Proteomic Profiling Identifies Predictive Signatures for Progression Risk in Patients with Advanced-Stage Follicular...

Background: Follicular lymphoma (FL) is characterized by an indolent nature and generally favorable prognosis, yet poses a particular clinical challenge, since disease progression is observed in a notable subset of patients. Currently, it is not possible to anticipate which patients will be at risk of progression, highlighting the need for reliable predictive biomarkers that can be detected early in the disease. Methods: We applied tandem-mass-tag labelled nano-liquid chromatography tandem mass spectrometry (nLC-MS/MS) on 48 diagnostic formalin-fixed, paraffin-embedded tumor samples from patients with advanced-stage FL. Of these, 17 experienced subsequent progression (subsequently-progressing, sp-FL) while 31 did not (non-progressing, np-FL). Results: We identified 99 proteins that were significantly differentially expressed between sp-FL samples and np-FL samples (p < 0.05; log₂-fold changes between 0.2 and -1.3). Based on this subset of proteins, we classified patients into high-risk and low-risk subgroups using unsupervised machine learning techniques. Pathway analyses of the identified proteins revealed aberrancies within the immune system and cellular energy metabolism. In addition, two proteins were selected for immunohistochemical evaluation, namely stimulator of interferon genes 1 (STING1) and isocitrate dehydrogenase 2 (IDH2). Notably, IDH2 retained significantly lower expression levels in sp-FL samples compared with np-FL samples (p = 0.034). Low IDH2 expression correlated with shorter progression-free survival (PFS, p = 0.020). Conclusions: This study provides evidence for some of the biological mechanisms likely to be involved in FL progression and, importantly, identifies potential predictive biomarkers for improvement of risk stratification up-front at time of FL diagnosis.