The material circulated for the HER-2 ISH assessment run comprised one normal breast tissue & four breast ductal carcinomas showing the HER-2 gene/chromosome 17 (HER-2/chr17) ratios as follows:

*PATHWAY®, Ventana (data from one reference lab.), **Inform HER-2 Dual SISH kit, Ventana (range of data from two reference labs.), ***HER2 FISH pharmDX™ Kit, Dako (range of data from two reference labs.).
All tissues were fixed for 24 - 48 h. in 10 % neutral buffered formalin (NBF).

HER-2 BRISH, technical assessment
The main criteria for assessing a BRISH HER-2 analysis as technically optimal was the ability to interpret
and evaluate the HER-2/chr17 ratios in all five tissues.

A staining was assessed as good, if the HER-2/chr17 ratios could be evaluated in all five tissues, but the interpretation was slightly compromised to e.g. due to a weak or excessive counterstaining, excessive retrieval or similar.

A staining was assessed as borderline if one of the tissue cores could not be properly evaluated e.g. because of too weak signals, no signals and/or a low signal-to-noise ratio.

A staining was assessed as poor if two or more of the tissue cores could not be properly evaluated.

HER-2 BRISH and FISH interpretation
For both BRISH and FISH the participating laboratories were asked to submit a scoring sheet with the interpretation of the HER-2/chr17 ratio for all the five tissues. The NordiQC FISH data was used as reference in order to evaluate the scoring consensus between the participating laboratories and NordiQC. A concordant interpretation was seen if the five tissues were evaluated as listed below:

• Evaluation of the normal breast tissue no. 1. corresponding a non-amplified status.

• Evaluation of the breast ductal carcinoma no. 2 corresponding an amplified status

• Evaluation of the breast ductal carcinoma no. 3 corresponding a non-amplified status.

• Evaluation of the breast ductal carcinoma no. 4 corresponding an equivocal (borderline) status.

• Evaluation of the breast ductal carcinoma no. 5 corresponding a non-amplified or equivocal (borderline)status.

Results BRISH
In total 105 laboratories participated in this assessment. 84 laboratories performed BRISH and out of these 68 (81 %) achieved a sufficient mark. The results are summarized in Table 1.
 

Table 1. Systems and assessment marks for BRISH HER-2, run B12

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

Comments
In this assessment a sufficient demonstration and evaluation of the HER-2 gene amplification status in all the tissues included in the multitissue block could be obtained by all the different BRISH systems used by the laboratories. All the included tissues were fixed in 10 % neutral buffered formalin for 24-48 hours according to the ASCO/CAP guidelines for the tissue preparation of breast tissue. However the breast ductal carcinoma, tissue no. 4 was slightly more challenging than the other tissues regarding the protocol settings as this tumour was found to be less robust as excessive retrieval typically impaired the morphology and thus complicating the identification of the BRISH signals. This pattern was seen for all systems used.

Optimal protocol settings

Two-colour HER-2 systems
For the INFORM™ Dual ISH system, Ventana, an optimal demonstration for HER-2 BRISH was in brief typically based upon HIER in Cell Conditioning 2 (CC2) for 24-32 min. at 86-90˚C and proteolysis in P3 for 8 - 16 min at 36-37˚C. The HER-2 SISH probe was typically applied for 6 hours at 50-52˚C, while the chr17 probe was applied for 2 hours at 42 - 44˚C.
Using these protocol settings a sufficient result was seen in 76 % of the submitted protocols (35 out of 46). 11 laboratories used a protocol with optimal settings, but for unexplained reasons a complete false negative staining reaction or a staining result with an excessive background staining e.g. due to silver precipitates was seen. The remaining insufficient results were characterized by an impaired morphology. This pattern was typically caused by excessive retrieval hampering the interpretation as the nuclei were almost totally digested complicating the identification and interpretation of the BRISH signals.

For the DuoCISH™ system SK109, Dako, the main protocol settings giving an optimal result were based on HIER for 10 min in the pre-treatment buffer at 95 - 98°C and proteolysis in Pepsin for 2-3 min. at 37°C or 7 - 10 min. at room temp. (both reagents included in the HER2 CISH pharmDX kit SK109). The HER-2 and the chr17 probes were applied for 14 – 20 hours at 45˚C and visualized by the DuoCISH™ kit SK109, Dako.
Using these protocol settings a sufficient result was seen in 75 % of the submitted protocols (6 out of 8). In the insufficient results typically a too weak or false negative staining for the HER-2 signals in both the neoplastic cells and in the normal stromal cells was seen. This observation might be related to a too low sensitivity of the reagents used for the immunohistochemical demonstration of the HER-2 genes. In this context it is of utmost importance that the Red chromogene used for the visualization of the HER-2 genes in the DuoCISH™ system is prepared immediately before use and that Pepsin is stored at appropriate conditions in order to maintain the proteolytic capacity.

One-colour HER-2 systems
For the INFORM™ HER-2 SISH, Ventana, an optimal result typically was based upon HIER in CC2 or Reaction buffer (RB) for 24-32 min. at 90-95˚C and proteolysis in P3 for 4 - 8 min at 37˚C. The HER-2 SISH probe was typically applied for 6 hours at 50-52˚C. Using these protocol settings a sufficient result was seen in 100 % of the submitted protocols (7 out of 7).

For the ZytoDot® CISH system C-3003, ZytoVision, an optimal result typically was obtained by using proteolysis in Pepsin for 3 min at room temp, HIER in EDTA for 10-15 min. at 98˚C, hybridization at 37˚C for 14-16 hours and visualized by the ZytoVision detection kit C-3003.

For the SPOT-Light® CISH system 84-0150, Invitrogen, an optimal result was obtained by using proteolysis in Pepsin for 5-7 min. at room temp, HIER in Tris-EDTA for 15 min. at 98˚C, hybridization at 37˚C for 14 hours and visualized by the Invitrogen detection kit 84-0150.

HER-2 interpretation and scoring consensus
Both the laboratories performing BRISH and FISH were requested to send in their own interpretation on the stained sections, which was completed by 98 out of the 105 laboratories participating in this run. The participants evaluations were compared to the HER2 FISH amplification status generated by the NordiQC reference laboratories. The overall  results are summarized in table 2. No significant difference in concordance rates was seen between the participants using FISH or BRISH.

Table 2: Interpretation and scoring consensus between the participants and the NordiQC FISH data

A high consensus rate (99-100%) between the participants and NordiQC regarding the determination of the HER-2 amplification status was seen for the normal breast tissue no.1 and the highly amplified breast carcinoma, tissue no. 2, whereas a lower consensus rate was seen in the breast carcinomas. tissues no. 3 and 5 (71–90%). For the breast carcinoma, tissue no. 4 with a HER-2/chr17 ratio in the range of 1.8 – 2.2 based on FISH (thus to be categorized as equivocal/borderline) a low consensus rate (43 %) was seen. However, at stated above, even one of the reference labs. obtained a score of 1.5 with dual-SISH (i.e., no consensus). This emphasizes the need both to focus on the technical set-up of the ISH procedure and to establish a better standardization to support and harmonize the interpretation of the ISH results. The recommendations for HER-2 testing in the UK¹ as well as digital image analysis may aid in more standardized and reproducible results.

1. Bartlett JM, Starczynski J, Atkey N, Kay E, O'Grady A, Gandy M, Ibrahim M, Jasani B, Ellis IO, Pinder SE, Walker RA. HER2 testing in the UK: recommendations for breast and gastric in-situ hybridisation methods. J Clin Pathol. 2011 Aug;64(8):649-53. Epub 2011 Jun 20. PubMed PMID: 21690244.

Fig. 1. Graphic illustrations showing the HER-2 amplifications in the five tissue cores as assessed by the laboratories. Each circle represents the HER-2/chr17 ratio given by a laboratory. The box shows the 25/75 percentile and the line within the box the median value. Whiskers show the 10/90 percentile. In each graph, left columns represent laboratories performing BRISH, right columns laboratories performing FISH.

This was the 7th NordiQC assessment of HER-2 BRISH/ISH. As seen in table 3, similar pass rates have been obtained in the last two assessments.

Table 3: Proportion of sufficient results for HER-2 BRISH in the seven NordiQC runs performed

Conclusion
In this assessment an optimal demonstration of HER-2 BRISH could be obtained by the two commercially available two-colour HER-2 systems INFORM™ HER-2 Dual ISH, Ventana and DuoCISH™,Dako.
Also the single-colour HER-2 systems, INFORM™ HER-2 SISH, Ventana, ZytoDot®, ZytoVision and SPOT-Light®, Invitrogen could be used to obtain an optimal demonstration.
For an optimal performance the retrieval settings – HIER + proteolysis – must be carefully balanced to provide both a high sensitivity and a preserved morphology. Attention must also be addressed to the interpretation as a high inter-observer variation was seen.