Objectives: No immunohistochemical marker has yet been established to reliably differentiate adrenocortical tumors from other adrenal masses (e.g. metastases). Thus, a panel of several markers like melan A and inhibin is currently used for this purpose, but suffers from limited diagnostic accuracy. We hypothesized that expression of steroidogenic factor 1 (SF-1), a nuclear transcription factor involved in adrenal development and steroidogenesis, might hold significant diagnostic potential for the differential diagnosis of adrenal masses. Moreover, SF-1 overexpression has been associated with increased cell proliferation and tumorigenesis in an adrenocortical mouse model. Therefore, the prognostic value of SF-1 expression in human ACC was also investigated.
Methods: SF-1 protein expression was assessed by immunohistochemistry with a commercially available specific monoclonal antibody (Perseus Proteomics, Japan) using tissue microarrays and regular tissue slides of formalin-fixed paraffin-embedded tissue samples from 156 ACCs, 15 adrenocortical adenomas, four normal adrenal glands, 50 malignant non-adrenal tissues (including colon, breast, kidney and lung cancer) and normal ovary. We also correlated SF-1 protein expression with clinical outcome in patients with ACC using KaplanMaier and Cox-regression analysis.
Results: SF-1 expression was detectable in 150/156 (96%) of ACC samples including 43 (28%) with strong SF1 staining and in 15/15 adrenocortical adenomas. In contrast, SF-1 expression was absent in all of the non-adrenocortical tissues except in granulosa cells of the ovary. In addition, strong SF-1 expression significantly correlated with poor clinical outcome (tumor stage adjusted hazard ratio for death 2.0 (95%CI 1.33.3; P<0.01).
Conclusion: Our study indicates that SF-1 is a promising immunohistological marker to determine the adrenocortical origin of an adrenal mass with high specificity. In addition, SF-1 expression is of prognostic value in patients with ACC.
25 - 29 Apr 2009
European Society of Endocrinology