Advertisement
Mayo Clinic Proceedings Home

Antineutrophil Cytoplasmic Antibody-Associated Renal Vasculitis Treated With Autologous Mesenchymal Stromal Cells: Evaluation of the Contribution of Immune-Mediated Mechanisms

      Abstract

      We report the first case of renal antineutrophil cytoplasmic antibody (ANCA)–associated vasculitis treated with autologous mesenchymal stromal cells (MSCs). A 73-year-old man was admitted to the hospital for malaise, weight loss, and oliguria. His serum creatinine level was 2.7 mg/dL but it rapidly increased to 7.8 mg/dL; urinalysis showed proteinuria and hematuria, and the ANCA to myeloperoxidase with a perinuclear pattern (pANCA) titer was high (132 IU/mL). Renal biopsy showed necrotizing crescentic glomerulonephritis. Standard immunosuppressive therapy (cyclophosphamide and corticosteroids) was ineffective. Rituximab therapy was started, but it was discontinued after the third dose to minimize the risk of systemic spread of a severe oral Candida infection and to prevent superinfections that were facilitated by leukopenia. The patient received autologous MSCs, 1.5 × 106 cells/kg body weight, intravenously. After 7 days, his serum creatinine level decreased to 2.2 mg/dL, pANCA titer decreased to 75 IU/mL, and urinalysis findings normalized. Eight months later, he received a second MSC infusion because his serum creatinine level increased. In 1 week, his creatinine level decreased to 1.9 mg/dL and his pANCA titer decreased to 14 IU/mL. Immunosuppressive therapy was subsequently withdrawn. At the last follow-up visit, 12 months after the second MSC infusion, the patient remained in clinical remission without any therapy. Infusion of MSCs induced expansion of the T-lymphocyte subset expressing a regulatory T-cell phenotype (CD4+CD25+Foxp3+) and a notable reduction in interferon-γ, interleukin 6, and tumor necrosis factor serum levels.

      Abbreviations and Acronyms:

      AAV (ANCA-associated vasculitis), ANCA (antineutrophil cytoplasmic antibody), BVAS (Birmingham Vasculitis Activity Score), CTL (cytotoxic T lymphocyte), CYC (cyclophosphamide), IL (interleukin), MSC (mesenchymal stromal cell), NK (natural killer), pANCA (ANCA with a perinuclear staining pattern), PBMC (peripheral blood mononuclear cell), PHA (phytohemagglutin), RTX (rituximab), TNF (tumor necrosis factor), Treg (regulatory T cell)
      The standard therapy for antineutrophil cytoplasmic antibody (ANCA)–associated vasculitis (AAV) is cyclophosphamide (CYC) and corticosteroids, which is typically successful; however, some patients may relapse or have a refractory course. Recently, rituximab (RTX) was tested in cases resistant to standard therapy on the assumption that B-lymphocyte depletion would suppress pathogenic autoantibodies.
      • Falk R.J.
      • Jennette J.C.
      Rituximab in ANCA-associated disease.
      • Flint J.
      • Morgan M.D.
      • Savage C.O.
      Pathogenesis of ANCA-associated vasculitis.
      In fact, RTX reduces circulating CD20+ B-cell numbers dramatically, but the decrease in ANCA levels is transient, and the survival of plasma cells in bone marrow and memory B cells in lymphoid tissues seems to play a crucial role in RTX failure.
      Evidence that tumor necrosis factor (TNF) has a pathogenetic role in ANCA-associated disease led to testing for treatment with TNF blockers, but their use causes concerns about excessive treatment-related morbidities, eg, a higher rate of infections and cancer development.
      • Huugen D.
      • Tervaert J.W.
      • Heeringa P.
      TNF-α bioactivity-inhibiting therapy in ANCA-associated vasculitis: clinical and experimental considerations.
      Although studies point to ANCA as the primary pathologic agent in AAV, recent studies indicate the pathogenic role of T cell–mediated immunity, and limited evidence suggests that defects in regulatory T-cell (Treg) number or function may account for slower remission.
      • Flint J.
      • Morgan M.D.
      • Savage C.O.
      Pathogenesis of ANCA-associated vasculitis.
      • Abdulahad W.H.
      • Stegeman C.A.
      • van der Geld Y.M.
      • et al.
      Functional defect of circulating regulatory CD4+ T cells in patients with Wegener's granulomatosis in remission.
      • Morgan M.D.
      • Day C.J.
      • Piper K.P.
      • et al.
      Patients with Wegener's granulomatosis demonstrate a relative deficiency and functional impairment of T-regulatory cells.
      Taken together, these data underline the need for a new approach to AAV treatment, and immunity modulation could be a feasible option. Mesenchymal stromal cells (MSCs) are multipotent cells that can differentiate into several types of adult cells, some of which promote tissue repair. In addition, MSCs have an array of effects on the immune system, including inhibition of B- and T-cell functions, redirection of type 1/type 2 helper T-cell balance toward type 2 helper T-cells, and expansion of Treg populations.
      • De Miguel M.P.
      • Fuentes-Julián S.
      • Blázquez-Martínez A.
      • et al.
      Immunosuppressive properties of mesenchymal stem cells: advances and applications.
      All these properties make MSCs a promising therapeutic option for a few patients who are identified by stringent criteria, ie, patients with refractory autoimmune disease (European Vasculitis Study Group definition), those in a life-threatening setting, or those who have contraindications to other treatments.

      Case Report

      A 73-year-old man was admitted to the hospital after experiencing malaise for 4 weeks and oliguria for the past 24 hours. On physical examination he was pale and afebrile, with a blood pressure of 140/90 mm Hg. Laboratory analyses detected a high serum creatinine level (2.7 mg/dL [to convert to μmol/L, multiply by 88.4]), a low Modification of Diet in Renal Disease–calculated glomerular filtration rate (22 mL/min per 1.73 m2), a plasma urea level of 151 mg/dL, and blood (3+), protein (1+), and erythrocyte casts in urine. Serum levels of complement fractions and immunoglobulins were normal. Test results were positive for antinuclear antibody, anti-extractable nuclear antigen and anti-Sjogrens Syndrome A and negative for anti-Sjogrens Syndrome B, anti-ribonucleoprotein/smooth muscle antibodies, double-stranded DNA, and anticardiolipin antibodies. We detected ANCAs to myeloperoxidase with a perinuclear staining pattern (pANCAs) by enzyme-linked immunosorbent assay (132 IU/mL) and indirect immunofluorescence (titer >1:640). Anti–proteinase 3 antibodies were undetectable. A computed tomographic chest scan yielded no evidence of pulmonary vasculitis. Renal biopsy revealed a necrotizing crescentic glomerulonephritis.
      The specimen evaluated by light microscopy was renal cortex with 14 glomeruli, 3 of which were globally sclerotic. Eight glomeruli had epithelial crescents with segmental glomerular capillary necrosis, disruption of the Bowman capsule, fibrin, and cells extending to the adjacent interstitium. Tubular lumina contained erythrocytes, and the interstitium was edematous with lymphocytic infiltrates (Figure 1, A). The specimen evaluated by immunofluorescence revealed linear glomerular capillary walls staining positive for IgG, IgA, and IgM, with moderate granular mesangial C3 staining. No immunoglobulin deposits in the form of immune complexes were present. Crescents stained for fibrin. Disease activity measured using the Birmingham Vasculitis Activity Score (BVAS) was 15.
      Figure thumbnail gr1
      Figure 1Representative sections of renal biopsies performed before the first (A) and second (B) mesenchymal stromal cell infusions (periodic acid–Schiff, original magnification ×200).
      Methylprednisolone pulse therapy (500 mg for 3 days) was started, followed by oral CYC (2 mg/kg daily) and prednisone (1 mg/kg daily), but the corticosteroid was reduced to 5 mg/d after only 6 weeks owing to the onset of severe diabetes mellitus. Within 1 month, his serum creatinine level peaked at 7.8 mg/dL, plasma urea level at 242 mg/dL, and potassium level at 7.6 mmol/L. The ANCA titer was persistently high. Treatment with RTX (375 mg/m2 per week) was started, but soon after administration of the first dose an oral Candida infection occurred that worsened during the next 2 weeks despite prompt initiation of fluconazole therapy. Furthermore, the patient presented with severe leukopenia (white blood cell count, 1000/μL [to convert to ×109/L, multiply by 0.001]). The risk of progressive fungal infection and potential superinfections associated with immunosuppressive therapy was further enhanced by the immunodeficiency caused by uremia
      • Libetta C.
      • Esposito P.
      • Sepe V.
      • et al.
      Dialysis treatment and regulatory T cells.
      • Libetta C.
      • Rampino T.
      • Dal Canton A.
      Polarization of T-helper lymphocytes toward the Th2 phenotype in uremic patients.
      • Libetta C.
      • De Nicola L.
      • Rampino T.
      • et al.
      Inflammatory effects of peritoneal dialysis: evidence of systemic monocyte activation.
      • Memoli B.
      • Libetta C.
      • De Nicola L.
      • et al.
      Hemodialysis related interleukin-2 receptor release by peripheral blood mononuclear cells.
      • Memoli B.
      • Libetta C.
      • Rampino T.
      • et al.
      Hemodialysis related induction of interleukin-6 production by peripheral blood mononuclear cells.
      and diabetes. Consequently, we canceled the fourth programmed RTX dose and withdrew CYC therapy. At this point, immunosuppressive therapy was contraindicated due to the presence of acute autoimmune renal disease with critical histopathologic lesions: necrotizing glomerulitis and extracapillary proliferation, indicating irreversible renal injury.
      In view of the severe renal disease, we decided to treat the patient with autologous MSCs on a compassionate basis. After obtaining written informed consent from the patient and approval from the local ethics committee, we isolated and ex vivo expanded MSCs derived from the patient’s bone marrow following Good Manufacturing Practice procedures using 5% platelet lysate as a growth factor source.
      • Bernardo M.E.
      • Avanzini M.A.
      • Perotti C.
      • et al.
      Optimization of in vitro expansion of human multipotent mesenchymal stromal cells for cell-therapy approaches: further insights in the search for a fetal calf serum substitute.
      • Bernardo M.E.
      • Zaffaroni N.
      • Novara F.
      • et al.
      Human bone marrow-derived mesenchymal stem cells do not undergo transformation after long-term in vitro culture and do not exhibit telomere maintenance mechanisms.
      • Bernardo M.E.
      • Avanzini M.A.
      • Ciccocioppo R.
      • et al.
      Phenotypical/functional characterization of in vitro-expanded mesenchymal stromal cells from patients with Crohn's disease.
      • Ciccocioppo R.
      • Bernardo M.E.
      • Sgarella A.
      • et al.
      Autologous bone marrow-derived mesenchymal stromal cells in the treatment of fistulising Crohn's disease.
      The patient's MSCs expressed high levels (>95% positive cells) of CD90, CD73, CD105, CD13, and HLA-A, HLA-B, and HLA-C surface antigens, whereas they were negative for CD34, CD45, CD14, CD80, CD31, and HLA-DR molecules. This pattern, as well as the clonogenic efficiency, proliferative capacity, morphologic features, differentiation potential, and genetic stability, reflected features observed in healthy donor MSCs.
      • Bernardo M.E.
      • Avanzini M.A.
      • Perotti C.
      • et al.
      Optimization of in vitro expansion of human multipotent mesenchymal stromal cells for cell-therapy approaches: further insights in the search for a fetal calf serum substitute.
      • Bernardo M.E.
      • Zaffaroni N.
      • Novara F.
      • et al.
      Human bone marrow-derived mesenchymal stem cells do not undergo transformation after long-term in vitro culture and do not exhibit telomere maintenance mechanisms.
      Moreover, MSCs suppressed in vitro lymphocyte proliferation to phytohemagglutin (PHA) in a dose-dependent manner (Figure 2, A) and CD3-redirected cytotoxic T-lymphocyte (CTL) activity (Figure 2, B). The PHA-induced lymphocyte proliferation was evaluated as previously described.
      • Bernardo M.E.
      • Avanzini M.A.
      • Ciccocioppo R.
      • et al.
      Phenotypical/functional characterization of in vitro-expanded mesenchymal stromal cells from patients with Crohn's disease.
      The CD3-redirected CTL activity was evaluated in effectors (peripheral blood mononuclear cells [PBMCs] alone or co-cultured with MSCs at a PBMC:MSC ratio of 10:1) recovered from 24-hour cultures by a chromium-51 release cytotoxicity assay following a previously described method.
      • Moretta A.
      • Tambussi G.
      • Bottino C.
      • et al.
      A novel surface antigen expressed by a subset of human CD3- CD16+ natural killer cells: role in cell activation and regulation of cytolytic function.
      Figure thumbnail gr2
      Figure 2A, Phytohemagglutin (PHA)-induced in vitro proliferation of patient's peripheral blood mononuclear cells (PBMCs) before and at different time points after mesenchymal stromal cell (MSC) infusions (PBMCs alone). In vitro inhibition of PHA-induced PBMC proliferation by MSCs is reported in the same panel; MSCs were added to in vitro cultures at different MSC:PBMC ratios. ∗Results are expressed as stimulation index = count per minute stimulated/count per minute unstimulated. Orange bar = before first MSC infusion (I MSC); blue bar = 1 month after I MSC infusion; brown bar = before second MSC infusion (II MSC); yellow bar = 1 month after II MSC infusion. B, CD3-redirected cytotoxic T-lymphocyte activity of patient's PBMCs alone (orange bar) or co-cultured for 24 hours with MSCs (blue bar), measured at different time points. ∗Results are expressed as percentage of specific lysis of target cells at an effector to target ratio of 10:1. C, Immunoglobulin serum levels at different time points. To convert IgA and IgM to mg/L, multiply by 10; to convert IgG to g/L, multiply by 0.01. rr = reference range; RTX = rituximab. D, Percentage of activated T lymphocytes (CD4+CD25+, orange line) and regulatory T cells (CD4+CD25+Foxp3+, blue line) at different time points. ∗CD4+CD25+ cells are expressed as a percentage of gated CD4+ T lymphocytes. CD4+CD25+Foxp3+ cells are expressed as a percentage of gated CD4+CD25+ cells. E, Serum interferon-γ (orange line), interleukin 6 (blue line), and tumor necrosis factor (brown line) levels at different time points.
      We obtained enough MSCs, expanded until passage 3, for 2 infusions of 1.5 × 106 cells/kg body weight. The MSCs were infused soon after preparation, ie, 1 month after the last RTX dose had been administered. Levels of pANCA were persistently high (101 IU/mL), and urinary sediment still indicated active renal disease (red blood cell count, >20 per high-power field), red blood cell casts, proteinuria greater than 1 g/24 h, and a BVAS of 15. The MSC infusion was well tolerated, but after 1 week a Doppler echocardiography–confirmed left venous femoropopliteal thrombosis occurred. Coagulation and hemostasis were normal, and low-molecular-weight heparin therapy was started. Serum analyses showed a decrease in the creatinine level to 2.2 mg/dL and in the pANCA titer to 75 IU/mL. Urinary sediment was also normalized.
      Four months after MSC infusion, the pANCA titer further decreased to 30 IU/mL, the serum creatinine level was 1.8 mg/dL, and the BVAS was 3. Three months later, the patient was readmitted to the hospital for malaise and weight loss. His serum creatinine level was 2.4 mg/dL; pANCA titer was greater than 1:640; urinary sediment showed red blood cell casts, leukocytes, and protein casts; and the BVAS was 6. Infections were excluded as the cause on the basis of an absence of physical signs and negative culture results. Renal biopsy revealed moderate interstitial cellular infiltrates, some glomeruli with segmental sclerosis and 3 out of 7 glomeruli with extracapillary proliferation, but no necrotizing capillaritis or arteriolitis (Figure 1, B). A second MSC infusion (1.5 × 106 cells/kg body weight intravenously) was administered 8 months after the first while continuing prednisone, 5 mg/d, therapy. Low-molecular-weight heparin treatment was started as prophylaxis. One week later, the pANCA titer decreased to 14 IU/mL, the creatinine level was 1.9 mg/dL, urinary sediment was normal, and the BVAS was 1. One month after the second MSC infusion, prednisone use was discontinued. At the last visit, 12 months after the second MSC infusion, the patient was in good health, was in clinical remission, and had a serum creatinine level of 1.6 mg/dL, a pANCA titer of 11 IU/mL, and a BVAS of 1. In the past 11 months, the patient did not undergo any therapy. The distribution of circulating T and natural killer (NK) lymphocytes and the levels of immunoglobulin serum (Figure 2, C) were not substantially changed, whereas an increase in the percentage of cells expressing a Treg phenotype (CD4+CD25+Foxp3+) with respect to conventional activated T cells (CD4+CD25+) was observed after each MSC infusion (Figure 2, D).
      After RTX administration, circulating CD19+ B-lymphocyte levels remained less than 1% for at least 8 months and reached 2% after 20 months. Serum interferon-γ, interleukin (IL) 6, and TNF levels decreased after the MSC infusions (Figure 2, E), and IL-10 levels remained unchanged. In vitro lymphocyte proliferation to PHA was not inhibited by the MSC infusions (Figure 2, A); the increase in the stimulation index observed after the MSC infusions was possibly due to the withdrawal of immunosuppressive therapy. In vitro cytotoxicity assays documented that spontaneously activated and MSC-stimulated patient PBMCs were unable to mediate cytolysis of autologous MSCs (data not shown); patient PBMCs mediated CD3-redirected CTL activity (Figure 2, B) and NK cell cytotoxicity (data not shown) at all the evaluated time points before and after the MSC infusions.

      Discussion

      To our knowledge, this is the first report of a patient with severe renal AAV treated with MSCs. We excluded alternative therapy with TNF and plasmapheresis, whose risks are as great as those of immunosuppressive therapy. We, therefore, considered the MSC option; this decision was based on a recent meta-analysis of selected randomized controlled trials in which a significant association between MSCs and infections or acute infusional toxicity were not observed,
      • Lalu M.M.
      • McIntyre L.
      • Pugliese C.
      • et al.
      Safety of cell therapy with mesenchymal stromal cells (SafeCell): a systematic review and meta-analysis of clinical trials.
      whereas there was evidence of a high infection incidence associated with RTX and CYC use.
      • Kelesidis T.
      • Daikos G.
      • Boumpas D.
      • et al.
      Does rituximab increase the incidence of infectious complications? a narrative review.
      • Wall N.
      • Harper L.
      Complications of long-term therapy for ANCA-associated systemic vasculitis.
      When MSCs were infused, ANCA levels were persistently elevated, and urinary indices of active disease were present. Thus, the remarkable decrease in the serum creatinine level to 2.2 mg/dL just 1 week after MSC administration suggests that it was the effect of the MSCs. However, renal function recovery, which occurred 5 weeks after the last RTX administration, ie, within the time limit for RTX activity, overlapped with the MSC infusion, making it impossible to determine which was responsible. Despite the overlap in treatments, a causal relationship between MSC administration and renal function recovery is suggested by the abrupt and striking decrease in ANCA levels and by the normalization of urine values, in contrast with the apparent persistence of these signs of active disease during RTX therapy. Indeed, the possible, but uncorroborated, therapeutic role of MSCs in the first episode of acute renal failure was soundly confirmed in the second flare, which occurred after 8 months of remission. The relapse included elevation of the serum creatinine level, and renal biopsy confirmed active disease, although it was less severe than that seen in the first biopsy. A second MSC infusion was followed 1 week later by a decrease in creatinine and ANCA levels. In view of this response, prednisone therapy was interrupted without any effect on the state of remission, which persisted until the last follow-up visit. Note that the response to the second MSC infusion was not affected by the overlap with previous immunosuppressive therapy. There were no immediate/short-term adverse events associated with the first MSC infusion. However, 1 week after the first infusion, a venous thrombosis occurred in the left leg. To date, similar complications have not been reported in the literature; moreover, AAV itself is associated with thromboembolic events, and a single episode does not prove a causal relationship. Nonetheless, we administered heparin with the second dose of MSCs because we believed it was reasonable to prevent thrombosis in a patient whose high risk was revealed by a recent episode.
      Within the limits of information that can be obtained from a case study, we explored some MSC effects on the distribution and function of immune cells and their products. We found that MSC therapy did not hinder the pattern of T-, B-, and NK-cell distribution; PHA-induced proliferation; NK-cell and CD3-redirected CTL activity; or serum immunoglobulin levels but that it induced a relevant expansion of tolerogenic Tregs. Serum interferon-γ and IL-6 levels were lower after MSC infusion than before MSC administration, suggesting that MSCs suppress the expansion of aggressive immune T-cell subtypes. Most important, MSCs considerably decreased serum TNF levels, indicating that MSCs could replace TNF blocker therapy.

      Conclusion

      Therapy with MSCs induced prompt recovery of renal function in a patient with severe AAV in whom standard therapy was not feasible. The MSC infusions were well tolerated. This case report suggests MSCs as a possibile treatment option that merits further investigation in selected cases of AAV that are resistant or have contraindications to current therapy.

      Acknowledgments

      We are grateful to Maurizio Bonfichi, Cesare Perotti, Laura Catenacci, Melissa Mantelli, Elisa Lenta, Claudia Alpini, and Laurene Kelly for their essential contributions to the completion of this work.

      Supplemental Online Material

      References

        • Falk R.J.
        • Jennette J.C.
        Rituximab in ANCA-associated disease.
        N Engl J Med. 2010; 363: 285-286
        • Flint J.
        • Morgan M.D.
        • Savage C.O.
        Pathogenesis of ANCA-associated vasculitis.
        Rheum Dis Clin North Am. 2010; 36: 463-477
        • Huugen D.
        • Tervaert J.W.
        • Heeringa P.
        TNF-α bioactivity-inhibiting therapy in ANCA-associated vasculitis: clinical and experimental considerations.
        Clin J Am Soc Nephrol. 2006; 1: 1100-1107
        • Abdulahad W.H.
        • Stegeman C.A.
        • van der Geld Y.M.
        • et al.
        Functional defect of circulating regulatory CD4+ T cells in patients with Wegener's granulomatosis in remission.
        Arthritis Rheum. 2007; 56: 2080-2091
        • Morgan M.D.
        • Day C.J.
        • Piper K.P.
        • et al.
        Patients with Wegener's granulomatosis demonstrate a relative deficiency and functional impairment of T-regulatory cells.
        Immunology. 2010; 130: 64-73
        • De Miguel M.P.
        • Fuentes-Julián S.
        • Blázquez-Martínez A.
        • et al.
        Immunosuppressive properties of mesenchymal stem cells: advances and applications.
        Curr Mol Med. 2012; 12: 574-591
        • Libetta C.
        • Esposito P.
        • Sepe V.
        • et al.
        Dialysis treatment and regulatory T cells.
        Nephrol Dial Transplant. 2010; 25: 1723-1727
        • Libetta C.
        • Rampino T.
        • Dal Canton A.
        Polarization of T-helper lymphocytes toward the Th2 phenotype in uremic patients.
        Am J Kidney Dis. 2001; 38: 286-295
        • Libetta C.
        • De Nicola L.
        • Rampino T.
        • et al.
        Inflammatory effects of peritoneal dialysis: evidence of systemic monocyte activation.
        Kidney Int. 1996; 49: 506-511
        • Memoli B.
        • Libetta C.
        • De Nicola L.
        • et al.
        Hemodialysis related interleukin-2 receptor release by peripheral blood mononuclear cells.
        ASAIO J. 1996; 42: 60-63
        • Memoli B.
        • Libetta C.
        • Rampino T.
        • et al.
        Hemodialysis related induction of interleukin-6 production by peripheral blood mononuclear cells.
        Kidney Int. 1992; 42: 320-326
        • Bernardo M.E.
        • Avanzini M.A.
        • Perotti C.
        • et al.
        Optimization of in vitro expansion of human multipotent mesenchymal stromal cells for cell-therapy approaches: further insights in the search for a fetal calf serum substitute.
        J Cell Physiol. 2007; 211: 121-130
        • Bernardo M.E.
        • Zaffaroni N.
        • Novara F.
        • et al.
        Human bone marrow-derived mesenchymal stem cells do not undergo transformation after long-term in vitro culture and do not exhibit telomere maintenance mechanisms.
        Cancer Res. 2007; 67: 9142-9149
        • Bernardo M.E.
        • Avanzini M.A.
        • Ciccocioppo R.
        • et al.
        Phenotypical/functional characterization of in vitro-expanded mesenchymal stromal cells from patients with Crohn's disease.
        Cytotherapy. 2009; 11: 825-836
        • Ciccocioppo R.
        • Bernardo M.E.
        • Sgarella A.
        • et al.
        Autologous bone marrow-derived mesenchymal stromal cells in the treatment of fistulising Crohn's disease.
        Gut. 2011; 60: 788-798
        • Moretta A.
        • Tambussi G.
        • Bottino C.
        • et al.
        A novel surface antigen expressed by a subset of human CD3- CD16+ natural killer cells: role in cell activation and regulation of cytolytic function.
        J Exp Med. 1990; 171: 695-714
        • Lalu M.M.
        • McIntyre L.
        • Pugliese C.
        • et al.
        Safety of cell therapy with mesenchymal stromal cells (SafeCell): a systematic review and meta-analysis of clinical trials.
        PLoS One. 2012; 7 ([published online ahead of print October 25, 2012]): e47559
        • Kelesidis T.
        • Daikos G.
        • Boumpas D.
        • et al.
        Does rituximab increase the incidence of infectious complications? a narrative review.
        Int J Infect Dis. 2011; 15: e2-e16
        • Wall N.
        • Harper L.
        Complications of long-term therapy for ANCA-associated systemic vasculitis.
        Nat Rev Nephrol. 2012; 8: 523-532