[799] Novel Dual Color Immunohistochemical Analysis for Detecting ERG and SPINK1 Status in Prostate Carcinoma

Ritu Bhalla, Lakshmi P Kunju, Scott A Tomlins, Kelly Christopherson, Connie Cortez, Juan M Mosquera, Gary Pestano, Arul Chinnaiyan, Nallasivam Palanisamy. University of Michigan, Ann Arbor, MI; Ventana Medical System, Tucson, AZ; Weill Cornell Medical College, New York, NY; Michigan Center for Translational Pathology, Ann Arbor, MI; Howard Hughes Medical Institute, Ann Arbor, MI

Background: ETS gene fusions occur in around 50% of prostate cancers (PCA) and result in over-expression of a truncated ERG protein. Overexpression of SPINK1 defines an aggressive molecular subtype of ETS fusion-negative PCA (SPINK1+/ETS-, 10% of all PCA). Previously, we characterized an antibody against ERG as showing diagnostic utility for the detection of ERG rearranged PCA. The aim of our study was to develop a dual color immunohistochemistry based assay for reliable simultaneous assessment of ERG and SPINK1 proteins in PCA.
Design: ERG and SPINK1 protein expression was evaluated utilizing an automated dual color immunohistochemistry protocol (Ventana-Roche, Discovery XT) using monoclonal antibodies against ERG and SPINK1 proteins on 8 tissue microarrays (TMAs) including a spectrum of localized PCAs (including variant histology) and castrate resistant metastatic PCA. A total of 218 PCA cases -150 localized (loc) and 68 metastatic (met) PCA were examined. The percentage of SPINK1 positive tumor cells in each core was estimated and assigned values of 0%, 5% or multiples of 10%. Any positivity was considered as expression of SPINK1 protein. Nuclear ERG staining was considered positive.
Results: Of the 218 cases evaluated 77 (35%) cases (58 loc and 19 met) were positive for ERG. In the remaining 141 ERG negative cases, SPINK1 was positive in 20 (9%) cases (4 loc and 16 met). SPINK1 expression was found to be mutually exclusive with ERG expression, however, we identified one case, which demonstrated ERG as well as SPINK1 positivity in two independent foci. Unlike the homogenous staining of ERG in cancer tissues, heterogeneous SPINK1 staining was observed in the majority of the cases. Further studies are required to understand the molecular heterogeneity of PCA in SPINK1 positive cases.
Conclusions: This study confirms the mutual exclusivity of ERG and SPINK1, suggesting that ERG+/SPINK1- and ERG-/SPINK1+ PCA may represent different molecular subtypes of PCA. We have shown that the dual ERG/SPINK1 staining is easily implemented, reproducible and easy to interpret. Also, as it can be done on a single unstained level, it enables the routine subtyping of PCA. This may have utility in retrospective or prospective stratification for therapy, prognosis or risk stratification of precursor/atypical lesions.
Category: Genitourinary (including renal tumors)

Monday, March 19, 2012 1:00 PM

Poster Session II # 150, Monday Afternoon

 

Close Window