Serum platelet-derived growth factor-BB as a noninvasive biomarker of significant liver fibrosis in patients with chronic hepatitis B: a cross-sectional study

Background Chronic hepatitis B (CHB) is a major cause of progressive liver fibrosis (LF), cirrhosis, and hepatocellular carcinoma. Although noninvasive fibrosis assessment is essential, current tools have practical and diagnostic limitations. Platelet-derived growth factor-BB (PDGF-BB)—a profibrogenic mediator of hepatic stellate cell activation—may be a biologically relevant fibrosis biomarker. This study examined the ability of serum PDGF-BB to detect significant fibrosis in CHB patients which is defined as liver stiffness > 7 kPa by vibration-controlled transient elastography (VCTE; FibroScan). Methods This cross-sectional exploratory diagnostic accuracy study included 90 treatment-naive adults with CHB attending Tanta University Hospitals and Kafr El-Sheikh University Hospitals. All participants underwent clinical evaluation; routine laboratory testing, including HBsAg, HBeAg, and serum HBV DNA; abdominal ultrasonography; serum PDGF-BB measurement by ELISA; and FibroScan examination. VCTE-defined significant fibrosis (F2–F4) was set as liver stiffness > 7 kPa. A non-randomized convenience sampling approach was used to enroll participants, and the optimal PDGF-BB cutoff was established based on the ROC curve derived from this cohort. Results Among 90 patients, 69 (76.7%) had F0-F1 fibrosis and 21 (23.3%) had F2-F4 fibrosis. HBeAg was positive in 21.1% of patients and median HBV DNA was 4.47 (3.68–5.31) log10 IU/mL Median serum PDGF-BB was significantly increased in patients with F2-F4 than in those with F0-F1 fibrosis: 2500.8 (2028.25-2721.8) vs. 777.13 (479.21-1743.15) pg/mL, respectively ( p < 0.001). Serum PDGF-BB discriminated VCTE-defined significant fibrosis at an exploratory, internally derived cutoff of ≥ 1885.9 pg/mL, with an AUC of 0.905 (95% CI 0.816–0.994), sensitivity of 85.7%, specificity of 84.1%, positive predictive value of 62.1%, negative predictive value of 95.1% (at the 23.3% prevalence of this cohort), and overall accuracy of 84.4% (Youden index 0.698; p < 0.001). PDGF-BB had a numerically higher AUC than APRI (0.717) and GPR (0.715), but these differences did not reach statistical significance after Bonferroni correction for multiple comparisons. It did not differ significantly from FIB-4 (0.780) and did not outperform platelet count (0.980) or INR (0.931). Conclusions Serum PDGF-BB demonstrated good discrimination for VCTE-defined significant fibrosis in treatment-naive patients with CHB. Its negative predictive value of 95%, though prevalence-dependent, suggests potential utility as a triage or adjunctive biomarker. However, these findings are preliminary and require external validation before clinical implementation or substitution for elastography.