
The slide to be stained for cytokeratin 7 (CK7)
comprised:
1:
liver, 2-3:
colon
adeno-carcinoma,
4:
breast
ductal carcinoma, 5:
Merkel cell carcinoma,
6:
stomach
fundal mucosa.
Criteria for assessing a CK7 staining as optimal
included: A strong and distinct
cytoplasmic reaction in the cells expected to stain: tumour cells of
the ductal breast carcinoma, bile duct epithelium in the liver,
fundic gastric epithelium (heterogeneous staining, particularly
chief cells in base of glands, focal staining in surface and
foveolar epithelium), the low differentiated colon adenocarcinoma
(widespread staining). The well differentiated colon adenocarcinoma
should be negative or give a weak focal staining only. Normal liver
cells and the Merkel cell tumour should be negative. Normal
endothelial cells may be focally positive.
71 laboratories submitted stainings. At the
assessment 36 achieved optimal staining (50 %), 26 good (37 %), 7
borderline (10 %) and 2 poor staining (3 %).
The most used
Ab for
CK7 was mAb OV-TL 12/30 obtained from either DakoCytomation (63), Novocastra (1), BioGenex
(1) or Immunotech (1). 3 used mAb clone K72.7 (NeoMarkers),
2 mAb clone LP5K (Novocastra, Ventana), and 1 used mAb clone K72 (Ventana).
Optimal results were achived
with mAbs OV-TL 12/30, K72.7 and K72.
Optimal
results with mAb OV-TL 12/30 was obtained with proteolytic pre-treatment,
HIER and a combination of these two retrieval techniques.
Using proteolytic
pre-treatment, optimal stainings were achieved with Protease 1 (Ventana),
Pronase E (Sigma) and Proteinase K (DakoCytomation).
Using HIER, optimal
stainings were achieved with MWO, pressure cooker, water bath and
autoclave, provided the use of Tris-EDTA/EGTA pH 9 or EDTA pH 8 as
the heating buffer.
HIER with Citrate pH 6 as
the heating buffer only resulted in an optimal result, if the HIER
was followed by a (gentle) proteolytic pre-treatment.
The optimal dilution of mAb
OV-TL 12/30 was in the range of 1:50 – 500 for both proteolytic
pre-treatment and HIER.
mAb clone K72.7 gave an
optimal staining using HIER with Tris-EDTA/EGTA pH 9 (2 of 2), while
proteolytic pre-treatment appeared inferior (1 of 1).
mAb clone K72 gave an
optimal staining using HIER with Citrate pH 6 (1 of 1).
The most frequent causes of
insufficient stainings were:
-No pre-treatment used
-HIER with Citrate pH 6 (mAb
OV-TL 12/30)
-False positive reaction due to endogenous biotin (HIER
combined with a
biotin based detection system without biotin
blocking)
-Inappropriate choice of antibody for CK7 demonstration.
|