Abstract C11: Impact of pathologic specimen quality on classifying molecular subtypes of breast cancer: A pilot study from three hospitals in Ghana, Africa

Background: Breast cancer is a heterogeneous disease with clinically, molecularly and pathologically defined subtypes that have different etiologies, clinical presentations and outcomes. African women reportedly have especially high rates of breast cancers that are estrogen receptor (ER) or progesterone receptor (PR) negative, or triple negative [ER, PR and human epidermal growth factor receptor-2 (HER-2) negative], with the proportion reported for triple negative breast cancers ranging from 28-82%, compared to 11-20% among Caucasian populations. However, it is unclear to what extent pathology and specimen handling affect receptor measurements in Africa. Here we sought to determine pathology specimen quality of breast tumor tissues from three hospitals in Ghana where we are conducting a molecular epidemiologic study of breast cancer. Methods: We acquired 15 formalin-fixed paraffin-embedded (FFPE) archived breast cancer tissue blocks with specimens obtained at biopsy (N=5) or surgery (N=10) from three hospitals in Ghana. Blocks were re-embedded and four 0.5um whole sections were cut for hematoxylin and eosin stains, and immunohistochemistry (IHC) using previously validated protocols for ER (clone 1D5), PR (clone PgR1294) and HER2 (polyclonal), all obtained from Dako (Carpinteria CA). Specimens were considered ER or PR positive if 10% or more cells stained positive, while HER2 was considered positive if intensity of staining was 3+. Results: Specimens were noted to be embedded in high-melting temperature paraffin. To perform ER, PR, and HER2 IHC staining, deparaffinization procedures were optimized for hard paraffin. Zonal and gradient staining patterns were noted in surgical specimens, features that are likely due to delayed/prolonged fixation. Zonal/gradient staining issues were not identified in the needle biopsies, which were smaller compared to specimens collected at surgery and generally placed in buffered formalin immediately after collection. IHC assays of the tumor specimens revealed that 67% were ER-positive, 20% PR-positive, and 66% HER2 positive. The proportion negative for all three markers was 13%. Conclusion: In contrast to previous reports for African breast cancer patients, within our sample of FFPE breast cancers from Ghana we found a higher prevalence of ER positive tumors and a lower prevalence of tumors negative for all three markers. These results suggest that the high rates of hormone receptor negativity previously reported in African populations may be inaccurate due to various sources of error, particularly fixation time and IHC staining protocols. Future studies focusing on proper documentation of sample collection procedures and optimization of IHC methods are required to determine the true prevalence of molecular subtypes of breast cancer among African women. Such efforts are essential for breast cancer surveillance, identifying etiologic factors for prevention efforts and developing more effective treatment programs. Citation Format: Jonine D. Figueroa, Lawrence Edusei, Ernest Adjei, Nicholas Titiloye, Kris Ylaya, Beatrice Addai, Joe Nat Clegg-Lamptey, Baffour Awuah, Kofi Nyarko, Daniel Ansong, Seth Wiafe, Louise A. Brinton, Stephen Hewitt. Impact of pathologic specimen quality on classifying molecular subtypes of breast cancer: A pilot study from three hospitals in Ghana, Africa. [abstract]. In: Proceedings of the Sixth AACR Conference: The Science of Cancer Health Disparities; Dec 6–9, 2013; Atlanta, GA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2014;23(11 Suppl):Abstract nr C11. doi:10.1158/1538-7755.DISP13-C11