We conduct ongoing clinical studies and outcomes research, as part of our company's dedication to advancing healthcare and our commitment to clinical excellence. Our HART blood tests are backed by robust clinical studies, outcomes research, and over 32 peer-reviewed presentations and publications.
We invite you to explore the latest evidence and see the impact our tests can have in your patient's cardiovascular health management.
Application of a Machine Learning-Driven Multibiomarker Panel for Prediction of Incident Cardiovascular Events in Patients with Suspected Myocardial Infarction. June, 2020
This is a study aimed at externally validating a multiple biomarker panel for predicting future cardiovascular events (CVEs) in patients with suspected myocardial infarction (MI) presenting to the emergency department. The study recruited 748 patients with suspected MI and used a multiplex assay including 4 biomarkers (NT-proBNP, KIM-1, osteopontin, TIMP-1). A predictive score ranging from 0 to 10 was derived in the CASABLANCA trial with 649 stable and unstable patients with suspected coronary artery
disease or MI. The study followed all patients for one year to assess incident CVEs. The predictive score had an AUC of 0.86, high negative predictive value of 99.4% and sensitivity of 86.4% for incident CVE.
This study aimed to identify biomarkers to predict cardiovascular events (CVE) in patients with chronic kidney disease (CKD) undergoing coronary catheterization. The study used a sample of 927 patients from the Catheter Sampled Blood Archive in Cardiovascular Diseases (CASABLANCA) and followed them up for two years. Using machine learning and targeted proteomics, the study found four biomarkers (kidney injury molecule-1, N-terminal pro B-type natriuretic peptide, osteopontin, and tissue inhibitor of metalloproteinase-1) that were integrated into a prognostic algorithm to predict CVE. The algorithm was able to accurately predict 2-year cardiovascular events with a C-statistic of 0.77, and patients with stage 3-5 CKD and higher risk of CVE had an 8.32 times higher hazard ratio for cardiovascular events compared to those at lower risk. The study concluded that measuring these biomarkers prior to coronary catheterization may be useful in individualizing CVE risk assessment for patients with CKD.
A study performed on 446 patients with chronic kidney disease (CKD) undergoing coronary or peripheral angiography aimed to predict incident cardiovascular events using a multi-biomarker panel. The panel consisted of four biomarkers (NT-proBNP, kidney injury molecule-1, osteopontin, and tissue inhibitor of matrix metalloproteinase-1) and was used to predict incident cardiovascular events (myocardial infarction, stroke, or cardiovascular death). Results showed that patients with CKD are at high risk for incident cardiovascular events, and the use of a multiple-protein biomarker panel could assist in identifying patients at higher risk, particularly those at high stages of CKD. The study concluded that biomarker panels such as the Heart CVE™ panel may be useful in individualizing CVE risk assessment and lowering the incidence of CVE among patients with CKD.
Performance of Novel Multi-Biomarker Based Scoring Model for the Prediction of Incident Cardiovascular Events: A Pooled Multi-national Validation Study American College of Cardiology (ACC) Scientific Sessions. May 15, 2021
A multinational pooled analysis of 1522 patients found that a multiple biomarker scoring model can accurately identify patients at low risk for Major Adverse Cardiac Event (MACE) consisting of Myocardial Infarction (MI), stroke, or cardiovascular death. The biomarker scoring model had a high accuracy of 71% sensitivity, 74% specificity and a negative predictive value of 98% for MACE. This study highlights the potential of biomarkers in providing early signals for ischemic heart disease.
The Correlation Between Changes in Cardiac Biomarkers and Cardiac Events in Patients with Non-small Cell Lung Cancer (NSCLC) Treated with Stereotactic Body Radiation Therapy (SBRT)-Exploratory Analysis of a Phase II Study American Society for Radiation Oncology (ASTRO) October 26, 2021
This study analyzed 62 samples from 17 patients receiving thoracic radiotherapy (RT) to assess the correlation between cardiac risk score (CRS) and cardiovascular events (CVE). A previously established CRS determined by a validated cardiac biomarker panel of 4 proteins was used to evaluate changes in CRS before, during and after RT. The results showed a significant correlation between an increase in CRS during RT and the development of CVE in follow-up after RT. The study suggests that the increase in CRS during RT may be a predictive marker for RT-induced cardiotoxicity.
Performance of Clinical/Proteomic Biomarker Panels to Predict Coronary Artery Disease Presence or Cardiovascular Prognosis in Patients with and without Diabetes Mellitus American Diabetes Association (ADA) Scientific Sessions. June 24, 2018
The study found that the HART CAD and HART CVE™ biomarker panels are effective in predicting the presence of prevalent coronary artery disease (CAD) and the risk for incident major adverse cardiac events (MACE) in patients with diabetes mellitus. The HART CAD panel had excellent performance for the prediction of ≥ 70% coronary obstruction and the HART CVE™ panel was highly predictive of incident 1-year MACE. The results were comparable to those in patients without diabetes mellitus.
Multiple biomarkers for predicting major adverse cardiac events in patients undergoing diagnostic coronary angiography: results from the Catheter Sampled Blood Archive in Cardiovascular Diseases (CASABLANCA) Study. March 2017
The CASABLANCA study aimed to develop a non-invasive, multi-biomarker based score for predicting incident major adverse cardiac events (MACE) in a moderate to high-risk population undergoing diagnostic coronary angiography. The final score, comprising of four biomarkers (Kidney Injury Molecule-1, NT-proBNP, osteopontin, and Tissue Inhibitor of Metalloproteinase-1), was generated using machine learning techniques. Lower risk patients (score 0-4.5) had a high negative predictive value (NPV) of 97% for predicting MACE, while higher risk patients (score 6-10) had a positive predictive value (PPV) of 36%. The risk score may be useful for patients with suspected or known stable coronary artery disease and has potential in enriching clinical trials of therapies to prevent MACE. The
score can help identify higher-risk patients who require more aggressive medical therapy while avoiding unnecessary measures in lower-risk patients.
“Derivation and External Validation of a High-Sensitivity Cardiac Troponin-Based Proteomic Model to Predict the Presence of Obstructive Coronary Artery Disease” Journal of the American Heart Association (JAHA). August 6, 2020 (Print Issue).
This study used machine learning to develop a novel scoring system (HART CADhs™ panel) that predicts the presence of significant Coronary Artery Disease (CAD) stenosis in at-risk patients. The panel consisted of three clinical variables (age, male sex, prior percutaneous coronary intervention) and three biomarkers (hs-cTnI, adiponectin, and kidney injury molecule-1). The study validated the HART CADhs™ panel in two separate patient cohorts, with an AUC of 0.85 and 0.86 for coronary stenosis ≥ 70% in the internal and external validation cohorts, respectively. The results showed that the HART CADhs™ panel is more accurate than cardiac stress testing for predicting severe CAD, with an AUC of 0.85 compared to 0.52 for stress testing. In patients who could not be ruled in nor ruled out for myocardial infarction via hs-cTnI, the HART CADhs™ panel had an area under the receiver operating characteristic curve of 0.88 (P<0.001).
Kawasaki disease (KD) is a condition that affects mainly children under 5 years of age and is characterized by fever, rash, swollen lymph nodes, and red eyes. The cause of KD is unknown and there is currently no diagnostic test available. The most serious complication of KD is damage to the coronary arteries which requires long-term management. Seattle Children’s Hospital and Prevencio have formed a partnership to develop a highly accurate blood test based on multiple protein panels and AI algorithms for the diagnosis of KD, addressing the need for methods to diagnose and monitor this disease.
“An Artificial Intelligence Derived Blood Test to Diagnose Kawasaki Disease.” Hospital Pediatrics February 2023 (Online Issue) March 2023 (Print Issue). A study aimed at developing an accurate blood test for Kawasaki Disease (KD) was conducted using blood samples from children presenting with symptoms of fever ≥ 38.1 0 C. The study used a proteomics and machine learning approach, using a machine learning method to select protein panels for KD diagnosis and then trained a diagnostic model using 150 patient samples. The final panel identified had three biomarkers: NT- proBNP, C-reactive protein (CRP) and T Uptake (TU). The model had an AUC of 0.92 for accurate diagnosis of KD with a NPV of 96% for low-risk patients and a PPV of 86% for high-risk patients. The results indicate that this protein-based approach using AI and machine learning methods can provide a rapid and straightforward confirmation of KD diagnosis, reducing time to diagnose and treat patients.
Artificial Intelligence Derived Proteomic Panel to Diagnose Kawasaki Disease. October 2021.
A study aimed at developing an accurate blood test for Kawasaki Disease (KD) was conducted using blood samples from children presenting with symptoms of fever ≥ 38.1 0 C. The study used a proteomics and machine learning approach, using a machine learning method to select protein panels for KD diagnosis and then trained a diagnostic model using 150 patient samples. The final panel identified had three biomarkers: NT- proBNP, C-reactive protein (CRP) and T Uptake (TU). The model had an AUC of 0.92 for accurate diagnosis of KD with a NPV of 96% for low-risk patients and a PPV of 86% for high-risk patients. The results indicate that this protein-based approach using AI and machine learning methods can provide a rapid and straightforward confirmation of KD diagnosis, reducing time to diagnose and treat patients.
Performance of Clinical/Proteomic panel to predict obstructive peripheral artery disease in patients with and without diabetes mellitus
European Society of Cardiology (ESC) Scientific Sessions, Munich, Germany. August 25, 2018 - A study was conducted to validate a biomarker clinical/proteomic panel, using proteomics and machine learning, to predict peripheral artery disease (PAD) in patients undergoing diagnostic peripheral angiography and/or coronary angiography. The panel consists of one clinical variable (history of hypertension) and six biomarkers and was tested on a prospective cohort of 354 patients referred for diagnostic peripheral angiography and/or coronary angiography. The results showed that the panel had excellent performance for the diagnosis of PAD in patients with diabetes mellitus (DM) and was highly accurate in predicting the need for revascularization in both patients with and without DM. The study concluded that a clinical/biomarker score with high accuracy for predicting the presence of PAD and need for revascularization was described in patients with and without DM. For commercialization, the HART PAD panel was optimized to consist of one clinical variable (history of hypertension) and three biomarkers (Kidney Injury Molecule-1, osteopontin, and adiponectin).
A Clinical and Biomarker Scoring System to Predict the Presence of Obstructive Peripheral Artery Disease: Results from the Catheter Sampled Blood Archive in Cardiovascular Diseases (CASABLANCA) Study. Accepted for Publication on March 4, 2018.
The study aimed to identify alternative means for evaluating peripheral artery disease (PAD) by using machine learning and big data techniques on a prospective cohort of 355 patients referred for diagnostic peripheral angiography and/or coronary angiography. The
study generated a final test panel and scoring system consisting of one clinical variable (history of hypertension) and six biomarkers with excellent performance for diagnosing PAD and predicting the need for revascularization in patients with and without diabetes mellitus. Low-risk patients had a negative predictive value of 98% and high-risk patients had a positive predictive value of 86%. The score could act as a gatekeeper prior to imaging or invasive testing, reducing cost and exposure to radiation, and also be used to evaluate at-risk patients for vascular complications or adverse effects of drug therapies. For commercialization, the HART PAD panel was optimized to consist of one clinical variable (history of hypertension) and three biomarkers (Kidney Injury Molecule-1, osteopontin, and adiponectin).
The study aims to develop a scoring system to predict the presence of obstructive peripheral artery disease (PAD) in patients referred for diagnostic peripheral angiography. A combination of over 50 clinical variables and 109 biomarkers was analyzed among 258 patients and a score was derived using logistic regression with LASSO. The scoring system consisted of body-mass index, history of hypertension and 6 biomarkers, with an AUC of 0.76 for PAD. The score had 63% sensitivity, 75% specificity, 84% positive predictive value and 50% negative predictive value for PAD. A high score (≥7/10) predicted a higher risk of revascularization within 1 year and up to 4 years, while a low score (≤3/10) had a negative predictive value of 67% for PAD. The study concludes that a clinical and biomarker score can accurately predict the presence of anatomically significant PAD. For commercialization, the HART PAD panel was optimized to consist of one clinical variable (history of hypertension) and three biomarkers (Kidney Injury Molecule-1, osteopontin, and adiponectin).
Open Heart. November 2018. The CASABLANCA Study aimed to identify biomarkers and clinical parameters associated with severe Calcific Aortic Stenosis (AS), a common cause of valvular heart disease in the Western world. A diagnostic model was developed using machine learning and big data, which incorporated age and three biomarkers (N-terminal pro-B-type natriuretic peptide, von Willebrand Factor, and fetuin-A). The model showed good discrimination for severe AS with a high sensitivity, specificity, and negative predictive value. This approach may supplement current diagnostic methods and facilitate the diagnosis of severe AS.
Clinical Cardiology. January 2019. This study analyzed data from 2000-2014 from the National Inpatient Sample (NIS) and the National Health Interview Survey (NHIS) to estimate trends in acute kidney injury (AKI) in adults with and without diabetes. Results showed that the age-standardized rates of AKI hospitalizations increased by 139% among adults with diagnosed diabetes and by 230% among those without diabetes. The study also assessed the accuracy of the AI- driven HART AKI diagnostic test, which assesses variables associated with AKI risk, on patients referred for angiography. The results showed that 4.8% of the 889 patients developed AKI and that HART AKI had a high accuracy (AUC=0.82) for predicting procedural AKI, with a negative predictive value of 98%.
The study focuses on developing a biomarker model to predict Acute Kidney Injury (AKI) in diabetic patients undergoing coronary angiography procedures. The study involved 217 diabetic patients and found that 18 of them (8.3%) developed AKI. The final model consisted of five predictors, three of which had a positive association with AKI, and two had an inverse association. The model had an Area Under the Receiver Operating Characteristic Curve (AUC) of 0.87, with a 100% sensitivity and negative predictive value (NPV) for AKI. The study suggests that the biomarker model is highly accurate in predicting procedural AKI in diabetic patients undergoing coronary angiography and can aid in early detection of risk and alteration of management approach.
HART CVE™ and HART CADhs™ received proprietary CPT PLA billing codes by the American Medical Association (AMA) in 2022. Effective January 1, 2023, the proprietary CPT PLA billing codes were included on the National Clinical Laboratory Fee Schedule by Centers for Medicare and Medicaid Services (CMS), who establish National pricing for laboratory tests. Prevencio is currently working to broaden reimbursement with insurance providers. We also offer other affordable options.
CACs test is a 10-year risk assessment test that assesses total plaque (calcium) burden in a portion of the coronary arteries. CAC measures calcium and does not differentiate between hard/calcified plaque and soft/lipid rich/vulnerable plaque. The lipid rich/vulnerable plaque can rupture and cause obstruction. The CAC test does not provide a 1-year risk of heart attack, stroke, or cardiac death. The HART tests allow the healthcare provider to make informed decisions about prescribing lipid lowering drugs, intensity of lipid lowering drugs or other direct therapies for atherosclerosis based on the score.
We used machine learning, a subset of artificial intelligence, to assess and select the biomarkers and optionally clinical variables. We started with >100 biomarkers and >200 clinical variables. Based on the outcomes, the machine learning platform chose the protein biomarkers in each HART panel. The panels were then internally and externally validated.
There are three proteins and three clinical parameters in HART CADhs™: Troponin (heart stress); Kidney Injury Molecule (cardio-renal syndrome, vascular inflammation); Adiponectin (fatty acid & glucose metabolism); Age; Sex; and history of cardiac intervention (e.g., stent).
There are four proteins in HART CVE™: NT-proBNP (heart stress); Osteopontin (calcification, plaque); Kidney Injury Molecule (cardio-renal syndrome, vascular inflammation); and TIMP-1 (plaque rupture).
No. Fasting is not required for HART tests.
The HART test results provide an actionable and simple-to-understand personalized risk score identifying patients at greater risk of currently having obstructive coronary artery disease (CAD) and patients at greater risk of experiencing a heart attack, stroke or cardiac death (MACE) within 1-4 years. Our clinical team will assist you in interpretation if needed.
The HART CVE™ test yields an algorithmically derived quantified risk score for heart attack, stroke, or cardiac death within the next 1-4 years. The HART CADhs™ test yields an algorithmically derived quantified risk score for detecting plaque and obstructive coronary artery disease (CAD). Both HART CVE™ and HART CADhs™ patient reports provide risk score explanation and individual protein values which can help facilitate highly personalized treatment plans.
Studies have shown the accuracy (AUC) rate is 85-88%, which is higher than standard-of- care stress tests, Coronary Artery Calcium scores (CACs), genetic risk tests and clinical risk scores.
Yes. HART CADhs™ represents “the here and now.” It is a simple diagnostic blood test for detecting plaque and obstructive coronary artery disease (CAD). HART CVE™ represents “the future.” It is a simple prognostic blood test that provides an immediate to short term risk (1–4 year risk) of having a heart attack, stroke, or cardiac death.
A LDT is a regulated type of in-vitro diagnostic test manufactured and validated in a single laboratory. There are thousands of LDT tests performed daily in the U.S. LDT tests are available for doctors to utilize for patient care. LDT’s are important to the continued development of personalized medicine and have been utilized for decades.
HART CVE™ and HART CADhs™ are multi-protein, AI-derived (machine learning), affordable, non-invasive, and highly accurate blood tests. HART CVE™ is the only marketed prognostic (risk) blood test that provides an immediate to short term risk (1–4 year risk) of having a heart attack, stroke, or cardiovascular death.
Prevencio, Inc./Atlas Genomics
2296 W. Commodore Way, #220
Seattle, WA 98199
Please reach out to us at firstname.lastname@example.org to learn more.
1. 2016 European Society of Cardiology (ESC)–HART CAD MGH Late Breaking Science
2. 2017 American College of Cardiology (ACC) Scientific Sessions–HART CVE™ MGH Validation
3. 2017 American Heart Association (AHA) Scientific Sessions–HART PAD Internal Validation
4. 2018 ACC Scientific Sessions–HART AS Internal Validation
5. 2018 American Diabetes Association (ADA) Scientific Sessions–HART CVE™/CAD in Diabetic Patients
6. 2018 ESC Congress–HART PAD in Diabetic Patients
7. 2018 AHA Scientific Sessions–HART AKI Acute Kidney Injury Prognostic Data
8. 2019 ACC Scientific Sessions–HART CVE™ Hamburg Ext Validation
9. 2019 ACC Scientific Sessions–HART CAD Inova External Validation
10. 2019 ADA Scientific Sessions–HART AKI in Diabetics Validation
11. 2019 AHA Scientific Sessions–HART CADhs™ Intl/External Validation
12. 2020 ACC Scientific Sessions–HART CAD vs Stress Nuclear/Echo
13. 2021 Pediatric Academic Societies International Sessions–HART KD™ for Kawasaki Disease
14. 2021 ACC Scientific Sessions–HART CVE™ Multinational Validation
15. 2021 ESC Congress–HART CADhs™ Multinational Validation
16. 2021 International Spinal Cord (SC) Society Scientific Sessions–HART CVE™ for SC Injury
17. 2021 ASTRO Scientific Sessions–HART CVE™ for CV Events in Lung Oncology Radiation
18. 2021 International Kawasaki Disease Symposium–HART KD™ for Kawasaki Disease
19. 2021 AHA Scientific Sessions–HART CADhs™ Diabetic Multinational Validation
20. 2022 ACC Scientific Sessions—HART CVE ™in Diabetic Patients
21. 2022 AHA Scientific Sessions–HART CVE in Chronic Kidney Disease
1. 2017 Journal of American College of Cardiology–HART CAD MGH Internal Validation
2. 2017 American Journal of Cardiology–HART CVE™ MGH Internal Validation
3. 2018 Open Heart–HART AS Diagnostic Data
4. 2018 Clinical Cardiology–HART PAD Diagnostic Data
5. 2019 Open Heart–HART PAD Diabetic Diagnostic Data
6. 2019 Clinical Cardiology–HART AKI Acute Kidney Injury Prognostic Data
7. 2020 Biomarkers in Medicine–HART CVE™ External Validation
8. 2020 Journal of American Heart Association–HART CADhs™ Int/External Validation
9. 2022 European Journal of Preventive Cardiology—HART CVE™ in Diabetic Patients
10. 2022 International Journal of Cardiology—HART CVE™ in Chronic Kidney Disease (CKD) Patients
11. 2023 Hospital Pediatrics—HART KD™ for Kawasaki disease diagnosis