Abstract P2-06-09: Prediction of relapse in patients with locally advanced breast cancer after neoadjuvant treatment
Bibliographic record
Abstract
Abstract BACKGROUND. Despite advances in cancer treatment, over 25% of patients (pts) with locally advanced breast cancer (LABC) relapse (DR) during first 5 years after treatment (Trmt). OBJECTIVES. The primary objective was to construct a prediction tool for risk of relapse (RoR) in pts with LABC after neoadjuvant therapy (NAT). Previously published works (Matsuda N. et al, 2014; Keam B. et al, 2011; Katz A. et al. 2008) have also examined this issue. MATERIAL AND METHODS. This was single center, retrospective study of 546 pts with LABC who received NAT at the Ottawa Hospital Cancer Center between 2005 and 2015. Median follow-up (FU) was 49 months. The following data collected: demographics, tumor size, nodal and receptor status, grade, HER-2, stage of disease, cancer Trmt and clinical outcomes. Primary endpoints were local (L) and/or distant (D) DR rate during first 5 years and time to DR during the first 5 years. A prediction tool was devised based on the Cox regression model. RESULTS. In 545 pts NAT was prescribed as follows: FEC-D – 91 (17%), AC-Docetaxel – 330 (60%), other regimens (AC, AC-Paclitaxel, TC, TCH)– 124 (23%). Breast conserving surgery was performed in 67 (12%) pts, mastectomy in 440 (81%) pts. Adjuvant radiotherapy was given in 485 (89%). All patients had trastuzumab – 173 pts (34%) for Her2-positive disease and endocrine Trmt (tamoxifen and/or AI) – 356 (44%) pts – for endocrine-sensitive disease. DR rate during first 5 years of FU was 17.3% (L DR – 3.2%, D DR – 13.2%, L+D DR – 0.9%). Over 60 variables were included in primary analysis. Cox regression proportional hazards model resulted in only 5 factors with significant influence on RoR during first 5 years of FU. Risk factors and their risk prediction value are: 1) residual disease (yes- 4; no-0), (HR = 4.25; p-value=0.000), 2) lymph nodes status (positive-3; negative-0), (HR = 2.27; p-value=0.006), 3) Inflammatory histology (yes-2; no-0), (HR = 1.90; p-value=0.003) 4) estrogen receptors status (positive-2; negative-0), (HR = 2.07; p-value=0.001), 5) Adjuvant radiotherapy (yes-0; no-1), (HR = 1.76; p-value=0.036). When these factors are combined the following Relapse Prediction (RP) Score can be constructed (table 1). Internal validation of proposed model was performed. ROC analysis revealed a sensitivity of 75%. According to this simple RP score, pts can be classified into to three groups (RP score – 0-5; 6-7; 8-12). RoR was 7 times higher in patients with RP Score 8-12 vs pts with score 0-5 (p-value<0.0001). CONCLUSIONS. Pts with LABC represent a heterogeneous group with diverse risk of DR. Our prognostic tool based on 5 risk factors can be used to predict RoR after NAT with a sensitivity of 75%. Pts with high risk may require additional Trmt and/or more active FU strategies and this simple model may be used to design unique studies in LABC based on RP score. We intend to further validate this model on a larger multi center /provincial population.BACKGROUND. Despite advances in cancer treatment, over 25% of patients (pts) with locally advanced breast cancer (LABC) relapse (DR) during first 5 years after treatment (Trmt). OBJECTIVES. The primary objective was to construct a prediction tool for risk of relapse (RoR) in pts with LABC after neoadjuvant therapy (NAT). Previously published works (Matsuda N. et al, 2014; Keam B. et al, 2011; Katz A. et al. 2008) have also examined this issue. MATERIAL AND METHODS. This was single center, retrospective study of 546 pts with LABC who received NAT at the Ottawa Hospital Cancer Center between 2005 and 2015. Median follow-up (FU) was 49 months. The following data collected: demographics, tumor size, nodal and receptor status, grade, HER-2, stage of disease, cancer Trmt and clinical outcomes. Primary endpoints were local (L) and/or distant (D) DR rate during first 5 years and time to DR during the first 5 years. A prediction tool was devised based on the Cox regression model. RESULTS. In 545 pts NAT was prescribed as follows: FEC-D – 91 (17%), AC-Docetaxel – 330 (60%), other regimens (AC, AC-Paclitaxel, TC, TCH)– 124 (23%). Breast conserving surgery was performed in 67 (12%) pts, mastectomy in 440 (81%) pts. Adjuvant radiotherapy was given in 485 (89%). All patients had trastuzumab – 173 pts (34%) for Her2-positive disease and endocrine Trmt (tamoxifen and/or AI) – 356 (44%) pts – for endocrine-sensitive disease. DR rate during first 5 years of FU was 17.3% (L DR – 3.2%, D DR – 13.2%, L+D DR – 0.9%). Over 60 variables were included in primary analysis. Cox regression proportional hazards model resulted in only 5 factors with significant influence on RoR during first 5 years of FU. Risk factors and their risk prediction value are: 1) residual disease (yes- 4; no-0), (HR = 4.25; p-value=0.000), 2) lymph nodes status (positive-3; negative-0), (HR = 2.27; p-value=0.006), 3) Inflammatory histology (yes-2; no-0), (HR = 1.90; p-value=0.003) 4) estrogen receptors status (positive-2; negative-0), (HR = 2.07; p-value=0.001), 5) Adjuvant radiotherapy (yes-0; no-1), (HR = 1.76; p-value=0.036). When these factors are combined the following Relapse Prediction (RP) Score can be constructed (table 1). Internal validation of proposed model was performed. ROC analysis revealed a sensitivity of 75%. According to this simple RP score, pts can be classified into to three groups (RP score – 0-5; 6-7; 8-12). RoR was 7 times higher in patients with RP Score 8-12 vs pts with score 0-5 (p-value<0.0001). Table 1. Risk prediction scoreScoreRisk of relapse (5 years)No of patientsNo of pts with DR0-5Low – 7%153 (28%) Censored (C): 77 Analysed (A): 76L:3 (4%) D:2(3%) L+D:06-7Intermediate – 26%220 (40%) C: 96 A: 124L:5 (4%) D:27 (22%) L+D:08-12High – 51%172 (32%) C: 59 A: 113L:9 (8%) D:43 (38%) L+D:5 (4.5%) CONCLUSIONS. Pts with LABC represent a heterogeneous group with diverse risk of DR. Our prognostic tool based on 5 risk factors can be used to predict RoR after NAT with a sensitivity of 75%. Pts with high risk may require additional Trmt and/or more active FU strategies and this simple model may be used to design unique studies in LABC based on RP score. We intend to further validate this model on a larger multi center /provincial population. Citation Format: Aseyev O, Simmonds L, Gertler M, Dent S, Verma S. Prediction of relapse in patients with locally advanced breast cancer after neoadjuvant treatment [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P2-06-09.
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 | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| 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".