A Computational Pathology Model to Predict Docetaxel Benefit in Localized High-Risk and Metastatic Prostate Cancer
Bibliographic record
Abstract
PURPOSE: Docetaxel improves survival in metastatic hormone-sensitive prostate cancer (mHSPC) and high-risk localized disease, but benefits vary substantially among patients. Without predictive biomarkers, clinicians cannot identify patients who will benefit, exposing many to unnecessary toxicity. We developed and validated an artificial intelligence-based pathology image classifier (APIC) to predict docetaxel benefit. EXPERIMENTAL DESIGN: We analyzed digitized hematoxylin and eosin-stained biopsy specimens from two phase 3 trials: CHAARTED (286/790 patients with mHSPC) and NRG/RTOG 0521 (350/563 patients with high-risk localized disease). APIC used features capturing tumor-immune spatial interactions and nuclear heterogeneity. We evaluated the predictive value of APIC for docetaxel benefit on overall survival (OS) and castration resistance using Cox proportional hazards with interaction terms. RESULTS: In CHAARTED, APIC-positive patients (56.7%) showed significant OS improvement with docetaxel [HR, 0.52; 95% confidence interval (CI), 0.31-0.85; P = 0.008] and delayed castration resistance (HR, 0.48; 95% CI, 0.33-0.71; P < 0.001), whereas APIC-negative patients (43.3%) showed no benefit (HR, 1.31; 95% CI, 0.71-2.44; P = 0.39). Treatment-APIC interactions were significant (P = 0.022 and P = 0.031). In NRG/RTOG 0521, APIC-positive patients (44.7%) demonstrated survival benefit (HR, 0.49; 95% CI, 0.26-0.92; P = 0.023), whereas APIC-negative patients (55.3%) showed no benefit. Treatment-APIC interaction was significant (P = 0.024). Predictive value remained significant after adjusting for clinical variables. Limitations include retrospective analysis and need for prospective validation. CONCLUSIONS: APIC predicts docetaxel benefit in both metastatic and localized prostate cancers, independent of clinical factors. Validation in triplet therapy with androgen receptor pathway inhibitors is needed.
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 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 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 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".