Baseline PET-CT Scans May Flag High-Risk B-Cell Lymphoma

 Image credit: Baseline PET-CT Scans May Flag High-Risk B-Cell Lymphoma

A recent study has identified a composite radiomic score derived from FDG PET-CT imaging at baseline, which can predict survival outcomes for patients with newly diagnosed primary mediastinal B-cell lymphoma (PMBL). The research, published in Blood Advances, showed that this composite score, which includes total metabolic tumor volume (TMTV), maximal interlesion distance (Dmax), and median peripheral centroid distance (medPCD), serves as an independent predictor of overall survival.

The analysis utilized data from the LYSA cohort, involving 180 treatment-naive PMBL patients treated in France between 2007 and 2017, with a median age of 35.7 years. Researchers conducted a post hoc evaluation of 18F-FDG PET-CT scans, focusing on 12 three-dimensional PET features, and singled out TMTV, Dmax, and medPCD as significant predictors of survival.

Patients were classified based on these features into high or low categories, with a higher accumulation of high-risk indicators correlating with poorer survival outcomes. Specifically, patients exhibiting 2-3 high-risk features had notably worse 5-year progression-free and overall survival rates compared to those showing 0-1 features. The composite score was shown to outperform individual parameters and traditional clinical factors in prognostic assessments.

The study highlights the potential of this radiomic model to identify patients with unfavorable prognosis at diagnosis. As stated by the research team led by Pierre Decazes from the Centre Henri Becquerel in Rouen, France, the integration of tumor burden, spatial dissemination, and structural compactness measures could potentially guide future trial-based risk-adapted strategies and inform treatment decisions, particularly in exploring de-escalation approaches for low-risk patients.

Despite its promising findings, the study acknowledges limitations, such as its retrospective nature, treatment heterogeneity among participants, and the absence of an external cohort for validation. These factors may affect the generalizability of the results, and further research is warranted.