Author Affiliations: Department of Neurology (Drs Monson, Cravens, Frohman, Hawker, and Racke) and Center for Immunology (Drs Monson and Racke), The University of Texas Southwestern Medical Center, Dallas.
Rituximab, an anti-CD20 monoclonal antibody that depletes CD20+ B cells, has demonstrated efficacy in peripheral neurological diseases. Whether this efficacy can be translated to neurological diseases of the central nervous system with possible autoimmune B-cell involvement remains unknown.
To determine the effect of rituximab on cerebrospinal fluid B cells in patients with multiple sclerosis.
Four patients with primary progressive multiple sclerosis were treated with rituximab. Cerebrospinal fluid and peripheral blood B-cell subsets were identified by flow cytometry from each patient before and after rituximab treatment.
The B cells in cerebrospinal fluid were not as effectively depleted as their peripheral blood counterparts. Rituximab treatment temporarily suppressed the activation state of B cells in cerebrospinal fluid. The residual B cells underwent expansion after rituximab treatment.
The effect(s) of rituximab on the cerebrospinal fluid B-cell compartment is limited in comparison with the effect(s) on the B cells in the periphery, but this finding will need to be confirmed in a larger group of MS patients.
Multiple sclerosis (MS) is an inflammatory, demyelinating disease of the central nervous system (CNS) that likely involves an autoimmune response directed against myelin-associated antigens. Evidence exists that possibly implicates B cells in the development and perpetuation of MS.1- 11 Thus, efforts to address whether B-cell–depleting therapies may have a place in therapeutic regimens for patients with MS (hereafter referred to as MS patients) have begun.
Rituximab (Genentech, Inc, San Francisco, Calif; IDEC Pharmaceuticals Corp, San Diego, Calif) is a chimeric murine/human IgG1 κ monoclonal antibody. It targets CD20, a transmembrane phosphoprotein expressed on most B cells, except for plasma B cells.12- 14 Rituximab depletes B cells by binding to the CD20 molecule and thereby initiating complement-dependent cytolysis or antibody-dependent cell-mediated cytotoxicity. Rituximab has clear clinical benefit for patients with neoplastic B-cell–mediated diseases15 and treatment of anti-GM1 polyneuropathy.16 Whether this efficacy of rituximab in peripheral neurological diseases with autoimmune B-cell involvement can be translated to neurological diseases of the CNS with possible autoimmune B-cell involvement remains unknown.
The objective of this study was to determine the effect of rituximab on B cells in the peripheral blood (PB) and cerebrospinal fluid (CSF) of MS patients.
Five patients were recruited for rituximab therapy at the University of Texas Southwestern Medical Center at Dallas on the basis of previous response to plasmapheresis and continued disease progression. All patients signed a consent form approved by the institutional review board. Their clinical presentation is summarized in Table 1.
All 5 patients were treated with a weekly dose of intravenous rituximab (375 mg/m2) for 4 consecutive weeks.
Four of the 5 patients consented to donate a PB sample before and after completion of rituximab therapy. Approximately 10 mL of heparinized PB was collected at each sampling. Mononuclear cells from the PB were isolated by Ficoll-Hypaque sedimentation as described.17 Two of these patients also consented to donate CSF at the same time points as the blood samplings. The CSF cells were isolated by centrifugation of the CSF. The PB and CSF cells were stained with a phycoerythrin-labeled anti-CD19 monoclonal antibody and other markers that identify B-cell subsets (Table 2). The cells were collected using a flow cytometer (FACStar Plus; Becton Dickinson, San Jose, Calif) and the CD19+ B-cell frequencies and light-scatter characteristics analyzed using CellQuest software (Becton Dickinson). We used the χ2 test to compare the frequencies of CD19+ B cells (before and after rituximab treatment) in the PB and CSF.
The PB and CSF samples were collected from MS patients M116 and M154 before and after rituximab treatment. Only PB was sampled from MS patient M602 and a single patient with progressive myelopathy (patient M191). The samples were stained for CD19, a B-cell marker whose expression parallels that of CD20.18
The CD19+ B cells were effectively depleted in the periphery after rituximab therapy in all 4 patients (Table 3), as expected. Patient M116 had 18.43% CD19+ PB B cells before rituximab therapy, which decreased to 0.14% 1 month after rituximab therapy (Table 3). The other 3 patients demonstrated similar significant decreases in CD19+ PB B-cell frequencies after rituximab therapy. Patient M116 had a second course of rituximab, and 2 months after this second course, the CD19+ B-cell population was not depleted (4.42% vs 7.91%). This resistance to depletion on retreatment with rituximab has been reported previously.19
The time frame in which CD19+ PB B-cell frequencies recovered varied between the patients (Table 3). The CD19+ B-cell frequency in patient M116 had increased to 4.42% 18 months after rituximab treatment, but did not fully recover to the pretreatment B-cell frequency of 18.43%. Patient M191 also did not recover to the pretreatment B-cell frequency at 14 months after rituximab treatment. In contrast, patient M154 had a larger posttreatment frequency than that observed before rituximab treatment. Thus, there will likely be individual variability in the time frame and extent of CD19+ PB B-cell frequency recovery after rituximab therapy.
Only 1 of the patients demonstrated a significant decrease in CD19+ CSF B-cell frequency after rituximab therapy (1.03% vs 0.63% in patient M116; P < .001) (Table 4). In contrast, the CD19+ CSF B-cell frequency in the CSF of patient M154 after rituximab therapy increased (1.74% vs 3.44%; P = .14). Thus, depletion of CSF B-cells after rituximab therapy cannot be achieved as consistently as in the periphery.
The CD19+ B-cell frequency in patient M116 at 18 and 20 months after rituximab therapy was significantly greater than that in the pretreatment sampling, whereas the CD19+ B-cell frequency of patient M154 at 14 months after rituximab therapy remained the same as the posttreatment sampling results 2 months after rituximab therapy (Table 4). Thus, recovery of the CD19+ CSF B-cell frequencies after rituximab therapy is likely to vary among individuals. To provide an accurate time frame in which rituximab depletes B cells most effectively in the CSF, a thorough kinetic analysis of B-cell frequencies in a larger patient population would be required.
There was no decrease in the CD19+ B-cell frequency 2 months after patient M116 had received a second course of rituximab (Tables 3 and 4). To discern what B-cell subtypes survived the second rituximab course, we stained the PB and CSF with CD19, CD38, IgM, and IgD. Recently activated B cells express CD19/CD20 very brightly and are CD38−IgM+IgD−, whereas advanced memory (postswitch and preplasma B cells) dimly express CD19/CD20 and are CD38+IgM−IgD− (Table 2).
The CD19+ PB B-cell population of patient M116 could be subdivided into 2 separate populations: CD19bright and CD19dim. Most of the CD19+ B cells that remained after rituximab therapy were CD19dim (87%) (Figure 1A) and IgM−IgD− (Figure 1B) and did not express CD38 (see rectangular gate in Figure 1C). Thus, the residual B cells were postswitch B cells that had not yet differentiated to a plasma B-cell phenotype. There was also a population of cells that were CD19bright CD38+IgM+IgD+ (see oval gate in Figure 1C), which are likely immature B cells (Table 1) that have recently emerged from the bone marrow to begin repopulating the peripheral B-cell pool.
Representation of B cells in patient M116 after retreatment with rituximab. Peripheral blood leukocytes (PBL) (A-C) and cerebrospinal fluid (CSF) (D-F) B cells were stained for CD19, IgM, IgD, and CD38. Panels A and D are histograms, demonstrating the position of CD19dim and CD19bright populations within the lymphocyte population. The y-axis is cell count and x-axis is log fluorescent intensity. Panels B and E illustrate the IgM and IgD expression of the CD19+ B cells. Panels C and F illustrate the coexpression of CD38 on the CD19+ B cells. The rectangular gate indicates the CD19dim B-cell population discussed in the “Analysis of the Residual B-Cell Population in Patient M116” subsection of the “Results” section. The oval gate in panel C indicates the CD19bright CD38bright population. The square gate in panels C and F indicates cells negative for CD19 and CD38. In panels B, C, E, and F, the axis values for the indicated markers are given in fluorescence intensity.
Approximately 80% of the CD19dim CSF B cells were IgM−IgD−CD38+ (Figure 1E and rectangular gate in Figure 1F), which stratified them as plasma B cells. The other approximately 20% of the CD19dim B cells were IgM−IgD−CD38−, and had thus not differentiated to a plasma B-cell phenotype. The CD19−CD38+ population would include plasma B cells as well, but this population also contains T cells, so an accurate analysis of the plasma B cells in this population could not be ascertained.
Activated B cells (such as plasma B cells) are generally larger and more granular than their naive counterparts.20- 23 This phenomenon can be observed by flow cytometric light-scatter characteristics, looking at the placement of B cells in the forward scatter vs side scatter (FSC/SSC) plot, which separates cells by size and granularity, respectively. To identify B-cell subpopulations after rituximab therapy, we reanalyzed the B-cell compartments in the CSF and PB from the time point before and 2 time points after rituximab therapy in patients M116 and M154.
There were 2 populations of CD19+ PB B cells (bright and dim) in patient M116 (Figure 1A), so we analyzed them separately for their light-scatter profiles. Most of the B cells in the PB of patient M116 before rituximab therapy were CD19bright (78.8%) (Figure 1A), and were mostly contained in the typical lymphocyte gate (ie, low FSC/SSC) (red dots) (Figure 2A). The CD19dim B cells (green dots) (Figure 2A) were also mostly contained in the typical lymphocyte gate (ie, low FSC/SSC). This indicated that the CD19bright and CD19dim B cells were largely in a resting state.
B-cell activation states in the peripheral blood leukocytes (PBL) and cerebrospinal fluid (CSF) of patient M116. Peripheral blood (A-C) and CSF (D-F) were stained for CD19. Red dots indicate the location of CD19bright B cells; green dots, the location of CD19dim B cells. Panels A and D indicate the forward scatter vs side scatter (FSC/SSC) designations for the PB and CSF cells before rituximab therapy. Panels B and E indicate the FSC/SSC designations for the PB and CSF cells 1 month after rituximab therapy. Parts C and F indicate the FSC/SSC designations for the PB and CSF cells 18 months after rituximab therapy. Axis values are given in fluorescence intensity. The y-axis scale for panels C and F is linear, rather than logarithmic.
One month after rituximab therapy, the CD19bright B cells (red dots) were largely depleted (Figure 2B), as expected. However, the CD19dim B cells (green dots) remained and maintained their resting state (low FSC/SSC). Eighteeen months later, the CD19bright B cells (red dots) had repopulated the B-cell pool (Figure 2C) and were in a resting state, as had been observed before rituximab therapy. Interestingly, 18 months after rituximab therapy, the CD19dim B-cell population (green dots) had not only expanded (3.9% of CD19+ B cells before vs 96.4% of CD19+ B cells after rituximab treatment), but had increased their FSC (ie, were larger in size) to a position above the lymphocyte region (Figure 2C). This light-scatter profile is consistent with a more activated state of the CD19dim B cells than that observed before rituximab therapy.
The peripheral B cells from patient M154 behaved in a similar manner (Figure 3), with the CD19bright B cells (red dots) located in the typical lymphocyte region before rituximab treatment (Figure 3B), followed by depletion of the CD19bright B cells 2 months after rituximab therapy (Figure 3C), and finally, repopulation of the typical lymphocyte region 14 months after rituximab therapy with CD19bright B cells (Figure 3D). The CD19dim B cells (green dots) were also present before rituximab therapy (Figure 3B), but remained largely in the typical lymphocyte region (low FSC/SSC), and remained so 2 months after rituximab treatment (Figure 3C). Fourteen months later, the CD19dim B-cell population had expanded (20.1% of CD19+ B cells before vs 91.9% of CD19+ B cells after rituximab therapy) and were in a more activated state.
B-cell activation states in the peripheral blood leukocytes (PBL) and cerebrospinal fluid (CSF) of patient M154. Peripheral blood (A-D) and CSF (E-H) were stained for CD19. Red dots indicate the location of CD19bright B cells; green dots, the location of CD19dim B cells (B-D and F-H). Panels A and E are histograms indicating the CD19dim and CD19bright designations. Panels B and F indicate the forward scatter vs side scatter (FSC/SSC) designations for the PB and CSF cells before receiving rituximab. Panels C and G indicate the FSC/SSC designations for the PB and CSF cells 2 months after rituximab therapy. Parts D and H indicate the FSC/SSC designations for the PB and CSF cells 14 months after rituximab therapy. The y-axis scale for panels D and H is linear, rather than logarithmic.
We have observed that most of the CSF B cells are not located within the typical lymphocyte region, but instead have increased light-scatter profiles to reflect a more activated phenotype. The CSF B cells from patient M116 are a typical example of that, having few CD19bright B cells (red dots) relative to PB (compare Figure 2A and Figure 2D), and most CD19dim B cells (green dots) were located above the resting lymphocyte region (Figure 2D) before rituximab treatment. One month after rituximab therapy, most of the CSF B cells were CD19dim (green dots), and had relocated to the resting lymphocyte region (Figure 2E). Twenty months after rituximab therapy, some CD19bright CSF B cells (red dots) were observed within the typical lymphocyte region. However, the CD19dim B cells (green dots) had increased light-scatter profiles, indicating a heightened state of activation (Figure 2F) similar to their light-scatter profile before rituximab therapy. In addition, the CD19dim B cells (green dots) had expanded in comparison with the population before rituximab treatment (55% of CD19+ B cells before vs 77% CD19+ B cells after rituximab therapy).
Most of the CSF B cells from patient M154 before rituximab therapy were CD19dim (77.8%), and had a high light-scatter profile (Figure 3F). We had observed the same profile of CSF B cells from patient M116, and consider this pattern typical for CSF B cells, which are more activated than their peripheral counterparts. Two months after rituximab therapy, some CD19bright CSF B cells (red dots) were still detectable (Figure 3G), but the CD19dim B cells (green dots) remained and, interestingly, displayed a low light-scatter profile, indicative of a resting state (Figure 3G). However, 14 months after rituximab therapy, some CD19bright CSF B cells (red dots) had repopulated. The CD19dim B cells (green dots) had expanded significantly (20.2% of CD19+ B cells before vs 94.4% of CD19+ B cells after rituximab therapy) and had assumed their activated state (high light-scatter profile) (compare Figure 3H and Figure 3F).
It is evident that CD19+ B cells were depleted in the periphery within 1 to 2 months after rituximab therapy in the PB of all 3 patients. Even patient M191 maintained a depleted CD19+ B-cell population in the periphery 14 months after rituximab therapy. However, the PB CD19+ B-cell population was partially or fully reconstituted between 14 and 18 months after rituximab therapy, indicating that the inhibitory effect of rituximab on B cells no longer exists.
The effect of rituximab on the B-cell pool within the CNS tissue is not as well-defined, but it becomes an issue when considering rituximab as a therapeutic agent in B-cell–mediated diseases of the CNS. We have examined the effect of rituximab on CD19+ CSF B cells in 2 MS patients, which likely represents a subset of B cells that populate the CNS tissue. The CD19+ CSF B cells showed a mild decrease 1 month after rituximab therapy, followed by a significant expansion 18 and 20 months after rituximab therapy that were 5- to 9-fold the CD19+ B-cell frequency before rituximab therapy. However, the CD19+ CSF B cells in patient M154 were not affected by rituximab. This was in direct contrast to what had been routinely observed in PB B cells in response to rituximab, and prompts the question of whether this lack of CSF B-cell depletion has any repercussions on the potential efficacy of rituximab treatment of MS.
There are at least 2 possible scenarios that could account for the lack of CSF B-cell depletion. First, the concentration of rituximab that reaches the CSF can be nearly 10-fold less than the concentration readily achieved in the PB.24 Thus, if therapeutic levels of rituximab cannot reach the CSF, depletion of CSF B cells should not be expected. However, the observation that only 10% or less of the rituximab concentration that is reached in the PB is obtainable in the CSF is based on patients whose blood-brain barrier is not compromised, as it is in most MS patients. Thus, perhaps higher CSF rituximab concentrations can be achieved in MS patients whose blood-brain barrier has been compromised. This concept has not been rigorously tested. Second, rituximab is more effective at depleting B cells that have high expression of CD20, such as naïve B cells. However, most CSF B cells are advanced memory or plasma B cells, which express CD19/20 dimly,18 rendering them resistant to rituximab.
Rituximab also appears to affect CD19dim B-cell activation states, because the initial activated phenotype of the CD19dim CSF B cells was suppressed after rituximab therapy. However, these CD19dim CSF B cells had reestablished their activation phenotype 14 and 20 months after rituximab therapy. It is unclear how the suppression of this activation state shortly after treatment occurs. It is possible that rituximab itself asserts some form of suppression on residual B cells, or that some other changes in the CSF environment in response to rituximab causes CD19dim B cells to enter a resting state. What effect this resting state of the residual CD19dim CSF B cells has on disease status remains unknown, although it is compelling to predict that this suppression of the residual B cells by rituximab (directly or indirectly) will translate to a clinical improvement for the patients. However, it is clear that at some point (presumably when rituximab levels drop below a certain concentration), the surviving CSF B cells resume their activated phenotype.
This activation suppression phenomenon we had observed with the CSF CD19dim B cells was not observed in the peripheral CD19dim B-cell population because these cells are typically resting. However, the CD19dim B cells that survived in the periphery had acquired an activated state after rituximab therapy. Whether this activation state of CD19dim peripheral B cells parallels that of the CD19dim CSF B cells will need to be investigated, as well as what effect this might have on the reconstituted B-cell pool.
We also observed an expansion of PB and CSF CD19dim B cells after rituximab therapy. Expansion of residual CD19dim B cells may be based on the theory of space threshold, which suggests that there is a finite amount of space where cells are allowed to populate, and once the CD19bright B cells were depleted, the CD19dim B cells took advantage of the extra space and expanded as observed in these 2 patients. Alternatively, the chemokine milieu may have shifted during rituximab therapy, resulting in an influx of CD19dim B cells from other compartments (such as the bone marrow), and subsequent expansion. Third, perhaps residual CD19bright B cells that survived rituximab treatment were able to rapidly become activated and differentiate. However, for this hypothesis to apply to the CSF B cells, the concept that B cells can differentiate and become activated outside the context of germinal centers must be substantiated.
We have observed typical depletion of PB B cells after rituximab therapy in patients with primary progressive MS. Rituximab does not deplete CSF B cells efficiently because most CSF B cells are highly activated advanced memory or plasma B cells that express little or no CD20. Recovery of the initial B-cell frequencies will likely vary among individuals. Alterations in activation and expansion status of PB and CSF B cells in response to rituximab requires further study to resolve the mechanism(s) underlying these phenomena.
Correspondence: Nancy L. Monson, PhD, Department of Neurology and Center for Immunology, The University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd, Dallas, TX 75390 (email@example.com).
Accepted for Publication: June 24, 2004.
Author Contributions:Study concept and design: Monson, Frohman, and Racke. Acquisition of data: Monson, Frohman, Hawker, and Racke. Analysis and interpretation of data: Monson, Cravens, Frohman, and Racke. Drafting of the manuscript: Monson and Racke. Critical revision of the manuscript for important intellectual content: Monson, Cravens, Frohman, Hawker, and Racke. Statistical analysis: Monson. Obtained funding: Monson and Racke. Administrative, technical, and material support: Monson, Cravens, Frohman, Hawker, and Racke. Study supervision: Monson and Racke.
Funding/Support: This study was supported by grants K24 NS44250 (Dr Racke) and RO1 NS40993 (Dr Monson) from the National Institutes of Health, Bethesda, Md; grant RG 3267-A-1 from the National Multiple Sclerosis Society, New York, NY (Dr Monson); and by the Yellow Rose MS Foundation, Dallas, Tex (Drs Monson and Racke). Dr Monson is a Wadsworth Foundation Young Investigator.
Acknowledgment: We thank Gwen Vaughan and Andrea Jowdry for their critical review of this manuscript.
Thank you for submitting a comment on this article. It will be reviewed by JAMA Neurology editors. You will be notified when your comment has been published. Comments should not exceed 500 words of text and 10 references.
Do not submit personal medical questions or information that could identify a specific patient, questions about a particular case, or general inquiries to an author. Only content that has not been published, posted, or submitted elsewhere should be submitted. By submitting this Comment, you and any coauthors transfer copyright to the journal if your Comment is posted.
* = Required Field
Disclosure of Any Conflicts of Interest*
Indicate all relevant conflicts of interest of each author below, including all relevant financial interests, activities, and relationships within the past 3 years including, but not limited to, employment, affiliation, grants or funding, consultancies, honoraria or payment, speakers’ bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued. If all authors have none, check "No potential conflicts or relevant financial interests" in the box below. Please also indicate any funding received in support of this work. The information will be posted with your response.
Register and get free email Table of Contents alerts, saved searches, PowerPoint downloads, CME quizzes, and more
Subscribe for full-text access to content from 1998 forward and a host of useful features
Activate your current subscription (AMA members and current subscribers)
Purchase Online Access to this article for 24 hours
Some tools below are only available to our subscribers or users with an online account.
Download citation file:
Web of Science® Times Cited: 99
Customize your page view by dragging & repositioning the boxes below.
and access these and other features:
Enter your username and email address. We'll send you a link to reset your password.
Enter your username and email address. We'll send instructions on how to reset your password to the email address we have on record.
Athens and Shibboleth are access management services that provide single sign-on to protected resources. They replace the multiple user names and passwords necessary to access subscription-based content with a single user name and password that can be entered once per session. It operates independently of a user's location or IP address. If your institution uses Athens or Shibboleth authentication, please contact your site administrator to receive your user name and password.