Breast cancer survival stratified by automated versus visual analysis of ER and PR immunohistochemistry
Bibliographic record
Abstract
Purpose: A woman should be able to trust that the steroid receptor status of her cancer is reported correctly. Immunohistochemistry (IHC) allows for specific assessment of tumour cells and the role of automated image analysis (IA) in the assessment of receptor IHC in breast cancer is topical. We compared ER and PR status assessed visually and by IA, and their respective abilities to predict cancer-specific survival and recurrence in a cohort with mature follow up.Patients and methods: Patients (n=525) treated for primary symptomatic operable breast cancer in Glasgow during 1995-8 were studied in triplicate tissue microarray using ER (Dako 6F11/12) and PR (Leica R636) IHC. Allred and weighted Histoscores were assigned visually (v-H Scores) and weighted H-scores by IA (ia-H Scores; Slidepath nuclear protocol). Interclass correlation coefficients, univariate and multivariate analyses were performed in SPSS v18.Results: Minimum follow up was 11.8 years (median 13.8). There was excellent agreement between steroid receptor v-H and ia-H Scores and both methods predicted cancer-specific survival equally well overall and in the 384 patients who received endocrine treatment. Allred score was also effective.Conclusions: Automated analysis of ER and PR status gave results in excellent agreement with visual analysis and equally effective (but not better) prediction of survival and tumour recurrence. Visual confirmation that the cells being analysed are representative of the invasive carcinoma remains essential and no quality asssurance requirements are made redundant by image analysis if disasters like the Newfoundland ER testing debacle (Hede, K: JNCI 2008;100:837-) are to be avoided.
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 | 0.001 | 0.004 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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".