Validation of IHC4 algorithms for prediction of risk of recurrence in early breast cancer using both conventional and quantitative IHC approaches.
Bibliographic record
Abstract
517 Background: Hormone receptors, HER2 and Ki67 are residual risk markers in early breast cancer. Combining these markers into a unified algorithm (IHC4) provides information on residual recurrence risk of patients treated with hormone therapies. This study aimed to independently investigate the validity of the IHC4 algorithm for residual risk prediction using both conventional (DAB)-IHC and quantitative immunofluorescence (QIF-AQUA). Methods: The TEAM pathology study recruited >4500 samples from patients treated in the TEAM trial. TMAs were stained for ER, PgR, HER2 and Ki67 using QIF-AQUA technology or DAB-based immunohistochemistry (DAB-IHC). Central HER2 FISH was performed. Quantitative image analysis was used to generate expression scores that were normalized to produce “IHC4 algorithm” as well as novel algorithm scores. Algorithm scores were compared with disease recurrence in univariate and multivariate Cox Proportional Hazards models. Results: Both DAB-IHC and QIF-AQUA IHC4 continuous models were significant (P<0.0001) for prediction of disease recurrence with a continuous Hazard Ratio (HR) of 1.011 (1.010 – 1.013) for QIF-AQUA IHC4 versus 1.008 (1.007 – 1.010) for the DAB-IHC IHC4 model using the published IHC4 algorithm (Cuzick et al 2011). Binning continuous model scores (4 bins) by Kaplan-Meier survival analysis was used to graphically illustrate these effects. De novo models for both DAB-IHC and QIF-AQUA were also significantly (P<0.0001) predictive of residual risk in early breast cancer. Additionally, all 4 models were independent predictors of recurrence (P<0.0001) with other recognized clinical prognostic factors in multivariate analysis. Although results from DAB and QIF-AQUA were modestly correlated, the QIF-AQUA model showed enhanced prediction of recurrence in both Cox Proportional Hazards Modeling and C-index calculations. Conclusions: Either conventional DAB or QIF-AQUA methods of IHC provided evidence supporting the clinical utility of IHC4 algorithms in the context of the TEAM study. With careful standardization, either of these IHC4 assays should be considered for prediction of residual risk in early breast cancer.
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 | 0.019 | 0.030 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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".