GDF-15 levels increased progressively across cardiometabolic risk groups in young adults and showed high discriminatory ability between obese and control participants but only modest discrimination between low-risk and high-risk obese groups, identifying GDF-15 as a candidate biomarker requiring confirmation in larger prospective cohorts.
Key Findings
Results
GDF-15 levels rose progressively across control, low-risk, and high-risk cardiometabolic groups in young adults.
Control group mean GDF-15: 337.7 ± 82.6 pg/mL; low-risk group: 865.8 ± 76.9 pg/mL; high-risk group: 1130.7 ± 79.2 pg/mL
Differences across all three groups were statistically significant (p < 0.001)
GDF-15 levels remained significantly different across groups after adjustment for age
Study sample consisted of 90 adults aged 18–35 years classified into three groups using predefined criteria
Results
GDF-15 showed high discriminatory ability for distinguishing controls from obese groups but only modest discrimination between low-risk and high-risk obese groups.
AUC for controls versus obese groups ranged from 0.939 to 0.954, reflecting strong phenotypic separation
AUC for low-risk versus high-risk comparison (the clinically relevant comparison) was 0.677 (95% CI 0.535–0.819; p = 0.024)
The authors characterize the low-risk vs. high-risk discrimination as 'only modest'
ROC analysis was used for these discriminatory assessments
Results
Leptin, the leptin-to-adiponectin ratio, and adiponectin were elevated in both risk groups compared to controls, with no significant difference between the two risk groups.
Both low-risk and high-risk groups had significantly higher leptin, leptin-to-adiponectin ratio, and adiponectin than controls (p < 0.001)
No statistically significant difference in these adipokine measures was observed between the low-risk and high-risk groups
Parameters analyzed included anthropometric, glycemic, and lipid measures, hs-CRP, GDF-15, leptin, and adiponectin
ANCOVA was used for group comparisons
Results
Adiponectin was positively associated with GDF-15 in both low-risk and high-risk subgroups in exploratory regression analyses.
In the low-risk group, adiponectin showed a standardized regression coefficient of β = 0.465 (p = 0.011) with GDF-15
In the high-risk group, adiponectin showed a standardized regression coefficient of β = 0.551 (p = 0.038) with GDF-15
These analyses were described as 'exploratory subgroup regression'
Partial correlation and regression methods were used in the overall analysis
Results
Combining GDF-15 with waist-to-hip ratio produced a numerically higher AUC than GDF-15 alone for distinguishing risk groups.
The combined GDF-15 plus waist-to-hip ratio AUC was 0.708, compared to 0.677 for GDF-15 alone
The authors note the difference was 'not formally tested', limiting conclusions about statistical significance
Body Roundness Index (BRI) was among the anthropometric parameters evaluated in the study
This finding is described as a numeric observation rather than a confirmed result
Conclusions
The study design was cross-sectional, and the authors characterize findings as exploratory, requiring confirmation in larger prospective cohorts.
The sample size was 90 adults aged 18–35 years, classified into control, low-risk, and high-risk groups
The study used ANCOVA, partial correlation, regression, and ROC analysis
Authors explicitly state these are 'exploratory, cross-sectional findings'
The authors call for 'confirmation in larger prospective cohorts'
What This Means
This research suggests that a protein called Growth Differentiation Factor-15 (GDF-15), which is produced by cells under stress, may serve as a marker of cardiometabolic risk—the combined risk of heart disease and metabolic conditions like diabetes—in young adults with obesity. In a study of 90 adults aged 18–35, GDF-15 blood levels were much higher in people with obesity compared to normal-weight controls, and were even higher in those classified as being at high cardiometabolic risk compared to those at low risk. The study also found that adiponectin, a hormone involved in regulating metabolism, was linked to GDF-15 levels in both risk groups.
However, while GDF-15 was very good at distinguishing people with obesity from those without it, it was only modestly effective at telling apart those with low versus high cardiometabolic risk—which is the more clinically useful distinction. Combining GDF-15 with a waist-based body shape measurement slightly improved this discrimination, though this was not formally tested statistically. Fat-related hormones like leptin were elevated in both obese groups compared to controls but did not differ between the low- and high-risk obese groups.
This research suggests GDF-15 is a promising candidate biomarker for early detection of cardiometabolic risk in young adults, a population where such risks are often not yet clinically apparent. However, because this was a small, cross-sectional study—meaning it captured a snapshot in time rather than following people over time—the authors caution that these findings are preliminary and need to be confirmed in larger studies that track participants over time before any clinical conclusions can be drawn.
Mohammed O, Mehdi H. (2026). Growth differentiation factor-15 and body roundness index in relation to cardiometabolic risk in young adults with obesity.. Endocrine. https://doi.org/10.1007/s12020-026-04784-4