Cell-free RNA in Urine May Detect Bladder Cancer and Distinguish Response
A noninvasive test that enables cancer detection at high specificity and sensitivity would be valuable for the screening and monitoring of many malignancies, particularly those located in regions otherwise difficult to sample. Liquid biopsy represents a tractable approach, wherein tumor-derived signals are assessed in bodily fluids such as blood; for genitourinary malignancies, urine-based analytes may be particularly informative, given the potential for direct tumor cell contact. Liu and colleagues developed a method that analyzes urine cell-free RNA (cfRNA) for the noninvasive detection of bladder cancer. The workflow, termed urine random priming and affinity capture of cfRNA fragments for enrichment analysis by sequencing (uRARE-seq), builds on a hybrid capture-based sequencing approach for cfRNA in plasma and takes into account the high degree of fragmentation observed in cfRNA from the supernatant of urinary fluid. uRARE-seq was applied to noncancer controls or urine cfRNA from non–muscle-invasive bladder cancer (NMIBC) or muscle-invasive bladder cancer (MIBC) pretreatment samples, using a panel of over 3,800 genes determined to be rarely expressed in urine from healthy individuals. Machine learning was then used to develop a tumor-naïve bladder cancer detection model based on uRARE-seq data, achieving a sensitivity of 95% at 90% specificity in the training cohort with an in silico–estimated 95% limit of detection at 0.05%, with similar performance in the testing cohort. Importantly, the cfRNA-based method outperformed tumor DNA methods and was unaffected by the presence of field-effect mutations. cfRNA distinguished clinical grade (low-grade versus high-grade tumors) and invasiveness (NMIBC versus MIBC), with evidence suggesting that the models may also detect features that inform the likeliness of recurrence. Notably, application of uRARE-seq to samples from a cohort of patients with NMIBC (n = 36) collected before surgery, before Bacillus Calmette–Guérin (BCG) immunotherapy induction, and at first follow-up after BCG revealed that the assay detected minimal residual disease and found relatively higher levels of bladder cancer cfRNA after either surgery or BCG in patients who ultimately developed recurrence. Analysis of whole-transcriptome uRARE-seq data suggested that pretreatment profiles of patient samples differed whether the patients later exhibited a molecular response to BCG, prompting the development of a model that could predict whether patients would respond to BCG or intravesical chemotherapy. Together, these findings demonstrate that urine cfRNA profiling can enable detection, stratification, response prediction, and monitoring of residual disease.Liu KJ, Shi WY, Nesselbush MC, Phillips NA, Mach KE, Luca B, et al. Urine cell-free RNA for bladder cancer detection and treatment response prediction. Nat Med 2026 Oct 2 [Epub ahead of print].Note: Research Watch is written by Cancer Discovery editorial staff. Readers are encouraged to consult the original articles for full details. For more Research Watch, visit Cancer Discovery online at https://aacrjournals.org/cdnews.
Publication Details
- Journal
- Cancer Discovery
- Published
- 2026-11-20
- DOI
- https://doi.org/10.1158/2159-8290.cd-rw2026-104
- Primary Topic
- Cancer Genomics and Diagnostics
- Type
- article
- Field-Weighted Citation Impact
- 0.00