Creatine kinase-(MB) and hepcidin as candidate biomarkers for early diagnosis of pulmonary tuberculosis: a proof-of-concept study in Lambarene, Gabon

Background The present study aimed to evaluate the diagnostic utility of creatine kinase-MB (CK-MB), hepcidin (HEPC), phospholipase A2 group IIA (PLa2G2A), and myosin-binding protein C (MYBPC1) for tuberculosis (TB). These four biomarkers are differentially regulated between quiescent Mycobacterium tuberculosis (Mtb) infected individuals (non-progressors to TB disease) and Mtb-infected TB disease progressors 6 months before the onset of symptoms. Methods We enrolled samples from patients experiencing moderate-to-severe pulmonary infections diseases including 23 TB cases confirmed by smear microscopy and culture, and 34 TB-negative cases. For each participant, the serum levels of the four biomarkers were measured using ELISA. Results The levels of CK-MB and HEPC were significantly reduced in patients with active TB disease. CK-MB median level was 2045 pg/ml (1455-4000 pg/ml) in active TB cases and 3245 pg/ml (1645-4000 pg/ml) in non-TB pulmonary diseases. Using the receiver operating characteristic curve (ROC) analysis, HEPC and CK-MB had the Area Under the Curve (AUC) of 79% (95% CI 67-91%) and 81% (95% CI 69-93%), respectively. Both markers correlated with TB diagnosis as a single marker. PLa2G2A and MYBPC1 with AUCs of 48% (95% CI 36-65%) and 62% (95% CI 48-76%) did not performed well as single biomarkers. The three markers'model (CK-MB-HEPC-PLa2G2A) had the highest diagnostic accuracy at 82% (95% CI 56-82%) after cross-validation. Conclusion CK-MB and HEPC levels were statistically different between confirmed TB cases and non-TB cases. This study yields promising results for the rapid diagnosis of TB disease using a single marker or three biomarkers model.

Automated summary

This study evaluated the diagnostic utility of creatine kinase-MB (CK-MB) and hepcidin (HEPC) for tuberculosis (TB) and found that their levels were statistically different between confirmed TB cases and non-TB cases. Combining them with two other biomarkers can improve the diagnostic accuracy.