Abstract / Summary
To systematically assess the prognostic significance of baseline 18F-fluorodeoxyglucose positron emission tomography/computed tomography ([18F] FDG PET/CT) metabolic parameters in patients with non-small cell lung cancer (NSCLC) undergoing immune checkpoint inhibitor (ICI) therapy. A systematic search of five core databases was conducted up to March 2026. Relevant data concerning NSCLC patient characteristics, metabolic [18F] FDG PET/CT parameters, and survival outcomes were extracted. Hazard ratios (HRs) were pooled to assess the prognostic significance of maximum standardised uptake value (SUVmax), mean standardised uptake value (SUVmean), metabolic tumour volume (MTV), and total lesion glycolysis (TLG) regarding overall survival (OS) and progression-free survival (PFS). A total of 33 studies were eligible for inclusion. Univariable analysis showed no prognostic value for SUVmax and SUVmean. Conversely, elevated MTV and TLG were associated with inferior PFS (MTV: HR = 1.71,95% CI: 1.39-2.11; TLG: HR = 1.84, 95%CI: 1.31-2.57) and OS (MTV: HR = 1.80, 95%CI: 1.40-2.33; TLG: HR = 1.92, 95%CI: 1.18-3.12) (P < 0.01), particularly when cut-off values ranged of 50-100cm3 for MTV (PFS: HR = 1.71, 95%CI: 1.16-2.52; OS: HR = 2.23, 95%CI: 1.42-3.51). Multivariable Cox regression analysis further identified MTV as a robust independent predictor of both PFS (HR = 1.21, 95%CI: 1.07-1.36) and OS (HR = 1.14, 95%CI: 1.03-1.26). While TLG maintained independent prognostic significance for PFS (HR = 1.46, 95%CI: 1.10-1.94), it failed to reach statistical significance regarding OS (HR = 1.00, 95%CI: 0.99-1.01) (P > 0.01). Notably, subgroup analyses demonstrated that the prognostic efficacy of MTV and TLG was further pronounced in patients presenting with Stage IV disease, those undergoing first-line treatment, or those managed with chemoimmunotherapy regimens. Baseline MTV is an independent prognostic factor for PFS and OS in NSCLC patients treated with ICIs, with a cut-off value within the range of 50-100 cm³ demonstrating optimal predictive performance. In contrast, the predictive significance of TLG for OS was not maintained after multivariable adjustment. https://www.crd.york.ac.uk/PROSPERO/view/CRD420261375745, identifier CRD420261375745.
Topics
Primary Source
Frontiers in immunology
Ask Prognia AI
Have questions about this meta-analysis?
Prognia AI can search this source alongside 35M+ PubMed papers and current ESC, AHA, NICE, and ADA guidelines to give you a fully cited clinical answer.