Humanin Blood Levels Sharply Boost Heart Risk Prediction in Dialysis Patients
A mitochondrial micropeptide called Humanin dramatically improves cardiovascular risk forecasting in hemodialysis patients when added to existing models.
Summary
Researchers followed 83 chronic hemodialysis patients for 24 months and found that adding blood Humanin levels to established cardiovascular risk scoring models significantly improved their predictive accuracy. Cardiovascular events occurred in 40% of patients. Humanin showed a U-shaped relationship with risk — both very low (<462 pg/mL) and very high (>778 pg/mL) levels were associated with increased danger. When Humanin was incorporated into four validated risk scores and one internal model, discrimination improved by 8–11% (ΔAUC), explained variance rose 9–15%, and reclassification metrics improved substantially. This mitochondrial peptide may serve as a valuable additive biomarker for identifying highest-risk dialysis patients.
Detailed Summary
End-stage kidney disease (ESKD) patients on chronic hemodialysis (HD) face cardiovascular mortality rates nearly 50% higher than the general population, yet existing risk prediction tools — developed largely in general populations — perform poorly in this setting. Uremia introduces unique cardiovascular risk pathways, including mitochondrial dysfunction and oxidative stress, that standard models fail to capture. This multicenter prospective study investigated whether Humanin, a small mitochondrial-derived peptide with antioxidant properties, could meaningfully enhance cardiovascular risk prediction in HD patients.
Eighty-three chronic HD patients (mean age 67 years; 74.7% male) were enrolled across three centers in Italy and Greece and followed for 24 months. The primary endpoint — fatal or non-fatal cardiovascular events requiring hospitalization — occurred in 33 patients (39.7%). Baseline serum Humanin was measured by ELISA before a mid-week dialysis session. The study evaluated one internally derived prognostic model alongside four externally validated scores: AROii-2, jDOPPS, You et al., and Zhang et al.
A key finding was that Humanin's relationship with cardiovascular events was curvilinear rather than linear. Both low levels (<462 pg/mL) and high levels (>778 pg/mL) were associated with elevated risk, suggesting that the peptide reflects a complex physiological balance rather than a simple dose-response. Baseline models performed modestly: AUC ranged from 0.610 (jDOPPS) to 0.671 (internal model). After integrating Humanin, all models improved substantially — AUC gains ranged from 0.083 to 0.109, explained variance (R²) increased by 9–15 percentage points, Net Reclassification Improvement ranged from 8% to 27.9%, and Integrated Discrimination Index improved by 4.4–7.3%. All models remained well calibrated after Humanin addition.
These findings align with prior observations linking Humanin to coronary artery disease risk and with the authors' earlier work showing Humanin predicts overall mortality in HD patients. The U-shaped association is biologically plausible: very low Humanin may reflect severe mitochondrial failure and oxidative burden, while extremely elevated levels may represent a compensatory but ultimately insufficient response to overwhelming cellular stress.
Clinically, these results suggest that a single pre-dialysis Humanin measurement could be layered onto existing risk scoring frameworks to better identify patients who need more aggressive cardiovascular monitoring or intervention. The improvements in reclassification are particularly meaningful — they indicate that Humanin helps correctly move patients across risk categories that would otherwise be misclassified. However, the study is exploratory in nature, and larger multicenter validation trials are needed before clinical implementation.
Key Findings
- Cardiovascular events occurred in 39.7% of 83 HD patients over 24 months of follow-up.
- Humanin shows a U-shaped risk curve — both <462 pg/mL and >778 pg/mL linked to elevated CV risk.
- Adding Humanin boosted AUC of all five risk models by 0.083–0.109 across the board.
- Net Reclassification Improvement reached up to 27.9%, correctly repositioning high-risk patients.
- Explained variance (R²) increased by 9–15% in every model after Humanin integration.
Methodology
Prospective multicenter observational cohort study across three Italian and Greek dialysis centers enrolling 83 stable chronic HD patients followed for 24 months. Serum Humanin was measured by blinded duplicate ELISA before a mid-week HD session, and its additive value was assessed across five cardiovascular risk models using AUC, R², NRI, and IDI metrics.
Study Limitations
The study enrolled only 83 patients at three centers, limiting statistical power and generalizability to broader HD populations. The observational design cannot establish causality, and Humanin was measured at a single time point without tracking dynamic changes during follow-up. External validation in larger, ethnically diverse cohorts is required before clinical adoption.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.
Enter your email to subscribe: