Impact of concurrent (CON) and sequential (SEQ) radiotherapy (RT) with adjuvant aromatase inhibitors (AI) in early-stage breast cancer (EBC): NCIC CTG MA.27.
Bibliographic record
Abstract
500 Notice of Retraction: "Impact of concurrent (CON) and sequential (SEQ) radiotherapy (RT) with adjuvant aromatase inhibitors (AI) in early-stage breast cancer (EBC): NCIC CTG MA.27." Abstract 500, published in the 2012 Annual Meeting Proceedings Part I, a supplement to the Journal of Clinical Oncology, has been retracted by Wendy Parulekar, MD, and Timothy J. Whelan, BM, BCh, MSc, on behalf of all authors of the abstract. The abstract concluded by suggesting that concurrent administration of an AI during the period of radiation may improve event-free survival as compared to commencing AI after completing radiation therapy. After submitting the abstract for the 2012 ASCO Annual Meeting, the authors identified issues associated with the statistical analysis of this research, which led them to reanalyze the data and in so doing, they reached different conclusions from those described in the abstract. As opposed to the abstract, which reports a hazard ratio (HR) of AI administration that is concurrent with RT vs. sequential to RT of 0.78 (p=0.001), they have determined that using a more appropriate analysis, the hazard ratio is 0.84 (p=0.13). Multiple sensitivity analyses have been performed and yield hazard ratios of 0.81-0.84 and p values of 0.07 to 0.13. In view of these findings, the conclusions reported in the abstract cannot be supported. Background: Optimal timing of administration of adjuvant (adj) RT and AI in EBC is unknown. Methods: MA.27 was a phase III RCT of exemestane to anastrozole in postmenopausal women with hormone receptor positive EBC (Goss et al. Cancer Res. 70(24, Suppl):75s, 2010). The final trial database was used for this retrospective analysis. Median follow-up was 4.1 years. MA.27 patients received CON-AI [any overlap with AI; 4233 (57%) patients], SEQ-AI [RT preceded AI, no overlap with AI; 1010 (14%) patients] and No RT [AI only; 2128 (29%) patients]. Outcome measures for this analysis were: event free survival (EFS; time to locoregional or distant disease recurrence, new primary BC, or death from any cause), locoregional recurrence (LRFS), distant recurrence (DDFS) and overall survival (OS). RT groups were compared univariately (uni) with stratified log-rank tests, and multivariately (multi) with step-wise stratified Cox regression adjusted by stratification factors: nodal status, adj chemotherapy (chemo), celecoxib, aspirin, and trastuzumab. Results: 7371 eligible women received AI; were included in the analysis; and 71% (5243) received adj RT. CON-AI and SEQ-AI groups were comparable by median age (63 v 63), proportion T1 tumours (75% v 75 %), and mastectomy rate (10% v 11%). The frequency of axillary dissection for CON-AI and SEQ-AI was 48% v 44%, proportion N0 was 73% v 69%, and proportion receiving adj chemo 29% v 41%. CON-AI had similar uni results to SEQ-AI: EFS, HR=0.86, p=0.20; LRFS, HR=0.82, p=0.51; DDFS, HR=0.92, p=0.59; and OS, HR=1.04, p=0.80. In multi analyses, CON-AI had better EFS than SEQ-AI patients [stratified HR of CON-AI to SEQ-AI 0.78 (0.66 – 0.91), p=0.001]; as well, age≥70 (p<0.0001), ECOG PS≥1 (p<0.0001), L-sided tumours (p=0.02), T2-T4 (p<0.0001), N2/N3 (p<0.0001), and no adj chemo (p=0.01) had significantly shorter EFS. There was no multi difference between CON-AI and SEQ-AI for LRFS (p=0.50), DDFS (p=0.72), or OS (p=0.85). Conclusions: Patients receiving CON-AI had significantly better EFS than SEQ-AI suggesting timing of administration of AI and RT may affect patient outcomes. Further research is necessary to confirm these findings.
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 | 0.013 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.008 | 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".