Survival prediction in patients treated by FOLFIRI and bevacizumab for metastatic colorectal cancer (PRODIGE 9) using contrast-enhanced CT texture analysis (SPECTRA).
Bibliographic record
Abstract
3601 Background: Quantitative assessment of tumor architecture changes may help to early identify non-responder patients and propose a tailored treatment strategy. Our objective was to build and validate a radiomics signature able to predict early the lack of response to chemotherapy including FOLFIFRI and bevacizumab using baseline and first evaluation CT and to compare it to the RECIST and morphological criteria. Methods: For 230 patients of PRODIGE 9 study and treated by FOLFIRI and bevacizumab, a computed analysis (CA) was performed on the dominant liver lesion (DLL) at baseline and 2 months post-chemotherapy. RECIST evaluation was performed at 2 and 6 months. The sum of the target liver lesions (STL), the density of the DLL, CA parameters and their changes rates were correlated with the 2-year survival status. A radiomics signature combining 3 parameters was built in one arm and validated in the second arm. Survival was estimated with the Kaplan-Meier method and compared with log-rank test. Results: The strongest predictive factors for 2-year survival status were decrease in STL(AUC = .69±.05[95%CI:.60-.77]), change rate in kurtosis(ssf = 0) (AUC = .66±.05[95%CI:.57-.74]), and the baseline density of the DLL (AUC = .68±.05[95%CI:.59-.77]). Using multivariate analysis, predictive factors of 2-year survival status were the decrease in STL > 15%(HR = 1.92, P= .002), the increase in kurtosis value(ssf = 0) > 93% (HR = 2.16, P= .001), and baseline DLL > 64.3UH (HR = 1.70, P= .02). Then, the SPECTRA-score was built by according 1 point for each of the 3 criteria. Patients with a SPECTRA-score > 1 had a lower overall survival in the training ( P= .001) and in the validation cohort ( P= .002). Non-response according to RECIST at 6 months had the same prognostic value as SPECTRA-score>1 at 2 months. Conclusions: A radiomics signature combining STL, density and CA on baseline and first evaluation CT is be able to predict which patient will have a poor outcome with same performances than standard evaluation with RECIST1.1 at 6 months in mCRC patients. Clinical trial information: NCT00952029.
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 | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 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 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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".