PIVKA-II and hematological parameters in hepatocellular carcinoma and non-malignant chronic liver disease: A case-control biomarker comparison in a hepatitis-endemic population

Background Early detection of hepatocellular carcinoma (HCC) remains challenging in hepatitis-endemic settings such as Pakistan, where surveillance commonly relies on ultrasonography and alpha-fetoprotein (AFP), both of which have limited sensitivity for early-stage disease. This study evaluated the diagnostic performance of serum Protein Induced by Vitamin K Absence or Antagonist-II (PIVKA-II), complete blood count (CBC) parameters, and CBC-derived hematological indices for discriminating HCC from non-malignant chronic liver disease (CLD). Methods We conducted a case-control biomarker study involving 160 adults with CLD at Liaquat University Hospital, Sindh, Pakistan, between October 2025 and March 2026, including 74 patients with established HCC and 86 with non-malignant CLD. Serum PIVKA-II was measured using electrochemiluminescence immunoassay. CBC-derived neutrophil-to-lymphocyte ratio, platelet-to-lymphocyte ratio, monocyte-to-lymphocyte ratio, systemic immune-inflammation index, and red cell distribution width-to-platelet ratio were calculated. Between-group differences were assessed using non-parametric methods, and receiver operating characteristic (ROC) analysis evaluated discriminatory performance. Benjamini–Hochberg correction was applied to prespecified families of multiple comparisons. Results PIVKA-II demonstrated the strongest discriminatory performance, with an area under the ROC curve (AUC) of 0.753 (95% CI, 0.670–0.827). At a cutoff of >82.1 mAU/mL, sensitivity was 86.4% and specificity was 56.0%. Platelet count (AUC, 0.642) and the RDW-to-platelet ratio (AUC, 0.638) also demonstrated significant between-group differences that remained significant after false discovery rate correction, although their discriminatory performance was modest. PIVKA-II discrimination was substantially higher among patients with vascular invasion (AUC, 0.853) and advanced BCLC stage C/D disease (AUC, 0.853) than among those with early/intermediate BCLC A/B disease (AUC, 0.571). A combined model incorporating PIVKA-II, lymphocyte percentage, and platelet count yielded an AUC of 0.784, but did not significantly improve discrimination compared with PIVKA-II alone (p = 0.166). Bootstrap validation showed minimal optimism in the overall AUC; however, the optimal PIVKA-II cutoff demonstrated substantial variability, with a 95% percentile range of 58.7–715.8 mAU/mL. This was corroborated by an independent 10-fold cross-validation analysis, which yielded a comparable mean AUC of 0.751 for PIVKA-II (optimism = +0.002). Conclusion PIVKA-II demonstrated fair diagnostic discrimination between HCC and non-malignant CLD, with relatively high sensitivity but limited specificity. Its substantially lower performance in early/intermediate-stage disease and the marked instability of the optimal cutoff indicate that PIVKA-II should be considered a complementary biomarker rather than a standalone diagnostic test. Larger multicenter studies with standardized control populations and independent external validation are needed to establish robust population-specific thresholds and determine the incremental clinical value of combining PIVKA-II with hematological parameters.