Immature blood vessels in rheumatoid synovium are selectively depleted in response to anti-TNF therapy

Abstract

Background: Angiogenesis is considered an important factor in the pathogenesis of Rheumatoid Arthritis (RA) where it has been proposed as a therapeutic target. In other settings, active angiogenesis is characterized by pathologic, immature vessels that lack periendothelial cells. We searched for the presence of immature vessels in RA synovium and analyzed the dynamics of synovial vasculature along the course of the disease, particularly after therapeutic response to TNF antagonists.

Methodology/principal findings: Synovial arthroscopic biopsies from RA, osteoarthritis (OA) and normal controls were analyzed by double labeling of endothelium and pericytes/smooth muscle mural cells to identify and quantify mature/immature blood vessels. To analyze clinicopathological correlations, a cross-sectional study on 82 synovial biopsies from RA patients with variable disease duration and severity was performed. A longitudinal analysis was performed in 25 patients with active disease rebiopsied after anti-TNF-alpha therapy. We found that most RA synovial tissues contained a significant fraction of immature blood vessels lacking periendothelial coverage, whereas they were rare in OA, and inexistent in normal synovial tissues. Immature vessels were observed from the earliest phases of the disease but their presence or density was significantly increased in patients with longer disease duration, higher activity and severity, and stronger inflammatory cell infiltration. In patients that responded to anti-TNF-alpha therapy, immature vessels were selectively depleted. The mature vasculature was similarly expanded in early or late disease and unchanged by therapy.

Conclusion/significance: RA synovium contains a significant fraction of neoangiogenic, immature blood vessels. Progression of the disease increases the presence and density of immature but not mature vessels and only immature vessels are depleted in response to anti-TNFalpha therapy. The different dynamics of the mature and immature vascular fractions has important implications for the development of anti-angiogenic interventions in RA.

Figure 1. Detection of immature or mature blood vessels in RA synovial tissues.

Double immunoflurescent labeling of endothelium (CD31, red fluorescence) and pericytes/smooth muscle cells (aSMA, green fluorescence) in normal and RA synovial tissue is shown. Original magnification ×400. Right panels show the same area as in middle panels with higher magnification. Mature CD31+ vessels covered by aSMA+ periendothelial cells are marked by arrows, and immature CD31+ vessels lacking aSMA+ mural cells by arrow heads.

Figure 2. Double labeling of lymphatic and CD31-positive vessels in RA synovial tissues.

Lymphatic vessels were detected by immunoperoxidase (brown immunostaining) detection of podoplanin and double immunofluorescent labeling (red fluorescence) of CD31. The same field was photographed by light or fluorescent microscopy to show the position of CD31+ (arrowheads) and podoplanin+ vessels (arrows). Light microscopy image was inverted and merged with CD31 fluorescent image of the same field to show the relative position of podoplanin (blue) and CD31 (red) labeling.

Figure 3. Clinicopathological correlations of immature blood vessels in RA synovial tissues.

Disease duration, DAS28 score, erosive disease, and synovial tissue infiltration by CD3, CD20 or CD68 cells is shown in groups with (+) or without (−) immature vessels as indicated. Density of mature or immature vessels in patients stratified by disease duration and levels of activity (low: DAS28<3.2, moderate 3.2–5.1, or high>5.1). Spearman's correlation coefficients between immature vessels density and disease duration, DAS28, CD3 or CD20 infiltration are shown. (*) p<0.05 (see text). ¶ p = 0.04 (Kruskall Wallys test and post hoc Dunns test (low versus moderate or high activity groups).

Figure 4. Variation in the density of immature vessels stratified by the levels of response to anti-TNF-α therapy.

Decrease in immature (left graphics) or mature (right graphics) vessels density between the first and second biopsy after anti-TNF-α therapy is shown stratified by EULAR responses: 0 = : No response; 1: Moderate response; 2: Good response. (*) Kruskall Wallys test and post hoc Dunns test (non-responders versus good responders).