Institutional quality assurance for breast cancer HER2 immunohistochemical testing: identification of outlier results and impact of simultaneous fluorescence in situ hybridization cotesting.

The College of American Pathologists Accreditation Checklist requires comparison of laboratory predictive results with published benchmarks but does not require analysis of individual pathologists. With the availability of targeted human epidermal growth factor receptor 2 (HER2) protein therapy, uniform reporting of HER2 protein status by immunohistochemistry (IHC) is essential. Our aim was to compare HER2 IHC results among pathologists in routine clinical practice within a single institution and assess the impact of simultaneous IHC and fluorescence in situ hybridization (FISH) ordering. We reviewed reports from 928 consecutive breast needle biopsies from 2008 to 2012 at a tertiary academic medical center in which HER2 IHC and HER2 FISH were ordered. There was a significant association between breast pathologist and IHC result (negative, 49.8%-83.2%; positive, 8.7%-14.1%; equivocal, 5.2%-41.5%; P < .0001) but not breast pathologist and FISH result (P = .69). For 1 pathologist, IHC signed out with FISH had an equivocal rate nearly 2-fold lower than IHC results that were reported first (10.5% versus 20.9%) (P = .04). Institutions should be aware that although overall HER2 IHC reporting may be consistent with guidelines, there can be significant variation among practitioners. In addition to aggregate data, we recommend comparing the rates from individual pathologists to standards. Furthermore, routine simultaneous ordering of both IHC and FISH could impact interpretation of test results and may inappropriately encourage less confidence in IHC results among pathologists.

Do amount of variant differentiation and mitotic rate in bladder cancer change with neoadjuvant chemotherapy?

Neoadjuvant chemotherapy (NAC) is currently recommended to all candidate patients with muscularis propria-invasive bladder cancer. However, NAC is effective in only a subset of patients, and predictors of response are lacking. Our study aimed to characterize tumoral changes with NAC usage and to identify features at bladder biopsy/transurethral resection (Bx/TUR) that may predict response. A retrospective search was performed to identify patients with bladder cancer that were pT2 at Bx/TUR upon whom a radical cystectomy (RC) was performed from 2007 to 2010. A blinded slide review of the Bx/TUR and RC was conducted. Presence, type, percent of tumor variant morphology, and tumoral mitotic rate were assessed. Ninety RC patients with slides available were identified (46 NAC, 44 non-NAC). In NAC-treated patients, there was a significantly higher percentage of nonurothelial variant differentiation in the RC compared with Bx/TUR, whereas there was no difference in the non-NAC subgroup. Percent variant differentiation at Bx/TUR was not a predictor of response. There was a significant decrease in mitotic rate between Bx/TUR and RC in NAC patients, whereas there was no difference in the non-NAC subgroup, although mitotic rate was not a predictor of response. In conclusion, percent variant differentiation and mitotic rate changed significantly from Bx/TUR to RC with NAC usage, although neither predicted response. Pathologists should be aware that variant differentiation is common in bladder cancer, with increased presence after NAC, in order to improve recognition and documentation of these findings.

Trends in prostatic adenocarcinoma tumor volume by visual estimation in prostatectomy specimens.

We retrospectively reviewed 1792 consecutive radical prostatectomies (RP) from 2003 to 2006 at a single institution to establish tumor volume reference values, to determine current trends in visually estimated prostate adenocarcinoma tumor volume, and to characterize cases with no residual cancer on RP. Tumor volumes were recorded and subsequently stratified as very low, 0-1%; low, 1.1-10%; intermediate, 10.1-20%; high, 20.1-50%; and very high, >50%, with incidences of 11.7%, 52.1%, 21.5%, 13.2%, and 1.5%, respectively. The incidence of very low volume tumors increased within the time period (p=0.04). Seminal vesicle involvement was detected in 5.0% of cases and lymph node metastasis occurred in 1.4%. Volume categories statistically correlated with seminal vesicle invasion (p=0) and lymph nodes metastases (p=0). Eleven cases of no residual cancer (0.6%) were identified with a non-statically significant increase during the study (p=0.07). The rising incidence of very low volume tumors should be considered by clinicians when discussing treatment options with patients. A discrete tumor volume should be provided for RP specimens as it may be an important prognostic factor.