Serum Dickkopf-1 in Combined with CA 19-9 as a Biomarker of Intrahepatic Cholangiocarcinoma

SIMPLE SUMMARY: Cholangiocarcinoma (CCC) is a rare cancer, but its incidence and mortality have been increased in the past few decades worldwide, representing a global health problem. CCC is usually asymptomatic in early stages and, therefore, often diagnosed when the disease is already in advanced stages, which highly compromises therapeutic options, resulting in a dismal prognosis. The current diagnosis of CCC by non-invasive approaches such as serum biomarker, carbohydrate antigen 19-9 (CA 19-9), is not accurate enough due to the limitations in its low sensitivity, especially at the early stages of the disease. Therefore, new biomarkers with higher sensitivity and specificity are needed. As the clinical significance of dickkopf-related protein-1 (DKK-1) has been reported in various tumors including intrahepatic CCC (ICC), we aimed to identify the diagnostic and prognostic performance of the DKK-1 and its additive effect combined with CA 19-9 in patients with CCC. ABSTRACT: Dickkopf-related protein 1 (DKK-1) has a diagnostic and prognostic value in various malignant tumors. We investigated the diagnostic and prognostic performance of DKK-1 in combination with carbohydrate antigen 19-9 (CA 19-9) in cholangiocarcinoma (CCC) patients. Serum DKK-1 levels were measured using enzyme-linked immunosorbent assay. The receiver operating characteristic (ROC) curve, area under ROC (AUROC) analyses, Kaplan–Meier method, and Cox proportional hazard model were used to evaluate the diagnostic and prognostic performance of DKK-1 in combination with CA 19-9. We checked DKK-1 levels in 356 CCC patients and found that DKK-1 was significantly elevated only in 79 intrahepatic CCC (ICC) patients compared to controls (340.5 vs. 249.8 pg/mL, p = 0.002). The optimal cutoff level of DKK-1 used to identify ICC patients was 258.0 pg/mL (AUROC = 0.637, sensitivity = 59.5%, specificity = 56.9%, positive predictive value (PPV) = 40.5%, negative predictive value (NPV) = 74.0%, positive likelihood ratio (LR) = 1.38, and negative LR = 0.71). Using this cutoff, 47 (59.5%) patients were correctly diagnosed with ICC. DKK-1 in combination with CA 19-9 showed a better diagnostic performance (AUROC = 0.793, sensitivity = 74.7%, specificity = 56.3%, PPV = 45.7, NPV = 81.8, positive LR = 1.71, and negative LR = 0.45) than CA 19-9 alone. The low DKK-1 and CA 19-9 expression group had a significantly longer overall survival (OS) than the high expression group (p = 0.006). The higher level of DKK-1 and CA 19-9 was independently associated with shorter OS (hazard ratio = 3.077, 95% confidence interval 1.389–6.819, p = 0.006). The diagnostic and prognostic performance of DKK-1 in combination with CA 19-9 might be better than those of CA 19-9 alone in ICC patients.