Table 2.

Overview of principal studies on emerging novel exercise-related CVD biomarkers.

Biomarker category	Study population and intervention	Key finding	Clinical implications	Limitations	Ref.
H-FABP	47 patients (26 with >70% stenosis, 21 normal coronary anatomy), EST	From baseline to three hours after EST, H-FABP dropped (CAD: -2.79 ng/mL; control: -0.84 ng/mL). reduction linked to proteinuria brought on by exercise rather than heart injury.	Proteinuria is a confounding factor; a reduction in H-FABP during EST should not be mistaken for the lack of CAD.	Small sample (n=47); single center; no comparison with troponin; unexpected finding requires replication.	(Arı et al., 2011)
H-FABP, sST2	98 participants, 8 months of regular physical activity	After prolonged exercise, sST2 increased by 793 pg/mL (p=0.045), indicating adaptive vascular stress, whereas H-FABP decreased by 0.57 ng/mL (p<0.01), indicating better perfusion.	Frequent exercise improves indicators of subclinical ischemia; an increase in sST2 may indicate adaptive vascular remodeling.	No control group; single time point post-training; lack of imaging correlation; moderate sample size.	(Sponder et al., 2019)
cMyBP-C	158 participants (75 men, 83 women), exercise stress echocardiography + 1–1.5 year follow-up	Before stress, HR 8.1 for significant occurrences (p=0.041) is the MyBP-C threshold (96% sensitivity). AUC 0.91, HR 13.8 (p=0.000472) for important events (death, MI, stroke, and PE). EF was normal for the majority of crucial occurrences (6/7).	Even in individuals with normal EF, cMyBP-C is a very sensitive screening biomarker for severe CVD.	Small sample (n=158); short follow-up (1–1.5y); no external validation; no CMR comparison.	(Tong et al., 2017)
cMyBP-C	205 symptomatic HFrEF patients (SMARTEX post hoc analysis), 12 weeks HIIT or MCT vs. control	There was no discernible difference between the groups’ cMyBP-C alterations (Δ IG: -0.5 vs. CG: -0.7). There was a correlation between log cMyBP-C and log hs-cTnI (R = 0.52, 95% CI 0.37-0.66, P<0.001).	In HFrEF, cMyBP-C may indicate subclinical myocardial damage during ET.	Post hoc analysis; IG merged; no healthy control group; correlation moderate.	(Riveland et al., 2025)
miRNAs	20 CAD patients (10 male, 10 female), maximal cycle ergospirometry	After exercise, 33 miRNAs were altered, with 16 exhibiting sex differences. The responses to nine miRNAs (let-7e-5p, miR-1, miR-19b-1-5p, etc.) varied statistically by sex.	Personalized risk classification in CAD may be made possible by sex-specific miRNA responses to exercise.	Very small sample (n=20); single exercise session; no validation cohort; no long-term outcomes.	(Mayr et al., 2021)
miRNAs	20 sedentary individuals (HERITAGE Family Study), 20 weeks endurance training	Nine of the 14 c-miRNAs were downregulated, and five were upregulated. Target genes that are connected to more than 345 biological processes.	Widespread alterations in c-miRNA patterns associated with metabolic and cardiovascular pathways are produced by regular exercise.	Small sample (n=20); no control group; no correlation with clinical outcomes; descriptive only.	(Barber et al., 2019)
c-miRNAs	Healthy active runners, 10km, half-marathon, and marathon races	Exercise duration led to a dose-dependent increase in all cardiac biomarkers. After ten kilometers, five c-miRNAs were dysregulated, and after a marathon, nineteen. “Pseudo-disease” signatures were brought on by exercise.	Clinicians must identify exercise-induced pseudo-disease patterns, even if c-miRNAs may enhance the diagnostic utility of hs-cTnT/NT-proBNP.	Healthy subjects only; no pathological controls; single time point post-exercise; no follow-up.	(de Gonzalo-Calvo et al., 2018)
miR-15a-5p	ApoE-/- mice (high-fat diet) + moderate treadmill training, 12 weeks	In atherosclerotic aortas, exercise dramatically decreased miR-15a-5p. Through 3’-UTR targeting, miR-15a-5p inhibited Sema3A. Plaques were destabilized by VSMC-specific overexpression. Human carotid plaques are positively associated with plaque vulnerability.	A possible therapeutic target for plaque stabilization by AE is miR-15a-5p.	Animal model; little human validation (plaque samples only); no human exercise intervention.	(Hu et al., 2026)
Exosomal miRNAs	Review of exercise-induced EXO studies (HIIT, endurance training)	HIIT targets CTGF and lessens cardiac fibrosis by increasing circulating exosomal miR-133a (2-3×). Exosomal miRNAs control vascular flexibility, cardiac healing, and endothelial function.	Exosomal miRNAs are both promising precision diagnostic tools and mediators of EIC.	Review article; lack of clinical validation; inconsistent methods; absence of standardized tests.	(Yu et al., 2026)
Exosomal EVs + AI	Review of EV biomarker studies (AMI, HF)	AUCs of 0.80–0.92 are achieved for EV biomarker classification using AI ensemble techniques (Random Forest, gradient boosting). The majority of research lacks external validation and has small cohorts (n<50).	Although promising, AI-integrated EV analysis is still in the proof-of-concept stage; multicenter validation linked with MISEV is necessary for clinical adoption.	Exploratory studies only; small sample sizes, no comparisons with recognized biomarkers, and no standardized procedures.	(Dinc and Ardic, 2026)
LncRNA MALAT1	Rat model of CHF (left anterior descending ligation), 8 weeks AE post-surgery	Exercise enhanced autophagy and decreased ROS, inflammatory cytokines, MMPs, and apoptosis. Exercise benefits were eliminated by MALAT1 overexpression. MALAT1 suppresses PI3K/Akt and downregulates miR-150-5p.	A possible therapeutic target, MALAT1, is a crucial modulator of EIC in CHF.	Animal model only; no human validation; only one lncRNA was examined; mechanistic rather than translational.	(Hu et al., 2021)
LncRNAs and CircRNAs	Review of ncRNAs in cardiac hypertrophy and CHF	Autophagy, apoptosis, and hypertrophy are regulated by Bvht, Mhrt, Chast, Chaer, Carmen, Fendrr, Apf, MALAT1, Chrf, Hrcr, Cdr1as, and circ-Amotl1. Important research gap: function in physiological hypertrophy or EIC.	Although ncRNAs offer therapeutic potential for CHF, little is known about how they contribute to exercise-mediated protection.	Review article; mostly preclinical; no evidence on human exercise; descriptive rather than mechanistic for EIC.	(Ma and Liao, 2019)