Ki67 in Breast Cancer Assay: An Ad Hoc Testing Recommendation from the Canadian Association of Pathologists Task Force

Round 1

Reviewer 1 Report

Hello

Please provide clear pictures and brief explanations

In the end, the conclusion has not been made

Author Response

1. Please provide clear pictures and brief explanations:

Regarding figure 1(a, b and C) what constitutes positive, as per the IKWG any nuclear staining above background is considered positive and these images provides a spectrum to what constitutes positive and negative nuclei by agreement by at least 4 authors. We are not sure if Reviewer 1 had a chance to assess the examples of positive nuclei, including very low staining, negative nuclei and potential caveats, all provided in the appendix. Additional brief explanations have been added to appendix 1 as well.

2. In the end, the conclusion has not been made:

We feel that we have provided recommendations to the stated questions: when to test, what tissue to use, what antibody or platform to use, how to score, how many cells to count, what constitutes positive nuclei and how to report.

Reviewer 2 Report

A useful and well compiled document.

It may be better if you could specify what constitutes a very low case load and what may be taken as an appropriate case load. Even though there maybe no publication on this, a consensus among experts may be a useful starting point.

Author Response

1.It may be better if you could specify what constitutes a very low case load and what may be taken as an appropriate case load. Even though there maybe no publication on this, a consensus among experts may be a useful starting point.

In the absence of evidence supporting a specific case load the Task Force suggest using a broad directive over a specific number. We believe that pathologists are used to engage an expert consultant whenever they are uncomfortable with a case, and this consultation could be related to a diagnostic challenge or scoring of biomarkers. Pathologists with very few requests for Ki67 scoring in breast cancer may enrich their exposure through read out modules offered by CBQA (https://cbqareadout.ca/; Breast Ki67 module is under construction) and other proficiency testing modules such as "The Canadian Pathology Quality Assurance – Assurance qualité canadienne en pathologie" (CPQA-AQCP), a non-profit corporation dedicated to ensuring quality biomarker testing in medical diagnostic laboratories. Pathologists that are uncomfortable with scoring Ki67 in breast cancer are encouraged to send it to another lab with appropriate expertise.

Reviewer 3 Report

This is an interesting paper providing useful information and recommendations regarding current issues about the evaluation of Ki67 as biomarker and its importance for patient’s management.

These recommendations are mainly based on previous studies of the IKWG group, the Monarch trial and experts’ opinion.

It is well written and conveys clear information. Provides a wealth of images which help to understand the recommendations as well as web resources which can be very useful.

Minor issues:

- It is not clear why reflex testing is not recommended upfront at the initial biomarkers testing since its results can guide therapeutical decision in luminal cancers. It would save also some time. The authors say “…there is no need or capacity for reflex testing”. Please explain further.

- It is also not clear how to select the 4 high power fields for global Ki67 assessment “across the tumor”: random? Adjacent fields?

Author Response

1.It is not clear why reflex testing is not recommended upfront at the initial biomarkers testing since its results can guide therapeutical decision in luminal cancers. It would save also some time. The authors say “…there is no need or capacity for reflex testing”. Please explain further.

At present, based on ASCO/CAP guidelines the only clinical utility of testing Ki67 in breast carcinoma is when systemic treatment with abemaciclib is considered in patients with node positive clinically high-risk ER+/HER2- breast carcinoma. Several multigene assays are used to guide selection of systemic therapy in ER+/HER2- early breast cancer (<3 positive nodes). According to Statistics Canada the majority of newly diagnosed breast cancer patients are diagnosed at an early stage (https://www150.statcan.gc.ca/n1/pub/82-003-x/2018012/article/00003/c-g/c-g03-eng.htm) and in these patients KI67 results has no clinical rule.

Based on the experience of Sunnybrook Health Sciences Centre breast biomarkers were tested in over 2000 invasive breast carcinomas during 2022, Ki67 was requested in less than 10% of ER+/HER2- in-house cases when clinicians had access to the test and initiated a request on a case-by-case basis.

With growing workload and limited resources there is no justification to implement universal upfront testing.

2- It is also not clear how to select the 4 high power fields for global Ki67 assessment “across the tumor”: random? Adjacent fields?

1-Using a regular light microscope, review the H&E slide and note the areas with invasive carcinoma, mentally register areas of DCIS, benign breast, dense inflammatory cell infiltrate and other potential artefacts.

2-Review Ki67-stained breast cancer slide at low power and input estimates of the percent area with negligible, low, medium, or high Ki67 index in invasive carcinoma (these are determined with relation to the case to be scored, without absolute numerical values to account for heterogenous distribution of positive cells).

3.Select high power fields (40x) that represent the four areas of negligible, low, medium, or high Ki67 index, and score 100 nuclei negative or positive in each field type (preferably using the app). The fields are randomly selected to represent the low power estimate and if there are less than 100 tumor cells in the field, move the slide to an adjacent field.

4.Record “Weighted global score” output as the Ki67 index for that slide

Figure 4 is added to appendix 1 with these steps to select high power fields for global assessment of Ki67 across the tumor.