PATHOLOGY

Partial atrophy on prostate needle biopsy cores: a morphologic and immunohistochemical study
Wang W, Sun X, Epstein JI
Department of Pathology, The John Hopkins Medical Institutions, Baltimore, MD 21231, USA
Am J Surg Pathol. 2008; 32: 851-7

• Partial atrophy is the most common benign mimicker of prostate cancer on needle biopsy. Of 3916 prostate needle core biopsy cases received in our consultation service over a period of 3 months (March 1, 2007 to May 31, 2007), 170 cases (4.3%) with partial atrophy were diagnosed as atypical glands by outside pathologists and prospectively identified. We supplemented our material with 108 cases of partial atrophy sent to our consultation service in 2006 from a single institution, which frequently uses a triple cocktail stain [p63, high molecular weight cytokeratin (HMWCK), alpha-methyl acyl-Coa racemase (AMACR)]. The morphologic features of the 278 cases and immunohistochemistry of 236 cases (198 with prostate cocktail and 38 with only basal cell makers) were analyzed. Forty-eight of 278 (17.3%) partial atrophy cases were mixed with postatrophic hyperplasia. Enlarged nuclei were visible in 43/278 (15.5%) cases, with prominent nucleoli seen in 58/278 (20.9%) cases (30 cases associated with nuclear enlargement). Of 198 cases with a prostatic cocktail stain, 48 (24.2%) had a cancer pattern for both basal cells and AMACR (p63-, HMWCK-, and AMACR+), 14 (7.1%) had a cancer pattern for basal cells (p63-, HMWCK-, and AMACR-), 89 (44.9%) had a cancer pattern for AMACR (p63+, HMWCK+, and AMACR+), and 47 (23.7%) had a totally benign pattern (p63+, HMWCK+, and AMACR-). Of the 198 cases using the cocktail stain, 136 (68.7%) had positive basal cell staining. The percentage of basal cells labeled with the combination of p63/HMWCK was: < 5% in 42 (21.2%) cases, 5% to 75% in 58 (29.3%) cases, and > 75% in 36 (18.2%) cases. An additional 38 cases immunostained only for p63 and/or HMWCK was negative in 2 (5.2%) cases, < 5% (13.1%) in 5 cases, 5% to 75% in 19 (50%) cases, and > 75% in 12 (31.6%) cases. In conclusion, partial atrophy is a benign mimicker of adenocarcinoma both as a result of its routine morphologic features and its immunohistochemical profile. Recognition of the classic morphology of partial atrophy on routine hematoxylin and eosin-stained sections is critical to avoid misdiagnosing partial atrophy as adenocarcinoma.
  
  
• Editorial Comment
  The most common benign lesion that causes difficulty in the differential diagnosis with adenocarcinoma of the prostate is partial atrophy. This lesion was reported in the periodic literature in 1998 (1). Architecturally, partial atrophy consists of crowded glands often with a disorganized growth pattern. In contrast to complete atrophy, which can typically be diagnosed at scanning magnification owing to the presence of well-formed glands with a very basophilic appearance, partial atrophy has pale cytoplasm lateral to the nuclei giving rise to pale staining glands that more closely mimic cancer. Characteristically the basal cells are discontinuous and in some acini may be absent. An additional factor that contributes to the difficulty in distinguishing cancer from partial atrophy is the positivity for AMACR (a-methylacyl coenzyme A racemase) in some acini. In a recent study in our institution, we used the cocktail AMACR+34ßE12 for analyzing the immunohistochemistry expression of a total of 727 acini on needle prostatic biopsies corresponding to 324 adenocarcinoma acini, 213 normal acini, and 190 partial atrophy acini. Adenocarcinoma acini showed weak, or strong expression of AMACR in 73/324 (22.5%), and 251/324 (77.5%) acini, respectively; normal acini showed negative, weak, or strong expression in 167/213 (78.4%), 33/213 (15.5%), and 13/213 (6.1%) acini, respectively; and foci of partial atrophy showed negative, and weak expression in 143/190 (75.3%), and 47/190 (24.7%) acini, respectively. No acini in partial atrophy showed strong expression. The distribution of basal cells in partial atrophy was continuous, discontinuous, and absent in 42/190 (22.1%), 104/190 (54.7%), and 44/190 (23.2%) acini, respectively. The absence of basal cells in 44/190 (23.2%) of partial atrophy foci, makes the use of AMACR attractive for the differential diagnosis. No strong positivity was seen in partial atrophy acini, however, the weak positivity seen in approximately 25% of the acini may be a pitfall for the correct interpretation. Furthermore, normal acini may show strong expression of AMACR in approximately 5% of the acini.

Reference
1. Oppenheimer JR, Wills ML, Epstein JI: Partial atrophy in prostate needle cores: another diagnostic pitfall for the surgical pathologist. Am J Surg Pathol. 1998; 22: 440-5.

Dr. Athanase Billis
Full-Professor of Pathology
State University of Campinas, Unicamp
Campinas, São Paulo, Brazil
E-mail: athanase@fcm.unicamp.br