Objective: To prospectively determine the reliability and validity of serum fibroblast growth factor 21 (FGF-21) as a biomarker for mitochondrial disease in a cross-sectional cohort of adults with mitochondrial disease from a specialist primary care and tertiary referral clinic.
Methods: We recruited 140 subjects, including 54 adults with mitochondrial disease, 20 patients with nonmitochondrial neuromuscular disease, and 66 control subjects, between November 2011 and October 2012. We compared serum FGF-21 concentrations to classical biomarkers, serum creatine kinase, lactate, pyruvate, and lactate to pyruvate ratio, to determine its validity and reliability as a biomarker of mitochondrial disease. We determined the sensitivity, odds ratio (OR), and overall reliability of FGF-21 as a marker of mitochondrial disease using statistical analyses.
Results: Median serum FGF-21 concentrations were significantly elevated in patients with mitochondrial disease and differed significantly between all experimental groups. FGF-21 showed a markedly higher diagnostic OR (45.7 [95% confidence interval = 12.6-166.5], p < 0.0001) when compared to other biomarkers and was the best predictor of disease according to sensitivity and receiver operating characteristic curve analysis. After multivariate logistic regression analysis controlling for potential confounders, FGF-21 was the only measured parameter capable of predicting mitochondrial disease.
Conclusion: This prospective study establishes serum FGF-21 levels as a sensitive biomarker of mitochondrial disease and demonstrates that they are the best predictor of this disorder when compared to serum levels of classical indicators: creatine kinase, lactate, pyruvate, and the lactate to pyruvate ratio.