The abstract is read at the publisher; the summary is JClub's.
Quantitative proteomics with machine learning classifies all four renal cell carcinoma subtypes—ccRCC, chRCC, pRCC, and oncocytoma—with 95.5% patient-level accuracy, including robust chRCC versus oncocytoma discrimination.
- Why it matters: Morphological overlap and cellular heterogeneity make renal tumor subtype and benign-versus-malignant classification difficult, especially when advanced RCC shows complex histology.
- What they did: MALDI-MSI and LC-MS quantitative proteomics generated peptide/protein profiles from a tissue microarray of 541 cores from 126 patients, then trained patient-level cross-validated machine learning models for subtype classification and biomarker-panel discovery.
- The result: RCC peptide mapping reliably separated ccRCC from healthy tissue but not chRCC versus oncocytoma, while quantitative proteomics identified new ccRCC and pRCC biomarker candidates and enabled mass-spectrometry-based immunohistochemistry to distinguish oncocytoma from chRCC.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.