Abstract GS3-06: Dynamics of breast cancer relapse reveal molecularly defined late recurring ER-positive subgroups: Results from the METABRIC study
Bibliographic record
Abstract
Abstract Background: Recent studies have demonstrated that women with early stage ER-positive (ER+) and HER2-negative (HER2-) breast cancer have a persistent risk of recurrence and cancer related death up to 20 years post diagnosis, highlighting the chronic nature of ER+ breast cancer and critical need to identify tumor characteristics that are more predictive of risk of recurrence than standard clinical covariates. However, progress in delineating the dynamics of breast cancer relapse and biomarkers of late recurrence has been hindered by the lack of large cohorts with long-term clinical follow-up and molecular information. Methods: We report the results of a cohort of 3,240 breast cancer patients from the United Kingdom and Canada with 20 years of follow-up (median 9.75 years), including 1,980 with accompanying molecular data from the primary breast tumor. Information for each patient on loco-regional recurrence (LR), distant recurrence (DR), and site(s) of metastases was collected. We developed a non-homogenous Markov chain model that accounted for different clinical endpoints and timescales, as well as competing risks of mortality and the distinct baseline hazards that characterize different molecular subgroups. This approach enabled robust analysis of the spatio-temporal dynamics of breast cancer recurrence across the clinical subgroups, PAM50 subgroups and the integrative clusters, while also enabling individual risk of relapse predictions. Results: We employed our multistate model to compute the probability of experiencing a LR or DR, as well as the baseline transition probabilities from surgery, LR or DR at various time intervals for average individuals in each of the clinical/molecular subgroups. These analyses reveal four late-recurring ER+ (predominantly HER2-) subgroups, together accounting for 26% of all ER+ tumors, with high (median 42-55%) risk of recurrence up to 20 years post-diagnosis. Each of these four subgroups maps to one of the Integrative Clusters, defined based on genomic copy number alterations and gene expression, and is enriched for a characteristic copy number amplification events: 11q13 (CCND1, RSF1), 8p12 (FGFR1, ZNF703), 17q23 (RPS6KB1) and 8q24 (MYC). These four molecular subgroups are superior in predicting late DR than standard clinical variables. Conclusions: A detailed understanding of the rates and routes of metastasis and their variability across the distinct molecular subtypes is essential for devising personalized approaches to breast cancer care. We describe a molecularly characterized breast cancer cohort with long-term clinical follow-up and a statistical modeling framework, enabling delineation of the dynamics of breast cancer recurrence at unprecedented resolution. These analyses reveal four late recurring ER+ subgroups and accompanying biomarkers that collectively define the quarter of ER+ cases at highest risk of recurrence. Our findings highlight opportunities for improved patient stratification and biomarker-driven clinical trials directed at the subset of breast cancer patients with persistent risk of recurrence. Citation Format: Curtis C, Rueda OM, Sammut S-J, Chin S-F, Caswell-Jin JL, Seoane JA, Callari M, Batra R, Pereira B, Bruna A, Ali HR, Provenzano E, Liu B, Parisien M, Gillett C, McKinney S, Green A, Murphy L, Purushotham A, Ellis I, Pharoah P, Rueda C, Aparicio S, Caldas C. Dynamics of breast cancer relapse reveal molecularly defined late recurring ER-positive subgroups: Results from the METABRIC study [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr GS3-06.
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 | 0.004 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.001 |
| 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".