Two large new studies published in JAMA shed important new light on CRP and other biomarkers, but are unlikely to result in significant changes in clinical practice.
In the first report, a large genome-wide association and replication study, followed by a mendelian randomization study, failed to find a causal role for CRP in coronary heart disease. The researchers, led by by Paul Elliott, said that the “this study suggests that development of therapeutic strategies targeting specific reductions in plasma levels of CRP are unlikely to be fruitful.”
In a comment to CardioBrief, leading CRP expert and proponent Paul Ridker responded: “I have always felt that CRP is a terrific clinical biomarker for inflammation (and high vascular risk) but that it is inflammation that is likely to be causal for atherosclerosis, not CRP itself.” [Ridker's full comment is published in full below.]
The second report from a large Swedish cohort found that for novel biomarkers “the gains over conventional risk factors are minimal.” Olle Melander and colleagues wrote in their conclusion:
“What may be relevant to clinical care, however, is not whether changes in predicted probabilities are statistically significant but whether they result in reclassification of individuals to new, clinically meaningful risk categories. Our data indicate that a relatively small proportion of individuals are moved to new risk categories by the addition of biomarkers…”
An accompanying editorial by Svati Shah of Duke and James de Lemos, of the University of Texas Southwestern Medical Center, agreed with the Swedish researchers conclusion that “CRP does not cause CHD” but then asks “an important question”:
“…does it matter whether CRP is a risk marker instead of a risk factor? The answer is both yes and no. Certainly, these null findings make it much less likely that therapies specifically altering CRP levels alone will prove beneficial. However, the study does not negate the role of inflammation as causal in CHD, nor does it exclude a potential biological interaction between drugs acting on other inflammatory pathways and CRP levels. If CRP increases in response to other inflammatory triggers, it may still be a useful tool for personalizing selection of anti-inflammatory therapies, including statins. Alternatively, CRP may emerge as one of several actuarial tools to help identify individuals with sufficiently high risk of CHD for whom statins or other preventive therapies will have favorable risk-benefit and cost-benefit profiles.”
Here is Ridker’s full comment about the Swedish study:
“I am not sure there is much new here – we and others published more than a year ago large GWAS for CRP, so the core data (Figure 1) is essentially an identical replication of prior work (see reference 65 if you want to check this). Beyond that, several “Mendelian Randomization” studies have already been performed in this arena (including a NEJM one last year that appeared about a week or two before JUPITER came out). What is important to understand in this literature is that a “null” Mendelian Randomization may not mean very much; while it does not support causality, I don’t think most genetic statisticians believe it excludes a causal pathway either.”
“I suspect the issue is moot anyway; I have always felt that CRP is a terrific clinical biomarker for inflammation (and high vascular risk) but that it is inflammation that is likely to be causal for atherosclerosis, not CRP itself. So studies such as this really don’t tell us much, even if you believe the Mendelian Randomization concept. For example, you may be surprised to know that Mendelian Randomization has also failed to find HDL to be “causal”. Does that mean that we should abandon research on HDL? I don’t think so.”
“Last, there is a logical shortcoming here. For a biomarker to be useful in clinical practice, it does not need to be in the causal pathway anyway. (Temperature is not “causal” of pneumonia, but it is very useful to measure it!). So even if in the end CRP is proven non-causal, we would still measure it because (a) it predicts vascular risk even when cholesterol is low and other risk factors are absent; and (b) because we have clinical trial data demonstrating that if you have an increased level of RP, you will live longer and have fewer heart attacks and strokes if you take a statin. So my bigger point here is that there probably is no public health message here – this is a discussion of causal pathways and epidemiologic techniques, not of whether or not you want your CRP (or any other biomarker for that matter) measured, etc.”
Allan Jaffe from the Mayo Clinic sent the following comments to CardioBrief:
“CRP polymorphisms: The data presented by Elliott and colleagues make a persuasive case that CRP genetic polymorphisms and perhaps CRP may not be causally related to cardiovascular events. What it does not do is obviate the diagnostic importance of CRP that has derived over many years and a large number of trials. Indeed, it is likely that CRP works because it integrates a large number of inputs and thus provides a signal that can be detected and utilized to define patients at risk. Genetic effects are only one of these many inputs. Thus, these data should not be used to infer that CRP may not be a valuable marker of cardiovascular risk. The inferences are stronger in regard to the likelihood that CRP is etiologic but even so, it is conceivable that some of the other genetic or other inputs over may be more potent risk factors than a given CRP polymorphism and by doing so obscure its importance or it could be that only modest levels of CRP are necessary to facilitate the underlying pathophysiology. The clinical and scientific challenge is now to dissect why CRP seems to be a valuable marker as a way of elucidating the best therapeutic approach.”
“Novel markers: The data from Melander and colleagues are of interest. One of the challenges in this area is to figure out how to predict who is at risk with sufficient assurity that steps can be advocated to avoid morbidity. The C statistic as shown in this study is a very high bar to scale as it is fairly robust with just conventional risk factors alone. Reclassification has been suggested as an alternative and the group benefited in this analysis is exactly the group one might expect would be helped. For those who have no risk factors or those with a large number, new markers are unlikely to add very much. However, most patients are in the intermediate risk group and that group is the group which is hard for clinicians to triage. Thus, the use of biomarkers should be most helpful as indicated by Bayes law in this group and they are. How one uses the markers is key. This study used the markers are continuous variables which is the fairest way to compare markers in this investigator’s view. However, clinicians use cut off values more often since it is difficult to remember the key numbers with continuous data. They then found only modest improvement in predictive accuracy. Pessimists will say that these markers do very little but optimists will opine that eventually, these markers which are each robust in their own right absent “conventional risk factors” may eventually be used in place of some of those factors to provide a more facile assessment. Time will tell.”
Here are the JAMA press releases:
Findings of Genetics Study Does Not Support Causal Association of C-Reactive Protein With Coronary Heart Disease
CHICAGO – An analysis of the association between genetic variations of the inflammation biomarker C-reactive protein (CRP) with coronary heart disease failed to support a causal association, according to a study in the July 1 issue of JAMA.
Coronary heart disease (CHD) is the leading cause of death worldwide. Inflammation plays a key role in the development of CHD at every stage, from initiation to progression and rupture of plaque. CRP is currently the most widely used biomarker of inflammation, according to background information in the article. “There is considerable interest in establishing whether CRP has a causal role in CHD or whether CRP is merely a marker of underlying atherosclerosis,” the authors write.
Paul Elliott, F.R.C.P., of Imperial College London, and colleagues conducted a genetic association study to identify common genetic loci (the specific site of a particular gene on its chromosome) that influence CRP levels and used the concept of mendelian randomization (the randomized allocation of alleles—an alternative form of a gene at a locus—at conception) to examine the possible causal relationship of CRP levels with CHD. First a genome-wide association (n = 17,967) and replication study (n = 13,615) were conducted to identify genetic loci associated with plasma CRP concentrations. Data collection took place between 1989 and 2008 and genotyping between 2003 and 2008. The researchers then carried out a mendelian randomization study of the most closely associated single-nucleotide polymorphism (SNP) in the CRP locus and published data on other CRP variants involving a total of 28,112 cases and 100,823 controls, to investigate the association of CRP variants with coronary heart disease. These findings were compared with findings predicted from meta-analysis of observational studies of CRP levels and risk of coronary heart disease.
The researchers found: “The present genome-wide association study confirms the associations of common genetic variants in the LEPR, IL6R, CRP, and HNF1A loci and APOE-CI-CII cluster with CRP levels. However, the minor allele of SNP rs7553007 and other variants in the CRP locus included in our mendelian randomization study were not associated with CHD risk.”
The authors write that the variants included in their mendelian randomization study are associated with approximately 20 percent lower CRP levels, corresponding to a 6 percent reduction in CHD risk predicted by the meta-analysis of observational studies of CHD risk. “The lack of association with CHD of genetic variants in the CRP locus suggests that the observational data linking CRP levels and CHD may be confounded [factors that can influence outcomes] by association with other CHD risk factors, or reflect a secondary inflammatory response associated with atherosclerosis (reverse causation), rather than indicate a causal relationship.”
“In summary, our mendelian randomization study of more than 28,000 cases and 100,000 controls found no association of variants in the CRP locus and CHD, arguing against a causal role for CRP in atherosclerosis. Moreover, this study suggests that development of therapeutic strategies targeting specific reductions in plasma levels of CRP are unlikely to be fruitful,” the researchers conclude.
(JAMA. 2009;302[1]:37-48.)
Editorial: Biomarkers and Cardiovascular Disease
In an accompanying editorial, Svati H. Shah, M.D., M.H.S., of Duke University Medical Center, Durham, N.C., and James A. de Lemos, M.D., of the University of Texas Southwestern Medical Center, Dallas, comment on the two studies in this week’s JAMA that examine the use of biomarkers for predicting cardiovascular disease.
“What are the implications of these 2 important studies? Ideally, biomarkers would also be risk factors and could be used for both risk assessment and to individualize specific therapies. Large collaborative investigations incorporating genome-wide association study and mendelian randomization as highlighted by Elliott et al offer a blueprint for definitive evaluation of the causal role of intermediate traits such as biomarkers. Similarly, studies such as that by Melander et al exemplify the necessity of comprehensive appraisal of the value of novel biomarkers, including CRP, beyond standard risk factors in specific populations. Studies such as these will help determine which biomarkers are likely to be useful as specific drug targets but also whether they have a potential role in risk assessment or even therapeutic selection. In the future, better biomarkers and more creative strategies for combining them will be needed, along with comprehensive statistical and functional evaluation of causality, to fulfill the promise of biomarkers for personalized medicine.”
(JAMA. 2009;302[1]:92-93.)
Compared to Conventional Risk Factors, Certain Biomarkers May Have Limited Added Benefit For Predicting Cardiovascular Events
CHICAGO – Use of several older and newer biomarkers appears to offer minimal added benefit in the prediction of cardiovascular events compared to conventional risk factors such as high cholesterol and high blood pressure, according to a study in the July 1 issue of JAMA.
“Cost-effective cardiovascular prevention relies on the accurate identification of individuals at risk. However, a large proportion of individuals with cardiovascular events have 1 or fewer of the conventional risk factors, including smoking, diabetes, hypertension, or hyperlipidemia,” the authors write. As a result, the use of recently identified biomarkers to supplement standard risk algorithms has attracted increasing attention in recent years. However, prior studies have reached differing conclusions regarding the usefulness of biomarkers for cardiovascular risk prediction, according to background information in the article.
Olle Melander, M.D., Ph.D., of Lund University, Malmö, Sweden, and colleagues assessed several cardiovascular biomarkers, individually and in combination, regarding their usefulness in predicting future cardiovascular events, compared with conventional risk factors. The study included 5,067 participants (average age, 58 years; 60 percent women) without cardiovascular disease from Malmö, Sweden, who were examined at the beginning of the study, between 1991 and 1994. Participants underwent measurement of older biomarkers (C-reactive protein [CRP] and N-terminal pro-B-type natriuretic peptide [N-BNP]) and newer biomarkers (cystatin C, lipoprotein-associated phospholipase-2 [Lp-PLA2], midregional proadrenomedullin [MR-proADM], and midregional proatrial natriuretic peptide [MR-proANP]. There was follow-up until 2006, using the Swedish national hospital discharge and cause-of-death registers and the Stroke in Malmö register for first cardiovascular events (heart attack, stroke, coronary death). During median (midpoint) follow-up of 12.8 years, there were 418 cardiovascular events and 230 coronary events.
When considered individually, 5 of 6 biomarkers predicted future cardiovascular events and 3 (cystatin C, MR-proADM, and N-BNP) predicted future coronary events in models adjusting for conventional risk factors. “The best combinations of biomarkers were CRP and N-BNP for predicting cardiovascular events and MR-proADM and N-BNP for predicting coronary events. The use of multiple biomarkers minimally improved the accuracy of risk prediction models over and above conventional cardiovascular risk factors and did not reclassify a substantial proportion of individuals to higher or lower risk categories,” the authors write.
The researchers add that what may be relevant to clinical care, however, is not whether changes in predicted probabilities are statistically significant but whether they result in reclassification of individuals to new, clinically meaningful risk categories. “Our data indicate that a relatively small proportion of individuals are moved to new risk categories by the addition of biomarkers—8 percent or fewer when both upward and downward risk category movement are included and fewer than 1 percent when only the movements likely to lead to changes in therapy according to the Adult Treatment Panel III [cholesterol] guidelines are included. Furthermore, these reclassifications result in only modest improvements in the overall concordance between risk categories and actual event rates, as measured by the net reclassification improvement.”
“The challenge will be to find new cardiovascular biomarkers that alone or in combination with existing biomarkers can bring about improvements in risk assessment that are not just statistically significant but clinically significant as well,” the authors conclude.
(JAMA. 2009;302[1]:49-57.)
===========================================================
Don’t lose touch with CardioBrief. Click here to sign up for a daily email newsletter.
===========================================================
Filed under: Uncategorized
