Using Ki67 To Improve and Simplify Outcome Modeling for Breast Cancer.
Bibliographic record
Abstract
Abstract Ki67 is a marker of proliferation which has several advantages over histological grade. Ki67 is determined less subjectively and is a continuous rather than a categorical variable. Many studies that have looked at Ki67 have been small and underpowered. A large population based tissue microarray was used to: A) test the prognostic value of Ki67 in testing and validation sets, and B) construct an improved model including Ki67 and conventional prognostic variables to predict patient outcome. Methods: The cohort included 2,780 patients with early breast cancer diagnosed in British Columbia and a median follow up of 14.5yrs. Variables included were: tumor size (T), number of positive nodes (N), grade, lymphovascular invasion (LVI), estrogen receptors (ER), progesterone receptors (PR), Her2, Ki67, local treatment (surgery, radiation) and systemic treatment (chemotherapy, hormonal). Prognostic factors were balanced between the training and validation sets. Prognostic variables were identified in the testing set among ER positive and ER negative cohorts using Cox Regression analysis and tested in the validation set. Results: The inclusion of Ki67 in the Cox Regression analysis resulted in the elimination of grade as a predictor. For ER positive disease independent predictors were T, N, LVI, PR, HER2, Ki67, local treatment, chemotherapy and hormonal therapy. Independent predictors among ER- cases were T, N, LVI, Ki-67, and chemotherapy. Predicted 10-yr Breast Cancer Specific Survival in the validation set was 72.0% versus 72.4% [SE: 1.2] observed. As subtle prognostic differences may result in very disparate treatment recommendations for Stage I breast cancers, we specifically reviewed this group. In analysis of stage I patients, there were no statistically significant deviations between predictions and observation; agreement between the predicted and observed 10-yr BCSS was excellent (86.0% vs 87.6% (SE: 1.6) p = 0.3169). As well, elevated Ki67 was common (53%) and was a powerful prognostic variable with causing more than a doubling of the 10-yr BrCa mortality (elevated ≥ 10% vs low < 10%, 16.8% vs 6.8%) in this group (table). Conclusion: In this study the proliferation marker Ki67 replaced histologic grade as a predictor of outcome for patients with early breast cancer. Predictive models such as Adjuvant! could incorporate Ki67 as an input variable and this modification is being developed. If models that use Ki67 are validated they may be able to be used globally and be cost effective compared to more expensive genomic predictors.Table: Predicted versus observed 10yr BCSS, based on the new modelPatientsSubgroupNPredicted SurvivalObserved Survivalp-valueAll patientsOverall139772.072.40.75 ER+101277.276.40.56 ER-38558.461.20.27 HER2+19755.658.00.50 HER2-120074.874.80.99Stage IOverall45386.087.60.32 ER+35287.690.40.09 ER-10179.678.40.77 HER2+4378.882.40.55 HER2-41086.888.40.33 Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 4042.
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 | 0.004 | 0.008 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 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".