Can FAPI-PET be used to evaluate cardiac fibroblast activation and predict left ventricular function in patients with cardiovascular diseases? A systematic review and meta-analysis

Scritto il 26/02/2026
da Shu Wang

Eur J Med Res. 2026 Feb 26. doi: 10.1186/s40001-026-04040-5. Online ahead of print.

ABSTRACT

BACKGROUND: Fibroblast activation protein inhibitor (FAPI)-PET is a novel imaging modality that targets FAP. The purpose of this study was to conduct a meta-analysis and synthesize the available data about the sensitivity and specificity of FAPI-PET in assessing cardiac fibroblast activation and predicting left ventricular function in patients with cardiovascular disease (CVD).

METHODS: We searched the PubMed, Web of Science, and Embase databases from the time the databases were created until December 19, 2025. The diagnostic performance of FAPI-PET was meta-analyzed with Python (version 3.12). Data management and numerical computations were carried out with pandas and NumPy; meta-analytic calculations and statistical tests were implemented using SciPy; and all figures (forest plots, subgroup analyses, Deeks'funnel plot, SROC curve, and leave-one-out sensitivity analyses) were produced using Matplotlib.

RESULTS: There were 13 studies involving 256 patients, including 166 males with CVD. The age of the patients ranged from 23.0 to 76.1 years. The CVD patient-level pooled sensitivity, specificity and area under the curve (AUC) of FAPI-PET in evaluating cardiac fibroblast activation were 0.91, 0.75 and 0.9358, respectively. The pooled sensitivity and specificity of FAPI-PET in evaluating cardiac fibroblast activation in CAD patients were 0.96 and 0.82, respectively; those in non-CAD patients were 0.89 and 0.58, respectively. Multiple semiquantitative FAPI-PET uptake metrics were significantly correlated with LVEF. Among them, FAP volume (Pearson) demonstrated the best random effects pooled correlation estimate (-0.73), with substantial heterogeneity (I2 = 0%; τ2 = 0.0000, p = 0.41).

CONCLUSIONS: FAPI-PET has high sensitivity, specificity and AUC in evaluating cardiac fibroblast activation in CVD. It has significant predictive value for LV function and can be used successfully for the early detection and assessment of activated fibroblasts.

REGISTRATION: PROSPERO; No.: CRD42024561672; URL: https://www.crd.york.ac.uk/prospero.

PMID:41749239 | DOI:10.1186/s40001-026-04040-5