Rituximab treatment results in impaired secondary humoral immune responsiveness

Rituximab, the chimeric CD20 monoclonal antibody (mAb), has become an important treatment modality in B-cell malignancies.1,2 It targets the CD20 antigen, expressed on more than 95% of normal and malignant B cells. In patients with relapsed, low-grade non-Hodgkin lymphoma (NHL), peripheral blood B-cell depletion occurred within 24 to 48 hours after the first infusion of rituximab. Recovery of B-cell counts started 6 to 9 months after the completion of therapy, and normal levels were obtained after 9 to 12 months.1 

The prolonged period of rituximab-induced B-cell depletion might compromise the immune system. However, no studies directly addressing the immune responsiveness of patients treated with rituximab have been published. We investigated the humoral immune responses of patients with low-grade lymphoma to 2 primary antigens (keyhole limpet hemocyanin [KLH] and hepatitis A vaccine [HAV]) and 2 recall antigens (tetanus toxoid [TT] and poliomyelitis vaccine [PV]), both before and after treatment with rituximab.

The present immunization study was part of a phase I/II study evaluating the safety and efficacy of the combination of rituximab (375 mg/m² weekly for 4 weeks) and granulocyte–colony-stimulating factor (G-CSF; 5 μg/kg per day, administered on 3 consecutive days starting 2 days before each infusion).3 Inclusion criteria for the immunization study were those of the phase II study. Both studies were approved by the ethics committee, and all patients gave written informed consent according to the rules of the institute.

One thousand micrograms KLH (Calbiochem, San Diego, CA; dissolved in NaCl as described previously4) was administered subcutaneously. HAV (1440 IE of inactivated hepatitis A virus strain HM-175; Havrix) was obtained from SmithKline Beecham (Rijswijk, The Netherlands) and administered intramuscularly. TT (0.5 mL tetanus vaccine containing 80 IE/mL tetanus toxoid) and inactivated PV (type 1, at least 30 DE [PV1]; type 2, at least 6 DE [PV2]; type 3, at least 24 DE [PV3]) were obtained from the RIVM (Utrecht, The Netherlands) and were administered intramuscularly and subcutaneously, respectively.

Anti-KLH–specific immunoglobulin levels (IgG) and anti-TT antibodies were measured with an enzyme-linked immunosorbent assay (ELISA).4 Anti–hepatitis A virus–specific IgG levels were measured by ELISA (HAVAB 2.0 Quantitative; Abbott Laboratories, Diagnostics, Abbott Park, IL). Titers of neutralizing antibodies against PV1, PV2, and PV3 were measured by the poliovirus-neutralizing antibody test.5 For KLH, a positive response was defined as a ratio of IgG antibodies (after immunization versus before immunization) greater than 1.25.4 Seroconversion to HAV was defined as an increase in HAV titers to more than 20 mIU/mL.6 

Because it was expected that this group of patients with pretreated low-grade lymphoma would have impaired immune responsiveness when compared with healthy volunteers,7 8 it was considered more appropriate to use the patients as their own controls. To avoid boosting effects (ie, turning a primary antigen into a recall antigen and inappropriately amplifying recall responses), in each patient different vaccines were needed for the immunizations performedbefore and after rituximab treatment. Therefore, 2 primary antigens (KLH and HAV) and 2 recall antigens (TT and PV) were used. Each patient simultaneously received one primary and one recall antigen 2 weeks before treatment and crossed over to the other 2 vaccines 4 weeks after treatment. To correct for a possible influence of one of the vaccines on the response to the other vaccines, 4 different immunization schedules had to be used (Table1). Patients were sequentially assigned to 1 of the 4 groups. Before immunization and 2 weeks after immunization, serum samples were collected and stored at −20°C until use. Pretreatment immunizations were performed during the necessary pretreatment work-up. Therefore, the start of rituximab treatment was not delayed in any of the patients.

Antibody responses are expressed as the ratio between the antibody titers 14 days after immunization and those before immunization. To compare the responses to the 2 different primary antigens and recall antigens within one patient, for each vaccine the ratios of antibody responses were ranked. Each of the patients received a rank obtained for his or her response to, for example, TT (independent of the moment of immunization, before or after rituximab treatment), and a rank was obtained for his or her response to PV. The rank a patient obtained before rituximab (eg, for TT) and the rank that this same patient obtained after rituximab (for PV1, PV2, and PV3, or vice versa) formed pairs. The pairs of ranks achieved by all patients were analyzed using the Wilcoxon signed rank test. P <.05 was considered significant.

All patients had progressive disease at the time of first immunization. Median age was 53 years (range, 27-70 years), and the median number of prior chemotherapy regimens was 1 (range, 1-3). Not every patient received all 4 immunizations because of various reasons (Table 1).

In all patients, complete depletion of B cells from the peripheral blood was observed 72 hours after the first infusion. One month after the last rituximab treatment (ie, at time of second immunization), B cells were still absent in all but one patient (patient H). Immunoglobulin levels (IgG, IgA, and IgM) remained stable during treatment and follow-up (data not shown).

Nine patients were immunized with TT and PV. Antibody responses, expressed as the ratio of titers before and after immunization, are shown in Table 2. To allow for comparison of the responses to TT and PV within one patient, ratios to TT and ratios to PV were ranked (Table 2). Responses to the poliovirus subtypes PV1, PV2, and PV3 are separately documented and analyzed in combination with TT. For example, the 3 pairs of ranks obtained by patient A were 9-2, 9-2, and 7-2 for PV1-TT, PV2-TT, and PV3-TT, respectively.

Pairs of ranks obtained by each patient were analyzed using the Wilcoxon signed rank test (n = 9 paired samples). The response to recall antigens after rituximab treatment was significantly lower than the response before treatment (P = .078,P = .013, and P = .012 for TT in combination with PV1, PV2, and PV3, respectively).

In 11 patients the primary immune response could be evaluated—in 5 patients before and in 6 patients afterrituximab treatment (Table 1). None of these patients developed a humoral immune response to either primary antigen before or after rituximab treatment (data not shown). Five patients were found to have high anti-HAV antibody titers, indicating that these patients either had experienced hepatitis A infection or had been immunized against hepatitis A.

In the present study, we have demonstrated that the humoral immune response to recall antigens decreases significantly with rituximab treatment. Surprisingly, although all patients responded to recall antigens, none of the patients responded to the primary antigens before or after rituximab treatment. At time of vaccination, all patients had progressive disease requiring treatment. Disease status at the time of vaccination might explain this strongly impaired immune response to primary antigens observed in our patients.9 

The observed significantly decreased response to recall antigensafter rituximab treatment might be attributed to a decrease in the amount of memory B cells after rituximab treatment. Previous studies showed that although rituximab treatment induces B-cell depletion for 6 to 9 months, this does not lead to a decrease in immunoglobulin levels, possibly because of the presence of long-lived plasma cells.10 However, our observation predicts thatprolonged treatment with rituximab may eventually lead to a decrease in circulating plasma cells because of a decreased replenishment from the pool of memory B cells. This might have consequences for the immunoglobulin levels and, consequently, the incidence of infectious complications during rituximab maintenance therapy.

Recently, rituximab treatment was found to be effective in several antibody-mediated autoimmune diseases,11,12 with some responses ongoing for more than 9 to 14 months.12 We suggest that rituximab, by depleting (memory) B cells, interrupts the ongoing humoral autoimmune response in autoimmune diseases. Although plasma cells surviving rituximab treatment may continue to produce (auto-)antibodies for a certain period of time, deletion of the autoreactive B-cell clone eventually leads to a decrease in antibody production.

Our data might have implications for studies investigating the feasibility of rituximab as maintenance therapy. Furthermore, these findings provide a rationale for the treatment of antibody-mediated autoimmune diseases with rituximab.

We thank Dr T. Out and Prof Dr R.A.W. van Lier for critical reading of the manuscript and for helpful discussions and F. de Wilde for measurements of KLH- and tetanus-specific antibodies by ELISA.