Abstract P3-10-35: Using Automated Image Analysis To Validate Ki67 as a Clinical Prognostic Biomarker for Estrogen-Receptor Positive Breast Cancers
Bibliographic record
Abstract
Abstract Background: Ki67, an indicator of proliferation, has been shown to be a useful prognostic and predictive marker for breast cancer. Ki67 can be used to identify two distinct estrogen-receptor positive subtypes: luminal A and luminal B. Luminal A breast cancers have been identified as having a lower proliferation and better outcome compared to luminal B. Furthermore, early clinical trials suggest that Ki67 may be useful in identifying a subset of patients that are sensitive to adjuvant docetaxel treatment. Currently, only estrogen-receptor (ER), progesterone-receptor (PR), and human epidermal growth-factor (HER2) are routinely performed. We have optimized and validated an immunohistochemical (IHC) Ki67 assay and automated computerized image analysis platform for routine clinical testing. Materials and Methods: Immunohistochemical staining was quantitatively assessed using the ACIS® III platform on a cohort (N=761) of tamoxifen treated patients who were diagnosed with breast cancer in Calgary between 1990 and 2001. Tissue microarrays were constructed using three 0.6 mm cores. Ki67 results were available for 510 patients, 461 of which are ER/PR positive and HER2 negative. Staining was performed using the DAKO FLEX ready-to-use system. The percent nuclear area positive was calculated using ACIS III and the maximum value was used in statistical analysis Results: X-tile statistical software was used to identify an optimal Ki67 cut point to distinguish differential overall survival in node negative ER positive cancers of 18.75%. This cut point was then used to categorize the 461 ER/PR positive and HER2 negative breast cancers into luminal A (407; 88.3%) or luminal B (54; 11.7%) subtypes. The 8-year breast cancer specific survival was 85.2% (95% CI = 81.3% - 89.1%) for luminal A and 53.6% (95% CI = 39.3% - 67.9%) for luminal B (P<0.0001). Cox regression showed a hazard ratio of 1.63 (95% CI = 0.99 — 2.67, p=0.055), adjusting for age, tumor size, grade and lymph node status. Discussion: Quantification of Ki67 expression using automated image analysis can be used clinically to distinguish luminal A from luminal B in ER/PR positive and HER2 negative breast cancers. The purpose of this project was to develop a reliable Ki67 assay that can be easily adopted by other testing centers. Using a ready-to-use IHC system — such as DAKO FLEX — allows for consistent results between other clinical laboratories. Additionally, using an automated, quantitative imaging system — such as the ACIS® III — reduces inter-observer variation that can occur by human visual assessment. Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P3-10-35.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.003 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.006 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".