Abstract P4-15-09: Refined estimates of local recurrence risk in a clinical utility study: Integrating the DCIS score, patient age and DCIS tumor size
Bibliographic record
Abstract
Abstract Background:Better tools are needed to estimate the risk of local recurrence (LR; DCIS or invasive) after breast-conserving surgery (BCS) for pts with DCIS in order to inform treatment decisions. Traditional clinico-pathologic (CP) factors, e.g., age and tumor size, provide an average LR risk derived from clinical trials and population studies. The Oncotype DX 12-gene DCIS Score assay has been validated to provide individual 10 yr LR risk estimates (Solin JNCI 2013; Rakovitch BCRT 2015). Previously we reported the impact of the DCIS Score result on radiotherapy (RT) recommendations including the pre-assay LR risk and RT recommendation and the change in RT recommendation from pre- to post-assay (Manders Ann Surg Oncol 2016).Recently a patient specific meta-analysis (MA) combined data from E5194 and Ontario DCIS Cohort (ODC) adjusting for pertinent clinico-pathologic factors to provide refined prediction estimates of LR risk after BCS alone (Rakovitch ASCO 2017). Herein we applied these risk estimates integrating DS, tumor size and patient age with adjustment for diagnosis in the year 2000 or later to refine estimates of LR in DCIS patients from the Manders et al study. Methods: 13 U.S. sites enrolled pts with DCIS treated with BCS alone from 3/2014 to 5/2015. Pts with LCIS but no DCIS, invasive BC, or planned mastectomy were excluded. Data were prospectively collected on CP factors, physician estimates of LR risk, and DCIS Score. Refined estimates of 10-yr risk of LR are presented by DCIS Score result category (0-38; 39-54; 55-100), age group (≥50 vs <50 yr) and tumor size (≤1; >1-2.5; >2.5 cm). Results: Of the 127 pts enrolled, median age was 60 yr,79.5% were postmenopausal. Median size was 8mm & 39% were ≤5mm. Median margin width was 3.0mm. ER and PR by IHC were positive in 89% and 78% of pts, respectively. For patients ≥50 yr with tumors ≤1 cm and low risk DS, the 10-yr LR risk ranges from 5.3-10.0%. A high DS result is associated with a higher 10-yr median predicted risk of LR in all subsets (table 1). The DCIS Score integrated with tumor size and patient age and the adjustment for diagnosis in 2000 or later provided risk estimates that are often lower than those provided by the DCIS Score alone without adjustment for diagnostic year. Using DS alone the percentage of patients with risk of LR <8% was 0%; however, incorporating patient age and tumor size with the DS and adjusting for diagnosis in 2000 or later, it increased to 30.9% of patients. Conclusions: Integration of the DCIS Score assay, that provides individual risk estimates of LR, with patient age and DCIS tumor size and adjusting for diagnosis in 2000 or later, provides refined estimates of 10-yr LR risk after BCS alone for DCIS. This integration enhances prognostic LR risk estimates and frequently provides lower risk estimates with which to guide individualized treatment decisions. Distribution of 10-year risk of local recurrence using DCIS Score (DS), tumor size, and age, adjusting for diagnosis in 2000 or later. Low DS (<39)Inter DS (39-54)High DS (≥55)Tumor Size(cm)Age(Yr)NMedian (Min-Max)%NMedian (Min-Max)%NMedian (Min-Max)%≤1≥50457.0 (5.3-10)810.8 (10.2-11.8)1015.1 (12.9-18.6) <50810.3 (7.4-12.1)414.9 (14.1-15.4)0 >1-2.5≥50249.5 (7.3-12.6)914.2 (12.9-15.6)519.6 (16.5-20.4) < 50216.4 (16.1-16.7)220.4 (19.8-21.1)122.2 (22.2-22.2)>2.5≥50515.7 (14.9-23.7)0 138.4 (38.4-38.4) <500 141.2 (41.2-41.2)149.3 (49.3-49.3) Citation Format: Manders JB, Solin LJ, Leonard CE, Mamounas EP, Lu R, Turner M, Baehner FL, White J. Refined estimates of local recurrence risk in a clinical utility study: Integrating the DCIS score, patient age and DCIS tumor size [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P4-15-09.
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 | 0.042 | 0.082 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.010 |
| Bibliometrics | 0.002 | 0.004 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.005 | 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".