A127: Validating Popliteal Lymph Node Phenotype and Bin Expansion as Biomarkers of Rheumatoid Arthritis Knee Flare in Clinical Pilot Studies

Abstract

Background/Purpose:Factors precipitating joint flare in rheumatoid arthritis (RA) patients are poorly understood. Contrast enhanced (CE) MRI studies in murine models have identified popliteal lymph node (PLN) phenotypes that correlate with arthritic flare in the adjacent knee defined by a sudden increase in synovial volume and focal erosions. Prior to flare, the PLN expands in size and CE due to increased lymph volume and accumulation of CD23+CD21hiCD1dhi B cells (Bin). Arthritic flare following PLN collapse is characterized by decreased CE and translocation of Bin to the paracortical sinuses with loss of efferent lymphatic drainage. To validate lymphatic biomarkers in RA, we performed pilot studies to demonstrate the feasibility of detecting expanding and collapsed PLN via CE‐MRI. We analyzed PLN from RA patients undergoing total knee replacement or amputation for ischemic limbs in patients without RA.Methods:RA patients experiencing knee flare had a 3T CE‐MRI prior to initiation of biologic therapy. PLN and synovium volumes were measured with Amira. PLN from RA patients and amputees were harvested, cells were stained and examined by flow cytometry.Results:MRI and 3D volume rendering of a representative flaring knee from early vs. long‐standing RA (Fig. 1) show evidence of expanding PLN (n = 3; 3.57, 1.23, 2.2 cm3) adjacent to highly vascular synovitis (203.40 cm3). In contrast, knee flare in longstanding RA (>30 yrs) presents with collapsed PLN (n = 2; 0.24, 0.16 cm3) adjacent to pannus (159.88 cm3) that contains both bright (vascular) and dark (necrotic) regions on MRI. Flow cytometry confirmed large numbers of Bin‐like cells (CD23hiCD21hi) in RA vs. non‐RA PLN. Clinical endpoints measured by disease activity score (25% improvement) and C‐reactive protein levels (40% decrease) preliminarily show a correlation between effective therapy and CE‐MRI changes. imageConclusion:We present evidence that similar to murine inflammatory arthritis, knee flare in RA is modulated by altered dynamics of draining lymph nodes and lymphatic vessels. We found that the flare in early arthritis is characterized by extensive joint inflammation that appears to exceed the capacity of expanded lymphatics. In contrast, knee flare in long‐standing arthritis is associated with loss of lymphatic drainage, which in mice, is caused by loss of the lymphatic pulse and clogging of efferent lymphatics by Bin cells. Additional studies are warranted to demonstrate the reliability of these findings of knee flare in early vs. chronic RA. Figure 1. Representative images of MRI and corresponding 3D rendering from RA patients with knee flare shows inflamed synovium with bright PLN in the T2 fat‐suppressed MRI (A) and uniform inflammatory tissue and large PLN (B) in early RA. In contrast, note marked focal erosions in the proton density weighted MRI (C) and patchy synovium with collapsed PLN (D) in long‐standing RA.