Comparing The Clinicopathological Features Of Chinese Patients With Pancreatic Neuroendocrine Tumors With Americans’ And Identifying Diagnostic And Prognostic Biomarkers

Background and Aims Pancreatic neuroendocrine tumors (PNETs) are rare but the incidence is rising year by year both in Europe/America and in Asia. Chromogranin A (CgA) is an effective serum biomarker for diagnosing PNETs. In Europe/America, it has been widely used in clinical diagnosis, but few related researches have been reported in China and using CgA for clinical diagnosis has not been officially approved by China Food Drug Administration (CFDA). Accidentally, we found that the serum levels of CgA in insulinomas and non-insulinoma PNETs patients were apparently different, and this point has not been getting as much attention as it deserves in clinical practice. The aim of the research is to evaluate the values of serum CgA in diagnosing different subtypes of PNETs patients especially insulinomas patients.Methods Serum levels of CgA of 57 insulinomas patients,32 non-insulinoma PNETs patients and 86 healthy participants were measured by ELISA method. The diagnostic values were assessed by drawing receiver operating characteristic (ROC) curves. Expression of CgA protein was detected in 26 PNETs tissues by immunohistochemical staining (IHC). Tumor grading and staging are according to European Neuroendocrine Tumor Society (ENETS) system.Results The CgA levels in 89 PNETs patients were significantly higher than that in healthy controls (P=7.2×10-9). Serum levels of CgA in 57 patients with insulinomas (median 64.8 ng/ml, range 25.0-164.2) were slightly higher than the levels in healthy controls (median 53.4 ng/ml, range 39.1-94.1) but much lower than the levels in 32 patients with non-insulinoma PNETs (median 193 ng/ml, range 26.9-9020.7), P=0.001. ROC curves showed CgA values at 60.4 ng/ml distinguished patients with insulinomas from healthy controls but its sensitivity and specificity were 66.7% and 73.3%, respectively. In contrast, CgA values at 73.9 ng/ml distinguished patients with non-insulinoma PNETs from healthy controls, and the sensitivity and specificity were 65.6% and 91.9%, respectively.Conclusions CgA is a reliable serum diagnostic biomarker for non-insulinoma PNETs patients, but not for insulinomas.Background and AimsPancreatic neuroendocrine tumors (PNETs) are rare but the incidence is rising year by year both in Europe/America and in Asia. So far, wether the grade and stage systems in European/American guidelines apply to the Chinese PNETs patients and wether the clinicopathological characteristics between European/American and Chinese PNETs patients are different have not been reported. The aims of the research are to compare the clinicopathological characteristics and prognosis between Chinese and American PNETs patients and to identify reliable prognostic clinicopathological markers.MethodsTumor grading and staging are according to European Neuroendocrine Tumor Society (ENETS) system. The clinicopathological characteristics between 517 Chinese PNETs patients and 450 American patients were compared by t test, non-parametric test,χ2 test or Fisher’s exact test. The prognostic markers were identified by Kaplan-Meier survival analysis and Cox regression analysis.Results1. The median age of Chinese patients was younger than that of American patients, showing that non-functional PNETs,46 vs.57 years, P=1.2×10-15; functional PNETs,46 vs.53 years, P=0.003; insulinomas,46 vs.54 years, P= 0.033; non-insulinoma PNETs 46 vs.56 years, P=5.4×10-15.2. The median size of Chinese non-functional PNETs was bigger than that of American non-functional PNETs,4 cm vs.3 cm, P=2.1×10-9.3. The primary tumor location between Chinese PNETs and American PNETs was different.65.2% American non-functional PNETs occured in pancreatic body/tail while 55.9% Chinese non-functional PNETs occured in pancreatic head/neck (P= 2.8×10-6).73.9% American insulinomas occured in pancreatic body/tail while 52.1% Chinese insulinomas occured in pancreatic body/tail (P=0.006).4. We found that higher grade, higher stage, higher Ki-67 index, bigger tumor and metastasis were significantly related with worse overall survival and worse disease-free survival of PNETs patients by Kaplan-Meier survival analysis. Using Cox regression analysis, we found that stage and grade were independent prognostic factors of PNETs patients in both overall survival (P value 3.12×10-8 and 8.25×10-6, respectively) and disease-free survival (P value 5.37×10-16 and 3.54×10-5, respectively).5. For American PNETs patients, female had longer overall survival (P=0.042), but not for Chinese patients.ConclusionsThere are some apparent differences in the clinicopathological characteristics of PNETs patients between Chinese and Americans. Grade and stage are reliable prognostic markers for PNETs patients.Background and AimsSome clinicopathological markers were useful for predicting the outcome of PNETs patients, but they still have some limitations. Our previous proteomic study on insulinoma found α-internexin and UCH-L1 expressed in insulinomas, and our previous study found α-internexin could predict overall survival of PNETs patients, but not for disease-free survival. The first aim of the research is to detect the expression of both two proteins in PNETs and to identify whether they could predict the prognosis of PNETs patients better in two independent collectives. In 2011, a research in USA found that the mutation of DAXX could be related with the better prognosis of PNETs patients and the gene mutation could result in decreased expression of DAXX. However, another research in Sweden found that the decreased expression of DAXX caused by mutation could be related with the unfavourable prognosis of PNETs patients in 2014. Allowing for the two contradictory conclusions, the second aim of the research is to detect the expression of DAXX in larger samples and to identify the relationship between DAXX expression and prognosis. We also identify the relationship between the expression of both DAXX and α-internexin and prognosis.MethodsThe expression of UCH-L1 and α-internexin were detected in 286 PNETs by IHC. The expression of both two proteins was correlated with clinicopathological features and patients’ prognosis in 2 independent collectives of PNETs.The expression of DAXX was detected in 295 PNETs by IHC. The expression of DAXX was correlated with clinicopathological features and patients’prognosis. The expression of both DAXX and α-internexin was correlated with clinicopathological features and patients’ prognosis. The expression of three proteins above was correlated with clinicopathological features and patients’ prognosis. The mutations of DAXX exon were detected by Sanger sequencing.ResultsThe expression of both UCH-L1 and α-internexin in PNETs was significantly associated with favorable prognosis (log rank, P=8.77×10-5).The expression of DAXX was significantly related with a better disease-free survival of PNETs patients (log rank, P=4.779×10-5). The expression of both DAXX and a-internexin in PNETs was not only significantly associated with a favorable disease-free survival (log rank, P=0.030) but also with a better overall survival (log rank, P=0.009). The expression of UCH-L1, a-internexin and DAXX was significantly associated with a better disease-free survival (log rank, P= 0.002).ConclusionsDetecting two proteins could be better than detecting one protein on predicting the prognosis of PNETs patients. UCH-L1 and a-internexin are independent prognostic biomarkers for PNETs patients. The expression of both DAXX and a-internexin in PNETs is associated with favourable prognosis of PNETs patients. Detecting UCH-L1, α-internexin and DAXX at the same time can predict the prognosis of PNETs patients better.