aceUrinary volatilomics using liquid-liquid extraction and gas chromatography-mass spectrometry (GC-MS) combined with machine learning algorithms as a tool for diagnosis and surveillance of urothelial bladder cancer
Abstract: Background Bladder cancer is the 11th most common cancer in the United Kingdom, with approximately 10,500 new cases annually. Diagnosis and surveillance typically involve cystoscopy, an expensive, time-consuming, and uncomfortable procedure which has encouraged efforts to identify biomarkers, particularly in urine, given its direct contact with malignant tissue. Methods Urine collected from 100 participants (50 bladder cancer patients, 50 controls) was subjected to solvent extraction followed by gas chromatograp…
"Gas Chromatography Laboratory - (1)" by Hey Paul from Sacramento, CA, USA is licensed under CC BY 2.0. To view a copy of this license, visit https://creativecommons.org/licenses/by/2.0/.
On 2026-08-07, researchers reported a urine-based test for urothelial bladder cancer that combines solvent extraction, GC-MS profiling, and machine learning. In 100 participants, an XGBoost model using an 8-metabolite panel achieved AUROC 0.869, improving on classical statistics at 0.752, with 85% balanced sensitivity and specificity.
The approach matters because current diagnosis and surveillance rely on cystoscopy, described as expensive, time-consuming, and uncomfortable, creating demand for non-invasive urine biomarkers. While accuracy looks promising for clinical deployment, the results remain early-stage from a single small cohort and have not yet been validated in larger, diverse populations or routine care.
- Study analyzed urine from 100 participants, 50 bladder cancer patients and 50 controls, using solvent extraction followed by GC-MS.
- Five machine learning algorithms were evaluated with recursive feature elimination to identify optimal biomarker panels.
- XGBoost outperformed classical univariate statistics, which had AUROC 0.752, reaching AUROC 0.869.
- When optimised for screening, the 8-metabolite panel reached 95% sensitivity with 70% specificity.
Solvent extraction of urine analyzed by GC-MS and XGBoost distinguished bladder cancer patients from controls with AUROC 0.869 and 85% balanced sensitivity and specificity using an 8-metabolite panel.
The rundown
Researchers collected urine from 50 bladder cancer patients and 50 controls and performed liquid-liquid solvent extraction followed by gas chromatography-mass spectrometry to profile volatile and semi-volatile metabolites.
They compared classical univariate statistics to five machine learning algorithms, using recursive feature elimination to select biomarker panels, with XGBoost achieving the highest performance and an 8-metabolite panel enabling trade-offs between balanced accuracy and screening sensitivity.
Sources
- Peer-reviewedBJC Reports2026-08-07
How should this claim be treated?
The debate