Abstract P6-16-01: Refined local recurrence risk estimates based on a multigene expression assay combined with clinicopathological features significantly impacts radiotherapy recommendation in patients with low/intermediate risk DCIS treated with breast-conserving surgery
Bibliographic record
Abstract
Abstract Background: Guidelines recommend that breast radiation (RT) can be omitted for patients with a low risk of local recurrence (LR) after breast-conserving surgery (BCS) for DCIS. The inability to identify women at low risk of LR (<10%) at 10 years after BCS has hampered efforts to de-escalate therapy for DCIS. The Oncotype DX Breast DCIS Score® (DS), a 12-gene expression assay, predicts LR risk after BCS. The revised DS report adds information from the DS with clinicopathological features (CPF) (age, tumor size, year) (BCRT 2018;169;358-690) to provide refined estimates of LR risk and can better identify patients with low 10-year LR risk (<10%) after BCS. We evaluated the impact of refined LR risk estimates on its ability to change radiation oncologists’ recommendations for RT in women with low/moderate risk DCIS treated with BCS. Methods: Prospective cohort study of women with low/moderate risk pure DCIS treated with BCS. Cases with age <45 yrs, margins <1mm, tumor >2.5cm, multifocality, or prior breast cancer were excluded. Baseline CPFs, the DS and risk of LR were collected. Pre-assay, the radiation oncologist provided an estimate of 10-year LR risk without RT and a preliminary recommendation for RT. Post-assay, final recommendations were recorded. The primary outcome was change in treatment recommendation by the radiation oncologist. Target sample size was 220 to provide data on 200 evaluable patients with adequate precision. Results: 217 patients were evaluable: mean age, 63 years and mean tumor size, 1.1 cm. Nuclear grade was low in 26 (12%), intermediate in 116 (53%), and high in 75 (35%) of patients. Mean DS = 32; 140 (64%) low (<39), 32 (15%) intermediate (39-54), 45 (21%) high risk DS (≥5). The assay lead to a change in treatment recommendation in 35.2% (76/216) (95% CI: 29.1-41.8) of cases. The proportion of cases recommended to receive RT decreased from 79% (N=171) pre-assay to 50% post-assay (p<.001). This was due to a significant increase in the proportion of cases (46%) with a predicted low risk of LR (<10%) post-assay and recommendations to omit RT for those with low LR risk. Pre-assay, physician estimates of 10-yr LR risks after BCS based on CPFs alone were <10% in 13 (6%) cases, 10-15% in 76 (35%) cases and >15% in 128 (59%) cases. Post-assay, estimated 10-yr LR risk after BCS was <10% in 101 (46%) patients (83% recommended no RT), 10-15% in 67 (31%) (RT recommended in 66%) and >15% in 49 (22%) (RT recommended in 98%). Conclusion: The use of the DCIS Score combined with CPFs identifies more women with an estimated low (<10%) 10-yr LR risk after BCS leading to a significant change in treatment recommendations with a decrease in recommendations for RT following BCS compared to CPFs alone. Table 1.Post-assay 10-yr LR risk<10%10-15%>15%(N=101)(N= 67)(N=49)Pre-assay estimated 10-yr LR risk• <10% (N=13)922• 10-15% (N=76)50188• >15% (N=128)424739RT recommended• Yes17 (17%)44 (66%)48 (98%)• No83 (83%)23 (34%)1 (2%) Citation Format: Eileen Rakovitch, Anne Koch, Laval Grimard, Hany Soliman, Christiaan Stevens, Francisco Perera, Iwa Kong, Senti Senthelal, Margaret Anthes, Ericka Wiebe, Jeffrey Cao, Mira Goldberg, Sally Smith, Luciana Spadafora, Sameer Parpia, Timothy Whelan. Refined local recurrence risk estimates based on a multigene expression assay combined with clinicopathological features significantly impacts radiotherapy recommendation in patients with low/intermediate risk DCIS treated with breast-conserving surgery [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr P6-16-01.
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 | 0.003 | 0.010 |
| 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 | 0.001 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 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".