Background: Identification of biomarkers of pancreatic neuroendocrine tumors (PNETs) is important for stratification of the prognosis. Ki-67 index was the significant prognostic factor for PNETs. Recently, MMP-9, DJ-1, and α-1-B glycoprotein (A1BG) were shown to be the prognostic markers in some malignant tumors except for PNETs. The aim of this study was to compare these tumor markers for predicting disease-free survival in surgically resected PNETs.
Methods: A retrospective review of patients pathologically diagnosed with PNETs at our institution from January 2012 to January 2014 was conducted. Tumor specimens were stained by immunochemistry for MMP-9, DJ-1, A1BG, and Ki-67. Clinicopathologic findings and these 4 tumor markers of patients with PNETs were analyzed. Prognostic factors were determined by univariate and multivariate analyses.
Results: A total of 40 patients were selected. Of 40 tumors, 19 (47.5%) were positive for MMP-9, 21 (52.5%) for DJ-1, 18 (45%) for A1BG, and 24 (60%) for Ki-67 > 2%. Four (10%) tumors expressed none of the markers. MMP-9 and DJ-1 expression were more frequently observed in distant metastasis, higher WHO grade, and lymph node metastasis. Similarly, those with Ki-67 > 2% had larger tumor, higher WHO grade, lymph node invasion and larger surgical procedures. A1BG had no substantial effect on distant metastasis. Univariate analyses revealed that grade 2/3 histology, distant metastasis, lymph nodes metastasis, high expression of MMP-9, high expression of DJ-1, and Ki-67 > 2% were predictive of inferior disease-free survival for PNETs (all p < 0.05). In multivariable analysis, the significant poor factors associated with disease-free survival were grade 2/3 histology, distant metastasis, lymph node metastasis, and Ki-67 > 2% (all p < 0.05).
Conclusions: Of the 4 markers studied, only Ki-67 was the independent prognostic factor. Expression of MMP-9 and DJ-1 correlated with invasion, metastasis, disease progression, and poor prognosis in PNETs, while A1BG had no prognostic value in PNETs. It is anticipated that these findings can serve as useful clinical survival predictors, especially in the setting of resected disease.