Rituximab, an anti-CD20 chimeric antibody, is the first monoclonal agent to be used in the therapy of cancer. It has been hailed as one of the most important therapeutic developments of the decade. While transient peripheral B cell depletion is common after rituximab therapy, immunoglobulin levels are generally not affected. This is because CD20 is expressed on pre-B and mature B lymphocytes but not on stem cells or plasma cells. Two adult patients with pre-existing primary antibody deficiency who presented with recurrent infections immediately following rituximab use for the treatment of refractory idiopathic thrombocytopenic purpura (ITP) are described. Both were previously treated with various immunosuppressive agents without any notable infective problems. However, a few weeks after treatment with rituximab, these patients presented with clinically significant immunodeficiency requiring intravenous immunoglobulin replacement therapy. This striking temporal relationship between rituximab administration and onset of infections suggests that rituximab has accelerated the presentation of immune deficiency in these patients. Increased vigilance around the use of newer immunomodulatory agents such as rituximab is recommended.
- common variable immune deficiency
Statistics from Altmetric.com
A 43-year-old woman was referred to the immunology clinic in June 2007 with panhypogammaglobulinaemia (IgG: 2.24 g/l; see table 1). She had been diagnosed with idiopathic thrombocytopenic purpura (ITP) at 19 years of age in 1983. Following splenectomy in 1984, she developed two episodes of pneumococcal meningitis (in 1989 and 1991), which were attributed to non-compliance with antibiotic prophylaxis. Isohaemagglutinins were weakly positive at this time. Immunoglobulin levels were normal except for borderline decreased IgA (table 1). She remained relatively well in the period between 1991 and 2006.
Platelet count remained low despite treatment with steroids, azathioprine, cyclophosphamide, α-interferon, ciclosporin and high dose intravenous immunoglobulin. In March 2006 she responded to four doses of rituximab, and her platelet count has since remained normal. Total lymphocyte count prior to starting rituximab was slightly raised at 5.82×109/l (normal 1.5–4×109/l). However, immunoglobulins and lymphocyte subsets were not checked at this time.
Unfortunately she experienced frequent episodes of sinusitis, ear infections and lower respiratory tract infections with Haemophilus influenzae and H parainfluenzae since April 2006. Immunoglobulins were checked in January 2007 and this revealed panhypogammaglobulinaemia. Response to test vaccinations with pneumococcal polysaccharide and conjugated haemophilus vaccine was decreased, with no change and twofold increase respectively in antibody levels post-vaccination. While B cells were present (table 1), there was a marked depletion of the switched memory B cell fraction (CD27+ IgD− cells; data not shown). After ruling out other conditions, she was diagnosed with common variable immune deficiency (CVID) in the immunology clinic and intravenous immunoglobulin replacement therapy was commenced in September 2007. Her infection frequency has significantly reduced since initiation of intravenous immunoglobulin therapy.
A patient in the third decade was referred to the immunology clinic with hypogammaglobulinaemia. ITP had been diagnosed at 5 years of age and was treated with steroids, azathioprine, cyclophosphamide and cyclosporine with little success. Splenectomy was performed in 1993. The first cycle of rituximab in August 2003 induced prolonged remission; following a relapse in June 2006, a second cycle was also successfully administered. Serum immunoglobulins and B lymphocyte numbers were normal in December 1991, but IgG and IgA were low immediately before rituximab therapy was initiated in 2003 (table 1).
There was no history of increased infections in childhood. A severe lower respiratory tract infection in December 2006 initially responded to intravenous antibiotic therapy but there was persistent cough associated with poor appetite. CT chest revealed enlarged mediastinal lymph nodes that showed multiple granulomata with areas of central necrosis on biopsy. Multinuclear giant cells were absent and various special stains performed for microorganisms were negative. Immunostaining showed complete absence of CD20+ B cells. There was no evidence of lymphoma on bone marrow biopsy, but it was noted that B cells were virtually absent while the T cell populations were relatively well preserved. Empirical anti-TB therapy along with high dose steroids for 6 months did not alter the lymphadenopathy.
At presentation to the immunology clinic in August 2007, severe hypogammaglobulinaemia was noted (table 1). Lymphocyte analysis at presentation revealed absent B cells with normal CD4 cells and CD8+ T cells (table 1). There was no response to test vaccination with pneumococcal polysaccharide, conjugated Hib or tetanus toxoid vaccines (no change in antibody levels following vaccination). Intravenous immunoglobulin was started in September 2007.
Over the next few weeks, this patient developed progressive lymphopenia and neutrophilia. There was rapid deterioration with persistent diarrhoea and subsequently fatal fulminant hepatic failure, the underlying cause for which remained unclear. The postmortem examination showed necrotic areas in liver, lungs and the small bowel, suggesting chronic infection, but the organism responsible could not be identified. “Smudge cells” were seen in the bowel and lung, in keeping with adenovirus infection. B cells were absent in lymph nodes. The reason for this patient's profound immunodeficiency remains unresolved.
Initially developed to treat B cell neoplasms, rituximab has been successfully employed for immune modulation in autoimmune conditions including refractory rheumatoid arthritis, systemic lupus erythematosus, ITP and ANCA associated vasculitis. Peripheral B cell depletion is a consequence of rituximab treatment but this is usually transient. Prolonged hypogammaglobulinaemia has been reported with rituximab use in children.1 While persistent panhypogammaglobulinaemia following rituximab treatment has been reported in adults with B cell malignancies,2 it has not been described in autoimmune disease.
Primary immune deficiency is associated with an increased risk of autoimmune conditions. Autoimmune haematological cytopenias, especially ITP, occur in up to 10% of patients with CVID3 and usually precede the diagnosis of immunodeficiency by many years. While both our patients did not show clinical features of immune deficiency, patient 2 had reduced immunoglobulins prior to rituximab treatment. Although patient 1 did not have immunoglobulins checked prior to her treatment with rituximab, a decline in immunoglobulin levels over time is well documented in patients with CVID and thus this is a likely underlying diagnosis in both our patients. However, it is noteworthy that both of them experienced significant clinical deterioration shortly after receiving rituximab, raising the possibility that the pre-existing immunodeficiency was aggravated by this treatment.
It is well recognised that rituximab use results in decreased CD27, CD40, CD80 expressing memory B cells4 and also CD40L and ICOS expression on CD4+ T cells.5 Delayed onset neutropenia and lymphopenia have also been reported after the use of rituximab.6 Further, opportunistic infections such as Pneumocystis jiroveci, JC virus infection have been reported with rituximab use, suggesting an effect on T cell function.7 Impaired humoral immune response with less robust B cell proliferation to simple haptens and recall antigen challenge suggest an adverse effect on memory B cells.8 While patient 2 had no B cells at presentation to immunology, patient 1 had depleted memory B cells. Memory B cell depletion is associated with a subclass of patients with CVID and has been proposed previously as a possible mechanism for rituximab related hypogammaglobulinaemia.9 However, it is impossible to be certain about the importance of this finding given that no comparative data on B cell populations pre- and post-rituximab infusions have been reported so far.
There is, in fact, very little information thus far regarding the effect of rituximab on the clinical course of primary antibody deficiencies such as CVID. In view of this, the current consensus document on rituximab use in rheumatoid arthritis suggests that the drug should be avoided in patients with hypogammaglobulinaemia.10 Given that many CVID patients present with autoimmune cytopenias, it is important to exercise caution while using the drug in this group of patients as well.
As the therapeutic repertoire of rituximab expands, it is conceivable that more groups of patients will be exposed to the drug, leading to the occurrence of hitherto unobserved adverse effects. Adequate monitoring of patients should therefore be considered with rituximab (or indeed any other immunomodulatory therapy). We propose that immunoglobulin levels should be checked in all patients prior to rituximab therapy and further investigations such as lymphocyte subset analysis should be performed where appropriate.
Rituximab can aggravate pre-existing antibody deficiency.
Autoimmunity is relatively common in patients with primary antibody defects.
Immunoglobulins should be checked in all patients prior to rituximab therapy. Alternatives should be considered in patients with low immunoglobulins.
Abnormal results should be discussed with a clinical immunologist.
Competing interests None declared.
Patient consent Obtained.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.