The slide to be stained for CK20 comprised:

1. Colon adenocarcinoma, 2. Merkel cell carcinoma, 3. Colon adenocarcinoma, 4. Appendix 5. Gastric corpus, and 6. Urothelial carcinoma.

All tissues were fixed in 10% neutral buffered formalin.

Criteria for assessing a CK20 staining as optimal included:

• A strong, distinct cytoplasmic staining reaction of all the surface epithelial cells of the appendix and at least a weak to moderate staining reaction in most crypt cells.
• An at least moderate, distinct cytoplasmic staining reaction of the majority of the foveolar epithelial cells of the stomach.
• A moderate to strong, distinct cytoplasmic staining reaction in virtually all the neoplastic cells of the colon adenocarcinoma no. 3 and focally in the colon adenocarcinoma no. 1.
• A moderate to strong, distinct dot-like intracytoplasmic staining reaction in virtually all the neoplastic cells of the Merkel cell carcinoma.
• An at least weak to moderate, distinct cytoplasmic staining reaction in the majority of the neoplastic cells of the urothelial carcinoma.

196 laboratories participated in this assessment. One lab submitted an inappropriate antibody (Ab), namely CK-LMW. Out of the remaining 195 labs 85 % achieved a sufficient mark. In table 1 the Abs used and marks are summarized.

Table 1. Abs and assessment marks for CK20, run 35

1) Proportion of sufficient stains (optimal or good), 2) Proportion of sufficient stains with optimal protocol settings only, see below.

Following central protocol parameters were used to obtain an optimal staining:

Concentrated Abs
mAb clone Ks20.8: The protocols giving an optimal result were based on either heat induced epitope retrieval (HIER), enzymatic pre-treatment or a combination of HIER and enzymatic pre-treatment. In total 56 optimal stains were seen with the use of one of these three pre-treatment procedures.

51 out of the 56 laboratories obtaining an optimal staining result used HIER in either Target Retrieval Solution pH 9 (3-in-1) (TRS pH 9; Dako) (8/20)*, TRS pH 9 (Dako) (3/9), Cell Conditioning 1 (CC1; BenchMark, Ventana) (17/29), Bond Epitope Retrieval Solution 2 (BERS 2; Bond, Leica) (13/17), Borg Decloaker pH 9.5 (Biocare) (1/2), Tris-EDTA/EGTA pH 9 (8/16) or EDTA/EGTA pH 8 (1/1) as the retrieval buffer.
The mAb was typically diluted in the range of 1:25-1:400 depending on the total sensitivity of the protocol employed. Using these protocol settings 87 out of 96 (91 %) laboratories produced a sufficient staining (optimal or good).
* (number of optimal results/number of laboratories using this reagent)

3 laboratories obtaining an optimal staining result used enzymatic pre-treatment as Protease 1 (Benchmark, Ventana) (2/11) or Bond Enzyme 1 (1/2).
The mAb was typically diluted in the range of 1:50-1:150 depending on the total sensitivity of the protocol employed. Using these protocol settings 7 out of 8 (88 %) laboratories produced a sufficient staining.

2 laboratories obtaining an optimal staining result used a combined pre-treatment using Protease 3 (Benchmark, Ventana) and HIER in CC1 (Benchmark, Ventana) (2/2). The mAb was diluted 1:50-80. Using these protocol settings 2 out of 2 laboratories produced an optimal staining.

pAb E16444: The protocols giving an optimal result were all based on HIER using either CC1 (BenchMark, Ventana) (2/4), Tris-EDTA/EGTA pH 9 (1/1) or Citrate pH 6 (1/1) as the retrieval buffer.
The pAb was diluted in the range of 1:50-1:400 depending on the total sensitivity of the protocol employed. Using these protocol settings 7 out of 7 (100 %) laboratories produced a sufficient staining.

Ready-To-Use Abs
mAb clone Ks20.8 (prod. no. IS/IR777, Dako): The protocols giving an optimal result were all based on HIER in PT-Link (heating time for 10-20 min at 97°C) using TRS pH 9 (3-in-1) (Dako) or TRS pH 9 (Dako) as HIER buffer, an incubation time of 20 min in the primary Ab and EnVision Flex or EnVision Flex+ (K8000/K8002) as the detection system. Using these protocol settings 20 out of 20 (100 %) laboratories produced a sufficient staining.

rmAb clone SP33 (prod. no. 790-4431, Ventana): The protocols giving an optimal result were typically based on HIER using mild or standard CC1, an incubation time of 16-32 min in the primary Ab and UltraView (760-500) +/- amplification or OptiView (760-700) as the detection system. Using these protocol settings 27 out of 27 (100 %) laboratories produced a sufficient staining.

The most frequent causes of insufficient stains were:

- Too low concentration of the primary Ab
- Less successful primary Ab
- Use of biotin based detection systems (giving a false positive staining due to endogenous biotin).

In this assessment and in concordance with the previous NordiQC runs for CK20, the prevalent feature of an insufficient staining reaction was a general too weak staining or a false negative staining of the structures expected to be demonstrated. Virtually all laboratories could demonstrate CK20 in the luminal epithelial cells of the appendix and in the neoplastic cells of the colon adenocarcinomas and the Merkel cell carcinoma, whereas the demonstration of CK20 in the foveolar epithelial cells of the stomach and in the neoplastic cells in the urothelial carcinoma was more challenging and required a correctly calibrated protocol.
A too weak or false negative staining reaction was seen in 83 % of the insufficient results and was mainly caused by a too low concentration of the primary Ab or using a less successful clone as the mAb clone PW31, Leica, which gave an insufficient result in all 5 out 5 protocols based on this clone.
In 17 % both a too weak and a false positive staining reaction was observed. The false positive staining reaction was observed when a biotin based detection system was used in combination with efficient HIER and was mainly seen as a granular intracytoplasmic staining reaction in the epithelial cells of the gastric crypt epithelium.
In the previous NordiQC assessment for CK20, run 25, 2009, HIER showed to give a significantly higher pass rate compared to enzymatic pre-treatment for the most widely used mAb clone Ks20.8. If HIER was used a pass rate of 76 % was seen (76 out of 100 labs), compared to 19 % if enzymatic pre-treatment was used (5 out of 26 labs). In this assessment only 12 laboratories used enzymatic pre-treatment for the mAb clone Ks20.8 and for these laboratories a pass rate of 58 % was seen, whereas a pass rate of 91 % was seen if HIER was performed (114 out of 126 labs).

In this assessment gastric corpus mucosa was the most appropriate control for CK20, as both the sensitivity and specificity (regarding endogenous biotin) could be evaluated in this tissue. The majority of the foveolar epithelial cells must show an at least weak to moderate cytoplasmic staining reaction, while other epithelial cells must be negative (apart from neuroendocrine cells).

This was the 3rd assessment of CK20 in NordiQC (Table 2) and a significant increase in the pass rate was seen compared to run 25.

Table 2: Proportion of sufficient results for CK20 in the three NordiQC runs performed

The significant improvement of the pass rate for CK20 can by be influenced by many parameters including new and less challenging material circulated. However, as the most challenging tissues in the previous run and this run being the gastric corpus and the urothelial carcinoma both were incorporated in the multi-tissueblocks constructed for the two runs, this supports that an improvement was achieved. The improvement most likely was related to the reduced use of enzymatic pre-treatment, which in both this run and the previous run has shown to be less successful compared to HIER for the mAb clone Ks20.8. In run 25, 21 % of the laboratories used enzymatic pre-treatment compared to 8 % in this run for the mAb clone Ks20.8. It was also related to the high quality and extended use of the Ready-To-Use (RTU) systems for CK20 from the two main providers Dako and Ventana, as the RTU systems from these two companies in this assessment showed a pass-rate of 100 % thus being superior to the pass-rates for the in-house validated protocols for CK20. The recently launched RTU system from Ventana based on the rmAb clone SP33 showed a superior performance in this run compared to the previous Ventana RTU system based on the mAb clone Ks20.8 used by the laboratories in run 25, where a pass rate of only 50 % was seen.

Conclusion
The mAb clone Ks20.8 and the pAb E16444 are both recommendable Abs for CK20. For both Abs, HIER is preferred to obtain an optimal staining.
The RTU systems from Dako and Ventana, based on the mAb clone Ks20.8 and the rmAb clone SP33 respectively, gave a pass rate of 100 % with optimal protocol settings, thus being superior to the in-house validated assays.
Gastric corpus mucosa is recommended as positive control: The majority of the foveolar cells must show an at least weak to moderate cytoplasmic staining reaction. Alternatively appendix can be used: Virtually all the luminal epithelial cells must show a strong cytoplasmic staining reaction, while the majority of the crypt epithelial cells an at least weak cytoplasmic staining reaction.