We read this article [1] from Baltaci et al. with interest. In Belfast, there is similar interest in Bcl-2 as a molecular target in tumour cells, with most of our experience being the morphological and quantitative assessment of bladder tumour tissue for the Bcl-2 protein. We have some concerns about the methodology used to extract the Bcl-2 protein from the prostate tissue and the controls used in the detection of the antigen in this study. Other than haematological malignancies, where it is overexpressed because of a known chromosomal translocation, Bcl-2 is a labile protein and degrades with time. Ferrieres et al.[2] reported that when Bcl-2 expression was examined in frozen sections of benign breast tissue, half of the staining disappeared if there was a 30 min delay in freezing after removing the tissue from the patient. The Belfast Bladder Cancer Research Group assessed Bcl-2 expression in superficial bladder tumour by immunohistochemistry of archival material and found difficulties with reproducibility. The buffer used in the study was changed from citrate to EDTA to enhance reproducibility. Because of the difficulties involved with immunohistochemistry in paraffin-embedded material our group now advocates the assessment of Bcl-2 in bladder tumours using frozen-section immunohistochemistry. This technique has been reproducible and significantly correlates with flow cytometry data. Because of the possible variation induced by the slow process of paraffin embedding, would the authors not agree that frozen-tissue samples are more reflective of the true level of Bcl-2 expression? A second concern is the use of normal mouse serum as a control with no specific isotype antibody. Although normal mouse serum will contain some immunoglobulin similar to the test antibody, it is not as rigorous as the specific isotype antibody control. More discussion is needed about the best ways of assessing Bcl-2 in urological tumour tissue. Standardization is a major issue in immunocytochemistry [3–5] and there is a need for an optimal method to be agreed by different institutions so that results can be easily compared; this is not the case at present.


  1. Top of page
  2. References
  • 1
    Baltaci S, Orhan D, Ozer G, Tolunay O, Gogus O. Bcl-2 proto-oncogene expression in low- and high-grade prostatic intraepithelial neoplasia. BJU Int 2000; 85: 155 9
  • 2
    Ferrieres G, Cuny M, Simony-Lafontaine J, Jacquemier J, Rouleau C, Guilleux F. Variation of bcl-2 expression in breast ducts and lobules in relation to plasma progesterone levels: overexpression and absence of variation in fibroadenomas. J Pathol 1997; 183: 204 11
  • 3
    National Committee for Clinical Laboratory Standards. Quality Assurance for Immunohistochemistry: Proposed Guideline. MM4-P. Pennsylvania: NCCLS, 1997
  • 4
    Balaton AJ. Defining objectives for the technical quality in immunohistochemistry. Cell Pathol 1999; 4: 69 77
  • 5
    Maxwell P & McCluggage WG. Audit and Internal Quality Control in Immunohistochemistry. J Clin Pathol 2000; submitted: