Loading
Life sciences · Review
Frontiers in Oncology · November 5, 2026
No summary has been generated for this record yet. What follows is drawn from its source metadata only.
Review.
No findings were extractable from the material analysed.
Safety was not reported in the material analysed. Check the source before drawing any conclusion about harm.
The source did not state who this applies to in practice.
Graded across the dimensions that decide whether you should act, each from what the source actually supports. There is no single score, and where a dimension was not assessed it says so.
This record has not been graded across any dimension yet. Treat the label above as provisional and read the source.
What is missing. This record has no bottom line, key findings, reported figures, evidence dimensions. That is a gap in the analysis, not a judgement about the study.
Background Prostate cancer is one of the most common male malignancies. Androgen deprivation therapy (ADT) is the backbone of treatment for locally advanced and metastatic disease, yet many patients progress to castration-resistant prostate cancer (CRPC), historically within about 2 years of ADT monotherapy, and validated, widely generalisable tools to identify high-risk patients early remain limited. Methods PubMed and OpenAlex were searched from January 2021 to August 2026 and supplemented by citation tracking; peer-reviewed studies reporting prediction models for progression to CRPC after ADT were included and synthesised narratively, with a PROBAST+AI-style appraisal of representative models. Results Of 65 unique records screened, 53 met the eligibility criteria (12 excluded as outside the scope or superseded); 17 references were added during revision, so 70 references are cited. CRPC arises through androgen receptor pathway activation, AR-V7, and neuroendocrine differentiation. Gleason score, PSA nadir, bone metastasis, inflammatory blood indices (neutrophil-to-lymphocyte ratio, systemic immune-inflammation index), circulating tumor cells, circulating tumor DNA, and genetic signatures are established risk factors. Individual studies report higher discrimination for machine-learning models than for traditional nomograms, but the comparisons are within-study and at high risk of bias and optimism, and external validation was reported for only two of eight core models. Super Learner, an ensemble framework, has proved stable across clinical outcome predictions; its use for CRPC prediction is only beginning. Conclusion A multi-parameter Super Learner model could improve CRPC risk stratification after ADT, provided it is validated prospectively in multicenter cohorts and reported transparently.