Summary of clinical data1
Every physician has examples of patients with LDL at target, well-controlled blood pressure and good stress tests, yet they have an event. We have an urgent need to accurately risk stratify patients and personalize their treatment. The CVD Risk Score will be an important new tool in our armamentarium to help us identify the highest risk patients, intervene more deliberately, monitor our interventions and ultimately improve outcomes cost-effectively.
Objectives: The aim of the study was to derive and validate a multi-protein model to calculate a cardiovascular risk score that predicts cardiovascular outcomes, including heart attack, stroke, congestive heart failure or death among patients with apparently stable coronary heart disease. An additional objective was to analyze longitudinal samples collected nearly five years apart to determine whether the follow-up CVD Risk Score was greater in people who experienced a cardiovascular event after the second sample than in people who remained event-free.
Methods: The study involved 1,909 patients who had a history heart attack, bypass surgery, stent placement or documented blockage of coronary arteries. The SOMAscan assay was used to measure 1,130 proteins simultaneously in each plasma sample. Machine learning techniques were used to select a coherent set of biomarkers and a mathematical model was developed to accurately predict future cardiovascular events. The prediction algorithm was derived from 938 banked samples from the Heart and Soul study, a prospective cohort of patients with stable CHD enrolled from 12 clinics in the San Francisco Bay Area with approximately 11 years of follow-up. From this cohort, a prognostic 9-protein model was constructed and then validated on 971 samples from HUNT3, a prospective population-based cohort study in Norway with approximately 5 years follow-up.
Hazard Ratios, by Disease State
The hazard ratios are shown for the top and bottom quintiles (Q5:Q1). These strong hazard ratios demonstrate that the SomaLogic CVD Secondary Risk Panel can be used with confidence to risk stratify patients.
Event-free Survival, by CVD Risk Score
• The population was ranked by CVD Risk Score, then divided into 10 equal groups.
• Good separation in risk strata.
Responsiveness to Changing Hazard
CVD Risk Scores and Framingham secondary risk scores were analyzed in 514 paired samples. The Framingham formula was re-fit using clinical and laboratory data from the Heart and Soul cohort to optimize performance. The hypothesis was that risk score would be worse when the event is closer. Furthermore, the risk scores for the Event Group should be worse than the Non-Event Group.
Within-subject changes were greater for the Event group than the No Event group (p<0.001). The within-subject increases were significantly greater for CVD Risk Score than for Framingham (p=0.002).
The study showed that the 9-protein model performed better than traditional risk factors that are included in the Framingham secondary event model.
Let’s start a conversation about the SomaLogic CVD Secondary Risk Panel.
1- Ganz P et al. (2016). Development and Validation of a Protein-Based Risk Score for Cardiovascular Outcomes Among Patients with Stable Coronary Disease. Journal of the American Medical Association. 315(23):2532-2541