Draining PLN and Synovial Inflammation in RA Knee Joints Pre and Post Anti-TNF or B Cell Depletion Therapy

The specific aim of this study is to examine the morphologic and functional relationship
between the draining PLN and synovial inflammation in the knee joints of RA patients before
and after therapy with TNF antagonists or B cell depletion therapy. The extensive search
for informative and practical clinical biomarkers of RA has produced several candidates
including CRP, anti-CCP antibodies and matrix metalloproteinase-3 (MMP-3), but none of these
soluble molecules can predict disease progression with a high degree of certainty, and no
biomarker has been identified that can predict flare. Thus, our findings that PLN collapse
precedes arthritis flare in the murine model offers a novel biomarker that may be of great
value in RA, but requires validation in humans. To this end, we propose two clinical pilots
that are designed to test the validity and feasibility of PLN assessment and quantification
in active RA by 3T CE-MRI and Doppler US. The second goal of these pilots is to see if there
is a relationship between the clinical response to either a TNF antagonist or B cell
depletion and a change in the size or CE pattern of PLN. The pilot studies outlined below
are designed to test the hypotheses that an effective clinical response to anti-TNF or BCDT
is associated with an increase in draining LN function. While our overall goal is to
determine if Doppler US can predict flare, we must first establish the validity and
feasibility of this approach so that we can design a clinical trial that is properly
designed and adequately powered.

Aim A: To assess the effect of TNF inhibition on volume and CE in the PLN in RA patients by
MRI, and compare the effectiveness of Doppler US to achieve the same outcome measures.

Hypothesis: Rheumatoid synovitis in the knee is triggered by TNF over-expression by
monocytes/macrophages in the synovium and/or the draining PLN, which produces inflammation
that exceeds PLN draining capacity. This results in decreased inflammatory cell egress from
the joint and arthritic flare. Effective anti-TNF therapy in RA will result in increased
PLN function due to decreased cellularity and lymphatic flow from the synovium to the
draining node.

Rationale: To address our first hypothesis, we plan to determine the validity and
feasibility of draining LN function by analyzing the ability of 3T CE-MRI and Doppler US to
quantify PLN in inflamed RA knees. While CE-MRI (volume x CE) will be the primary outcome
measure of PLN function based on our pre-clinical studies, this approach is not adaptable to
standard clinical practice. Furthermore, serial MRIs to monitor flare in RA are not feasible
due to cost, time, convenience and availability of resources. Thus, parallel US studies
will be performed to determine if this instrument, which is gaining acceptance in
rheumatology practice, can generate accurate and quantifiable PLN measurements.

Aim B. To assess the effect of anti-CD20 therapy on volume and CE in the PLN in RA patients
who "flare" after a period of effective anti-TNF therapy.

Hypothesis: RA flare in patients who had been effectively managed by anti-TNF therapy is
caused by a local insult that triggers B-cell migration and precipitates LN shutdown and
decreased inflammatory cell egress from the joint. Effective anti-CD20 BCDT therapy in
these patients will correlate with increased PLN function and decreased inflammation in the
synovium.

Rationale: RA is a complex syndrome in which excess TNF production is central to disease
pathogenesis. While ~70% of these patients can be effectively managed by anti-TNF therapy,
an additional insult (i.e. local immune response or trauma) results in a flare that cannot
be adequately suppressed by standard anti-TNF therapy in some patients (80, 81). Many
patients who flare on one or more anti-TNF therapies do respond to BCDT (26). An explanation
for this that fits our preliminary findings in the TNF-Tg mice is that a local
immune-inflammatory response that shares lymphatics with RA synovium can induce CXCL13
expression in monocyte/macrophages in the draining LN and these cells and soluble CXCL13
enter the LN via the lymphatics and triggers mass migration of B-cells into the paracortical
sinuses. This B cell translocation results in decreased egress of inflammatory cells, which
had been maintained by anti-TNF therapy, and manifests as an arthritic flare. To provide a
human correlate for the murine experiments designed to test this hypothesis outlined in Aim
2, and to produce feasibility data for a clinical trial to test this novel mechanism of
action, we will evaluate PLN radiographic changes following anti-CD20 BCDT in RA patients
who flare on anti-TNF therapy. We will also assess DAS28 clinical responses.

Inclusion Criteria

- Signed, IRB-approved, written informed consent

- Subjects can be of either gender but must be more than 18 years old.

- Subjects must fulfill the disease activity criteria for RA and a DAS28 will be
assessed at baseline and at 2 months after rituximab therapy.

- Aim A - Eligible subjects must meet criteria for RA and have an inadequate response
to MTX defined as DAS28 >5.1. They must have been on a steady dose of MTX (between
15 and 20 mg /week for a minimum of 8 weeks). Subjects must have evidence of knee
synovitis on exam to enter the study.

- Aim B - Subjects must have demonstrated a response to a TNF antagonist as evidenced
by a DAS score <2.8 or <4 tender and swollen joints. Flare will be defined as a DAS
28 >5.1 of more than 8 swollen and tender joints. Subjects must have evidence of
knee synovitis to enter the study. Subjects will be off etanercept, infliximab or
adalimumab for 4 weeks before starting BCDT. All subjects must also be on a stable
dose of DMARD (MTX, leflunomide, azulfidine, hydroxychloroquine) for 8 weeks before
entry into the study.

Exclusion Criteria

- Patients will be excluded for medical or other reasons at the discretion of the
investigators. The reasons for the exclusion must be recorded, e.g. risk of
non-compliance, vulnerability, medically unstable, etc.

- Active systemic disorders or inflammatory conditions (i.e., chronic infection with
hepatitis B, hepatitis C or HIV) other than the conditions being studied.

- Patients with a plasma creatinine > 1.5 mg/dl

- Aim B - Subjects with an allergy to corticosteroids will be excluded from the study.

- Anyone answering yes to the following questions will be excluded:

Do you have a history of:

1. Cardiac (Heart) pacemaker or defibrillator?

2. Cardiac (Heart) valve replacement or prosthesis?

3. Aneurysm clips from brain surgery?

4. Ear prosthesis (cochlear or stapedial implant)? (hearing aids?)

5. Neurostimulator?

6. Biostimulator?

7. Any type of pumps in or on your body?

8. Shrapnel, gunshot wound, BB pellet?

9. Metallic penile prosthesis?

10. Metallic blood vessel filter or stent?

11. Orthopedic prosthesis?

12. Have you ever had an injury involving metal fragments in your eye?

13. Have you ever had neurosurgery (brain or skull surgery)?

Note: Subjects treated with rituximab will not be excluded from study based on their
immunosuppressive drugs but use of these agents will be recorded and discussed in the
analysis.