Cardiovascular disease (CVD) remains the leading cause of death in most industrialized and developing countries. Prevention of cardiovascular disease depends on its timely diagnosis until the start of cardioprotective therapy at the beginning of the disease stage. However, the lack of an accurate risk model for predicting individual susceptibility to CVD remains.
As new The sciences of translation medicine The study describes an innovative model of proteomics that predicts the risk of cardiovascular events over the next four years with higher accuracy than current clinical models.
Training: Proteomic surrogates for cardiovascular outcomes that are sensitive to multiple mechanisms of risk change. Image courtesy: alexacrib / Shutterstock.com
Endpoints of clinical trials of cardiovascular drugs include acute coronary events, hospitalization, and death. However, this has led to the development of some drugs before they increase the risk of cardiovascular disease. In contrast, other drugs with promising cardioprotective effects have not been approved for such indications, as these effects were shown too late in the development process.
Traditional cardiovascular risk factors are also not helpful in predicting risk in people with SUD, but controlled levels of cholesterol and blood pressure are not helpful in those with multiple chronic diseases and the elderly.
Finally, many of these risk factors, including age, gender, history of diabetes, and some descriptive factors do not change to reduce the calculated risk-reducing effect when using agents that act independently of these factors. Thus, the researchers in the current study sought to generate and test a new cardiovascular risk model that used newer biomarkers as a measure of outcome instead of the previous clinical endpoint.
“The idealized requirement of accurate and sensitive predictions, which respond to all changes in outcomes regardless of the intervention mechanism in an unconventional and reliable manner, are key features of the surrogate endpoint.. “
Here, the researchers developed a proteomic prognosis score that predicted the actual cardiovascular outcome in a relatively short period of time and at the same time included all known mechanisms and allowed the model to respond to the resulting changes. If successful, this score will be useful for a Phase II study for drugs used in the prevention and treatment of SUE and diabetes, as well as a final point for rapid approval of advanced drugs.
Finally, researchers also expect that their scores can be used to classify drugs into individuals at risk for CVD and to measure patient outcomes.
The researchers measured 5,000 proteins in each plasma sample and used a machine study to use the results to develop a prediction model. Using 27 protein models, it predicted the absolute risk of each of the multiple components forming the endpoint, some of which could lead to heart attack, stroke, hospitalization for heart failure, and death from any cause within four years. occur next. .
This was tested in several groups with several co-morbidities, and changes in parameters were measured over time. In total, more than 11,600 participants were included in the four-year study.
At the time, 22% of the population had experienced one or more of these incidents for a total of 2,500 incidents. These events included 622 hospitalizations for heart failure, 601 heart attacks and 345 strokes.
Of the proteins used in this model, 14 showed a positive ratio and 13 showed a negative ratio. These proteins correspond to ten or more biological processes, for example, in maintaining blood volume and sodium excretion, vesicle formation, angiogenesis, and glomerular filtration rate.
Mendel’s analysis was used to study the possible cause-and-effect relationships between the 16 of these proteins that were present in the PheWAS database. This showed that dozens of them are associated with one or more CVD-related traits.
The current model can also predict the speed of events over a wide range of values. The highest and lowest projected risk quintiles in the first two sets of confirmations showed an increase in the incidence rate from five to seven times in four years. Metacogort, which included all 11,600 participants, also showed an increase in the level of four-year events.
Scientists have also created four risk categories based on protein values. These had a 6%, 11%, 20%, and 43% four-year incidence rate in the six studies that constituted the meta-cohort, respectively. This corresponded to low-risk, low-medium, medium-high, and high-risk, respectively. Moreover, the median lag to the event was less than two years compared to 1.5 years in the highest quintiles.
The model also responded in the right direction to negative and beneficial changes in protein risk. For example, in the ACCORD test, part of the data set used here, the risk of CVD increased by 6% over two years, which correctly predicts future adverse events after taking a second sample. The PRADA test also showed a 6% increase in risk from baseline levels within three months after starting anthracycline chemotherapy.
Beneficial changes in response to exogenous peptide agonist receptor glucagon-like 1 (GLP-1) were seen in the EXSCEL test. The absolute risk of four-year events decreased by 0.8% year-on-year compared to the projected 1.5% decline with this model. In the DiRECT test, another approximately 50% of diabetic remissions were achieved within one year, in which the absolute risk was reduced by 6.7% compared with the standard diet group.
Finally, the model did not show any significant therapeutic effect in the ACCORD test subset, which had severe diabetes control, and for PRADA test patients in response to beta-blockers or angiotensin receptor blockers.
The model also predicted higher risks with a variety of conditions known to increase the incidence rate, such as breast cancer treatment, those with previous events, and those who are now smoking / diabetic / have a history of cancer. In the first case, in the PRADA study, the predicted risk was 14% higher than the predicted 5% of the other group of eligible women.
The model developed in the current study showed a correlation between the rate of event and the absolute risk predicted, which is higher than the existing prediction models. In addition, the current model doubled the dynamic range and better classified cardiovascular risk. Furthermore, this model is biologically compatible because the various biological processes involved in cardiovascular health are mediated and regulated by proteins.
“Reliable identification of individuals with an observed rate of> 50% and an average time to event of 18 months is of clinical and economic importance.. “
Proteins also change depending on the level of gene expression with environmental conditions. All 27 proteins used in the model were associated with processes that predicted high cardiovascular risk. Of these, 16 and 12, respectively, were part of a database that studied the relationship between these proteins and the genome and were associated with a genetic factor for CVD or one of its risk factors.
In the context of positive, negative, and neutral changes in risk factors, this protein-based model showed a true reduction, increase, or change in the predicted absolute risk. When other conditions related to increased cardiovascular events were included in the analysis, including smoking and diabetes, the model continued to predict high risk correctly. It has also been predicted that high systolic blood pressure and high lipid levels in an untreated group increase the risk.
This suggests that the surrogate is universal and will respond to changes in outcome regardless of the mechanism. This multi-protein model is also more sensitive to risk factors than individual biomarkers.
Further work on the same lines can provide the necessary universal endpoint for cardiovascular risk.
Excerpt from the magazine:
- Williams, S.A., Ostroff, R., Hinterberg, MA, and others. (2022). Proteomic surrogates for cardiovascular outcomes that are sensitive to multiple mechanisms of risk change. The sciences of translation medicine. doi: 10.1126 / scitranslmed.abj9625.