Sensitivity and Specificity of Immunohistochemistry in Pituitary Adenomas

Abstract

Abstract: There is a paucity of knowledge regarding the clinical usefulness of immunohistochemistry (IHC) and appropriate standards for interpreting positive IHC in patients with pituitary adenomas. This study evaluated the correlation between IHC and the independent clinical diagnosis of patients with pituitary adenomas. A retrospective review was performed of patients who underwent pituitary adenoma surgery at the University of Arkansas between 1999 and 2004. Retrospective clinical diagnoses were made based on available clinical and hormonal data. IHC of resected specimens was independently reviewed by 1 author (REM) and graded on a 5-point scale. Data was derived from 72 patients; 18 with prolactinoma, 13 with acromegaly, 7 with pituitary Cushing syndrome, and 34 with a nonfunctioning adenoma. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of IHC was calculated at various levels of reaction positivity and correlated with the clinical diagnosis. The best combination of sensitivity, specificity, PPV, and NPV in patients with secretory tumors was observed with a ≥3+ reaction (strong reaction in 10%–50% of cells, or weaker reaction in 50%–75% of cells). In addition, 14 of 34 nonfunctioning adenomas were positive at ≥3+ for one or more hormones. IHC of ≥3+ yields high sensitivity, specificity, PPV, and NPV, correlating with the clinical diagnosis in secretory tumors. Although discrepancies between IHC results and the clinical diagnosis are well known, a standardized method of semiquantitative IHC may be useful in comparing results between different studies and in retrospect when clinical and hormonal data are inconclusive.