Beyond Heart Failure: Personalized Heart-Disease Treatment Aims at Earliest Signs

September 23, 2013

ORLANDO, FL — "As much as we as heart-failure specialists would like to focus on heart failure, I think the time has come to broaden our net," said a founder of the Heart Failure Society of America and its first president, Dr Jay N Cohn (University of Minnesota, Minneapolis), here at the organization's annual scientific meeting[1].

"If we focus only on heart failure, we will influence only a small fraction of people who will eventually die of cardiovascular disease." Cohn pointed to data suggesting that only 7% of cardiovascular deaths are from heart failure. High blood pressure was about the same, at 7.5%. Stroke was responsible for 16.5%, and coronary heart disease accounted for about 50% of CV deaths.

Cohn wasn't resorting to pessimism to persuade audience members to switch specialties; rather, he was taking aim at some central tenets of cardiovascular prevention and care, in particular their large dependence on risk-factor modification to improve outcomes.

Traditional screening with the goal of disease prevention looks at, for example, blood pressure, LDL cholesterol, blood glucose, adiposity, family history of CV disease, and other markers that bear a statistical relationship to heart disease but are "a very poor and imprecise guide" to phenotypes of cardiovascular disease.

Cohn, by his description, has narrowed the origins of cardiovascular disease to five such phenotypes he says are recognizable early in the disease course and "are not necessarily identified by risk factors." Individuals may predominantly express one or another phenotype, which can then form the basis for interventions tailored to their specific profile. The appropriate interventions—typically "aggressive" lifestyle modification and pharmacologic therapy—are the same as those guided conventionally by surrogate markers like blood pressure or serum lipids.

That's what happens at his center for every patient evaluated for CV disease, he said. They undergo a series of standard, noninvasive tests designed to reveal expression of any of the five phenotypes, which he identified as endothelial dysfunction, structural and functional abnormalities of small arteries and of the "large conduit arteries," atherosclerotic plaque, and ventricular dysfunction and structural remodeling.

"We're talking about the true disease processes," Cohn said in an interview. "It's not blood pressure that can kill you; it's your arteries and your heart that can kill you. So focusing on the target organs is the rational thing to do."

For example, if patient shows dysfunction of the small vessels, "we always put them on an inhibitor of the renin-angiotensin system. If it's large-artery disease, we lower their blood pressure as aggressively as we can, and we usually also include a statin," Cohn said. "If they have plaques, we always give them a statin. It's individualized."

Importantly, the patients are not receiving the drugs according to their traditional indications. For example, ACE inhibitors may be used whether or not blood pressure is elevated, and "most of the patients we put on statins do not have high LDLs. We're treating the disease, not the risk factor."

"Two Divergent Paths"

Discussion at the symposium characterized the phenotype-guided approach to stemming cardiovascular disease in individuals as either a foil or a complement to population-based approaches. Dr Margaret M Redfield (Mayo Clinic, Rochester, MN) remarked on the dichotomy that exists in contemporary medicine. "It seems as if we're pursuing two divergent paths. Individualized medicine with the promise of [for example] therapies tailored to individual genotypes or phenotypes." Then there is population-based care, represented by such guideline-based recommendations as blood-pressure reduction to universal target thresholds, and, Redfield proposed, the development of a polypill

Dr Lee R Goldberg (University of Pennsylvania, Philadelphia), also a speaker at the symposium, characterized the two approaches as complementary. "Reducing sodium in the diet, [recommending] exercise and reduction of body weight, getting blood pressure checked—those are population-based approaches that probably are going to have a pretty big impact."

Then, he said, there are subgroups of the population who will require a more personalized approach. Some persons will never have a heart attack or stroke despite showing an abundance of risk factors, and others without risk factors go on to have clinical events. "I still think there'll be a role for personalized medicine, including genomics and whatnot, to help us further identify those populations that are high risk but we don't capture with the traditional risk factors."

Cohn replied that he thinks "public-health agencies should be emphasizing population health, that is, improving the overall health of the population, to reduce the risks that are, to some extent, associated with these abnormal risk factors. But the physician should not be practicing population medicine," which, he noted, "is of no value when you're dealing with an individual patient."

How Phenotype-Based Care Works

"Most people who come through our program get treated for their phenotype—not for their blood pressure, not for their cholesterol, not for their left ventricular hypertrophy, not for any of their risk factors," Cohn said. The battery of tests takes about one hour and includes measurements of large- and small-artery elasticity and resting and exercise blood pressure, a retinal scan, urinary microalbumin-creatinine ratio, measurement of carotid intima-media thickness, an ECG, echocardiographic assessment of chamber dimensions, wall thickness, myocardial mass, and plasma natriuretic peptides.

Results are categorized as normal, borderline, or abnormal and individually assigned a score of 0, 1, and 2, respectively. The scores are totaled, to a maximum of 20; phenotypes are determined according to the domains of any abnormalities. In a recent cohort of >600 asymptomatic persons who underwent the screening tests[2], Cohn said, 33% scored a total of 0, 1, or 2, signifying no risk or low risk; 36% scored 3, 4, or 5, indicating moderate risk, and 31% were considered to be high risk with scores of 6 or higher. Those who scored high risk, Cohn said, were considered to have advanced disease regardless of whether they were symptomatic and regardless of any conventional risk factors. In fact, he said, scores correlated poorly with mass index, LDL cholesterol, and blood pressure, for example.

With their treatment tailored to phenotype, Cohn said, patients showed reversal of their disease after nine to 12 months. Those who had been determined to be at low risk (score 0 to 2) had very low clinical event rates over six years; those who had been at moderate risk (score 3 to 5) started to show an increase in event rates after about four years, he said. "And those who scored 6 or above began having events immediately, and by six years their event rate approached 50%."

Cohn called the phenotype-based strategy "a comprehensive approach to individualized care."

Cohn discloses owning stock in Cardiovascular Centers and Hypertension Diagnostics (CVC-HS and CVC-HD) and receiving research support from Forest. Redfield had no disclosures. Goldberg discloses consulting for Thoratec, Medtronic, and Respicardia and receiving research grants from Medtronic and Respicardia.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: