Kaoru Hashizume

Quantitative immunohistochemical evaluation of HER2/neu expression with HercepTestTM in breast carcinoma by image analysis

HercepTestTM (DAKO A/S, Glostrup, Denmark) is an immunohistochemical assay that detects HER2/neu gene products, and evaluates the overexpression status of the HER2/neu protein in determining eligibility for the Trastuzumab (HerceptinR, Genentech, San Francisco, CA, USA) therapy. However, practically, interobserver variability of the HER2/neu interpretation of the immunostained results has caused marked disagreement with regard to the intensity of tumor staining. In this study, we quantitated HER2/neu expression by image analysis, and applied this analyzing system to help to minimize interobserver variability of the interpretation of the HercepTestTM. All the immunostained results were scored semiquantitatively on a range of 0 to 3+ in accordance with the criteria described as per the manufacturer’s instructions, and quantitatively evaluated using an image analyzing system with image processing software. Among the 92 cases, 15 were scored as 3+, six were 2+, and 32 were 1+ under intraobservers agreement. When the cases were quantitated, a high correlation was shown between the signal area extracted by image analysis and the corresponding score of staining intensity with the HercepTestTM. By converting the quantitatively extracted data into a scoring system based upon the criteria, the outcome demonstrated a strong concordance with the scoring data obtained from immunostaining. The results indicated that a quantitative scoring system performed by simple image analysis may provide to improve interobserver agreement of the interpretation of the HercepTest TM in clinical practice.

Cytometrical image analysis for immunohistochemical hormone receptor status in breast carcinomas

A cytometrical image analyzing method for nuclear protein was established using WinROOF, a commercially available, inexpensive software, to determine the status of both estrogen and progesterone receptors. Immunohistochemical evaluation of estrogen receptors (ER) and progesterone receptors (PR) was performed with the anti‐ER (clone 1D5) and the anti‐PR (clone PgR636), respectively, combined with dextran polymer reagent EnVision+, all of which are approved in vitro diagnostics in Japan. The immunostained results were captured as digital images in Windows, and then analyzed in WinROOF with macroinstructions for analyzing each captured area either immunolabeled with chromogen or counterstained with hematoxylin. This image analysis method graded the immunostained nuclei of carcinoma cells based on staining intensities, and calculated the labeling index (LI) for both ER and PR. Furthermore, the LI correlated highly with the results from a histology score (HSCORE) when 20 breast carcinomas were quantified. Regarding ER, when 20% in the LI was considered as the cut‐off point for positive, the positivity of ER in computer‐assisted analysis was 75% (15 of 20 cases), and was completely concordant with that of HSCORE‐based analysis. These results indicate that the cytometrical image analysis‐based quantification could be appropriately applied to the objective determination of the immunohistochemical status of both ER and PR.

Interlaboratory comparison in HercepTest assessment of HER2 protein status in invasive breast carcinoma fixed with various formalin-based fixatives.

Although formalin-based fixatives are used in pathologic laboratories, there is no strictly standardized fixation protocol in Japan. To examine interlaboratory variation caused by different conditions of fixation in the assessment of human epidermal growth factor receptor (HER) 2 status on pathologic tissues, 274 archival invasive breast carcinomas from 5 different laboratories were evaluated using the HercepTest. In 1 laboratory in which 10% neutral buffered formalin was used, as recommended by the manufacturer, the overexpression rate was 22.4% and fell within the statistical expected range (20%–30%) for HER2 overexpression in breast carcinomas. The overexpression rates in the other 4 laboratories, in which either 20% nonbuffered formalin or 15% neutral buffered formalin was used, were near the expected range for HER2 overexpression. To clarify the influence of prolonged formalin fixation on the HercepTest, we compared 1-day with 7-day fixations using 36 cases fixed with 20% nonbuffered formalin. Of the 36 cases, 7 showed 3+ staining with 1-day fixation and sustained the same scoring results with 7-day fixation, although the staining intensities in these cases were reduced with the prolonged fixation. These results indicated that the immunohistochemical assessment of HER2 status with the HercepTest was comparatively resistant to prolonged fixation conditions and provided stable staining results in positive cases, particularly 3+ patients.

Automated immunohistochemical staining of formalin-fixed and paraffin-embedded tissues using a catalyzed signal amplification method

An immunohistochemical assay using catalyzed signal amplification (CSA), which is based on the peroxidase catalyzed deposition of biotinylated tyramide, is a highly sensitive method to visualize weak immunohistochemical signals originating from rare antigens or masked antigens in formalin-fixed, paraffin-embedded (FFPE) tissues. However, CSA methods are hampered by poor reproducibility and the complexity of their staining procedures. In this study, we aimed to apply the CSA procedure to a capillary gap-based, automated immunostainer, TechMate Horizon, to perform immunohistochemical signal amplification effectively and reproducibly. A variety of cellular antigens previously considered to be undetectable in FFPE human specimens were selected and examined with the automated immunostainer. Compared with the manual CSA staining method that takes more than 2 hours, the automated CSA method took less than 2 hours to complete. The staining results from the automated CSA method presented higher reproducibility, as well as lower background owing to well-regulated, punctual staining and washing at every step of the procedure. Conclusively, the automation of the CSA method enabled us to perform the time-consuming and complicated CSA amplification technique with minimal effort in an accurate, consistent, and reproducible manner.

N-acetyl-l-cysteine suppresses constitutive expression of CD11a/LFA-1α protein in myeloid lineage

We investigated the possible involvement of redox regulation in constitutive expression of CD11a/LFA-1alpha, a leukocyte integrin alpha subunit, in myeloid cells using antioxidants. In unstimulated HL-60 cells, CD11a/LFA-1alpha was highly expressed, however, no expression of CD11b and CD11c proteins was detected. N-acetyl-L-cysteine (NAC) markedly down-regulated CD11a/LFA-1alpha expression in a dose-dependent manner. The down-regulated CD11a/LFA-1alpha expression was gradually recovered when NAC was deprived 24h after treatment. Pyrrolidine dithiocarbamate (PDTC) also suppressed the level of expression CD11a/LFA-1alpha protein, although the effect of PDTC was less potent than NAC. Both NAC and PDTC suppressed NF-kappaB binding activity to consensus DNA probe, and this result was correlated with a suppressive effect to CD11a/LFA-1alpha expression. Furthermore, NAC also down-regulated CD11a/LFA-1alpha expression in both U937 cells and peripheral blood monocytes. These results indicated that the constitutive CD11a/LFA-1alpha expression in the myeloid lineage is implicated in oxidative stress occurring spontaneously, suggesting that alteration of the intracellular redox state using antioxidants may be effective in the modulation of cell adhesion associated with extravasation in leukocytes, at least, in myeloid cells.

Up-Regulation of CD13/Aminopeptidase N Induced by Phorbol Ester Is Involved in Redox Regulation and Tumor Necrosis Factor α Production in HL-60 Cells

Here, we demonstrate the possible involvement of oxidative stress in altered CD13/aminopeptidase N (APN) expression during myeloid cell differentiation induced by TPA. In flow cytometrical analysis, CD13/APN protein was constitutively expressed in HL-60 cells. When the cells were treated with TPA, CD13/APN expression was up-regulated with increased intracellular peroxides and a morphological change into macrophage-like cells. This increase in CD13/APN expression was suppressed by treatment with N-acetylcysteine. Transfection of Mn-superoxide dismutase (MnSOD) gene to the cells also suppressed the up-regulated CD13/APN expression. Furthermore, a neutralizing antibody to TNFα partially blocked this up-regulation. These results indicate that the change in intracellular redox state could be involved in the up-regulation of CD13/APN expression during TPA-induced differentiation of HL-60 cells, suggesting that TNFα may serve as, at least, one of the signals stimulated by TPA.