# Metabolic Heterogeneity in High-Grade Glioma Assessed by Multi-Tracer PET and Ex Vivo Metabolomics: A Systematic Review and Meta-Analysis

## TL;DR

This study reviews how PET scans can help distinguish tumor progression from treatment effects in high-grade brain tumors and highlights the need for better integration with metabolomic data.

## Contribution

The study provides a systematic review and meta-analysis of multi-tracer PET accuracy in high-grade glioma and emphasizes the need for imaging-omics integration.

## Key findings

- Amino acid PET shows high sensitivity for pseudoprogression detection.
- FDOPA PET improves discrimination of treatment-related changes from tumor recurrence.
- Baseline PET metrics show weak and heterogeneous associations with survival outcomes.

## Abstract

Background/Objectives: High-grade glioma shows marked metabolic heterogeneity. We performed a PRISMA-guided systematic review and meta-analysis to quantify PET accuracy for pseudoprogression (PsP) and for recurrence/progression versus treatment-related change (TRC), assess pool baseline associations with overall (OS) and progression-free survival (PFS), summarize PET-based prediction of molecular markers, and assess the PET–stereotactic biopsy–ex vivo metabolomics workflow. Methods: We searched PubMed/MEDLINE and the Web of Science Core Collection (Clarivate) from inception to 1 September 2025 for HGG cohorts with baseline PET. Eligibility: Adults with HGG; diagnostic syntheses required per-patient 2 × 2; prognostic syntheses required for HR with 95% CI. Risk of bias: QUADAS-2 (diagnostic) and QUIPS (prognostic). Random-effects models pooled log-HRs and sensitivity/specificity; molecular studies were summarized by AUCs. Imaging-to-omics concordance was reviewed narratively owing to the absence of co-registered PET–metabolite maps in human HGG. Results: The results included the following: OS k = 10; PFS k = 3; PsP k = 2 (N = 76); and TRC k = 3 (N = 152). For PsP, two amino acid PET cohorts yielded a sensitivity of 0.943 and a specificity of 0.826. For TRC, pooled FDOPA across two cohorts gave rise to a sensitivity of 0.879 and a specificity of 0.771. OS meta-analyses were non-significant under Hartung–Knapp modification—FDG HR of 1.09 (95% CI 0.69–1.73) and amino acid HR of 1.03 (0.72–1.46)—with substantial heterogeneity. PFS effects varied by tracer/metric; examples include FDOPA HR of 7.92 (2.17–28.90) and MET metabolic tumor volume HR of 1.60 (1.20–2.30). Conclusions: Amino acid PET is sensitive to PsP and, with FDOPA, aids TRC discrimination when MRI is equivocal, whereas baseline PET–survival associations are weak and heterogeneous. Prospective co-registered PET/MR with stereotactic biopsies and HR-MAS NMR spectroscopy/MALDI-MSI is required to quantify imaging-to-omics concordance and standardize spatial endpoints. Study registration: PROSPERO CRD420251113416. Funding: none.