Novel RNA-Affinity Proteogenomics Dissects Tumor Heterogeneity for Revealing Personalized Markers in Precision Prognosis of Cancer

To discriminate the patient subpopulations with different clinical outcomes within each breast cancer (BC) subtype, we introduce a robust, clinical-practical, activity-based proteogenomic method that identifies, in their oncogenically active states, candidate biomarker genes bearing patient-specific transcriptomic/genomic alterations of prognostic value. First, we used the intronic splicing enhancer (ISE) probes to sort ISE-interacting trans-acting protein factors (trans-interactome) directly from a tumor tissue for subsequent mass spectrometry characterization. In the retrospective, proteogenomic analysis of patient datasets, we identified those ISE trans-factor-encoding genes showing interaction-correlated expression patterns (iCEPs) as new BC-subtypic genes. Further, patient-specific co-alterations in mRNA expression of select iCEP genes distinguished high-risk patient subsets/subpopulations from other patients within a single BC subtype. Function analysis further validated a tumor-phenotypic trans-interactome contained the drivers of oncogenic splicing switches, representing the predominant tumor cells in a tissue, from which novel personalized biomarkers were clinically characterized/validated for precise prognostic prediction and subsequent individualized alignment of optimal therapy. The interpatient tumor-phenotypic heterogeneity hinders discovery of biomarkers for individualized prognosis. Wang and colleagues introduce an alternative splicing activity-based proteogenomic method that dissects tumor heterogeneity for de novo discovery of individualized prognostic biomarkers. The resulting biomarkers distinguish high-risk patient subpopulations from other patients within each single breast cancer subtype.