Abstract P1-06-02: Comparative survival analysis of multiparametric tests in the TEAM pathology study: What to do when molecular tests disagree?
Bibliographic record
Abstract
Abstract Multiparametric assays for risk are increasingly used in the management of node-negative and node-positive hormone receptor-positive invasive breast cancer. Data from multiple sources suggests different tests may provide different risk estimates at the individual patient level1. Analysis from the TEAM pathology study (Bayani and Yao et al npjBreast Cancer, 2017) allows direct comparison of prognostic information from gene signatures in a clinical trial cohort of postmenopausal patients. Risk classifications using genes comprising the following multi-parametric tests: OncotypeDx® (Genomic Health Inc.)2,3, Prosigna™(NanoString Technologies, Inc.)4-6, Mammaprint® (Agendia Inc.)7,8 were performed. For the OncotypeDX-Like Recurrence Score (RS), RNA abundance was processed to fit the measurement range as described2,3, with classification into high, intermediate or low risk groups based the derived RS and modeled for DRFS. For the Prosigna-Like Risk of Recurrence Score (ROR), samples were processed as previously outlined9, then modelled against DRFS. For the MammaPrint-Like Risk Score, samples were processed by published methods8 and modelled for DRFS. Comparing OncotypeDx-Like with Prosigna-Like showed that 45% of cases were classified identically by both (3.3% low risk, 20.9% intermediate, 20.7% high). Of 3370 cases, 353 (10.5%) had scores differing by more than 1 classification (i.e. hi/low or low/high). Almost all (343) of these were cases classified high risk by OncotypeDX-Like RS/low risk by Prosigna-Like ROR (Table 1). Univariate Cox regression analysis, using low/low cases as a reference (relative risk of distant metastasis =1.0), suggested that cases called low risk by Prosigna-Like ROR/High risk by OncotypeDx-Like RS did not perform differently from cases called low risk by both tests (Table 2). However, all cases called intermediate by one test and high risk by another appeared to be high risk (Table 2). Comparisons between Prosigna-Like ROR and MammaPrint-Like scores showed similar concordance between low/low and high/high (52.5% of cases with concordant results). In Prosigna-Like ROR intermediate risk cases, MammaPrint-Like results divided cases between low and high risk, as predicted. Comparisons between these tests is challenging, and evidence on their discordance in risk stratification presents further dilemmas. Preliminary analysis of TEAM suggests a complex inter-relationship between test results in the same patient cohorts requiring careful evaluation. Table 1OncotypeDX-Like RSLowInt.HighTotalLow1126163431071Prosigna-Like RORInt.1677046151486High10106697813Total289142616553370 Table 2OncotypeDX-Like RSLowInt.HighLowRef1.26 (0.57-2.79)1.13 (0.49-2.62)Prosigna-Like RORInt.1.2 (0.47-3.05)2.22 (1.03-4.78)4.27 (2.01-9.08)High6.10 (1.58-23.6)4.15 (1.79-9.59)4.92 (2.32-10.42) Citation Format: Bartlett JMS, Bayani J, Kornaga E, Piper T, Mallon E, Yao CQ, Boutros PC, Hasenburg A, Kieback DG, Markopoulos C, Dirix L, Seynaeve C, Can de Velde CJH, Rea DW. Comparative survival analysis of multiparametric tests in the TEAM pathology study: What to do when molecular tests disagree? [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 P1-06-02.
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 | 0.122 | 0.191 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.003 | 0.005 |
| Bibliometrics | 0.003 | 0.005 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.003 |
| Open science | 0.004 | 0.002 |
| Research integrity | 0.003 | 0.003 |
| Insufficient payload (model declined to judge) | 0.008 | 0.002 |
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".