| Pancreatic cancer (PC) which leads the tenth incidence of all cancers, is a highly malignant tumor. Because of lack of diagnostic tools for early diagnosis, pancreatic cancer has a poor prognosis with a five year survival rate less than 10%. Thus the development of specific biomarkers is in urgent need to improve PC patients’survival. This study aims to look for valuable plasma diagnostic tools for pancreatic cancer and evaluate their diagnostic sensitivity and specificity for future application.This study recruited 407 plasma samples including PC, chronic pancreatitis(CP), colorectal cancer(CRC), gastric cancer(GC), liver cancer(LC) and healthy controls. Seven candidate biomarkers (miR-20A, miR-21, miR-25, miR-155, miR-196A, miR-210 and Macrophage Inhibitory Cytokine-1(MIC-1)), which were reported to be potential diagnostic biomarkers for PC, and CA19-9 were detected to select specific biomarkers. The study was designed to have a training group, a validation group and an independent double-blinded test.The training group was utilized to select specific biomarkers to establish diagnostic indexes. Our results showed that in the training group all candidate markers were elevated in PC patients’compared with those of healthy controls (P<0.001). However miR-155, miR-196A, miR-210 also elevated in CP patients with no significant difference with PC patients. MiR-20A, miR-21, miR-25, MIC-1 and CA19-9 could distinguish PC patients from CRC, GC and LC patients. With multivariable logistic regression we established two specific diagnostic indexes for pancreatic cancer.Index1=1.971×ln(MIC-1)+1.795×ln(miR-21)+1.020×ln(CA19-9)-42.305;Index2=2.138×ln(MIC-1)+1.772×ln(miR-25)+1.125×ln(CA19-9)-45.006.The cutoff value of both indexes for diagnosis of pancreatic cancer was 0.In the training group, when distinguish PC from non-PC, Index1 yield a value of area under curve (AUC) of 0.968(95%CI,0.947-0.989), Index2 yield a value of 0.967(95%CI, 0.945-0.989) and CA19-9 of 0.895(95%CI,0.838-0.952). Sensitivity for Indexl, Index2 and CA19-9 were 0.895,0.895 and 0.816 respectively; specificity were 0.909,0.915 and 0.933 respectively and accuracy were 0.904,0.908,0.892 respectively. These indicated that combination of biomarkers improved diagnostic sensitivity and accuracy compared with single marker.The validation group was applied to identify the diagnostic value of the selected indexes. In the validation group, the AUC is 0.933(95%CI,0.874-0.992) for indexl, 0.921(95%CI,0.856-0.986) for index2 and 0.793(95%CI,0.687-0.899) for CA19-9. The sensitivity was 0.892 for indexl,0.838 for index2 and 0.649 for CA19-9 and the specificities were 0.913,0.925 and 0.913 respectively. The results in the training group and validation group showed that both indexes performed better sensitivity and specificity than CA19-9.Double-blinded test was carried out to assess the applicability of those selected indexes. In the double-blinded test our two indexes and CA19-9 performed the similar diagnostic sensitivity and specificity.Correlation analysis found that most biomarkers were not related with clinical characteristics such as gender, age, hyperhensions, diabetes and so on, although there’s a weak relationship(p<0.5) of Indexl and Index2 to age and diabetes.The survival status of PC patients was followed-up for the analysis of association of biomarkers’ expression with prognosis. As for the overall survival rate of 79 PC patients, Kaplan-Meier analysis revealed that all six miRNAs, MIC-1, CA19-9 and two indexes have no association with PC patient survival.The innovations:combination of plasma miRNAs and protein marker to diagnose PC; conducting tissue specificity analysis of candidate biomarkers; a reasonable study design with a training group, a validation group and an independent double-blinded test.In conclusion, our findings provided two specific diagnostic indexes with satisfied sensitivity and specificity for pancreatic cancer. Further investigations are required to confirm their potential in clinical applications. |
PDF Full Text Request