Evaluating the performance of the BOADICEA model in predicting 10-year breast cancer risks in UK Biobank.
Bibliographic record
Abstract
BACKGROUND: The BOADICEA model predicts breast cancer risk using cancer family history, epidemiological and genetic data. We evaluated its validity in a large prospective cohort. METHODS: We assessed model calibration, discrimination and risk classification ability in 217,885 women (6,838 incident breast cancers) aged 40-70 years old of self-reported White ethnicity with no previous cancer from the UK Biobank. Age-specific risk classification was assessed using relative risk (RR) thresholds equivalent to the absolute lifetime risk categories of < 17%, 17-30% and ≥30%, recommended by the National Institute for Health and Care Excellence guidelines. We predicted 10-year risks using BOADICEA v.6 considering cancer family history, questionnaire-based risk factors, a 313-SNP polygenic score and pathogenic variants. Mammographic density data were not available. RESULTS: The PRS was the most discriminative risk factor (AUC=0.65). Discrimination was highest when considering all risk factors (AUC=0.66). The model was well calibrated overall (E/O=0.99, 95%CI=0.97-1.02; calibration slope=0.99, 95%CI:0.99-1.00), and in deciles of predicted risks. Discrimination was similar in women younger and older than 50 years. There was some underprediction in women under age 50 (E/O=0.89, 95%CI=0.84-0.94; calibration slope=0.96, 95%CI:0.94-0.97), which was explained by the higher breast cancer incidence in UK Biobank than the UK population incidence in this age group. The model classified 87.2%, 11.4% and 1.4% of women in RR categories <1.6, 1.6-3.1 and ≥3.1, identifying 25.6% of incident breast cancer cases in category RR ≥ 1.6. CONCLUSION: BOADICEA, implemented in CanRisk (www.canrisk.org), provides valid 10-year breast cancer risk which can facilitate risk-stratified screening and personalized breast cancer risk management.
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 | 0.023 | 0.057 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.002 | 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".