Differential Expression of SPINK1 in Urothelial Neoplasia: Clinical and Pathological Implications
Kathy R Kawaguchi, Kyung Park, Arul Chinnaiyan, Massimo Loda, Brian D Robinson, Rosina Lis, Douglas S Scherr, Jonathan E Rosenberg, Mark A Rubin, Juan Miguel Mosquera. Weill Cornell Medical College, New York, NY; University of Michigan, Ann Arbor, MI; Dana Farber Cancer Institute; Harvard Medical School, Boston, MA
Background: High serum level/tissue expression of SPINK1 (Serine protease inhibitor Kazal type I; Tumor associated trypsin inhibitor [TATI]) have been associated with poor outcome in epithelial malignancies. SPINK1 defines an aggressive ETS-negative prostate cancer (PCa) and transcripts can be detected in urine. Recent data suggest that loss of SPINK1 expression in urothelial carcinoma of bladder (UCB) may be associated with advanced stage. SPINK1 is also a potential therapeutic target, given structural similarities with EGF. We assessed the expression of SPINK1 in a wide selection of urothelial neoplasia and benign urothelium.
Design: Tissue microarrays (TMAs) of 390 tissue samples from bladder (n=373), kidney/ureter (n=9), prostatic urethra (n=6) and metastatic sites (n=2) were studied. These included biopsies and resections of high-grade urothelial carcinoma (HGUC) (n=134; 97 invasive) and lymph node metastases (n=2); low-grade urothelial carcinoma (LGUC) (n=77); papillary urothelial neoplasm of low malignant potential (PUNLMP) (n=1); inverted papilloma (IP) (n=4); urothelial carcinoma in situ (CIS) (n=83); neuroendocrine carcinoma of bladder (NECB) (n=4); bladder squamous cell carcinoma (n=3); urothelial dysplasia (UD) (n=13); urothelial atypia (UA) (n=18); papillary cystitis (n=1); radiation cystitis (n=1); and normal urothelium (n=49).
Semi-quantitative SPINK1 IHC evaluation was performed: >20% staining extent and 2-3 intensity (scale 0 to 3) was considered positive.
FISH was performed in 116 cases using SPINK1 locus/control and 5'/3' split probes.
Results: SPINK1 overexpression was present in 50% of HGUC with loss of expression in 55% of invasive tumors. SPINK1 overexpression was present in 53% of LGUC and in a PUNLMP, 82% of CIS, 31% of UD, 22% of UA, 50% of NECB, and 33% of SCC. The inverted papillomas were negative. SPINK1 expression was seen in umbrella cells of normal urothelium (76%) and papillary and radiation cystitis. SPINK1 overexpression was also noted in urothelial samples from kidney/ureter and prostatic urethra. SPINK1 amplification or gross rearrangements were not seen by FISH.
Conclusions: SPINK1 is expressed in a wide range of urothelial neoplasia and in umbrella cells of benign urothelium. This should be considered if urine-based tests for detection of SPINK1 in both urothelial and prostatic carcinomas are implemented. SPINK1 appears to be a very specific marker of CIS. Loss of SPINK1 expression, especially in invasive tumors, is noteworthy.
Category: Genitourinary (including renal tumors)
Wednesday, March 21, 2012 1:00 PM
Poster Session VI # 134, Wednesday Afternoon