BNP-Boosting Gene Variant Hints at Possible Screening, Cardioprotective Roles

September 25, 2009

September 25, 2009 (Boston, Massachusetts) — A gene variant associated with chronically elevated natriuretic-peptide levels doesn't seem to confer any overall survival advantage in the general population but may in people with certain specific conditions, such as LV diastolic dysfunction or type 2 diabetes, a cohort study suggests [1]. No such effect was seen in those with other disorders closely associated with natriuretic-peptide metabolism, such as hypertension, coronary heart disease, or poor LV systolic function.

The findings, reported here last week at the Heart Failure Society of America 2009 Scientific Meeting by Dr Guido Boerrigter (Mayo Clinic, Rochester, MN) and associates, are partly consistent with the protective role endogenous natriuretic peptides--which help regulate blood pressure and volume, among other functions--are considered to play in such conditions.

The chronic elevations of B-type natriuretic peptide (BNP) and its amino-terminal congener (NT-proBNP) associated with the gene variant, the T-381C single-nucleotide polymorphism (SNP) rs198389, don't reach the heights seen in, for example, acute heart failure. But they could potentially influence interpretation of the tests when used for diagnosis or risk stratification in some other settings.

Looking forward to a possible day when assays for BNP or NT-proBNP are used to screen the general population for potentially treatable asymptomatic LV dysfunction, the same research team, in an analysis led by Dr Lisa C Costello-Boerrigter (Mayo Clinic) [2], warns that the rs198389 variant is common and would likely influence interpretation of the tests done on such a broad scale.

In people with rs198389, natriuretic-peptide levels go up with the presence of its C allele compared with the T allele; the levels are lowest for the TT genotype, intermediate for TC, and highest for the CC genotype, the group found. Adjusting for an individual's rs198389 genotype, they concluded, would boost the sensitivity and specificity of screening assays for BNP and NT-proBNP.

"It's a way to optimize the test," Costello-Boerrigter told heartwire . The rs198389-related genotype does influence the assay results "and hence would be important to know when and if we get to the point where we're screening the general population."

Boerrigter added that genotyping could also potentially aid the interpretation of natriuretic-peptide test results that don't clearly distinguish low from high risk.

In the case of the BNP assay from Biosite, he observed (there is another BNP assay available from Shionogi USA that is indexed differently), levels below 100 pg/mL are considered to indicate low risk while those >400 point to high risk of heart failure or some other cardiac condition.

"Values that are in between are in a gray zone and would indicate a need for further studies to figure it out," Boerrigter said. But adjusting for the person's rs198389 genotype could potentially place them more firmly in either risk category, perhaps making other tests unnecessary.

Prevalence of rs198389 Genotypes and Associated Median Natriuretic-Peptide Levels (in Pg/Ml) By Assay in Random Sample of 1970 Persons

Assay TT: 32.7% TC: 49.9% CC: 17.4% p*
BNP (Biosite) 20 24 33 <0.0001
BNP (Shionogi) 12 15 20 <0.0001
NT-proBNP (Roche) 55 71 90 <0.0001
*Degree of significance of the C allele as independent predictor of elevated natriuretic peptide level

The group studied a random sample of 1970 people in the community who were at least 45 years old, genotyped for the rs198389 variant, and followed clinically and with the available natriuretic-peptide assays and echocardiography. About one-sixth of the population had the CC genotype and two-thirds had at least one C allele.

Using natriuretic-peptide cut points the group had previously identified for detection of LVEF <40% and <50%, respectively, without regard to genotype, they saw that the number of C alleles had a direct positive relationship with test sensitivity and an inverse relationship with test specificity for both LVEF thresholds. Sensitivity rose and specificity fell in sequence from TT to TC to CC, such that there were consistently differences of about 10% in sensitivity and specificity between TT and CC.

Sensitivity and Specificity of BNP, NT-Probnp for Detection of LVEF ≤40% by rs198389 Genotype in General Population Sample (and Resulting Utilization of Echocardiography)*

Parameter by natriuretic peptide test All patients (%) TT (%) TC (%) CC (%)
BNP (Biosite)        
Sensitivity 81.1 76.5 85.7 83.3
Specificity 82.2 86.1 81.7 75.9
Proportion screened needing echo 19.0 15.5 19.2 25.1
Propo rtion of echoes negative 92.0 87.0 93.7 94.2
NT-proBNP (Roche Diagnostics)        
Sensitivity 83.8 70.6 100 83.3
Specificity 86.1 89.0 86.1 81.0
Proportion screened needing echo 15.2 12.6 15.2 20.2
Propo rtion of echoes negative 89.6 85.2 90.1 92.8
*Natriuretic-peptide level cut points for identifying LVEF threshold: 66 pg/mL for BNP and 228 pg/mL for NT-proBNP, determined without regard to genotype

Survival in the entire population over a median follow-up of 5.6 years did not differ significantly by presence of the C allele--that is, persons with the TT genotype vs the combined group with either the TC or CC genotype. Nor was such a significant difference seen in subgroups with hypertension, coronary artery disease, a history of MI, or an LVEF of <40% or <50%. But among persons with moderate to severe diastolic dysfunction, those with either the TC or CC allele had a significant greater estimated survival at six years than those with the TT allele (85.5% vs 71.6%; p=0.02). A similar pattern was seen among those with type 2 diabetes, although it fell short of significance (90.7% vs 78.8%; p=0.065).

The group's findings follow a 2007 analysis [3], a retrospective case-control study combining several populations, in which the CC genotype was associated with a significant 15% lower risk (p=0.008) of type 2 diabetes compared with a TT/TC genotype and, in a cohort with systolic heart failure, significantly increased BNP levels (p=0.015). Meirhaeghe and colleagues speculated that "relatively higher BNP expression may protect against type 2 diabetes."

The rs198389 analyses were supported by the National Institutes of Health and the Mayo Foundation.