Protein-Based Risk Score Improves Prediction of Cardiovascular Events

NEW YORK - A new protein-based risk score outperforms the Framingham model for predicting cardiovascular outcomes in patients with stable coronary heart disease.

"Patients who carry the diagnosis of stable coronary heart disease have been viewed traditionally as a homogeneous population within which all individuals tend to be treated similarly," Dr. Peter Ganz, from the University of California, San Francisco, told Reuters Health by email.

"Instead, we found that individuals who all carried the same clinical diagnosis of stable coronary heart disease had a risk of adverse events (heart attacks, strokes, heart failure, and death) that varied by as much as 10-fold, as revealed by analysis of the levels of nine proteins in their blood," he said.

Dr. Ganz and colleagues sought to derive and validate a score to predict the risk of cardiovascular outcomes among patients with coronary heart disease, using modified aptamers to measure 1,130 proteins in plasma samples.

Aptamers are small nucleic acids that can form secondary and tertiary structures capable of specifically binding proteins and thus can be considered the chemical equivalent of antibodies.

The researchers' unbiased statistical approach identified nine proteins, from which they derived a risk score that reflects the probability of a cardiovascular event occurring within four years.

In both the derivation and validation cohorts, participants had four-year cumulative event rates of 60% to 80% in the highest risk score decile and less than 10% in the lowest risk score decile, according to the June 21 JAMA report.

Compared with the Framingham model, the protein-based risk score showed an absolute increase of 12% in average risk for participants with events compared with participants without events.

The protein-based risk score was within two percentage points of the observed event rate in the external validation cohort.

Moreover, the protein-based risk score changed more than the Framingham model among participants approaching new events, and the protein-based risk score at follow-up was a stronger predictor of subsequent outcomes than the preceding baseline risk score.

"We may now be able to tell individual patients with coronary heart disease, 'You are at a very high risk, medium risk, or a very low risk,' and they may opt to be treated differently from other patients with the same diagnosis," Dr. Ganz said.

"In addition to the results described in the JAMA paper that apply to patients with coronary heart disease, we have an ongoing discovery program to identify proteins that can predict the risk of cardiovascular disease in additional patient populations, including lower-risk individuals who appear healthy but may actually be at high risk of coronary heart disease due to high cholesterol, high blood pressure, diabetes, or smoking, or among individuals who may be at high risk due to kidney disease or HIV infection," Dr. Ganz said.

"Although more accurate risk prediction is always welcome, clinicians more readily embrace measuring a prognostic biomarker or calculating a risk score if the results could alter therapeutic decision making," writes Dr. Marc S. Sabatine from Brigham and Women's Hospital, Boston, in an accompanying editorial.

"To that end, it would be interesting to apply these arrays to samples from patients in randomized clinical trials of therapies," he said. "If a gradient of treatment benefit existed, such data would make measurement of the relevant proteins in clinical practice more compelling (which, for the current list, is impractical). Furthermore, part of the long-term value of this sort of proteomics work may come from exploring the basic pathways that underline some of the novel associations described."

Dr. Matthew Sherwood, from Duke University Medical Center, Durham, North Carolina, who recently described multimarker risk stratification in patients with acute myocardial infarction, told Reuters Health by email, "While the results are impressive, their scope is limited. Since the population studied is already at high risk for further cardiovascular events, more refined risk stratification may not have significant clinical import. These patients have indications for treatment of CAD at present, thus changes in medical management are unlikely."

"Our ability to use proteomic signatures to predict cardiovascular risk continues to expand, and may become available to a broad cohort of patients in the future," Dr. Sherwood said. "The clinical utility of these platforms remains uncertain, and further investigation is needed to determine if proteomic based risk scores could help to modify therapeutic management in lower risk populations."

SomaLogic provided funding for protein assays and employed two coauthors. Four coauthors and the editorialist reported disclosures.

SOURCE: http://bit.ly/28L6oEy and http://bit.ly/28NgaJg

JAMA 2016.

NEW YORK - A new protein-based risk score outperforms the Framingham model for predicting cardiovascular outcomes in patients with stable coronary heart disease.

"Patients who carry the diagnosis of stable coronary heart disease have been viewed traditionally as a homogeneous population within which all individuals tend to be treated similarly," Dr. Peter Ganz, from the University of California, San Francisco, told Reuters Health by email.

"Instead, we found that individuals who all carried the same clinical diagnosis of stable coronary heart disease had a risk of adverse events (heart attacks, strokes, heart failure, and death) that varied by as much as 10-fold, as revealed by analysis of the levels of nine proteins in their blood," he said.

Dr. Ganz and colleagues sought to derive and validate a score to predict the risk of cardiovascular outcomes among patients with coronary heart disease, using modified aptamers to measure 1,130 proteins in plasma samples.

Aptamers are small nucleic acids that can form secondary and tertiary structures capable of specifically binding proteins and thus can be considered the chemical equivalent of antibodies.

The researchers' unbiased statistical approach identified nine proteins, from which they derived a risk score that reflects the probability of a cardiovascular event occurring within four years.

In both the derivation and validation cohorts, participants had four-year cumulative event rates of 60% to 80% in the highest risk score decile and less than 10% in the lowest risk score decile, according to the June 21 JAMA report.

Compared with the Framingham model, the protein-based risk score showed an absolute increase of 12% in average risk for participants with events compared with participants without events.

The protein-based risk score was within two percentage points of the observed event rate in the external validation cohort.

Moreover, the protein-based risk score changed more than the Framingham model among participants approaching new events, and the protein-based risk score at follow-up was a stronger predictor of subsequent outcomes than the preceding baseline risk score.

"We may now be able to tell individual patients with coronary heart disease, 'You are at a very high risk, medium risk, or a very low risk,' and they may opt to be treated differently from other patients with the same diagnosis," Dr. Ganz said.

"In addition to the results described in the JAMA paper that apply to patients with coronary heart disease, we have an ongoing discovery program to identify proteins that can predict the risk of cardiovascular disease in additional patient populations, including lower-risk individuals who appear healthy but may actually be at high risk of coronary heart disease due to high cholesterol, high blood pressure, diabetes, or smoking, or among individuals who may be at high risk due to kidney disease or HIV infection," Dr. Ganz said.

"Although more accurate risk prediction is always welcome, clinicians more readily embrace measuring a prognostic biomarker or calculating a risk score if the results could alter therapeutic decision making," writes Dr. Marc S. Sabatine from Brigham and Women's Hospital, Boston, in an accompanying editorial.

"To that end, it would be interesting to apply these arrays to samples from patients in randomized clinical trials of therapies," he said. "If a gradient of treatment benefit existed, such data would make measurement of the relevant proteins in clinical practice more compelling (which, for the current list, is impractical). Furthermore, part of the long-term value of this sort of proteomics work may come from exploring the basic pathways that underline some of the novel associations described."

Dr. Matthew Sherwood, from Duke University Medical Center, Durham, North Carolina, who recently described multimarker risk stratification in patients with acute myocardial infarction, told Reuters Health by email, "While the results are impressive, their scope is limited. Since the population studied is already at high risk for further cardiovascular events, more refined risk stratification may not have significant clinical import. These patients have indications for treatment of CAD at present, thus changes in medical management are unlikely."

"Our ability to use proteomic signatures to predict cardiovascular risk continues to expand, and may become available to a broad cohort of patients in the future," Dr. Sherwood said. "The clinical utility of these platforms remains uncertain, and further investigation is needed to determine if proteomic based risk scores could help to modify therapeutic management in lower risk populations."

SomaLogic provided funding for protein assays and employed two coauthors. Four coauthors and the editorialist reported disclosures.

SOURCE: http://bit.ly/28L6oEy and http://bit.ly/28NgaJg

JAMA 2016.

NEW YORK - A new protein-based risk score outperforms the Framingham model for predicting cardiovascular outcomes in patients with stable coronary heart disease.

"Patients who carry the diagnosis of stable coronary heart disease have been viewed traditionally as a homogeneous population within which all individuals tend to be treated similarly," Dr. Peter Ganz, from the University of California, San Francisco, told Reuters Health by email.

"Instead, we found that individuals who all carried the same clinical diagnosis of stable coronary heart disease had a risk of adverse events (heart attacks, strokes, heart failure, and death) that varied by as much as 10-fold, as revealed by analysis of the levels of nine proteins in their blood," he said.

Dr. Ganz and colleagues sought to derive and validate a score to predict the risk of cardiovascular outcomes among patients with coronary heart disease, using modified aptamers to measure 1,130 proteins in plasma samples.

Aptamers are small nucleic acids that can form secondary and tertiary structures capable of specifically binding proteins and thus can be considered the chemical equivalent of antibodies.

The researchers' unbiased statistical approach identified nine proteins, from which they derived a risk score that reflects the probability of a cardiovascular event occurring within four years.

In both the derivation and validation cohorts, participants had four-year cumulative event rates of 60% to 80% in the highest risk score decile and less than 10% in the lowest risk score decile, according to the June 21 JAMA report.

Compared with the Framingham model, the protein-based risk score showed an absolute increase of 12% in average risk for participants with events compared with participants without events.

The protein-based risk score was within two percentage points of the observed event rate in the external validation cohort.

Moreover, the protein-based risk score changed more than the Framingham model among participants approaching new events, and the protein-based risk score at follow-up was a stronger predictor of subsequent outcomes than the preceding baseline risk score.

"We may now be able to tell individual patients with coronary heart disease, 'You are at a very high risk, medium risk, or a very low risk,' and they may opt to be treated differently from other patients with the same diagnosis," Dr. Ganz said.

"In addition to the results described in the JAMA paper that apply to patients with coronary heart disease, we have an ongoing discovery program to identify proteins that can predict the risk of cardiovascular disease in additional patient populations, including lower-risk individuals who appear healthy but may actually be at high risk of coronary heart disease due to high cholesterol, high blood pressure, diabetes, or smoking, or among individuals who may be at high risk due to kidney disease or HIV infection," Dr. Ganz said.

"Although more accurate risk prediction is always welcome, clinicians more readily embrace measuring a prognostic biomarker or calculating a risk score if the results could alter therapeutic decision making," writes Dr. Marc S. Sabatine from Brigham and Women's Hospital, Boston, in an accompanying editorial.

"To that end, it would be interesting to apply these arrays to samples from patients in randomized clinical trials of therapies," he said. "If a gradient of treatment benefit existed, such data would make measurement of the relevant proteins in clinical practice more compelling (which, for the current list, is impractical). Furthermore, part of the long-term value of this sort of proteomics work may come from exploring the basic pathways that underline some of the novel associations described."

Dr. Matthew Sherwood, from Duke University Medical Center, Durham, North Carolina, who recently described multimarker risk stratification in patients with acute myocardial infarction, told Reuters Health by email, "While the results are impressive, their scope is limited. Since the population studied is already at high risk for further cardiovascular events, more refined risk stratification may not have significant clinical import. These patients have indications for treatment of CAD at present, thus changes in medical management are unlikely."

"Our ability to use proteomic signatures to predict cardiovascular risk continues to expand, and may become available to a broad cohort of patients in the future," Dr. Sherwood said. "The clinical utility of these platforms remains uncertain, and further investigation is needed to determine if proteomic based risk scores could help to modify therapeutic management in lower risk populations."

SomaLogic provided funding for protein assays and employed two coauthors. Four coauthors and the editorialist reported disclosures.

SOURCE: http://bit.ly/28L6oEy and http://bit.ly/28NgaJg

JAMA 2016.