime, but the safety of both low LDL-C and rosuvastatin in a dose of 20 mg once daily was reassuring.

Although terminated early, there was almost a 50% reduction in the primary end point and a 20% reduction in allcause mortality. Does this degree of benefit surprise you?

Dr. Lepage: To me, the results were very surprising. In a primary prevention study, the magnitude of the relative risk reduction was more than I anticipated. I think the most impressive result was the overall mortality benefit, which we have not seen in many of the secondary prevention studies.

Dr. Dufour: I had predicted about a 20-30% reduction in the primary composite end point, particularly based on the baseline level of LDL-C and the degree of lipid lowering. The lar
to a reduction in the hsCRP as well as the LDL-C, but this has yet to be proven. The results were surprising, but they were very encouraging.

Figure 3.

<img2828|center>

Dr. Grover: The relative risk reduction was substantial, but previous studies have suggested about a 1% reduction in risk for every 1% reduction in LDL-C. As LDL-C was reduced by almost 50% on average in this study, the relative risk reduction is about what one would anticipate. However, I would have expected some curvilinear relationship at very low LDL-C levels, so this degree of risk reduction was somewhat surprising.

Dr. Honos: My appraisal of the primary prevention statin trials (WOSCOPS, AFCAPS/ TEXCAPS, ASCOT-LLA and JUPITER) suggests that the relative CV end point reduction achieved is proportional to the amount of LDL-C reduction. This reinforces the linear LDL-C reduction—CVD benefit relation previously described and may warrant a lowering of LDL-C target guidelines in primary prevention. The absolute risk reduction in these patients remains very small, however, and I believe that careful patient selection will continue to be required for treatment to be cost-effective.

Dr. Bergeron: I was very surprised by the magnitude of the relative benefit. The JUPITER population had a much lower absolute risk than many of the patients entered into previous secondary or primary prevention statin trials, so I did not expect so much relative benefit. I was particularly surprised by the significant reduction in all-cause mortality in this population, who would normally be presumed to be at relatively low risk, over such a short period.

Can results of JUPITER be applied immediately to clinical practice?

Dr. Dufour: The JUPITER trial was of sufficient size that I think it is possible to be confident of the results. We do not need another trial before we apply the results in clinical practice. It will be useful to have guidelines for selecting patients for treatment, but I think many physicians will be looking at hsCRP more routinely.

Dr. Lepage: JUPITER has established the importance of hsCRP as a risk marker, but I am not sure that we have enough information about when to measure it and the quality of current assays. I would like more information about how to incorporate hsCRP measurements into routine practice, but I expect this to move forward quickly.

Dr. Bergeron: The Canadian guidelines already recommend measuring hsCRP in moderate risk patients, so I think JUPITER basically confirms this approach. No doubt more physicians will be using hsCRP now that JUPITER provides a better idea of the degree of relative risk reduction that can be achieved by intensive lipid lowering in patients with elevated hsCRP. I do think that the basic conclusions of JUPITER can be applied to practice progressively.

Dr. Honos: The prevalence of an elevated hsCRP in an otherwise low CVD risk population with a low LDL-C is probably no greater than 10%, as attested to by the large number of patients screened relative to the number enrolled into the JUPITER trial. The lack of a low hsCRP control group in JUPITER makes it very difficult to interpret the role, if any, of this inflammatory marker in the selection of patients for aggressive primary prevention with high-dose statins. Moreover, it is not known from JUPITER whether hsCRP is simply a marker of CVD risk or involved in the pathophysiology of inflammatory vulnerable atherosclerotic plaque development. Finally, all statins lower hsCRP in proportion to their ability to reduce LDL-C. For all of these reasons, it is difficult to immediately apply the results of JUPITER to clinical practice and we await updated lipid guidelines for guidance.

Dr. Grover: JUPITER was a very large and well performed study that demonstrated that the benefit of intensive lipid lowering in patients with elevated hsCRP outweighed the risk. However, there is the question of absolute benefit, which was modest given that the underlying event rate was low even in the placebo group (<3% over two years). I think we have to consider other issues in the very low-risk individuals who were enrolled in JUPITER before we can apply them broadly. Not least of these is the additional costs, both in terms of dollars and manpower, of treating all those who might be candidates for intensive lipid lowering based on JUPITER. Given the serious shortage of primary care practitioners and extended waiting times to see a physician in Canada, these are important issues.

JUPITER included substantial numbers of women and minorities. Is it reassuring that the findings are applicable across all patient populations in the age ranges studied?

Dr. Honos: JUPITER’s most important contribution may be the proof that women definitely benefit from statin therapy in the primary prevention setting. This important result should finally silence the critics and strengthen the resolve of family physicians that all appropriately identified high-risk individuals, regardless of gender, will benefit from aggressive LDL-C reduction.

Dr. Grover: I have never been concerned that the effects of lipid lowering are different in females vs. males or among specific minorities. Nonetheless, it is always reassuring to have more data that is consistent with earlier results.

Dr. Lepage: It is good to have a study with a large number of women, particularly a primary prevention study, because women tend to have their events older so there may be more of an opportunity for primary prevention.

Dr. Dufour: I did not need reassurance. I have strongly believed that the lipid principle of treatment applied to both men and women. Perhaps having a large number of women makes it easier to convince those who have been withholding judgment.

Dr. Bergeron: The consistency of benefit across genders and minority groups was very compelling. I think it might be safest to say that the benefits are comparable in all individuals with Western lifestyles. In Asian countries where individuals have much lower levels of LDL-C and hsCRP to begin with, it may not be necessary to achieve the absolute risk reductions observed in JUPITER, but in Canada I think JUPITER tells us that benefits are relatively uniform.

Is hsCRP a reliable and consistent measure or are there circadian rhythms, dietary influences or other issues that must be considered? In other words, how easy is hsCRP as a clinical tool?

Dr. Grover: In labs that evaluate hsCRP routinely, these measures are fairly reliable. However, not all labs in Canada may have the same experience with this test. Moreover, I think that there is concern about high false positive hsCRP measures in patients with an acute or chronic inflammatory condition, so clinicians should be aware of these other factors that could undermine the value of an elevated result.

Dr. Dufour: We have been routinely measuring hsCRP at our centre for three or four years, and I can say that it is an easy test to perform and results are generally easy to interpret. There is some biological variability in the same subjects over time, and in presence of a high value, an infectious or inflammatory condition needs to be ruled out; but one always has to consider any laboratory result in context. Two consecutive values below 2 mg/L is pretty reassuring in terms of CV risk.

Dr. Honos: As stated peviously, hsCRP is a relatively expensive and not, at present, universally available inflammatory marker with high serial variability. It will therefore not be easy for clinicians to use it as a clinical tool. I believe that we are very likely to see the development of new, more specific and serially reproducible inflammatory markers in the future.

Dr. Bergeron: I think the precision of the available assays is now fairly reliable, but I also think that it is important to obtain at least two hsCRP results separated by two or three weeks. At our centre, we have obtained two or more values of hsCRP in more than 1000 patients over the last two years and we do see variability when values are taken six months apart if not adjusted for confounding factors, such as smoking or acute illness. Clinicians should [also] question patients about recent infections or other factors that affect results.