Somatostatin receptor 2a is a more sensitive diagnostic marker of meningioma than epithelial membrane antigen.

Abstract

following rare histologic variants of meningioma were included: 7 angiomatous, 3 secretory, 2 osseous metaplasia, 4 psammomatous, 2 microcystic, 2 chordoid, 1 lymphoplasmacyte rich, and 1 rhabdoid meningioma. The remaining 147 meningiomas represented more common subtypes, such as meningothelial, fibrous, and transitional forms. Two 2 mm cores were taken from each block to generate the TMA, along with control tissue from normal adult brain, normal adult meninges, 2 lung adenocarcinomas, and placenta. TMAs containing 13 cellular schwannomas, 23 schwannomas, and 33 malignant peripheral nerve sheath tumors, 38 metastatic carcinomas, 48 metastatic melanomas were also studied, along with 13 whole section meningeal hemangiopericytomas/solitary fibrous tumors. The following antibodies were used for IHC after optimization of pretreatment and primary antibody dilutions: EMA (DAKO, E29, 1:200); SSTR2a (Epitomics, UMB1, monoclonal, 1:2000); PR (DAKO, PgR 636, 1:500); and pSTAT3 (Cell Signaling, D3A7, 1:100). Each tissue core was scored based on percentage of tumor cells staining (0, 1+ for <25 % staining, 2+ for 26–75 % staining, or 3+ for more than 75 % staining) by two independent observers. All meningiomas stained strongly with SSTR2a; therefore, intensity was not scored. Somatostatin receptor 2a stained both duplicate cores from all 176 meningiomas (100 %). EMA stained both duplicate cores in 168 cases (95 %), one duplicate core in 4 cases (2 %), and did not stain 4 cases (2 %). PR stained both duplicate cores in 136 (77 %) cases, one duplicate core in 14 cases (8 %), and did not stain 26 cases (15 %). For the meningiomas with discrepant SSTR2a and EMA staining on small TMA cores (n = 8), whole tissue sections were stained with both SSTR2a and EMA and confirmed the above findings. When immunostaining scores were analyzed based on differences among normal meninges, Meningioma is the most common extraaxial primary central nervous system tumor. While most are diagnosed on routine hematoxylinand eosin-stained sections, the morphologic spectrum is wide and immunohistochemistry (IHC) is occasionally necessary for diagnosis. Currently, the most reliable IHC markers for meningioma are epithelial membrane antigen (EMA) and progesterone receptor (PR), though both have suboptimal sensitivities and specificities [4, 7]. Previous studies suggest that meningiomas express somatostatin receptors, especially somatostatin receptor 2a (SSTR2a) [1, 5, 6, 8], and one prior study found that a polyclonal SSTR2a antibody stained 87 % of WHO grade I and grade II meningiomas, although EMA stained 97 % of the same cohort [2]. A commercially available monoclonal SSTR2a antibody was recently developed [3]. STAT3 is a potential downstream target of SSTR2a in mouse embryonic stem cells [9]. Therefore, we studied a large meningioma tissue microarray (TMA) with SSTR2a, EMA, PR and phosphorylated STAT3 (pSTAT3) immunohistochemistry. Representative blocks from 176 meningiomas resected between 2002 and 2012 at UCSF were retrieved. Small biopsy material and decalcified or frozen tissues were excluded, yielding a study cohort of 128 WHO grade I, 36 WHO grade II, and 12 WHO grade III meningiomas. The