Head-to-Head Comparison of FAP Antibody Clones for Tissue Expression Analysis

Abstract

Background: Fibroblast activation protein (FAP) is a promising target for cancer diagnosis and therapy, but the specificity and sensitivity of anti-FAP antibodies used for immunohistochemistry (IHC) vary considerably. A head-to-head comparison of commercially available clones is needed to guide antibody selection, thereby standardizing FAP IHC protocols and ensuring reliable patient stratification for FAP-directed therapies. Methods: We evaluated three anti-FAP clones—FAP/4851 (mouse monoclonal), JA56-11 (recombinant rabbit), and EPR20021 (rabbit monoclonal)—in formalin-fixed, paraffin-embedded tissues from 137 patients with breast (n = 29), colorectal (n = 27), lung (n = 42), and head and neck squamous cell carcinoma (n = 39). FAP expression was assessed by manual semiquantitative scoring (H-score and percentage of positive cells) and, for a validation subset, by QuPath-based digital image analysis. The study included negative (normal liver/skeletal muscle) and positive (WI-38 cell block) controls. Results: Clones JA56-11 and EPR20021 showed specific stromal staining, whereas FAP/4851 exhibited intense non-specific reactivity and was excluded from quantitative analysis. EPR20021 demonstrated significantly higher H-scores than JA56-11 in all four cancer types (Wilcoxon test, p ≤ 0.0179), resulting in a higher overall detection rate under the tested conditions. Aberrant tumor cell expression was detected almost exclusively by EPR20021 in head and neck (59% of cases) and colorectal (59%) cancers. QuPath validation showed good correlation with manual scoring for EPR20021 in the tumor compartment of lung cancer (Spearman’s ρ = 0.701, p = 0.024) with small bias (Bland–Altman). Conclusions: Under the conditions tested, EPR20021 (Abcam) demonstrated higher staining intensity and detection rates than JA56-11 across several tumor types. Clone FAP/4851 showed nonspecific staining under the conditions used in this study and was therefore not suitable for quantitative analysis in the present experimental setting. Consequently, careful selection of well-validated antibody clones is essential for ensuring reproducible and reliable FAP immunohistochemical analysis.