Could plasma protein levels be a surrogate for VO2 max testing as a measure of cardiorespiratory fitness?
Cardiorespiratory fitness (CRF) has long been associated with improved mortality and morbidity outcomes, with increasing evidence demonstrating that higher CRF levels correlate positively with better health and longevity. Traditionally, the gold standard for assessing CRF is the VO2 max test, which measures the maximum rate of oxygen consumption during a maximal exercise test. However, despite its accuracy, VO2 max testing has significant overhead costs and practical limitations that make it infeasible for regular use in all patients. Most individuals must seek out and pay out-of-pocket for VO2 max tests unless there is a specific clinical need, and even if such a test is needed, physical limitations may preclude some patients from being able to perform it. In light of these challenges, researchers have begun exploring alternative methods to estimate VO2 max more feasibly, including a promising approach based on proteomics.
Proteomics and Its Role in CRF Estimation
Proteomics, part of the broader "-omics" fields, is the study of the totality of proteins in a biological sample. By quantifying the types and concentrations of proteins, researchers can associate specific protein profiles with outcomes such as CRF. Unlike genomics, which involves studying the stable genetic material of an individual, proteomics provides a dynamic snapshot of gene expression and cellular activity. Because protein concentrations are more changeable and responsive to environmental factors like exercise, proteomics has significant potential to serve as a proxy for measuring CRF over time.
A recent study sought to create a proteomic model that could predict CRF, providing an alternative to direct VO2 max testing. This study, conducted by Perry and colleagues, utilized proteomic data from the Coronary Artery Risk Development in Young Adults (CARDIA) trial to develop a model for estimating CRF. The team used 1569 participants to develop the model and 669 participants to validate it. To test the applicability of the proteomic CRF score, the researchers then applied the model to over 12,000 participants from three additional cohorts (Fenland, BLSA, and HERITAGE) and another 22,000 participants from the UK Biobank. They examined how well the proteomic CRF score correlated with all-cause mortality, cause-specific mortality, and the incidence of chronic diseases.
How Well Does the CRF Proteomics Score Predict VO2 Max?
In their analysis, the researchers used aptamers—short strands of nucleotides that bind to specific proteins—to measure between five to seven thousand proteins in plasma samples. These proteins were then analyzed to identify those most strongly correlated with CRF. A total of 307 proteins were selected for use in the model using penalized regression methods. These methods reduce the influence of proteins with weaker correlations in order to focus on those most closely tied to CRF, such as leptin (which is reflective of adiposity) and fatty acid-binding protein 4 (FABP4, a protein involved in energy metabolism).
The proteomic CRF scores developed from these protein markers showed a high correlation with VO2 max in the CARDIA validation set (Spearman correlation of 0.79). The correlation was moderately strong in subsequent cohorts, with Spearman correlations ranging from 0.67 to 0.71 in studies that measured VO2 max using maximal exercise tests. The correlation was weaker in the Fenland cohort (0.35), where VO2 max was estimated using submaximal exercise testing methods. This suggests that while the proteomics CRF score is a moderately reliable surrogate for VO2 max, it is not as accurate in predicting VO2 max derived from submaximal exercise testing, which can introduce greater variability.
CRF Proteomics Score and Mortality Prediction
While the proteomics CRF score may not match the accuracy of direct VO2 max testing, its primary value lies in its ability to predict long-term health outcomes such as mortality and morbidity. The study used regression models to link the proteomic CRF score with mortality data from the UK Biobank, with a median follow-up of 13.7 years. They found that each standard deviation increase in the proteomic CRF score was associated with an 89% increase in the risk of all-cause mortality (HR=1.89; 95% CI: 1.79-2.00). Additionally, lower CRF scores were linked to higher risks of disease-specific mortality, including deaths from cardiovascular disease, respiratory diseases, and cancer.
Beyond mortality risk, the study also found that lower proteomic CRF scores were associated with a higher incidence of chronic diseases, including cardiovascular disease, metabolic conditions, and neurodegenerative diseases such as Alzheimer’s disease. Interestingly, even in diseases with a known genetic component, higher CRF levels seemed to mitigate some of the risks associated with genetic predisposition. This was demonstrated in the context of polygenic risk scores for conditions like type 2 diabetes and Alzheimer’s disease. The results showed that high CRF could reduce the disease risk conferred by genetics, underscoring the protective role of CRF in preventing the onset and progression of chronic diseases.
Proteomic CRF Score and Longitudinal Exercise Interventions
A crucial aspect of any potential CRF estimation method is its ability to track changes over time. The researchers used data from the HERITAGE study, which involved a 20-week exercise intervention with regular VO2 max testing, to determine how well the proteomic CRF score could reflect changes in fitness. The results showed that each standard deviation change in the proteomic CRF score corresponded to an improvement of about 0.84 ml/kg/min in VO2 max. This suggests that the proteomic CRF score can capture changes in fitness over time, making it a useful tool for longitudinal monitoring, especially in individuals who cannot undergo maximal exercise testing.
However, the study also found that the proteomic CRF score was particularly responsive to short-term changes in fitness. The HERITAGE study involved a relatively intense exercise program, which likely resulted in large changes in protein levels. Longer-term data is needed to assess how well the proteomic CRF score continues to reflect incremental improvements in VO2 max over extended periods of regular exercise.
Feasibility and Clinical Application of the Proteomic CRF Score
The ultimate goal of the proteomic CRF score is to provide a practical and less invasive method for assessing cardiovascular fitness in clinical settings. One of the key advantages of using proteomics for this purpose is that blood-based tests are easier to implement than exercise tests, especially for patients with physical limitations or those who are unable to perform maximal exercise testing. The study found that adding the proteomic CRF score to standard risk factors (such as age, sex, and BMI) improved the prediction of mortality risk, making it a valuable tool for assessing health in a broader range of patients.
While the full proteomic CRF score relies on measurements of hundreds of proteins, the researchers also tested whether a smaller panel of 21 proteins could offer a reasonable substitute. Although using fewer proteins slightly reduced the effectiveness of the score, it still improved mortality prediction over traditional clinical factors and could be more feasible for widespread use in clinical practice.
Limitations and Future Directions
Despite its promising potential, there are several limitations to the current study that must be addressed before the proteomic CRF score can be widely implemented in clinical settings. One major issue is the time gap between the VO2 max tests and the proteomics analysis in the CARDIA study. This delay could introduce variability in fitness estimates, as changes in fitness during this period may not be fully captured by the proteomic data. Additionally, the study’s model was based on a relatively narrow age range, limiting its applicability to younger and older adults. To improve the predictive power of the model, future studies should aim to include more diverse age groups and collect concurrent proteomic and VO2 max data.
Conclusion: The Future of CRF Estimation
Cardiorespiratory fitness is one of the most important predictors of long-term health outcomes, and the ability to measure it accurately and efficiently has significant clinical implications. While traditional VO2 max testing remains the gold standard for assessing CRF, the proteomic CRF score offers a promising alternative that could provide valuable insights into fitness levels and health risks without the need for maximal exercise testing. As research progresses and the proteomic model is refined, it could become a routine tool for monitoring fitness, predicting disease risk, and guiding interventions to improve health outcomes.
Update from Peter Attia, on 2024-09-21Source