Abstract
B cells and their effector molecules, antibodies, are implicated in pathophysiology of the chronic graft-vs-host disease (cGVHD) and rituximab is effective cGVHD therapy. Here we investigate B cell reconstitution in bone marrow aspirates collected from 14 mantle cell lymphoma and 22 chronic lymphocytic leukemia patients receiving rituximab infusion 375mg/m2 weekly x 4 beginning 56 days after allogeneic hematopoietic cell transplantation (HCT) following total lymphoid irradiation 80cGy x10 daily fractions and anti-thymoglobulin (1.5 mg/Kg/day x 5). Primary GVHD prophylaxis was mycophenolic acid and cyclosporine tapered off by 6 months. We hypothesized rituximab would deplete alloreactive na•ve and memory B cells and result in less chronic GVHD. Here we present multi-parameter B cell FACS analysis characterizing extent of B cell depletion and developmental stage analysis and subsequent reconstitution kinetics. We collected bone marrow aspirates prior to rituximab, and then days 90, 180, and 365 following HCT.
Peripheral B cells were detected in 17 of 34 HCT patients prior to rituximab infusion day 56. Following rituximab, peripheral blood CD19+ B cells were detected in 4 by one year, 18 by 1.5 years, and 9 by 2 years post HCT.
Multi-parameter (12 colors-14 parameters) FACS analysis of bone marrow B cells using 2 different cocktails on the same bone marrow cells distinguished: common lymphoid progenitor (CD34+CD117+CD7+), Pro B cells (CD34+CD20−CD10−), pre B cells (CD34−CD20−, CD10+), immature B cell (CD20−CD38−IgM+ IgD low/neg), mature (CD20+CD38+IgD+, IgM+) B cells, and CD38+ CD138+ plasma cells.
Despite only modest reconstitution of PERIPHERAL B cells 2 months after HCT (17/32), bone marrow B cells expressing CD19 were present in 9 out of 9 patients at 56 days post HCT and were depleted to less than 0.05% of total lymphocytes after 4 rituximab infusions when measured 90 days post-HCT (below table). Following rituximab, CD19+ B cells were first detected in the bone marrow 180 days after HCT. The mature CD19+ B cells accounted for 2–5% by 365 days post HCT.
While rituximab depleted mature B cells, plasma cells remained unchanged. Furthermore, CD138+CD38+ plasma cells were FACS sorted shown by STR DNA polymorphism testing to be recipient derived (n=5).
Consistent with observed stable plasma cell frequency, total plasma IgG showed no significant change. Inherited polymorphisms in IgG heavy chain constant regions can be recognized by allotype-specific monoclonal antibodies and thereby distinguish donor and recipient antibodies. Such allotype detection of antimicrobial IgG confirmed stable anti-VZV and EBV as well as recipient origin of these plasma IgG up to 2 years post HCT.
In support of our hypothesis, alloreactive IgG responses against 5 minor histocompability antigens (mHA) encoded on Y chromosome (DBY, UTY, ZFY, RPS4Y, and EIF1AY) were decreased in TLI/ATG/rituximab treated patients. None of the 11 male patients with female donors treated with rituximab developed antibodies against H-Y proteins while 12 out of 24 (50%) F̂M undergoing TLI/ATG without rituximab developed allo-antibodies against H-Y proteins (p=0.09).
In summary, multi-parameter (12 colors-14 parameters) immunophenotyping of bone marrow shows rituximab treatment two months after allo-HCT causes delayed donor derived B cell reconstitution, persistent antimicrobial IgG from persistent recipient plasma cells, and undetectable allogeneic H-Y antibodies.
Summary table.
Days after HCT LYMPHOID PROGENITORS CD34+ CD117+ CD7+ PRO B CELLS CD34+ CD20− CD10− PRE B CELLS CD34+ CD20− CD10− MATURE B CELLS CD20+ Ig D+ Ig M+ PLASMA CELLS CD38+ CD138+ TOTAL IgG μg/dl(pre) 56 pre = ritux n = 9 20–25% 2–6% 0.1–4% 0.2–1% 0.7–1% 655 81% 90 n = 25 20–40% 2–9% 0.5–2% ND** 0.5–3% 910 101% 180 n = 28 13–20% 5–12% 0–0.7% ND** 0.5–2% 507 60% 365 n = 16 3–8% 2–10% 0–0.5% 1–5% 0.5–3.7% 642 78%
Analysis of the heterogeneity of the BCR H-CDR3 repertoire in the bone marrow and spleen of 3-, 12-, and 20-month old mice