Use of fast orthogonal search to predict chemotherapy-induced leukopenia
Bibliographic record
Abstract
706 Background: Myelosuppression is the predominant DLT of chemotherapy (CT). Growth factors (GF) can reduce the incidence and severity of myelosuppression, but are neither necessary nor cost-effective for all patients on CT. Attempts at developing a predictive model to identify patients at risk for severe myelosuppression have been unsuccessful to date. A reliable model to identify at risk patients early would allow closer monitoring and timely introduction of GF support. We propose using a nonlinear mathematical model, Fast Orthogonal Search (FOS), to achieve this goal. Methods: Women receiving adjuvant CMF (n=15), CEF (n=14), or CAF (n=6) for early breast cancer were selected from a single clinical practice. All were managed in a homogeneous fashion and GF support was introduced only as secondary prophylaxis. Using data from the electronic patient record, patients (pts) were retrospectively classified into high and low risk (HR/LR) categories. HR if: any hospitalization, ≥3 delays, any delay >40 days, delay at cycle 2, day 8 of any cycle not given, or any dose reduction in the first 3 cycles; LR if: no event, or delay beyond cycle 3. Pts meeting neither HR nor LR criteria were deemed medium risk and excluded from the analysis. Using complete blood count values from cycle 1 (baseline, day 8, day 28), the FOS model was trained on 14 randomly selected pts evenly split between HR and LR groups, and validated on a separate set of 14, together with an independent set of 7. Results: The model correctly classified 19 of the 21 pts. None of the LR and only 2 of the HR pts were misclassified, Fisher’s exact test p<0.00023 (2-tailed) and Matthews’ correlation coefficient φ = +0.83. The model was re-built switching the initial testing and training sets, leaving the independent 7 pts as part of the testing procedure. The model correctly classified 17 out of 21 pts. Four of the 10 LR pts were misclassified and none of the 11 HR were misclassified, Fisher’s exact test p<0.0039 (2-tailed) and Matthews’ correlation coefficient φ = +0.66. Conclusions: FOS might be applied to prospectively identify pts at high-risk for neutropenia. Further studies are needed to replicate this work on a separate larger data set, and define its reproducibility in the setting of other chemotherapy regimens. No significant financial relationships to disclose.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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 teacher head, 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".