, a breakthrough that could reshape how physicians approach diagnosis, prognosis, and treatment planning for this challenging form of brain cancer.){kind=link}
A study in Nature Precision Oncology has unveiled a new, noninvasive method for predicting immune-related biomarkers in low-grade gliomas (LGG), a breakthrough that could reshape how physicians approach diagnosis, prognosis, and treatment planning for this challenging form of brain cancer.
Low-grade gliomas, which include astrocytomas and oligodendrogliomas, are among the most common primary tumors of the central nervous system. While initially slow-growing, they frequently recur and progress into more aggressive forms. Despite advances in therapy, prognosis remains limited, and more accurate, individualized treatment strategies are urgently needed.
The research team, using data from the TCGA (The Cancer Genome Atlas) and TCIA (The Cancer Imaging Archive), focused on the pro-inflammatory cytokine Interleukin-18 (IL18), a protein involved in immune regulation that has recently drawn attention in cancer research. IL18 has been shown to play dual roles in cancer biology, at times supporting anti-tumor responses and at others promoting immune evasion and tumor progression. In glioma patients, the study found that IL18 expression was elevated and independently associated with poorer overall survival.
What makes this study stand out is its integration of transcriptomic, clinical, and radiologic data into a radiomics-based machine learning model. The model, developed using contrast-enhanced MRI scans, was designed to predict IL18 expression levels and stratify patients by prognostic outcome.
The researchers validated the model across three cohorts: a training set, a validation set, and an external test set derived from a single hospital center. The predictive performance, measured by the area under the receiver operating characteristic curve (AUC), ranged from 0.76 to 0.86, suggesting strong potential for clinical utility. Notably, patients with higher predicted IL18 levels had significantly worse survival outcomes, according to Kaplan-Meier analysis.
In addition to confirming the prognostic value of IL18, the study also delved into its relationship with the immune microenvironment. High IL18 expression correlated with increased infiltration of M2 macrophages, cells associated with tumor-promoting functions, and reduced response to immunotherapy, highlighting IL18’s complex role in shaping glioma progression and treatment resistance.
The implications of these findings are twofold. First, by enabling clinicians to estimate IL18 levels without the need for invasive biopsies, the model opens the door to more frequent and accessible monitoring of tumor biology. Second, it may help identify patients who are more likely to benefit from IL18-targeted therapies, an area of growing interest in immuno-oncology.
However, the authors caution that their model was validated with data from a single external site, and further studies across multiple institutions with more diverse populations are needed to confirm its generalizability. They also suggest future work could enhance model accuracy through the integration of additional imaging sequences and multi-omics data.