High interlaboratory and interobserver agreement of somatostatin receptor immunohistochemical determination and correlation with response to somatostatin analogs

Monoclonal antibodies to somatostatin receptors 2A (SSTR2A, UMB-1) and 5 (SSTR5, UMB-4) were reported to be highly reliable for immunohistochemical detection of these receptors in neuroendocrine neoplasms. However, the standardization of either the immunohistochemical procedure and the methods of evaluation has yet to be established. Fifty-two tissues from 38 patients with neuroendocrine neoplasm were retrieved from 2 institutions in Italy and Japan. The tissues were immunostained using 3 staining methodologies: 1 automated and 2 manual protocols from the Italian and Japanese institutions. The slides were independently evaluated by 3 observers (2 experienced pathologists and 1 medical student) using 3 scoring systems (Volante-Score, HER2-Score, and H-Score). The scores obtained from the staining methods were highly correlated with each other (r > 0.85, P < .0001). Especially, the Volante- and HER2-Scores were highly concordant (r ≥ 0.95, P ≤ .0001). Very high interobserver agreement was obtained irrespective of the method used and the experience of the evaluator, with the best concordance obtained by experienced pathologists evaluating automated system-stained slides (SSTR2A, r > 0.97; SSTR5, r > 0.96). HER2- and H-Scores were reliable to represent the characteristics of the patients. SSTR2A expression evaluated by the HER2-Score was significantly associated with clinical efficacy to somatostatin analogs (P = .04). SSTRs determination is an easily standardizable tool in different laboratories and is highly reproducible irrespective of the method of evaluation used. Given the positive association with clinical efficacy to somatostatin analogs, as well as the simple and widespread use, HER2-Score can be proposed as a standard evaluation procedure of SSTR2A and SSTR5 expression in neuroendocrine neoplasms.