[Expression and significance of INSM1 and SOX11 in pancreatic neuroendocrine tumor and solid pseudopapillary neoplasm].

OBJECTIVE: To investigate the expression and significance of insulinoma associated protein 1 (INSM1) and SRY-related high-mobility group box 11 (SOX11) in pancreatic neuroendocrine tumor (PNET) and solid pseudopapillary neoplasm (SPN). METHODS: To detect the expression of INSM1, SOX11, Syn, CgA, CD56, ß-catenin, and CD99 in 56 cases of PNET, 42 cases of SPN, 16 cases of ductal adenocarcinoma (DACC) and 8 cases of acinar cell carcinoma (ACC) by immunohistochemistry. The application value of combination of INSM1 and SOX11 was compared with conventional markers (Syn, CgA, CD56, ß-catenin, and CD99) in diagnosis and differential diagnosis of PNET and SPN. RESULTS: (1) In the 56 cases of PNET, the positive signals of INSM1 were located in the tumor and islet nucleus, the positive expression rate in the tumor tissues was 91.07% (51/56), whereas the signal was absent in 42 cases of SPN, 16 cases of DACC and 8 cases of ACC, and there were significant statistical difference between PNET with SPN, DACC, and ACC respectively (P < 0.001). (2) The positive signals of SOX11 were located in the tumor nucleus, with the positive expression rate was 92.86% (39/42) in SPN, however, the positive expression rate of SOX11 was 8.93% (5/56) in PNET, which included 3 cases of G1 and 2 cases of G3 types of PNET, the SOX11 positive signal was absent in 16 cases of DACC, 8 cases of ACC and peritumoral nomal pancreatic tissue, and the differences were statistically significant of positive rate between SPN with PNET, DACC and ACC, respectively (P < 0.001). (3) The sensitivity of INSM1(+)/SOX11(-) immunophenotype for PNET was 85.71%, vs. CD56 (57.14%), the difference was statistically significant (P=0.001); vs. Syn (80.36%) and CgA (71.43%), the difference was no statistically significant (P>0.05). The specificity of INSM1(+)/SOX11(-) for PNET was 100.00%, vs. Syn (42.86%) and CD56 (47.62%), the difference was statistically significant (P < 0.001); vs. CgA (92.86%), the difference was no statistically significant (P>0.05). The sensitivity of INSM1(-)/SOX11(+) immunophenotype for SPN was 92.86%, vs. ß-catenin (90.48%) and CD99 (85.71%), the difference was no statistically significant (P>0.05). The specificity of INSM1(-)/SOX11(+) for SPN was 96.43%, vs. CD99 (48.21%), the difference was statistically significant (P < 0.001); vs. ß-catenin (100.00%), the difference was no statistically significant (P>0.05). (4) The positive expression of INSM1 and SOX11 in PNET and SOX11 were not correlated with clinicopathological parameters (age, gender, tumor size, location, grade, and metastasis) (P>0.05). CONCLUSION: The positive expression patterns of INSM1 and SOX11 in PNET and SPN respectively are conductive to distinguish the both tumors. The combination of both take precedence over some corresponding conventional immunohistochemical markers in terms of sensitivity and specificity.

[Expression of CD200 and INSM1 in gastrointestinal and pancreatic neuroendocrine neoplasms and its diagnostic values].

Objective: To investigate the expression and diagnostic values of CD200 and insulinoma associated protein 1 (INSM1) in gastrointestinal and pancreatic neuroendocrine neoplasm (GIP-NEN). Methods: The expression of CD200, INSM1, Syn and CgA was detected in 69 cases of GIP-NEN, 66 cases of gastrointestinal and pancreatic non-neuroendocrine neoplasm (GIP-nonNEN) and 16 cases of metastatic neuroendocrine neoplasm by immunohistochemistry, to compare the values of CD200, INSM1, Syn, CgA and their combinations in diagnosing GIP-NEN. Receiver operating characteristics (ROC) curve was used. Results: The immunoreactivity of CD200 was present in the cytoplasma and/or membrane of the neoplasms cells, the positive expression rates in GIP-NEN and GIP-nonNEN were significantly different (P<0.01). The sensitivity and specificity of CD200 for diagnosing GIP-NEN were 95.7% and 78.8%, respectively. There was significant difference of the positive rates of CD200 between neuroendocrine tumor and neuroendocrine carcinoma (P=0.05). The immunoreactivity of INSM1 was present in the nuclei of neoplasms cells. The positive expression rates in GIP-NEN and GIP-nonNEN were significantly different (P<0.01). The sensitivity and specificity of INSM1 for diagnosis of GIP-NEN were 85.5% and 95.5%, respectively. There were also significantly different positive rates of INSM1 between neuroendocrine tumor and neuroendocrine carcinoma, as well as between G1 and G3 neuroendocrine tumors (P<0.05). There was no difference in the area under ROC curve (AUC) of single stain of CD200, INSM1, Syn or CgA (0.857, 0.907, 0.890 and 0.833, respectively, P>0.05). The sensitivity of combined CD200+INSM1 stains for diagnosing GIP-NEN was significantly higher than that of Syn+CgA (85.5% vs. 63.8%, P<0.05). The AUC of two combinations were 0.962 and 0.925, respectively, which were not statistically different (P>0.05). Conclusions: CD200 and INSM1 are two novel markers of neuroendocrine neoplasm, which aid to diagnosis for GIP-NEN and exclude its mimickers. They are associated with tumor grades. Combining both as an immunohistochemical panel shows high sensitivity and specificity. Thus, the combined panel can be utilized as useful supplement for Syn and CgA.