Article Text

Download PDFPDF

Rituximab in patients with chronic inflammatory demyelinating polyradiculoneuropathy: a report of 13 cases and review of the literature
  1. L Benedetti1,
  2. C Briani2,
  3. D Franciotta3,
  4. R Fazio4,
  5. I Paolasso5,
  6. C Comi6,
  7. M Luigetti7,
  8. M Sabatelli7,
  9. F Giannini8,
  10. G L Mancardi9,
  11. A Schenone9,
  12. E Nobile-Orazio10,
  13. D Cocito5
  1. 1Department of Neurology, Osp. S. Andrea, La Spezia, Genova, Italy
  2. 2Department of Neurosciences, University of Padova, Italy
  3. 3Laboratory of Neuroimmunology, IRCCS, Neurological Institute “C. Mondino”, University of Pavia, Pavia, Italy
  4. 4Department of Neurology, IRCCS San Raffaele, Milano, Italy
  5. 5Neurologia I, Department of Neuroscience, A.S.O. San Giovanni Battista, Torino, Italy
  6. 6Interdisciplinary Research Center of Autoimmune Disease (IRCAD), A. Avogadro, University of Eastern Piedmont, Novara, Italy
  7. 7Department of Neuroscience, Catholic University, Roma, Italy
  8. 8Department of Neuroscience, University of Siena, Siena, Italy
  9. 9Department of Neuroscience, Ophtalmology and Genetics, Genova, Italy
  10. 10Department Neurological Sciences, Milan University, IRCCS Humanitas Clinical Institute, Rozzano, Milano, Italy
  1. Correspondence to Luana Benedetti, Department of Neurology, Osp. S. Andrea, Via Vittorio Veneto 197, 19100 La Spezia, Italy; luanabenedetti{at}


Background A few case reports have shown controversial results of rituximab efficacy in patients with chronic inflammatory demyelinating polyradiculoneuropathy (CIDP).

Objective To analyse the efficacy of rituximab in a large CIDP cohort.

Methods A retrospective, observational and multicentre study on the use of rituximab in CIDP. 13 Italian CIDP patients were treated with rituximab after the partial or complete lack of efficacy of conventional therapies. Eight patients had co-occurring haematological diseases. Patients who improved by at least two points in standard clinical scales, or who reduced or discontinued the pre-rituximab therapies, were considered as responders.

Results Nine patients (seven with haematological diseases) responded to rituximab: six of them, who were non-responders to conventional therapies, improved clinically, and the other three maintained the improvement that they usually achieved with intravenous immunoglobulin or plasma exchange. Significantly associated with shorter disease duration, rituximab responses started after a median period of 2.0 months (range, 1–6) and lasted for a median period of 1 year (range, 1–5).

Conclusions Rituximab seems to be a promising therapeutic choice when it targets both CIDP and co-occurring haematological diseases. Timely post-onset administration of rituximab seems to be associated with better responses.

  • Rituximab
  • CIDP
  • polyradiculoneuropathy
  • neuroimmunology
  • neuropathy

Statistics from

Request Permissions

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.

Chronic inflammatory demyelinating polyradiculoneuropathy (CIDP) is a relapsing–remitting or chronic progressive disease of the peripheral nervous system.1 The aetiology of CIDP is unknown, but immunologic mechanisms, which involve both B- and T-cell-mediated responses, presumably play a central role in the pathogenesis of the damage to peripheral myelin. Said mechanisms offer a rationale for immunoactive therapies, such as intravenous immunoglobulin (IVIg), corticosteroids and plasma exchange (PE), but inadequate responses to these treatments lead to the use of immunosuppressant drugs.2 Rituximab, a monoclonal anti-CD20 antibody, has recently been used for the treatment of anti myelin-associated glycoprotein (MAG) antibody-associated polyneuropathy,3 4 another demyelinating neuropathy with immune-mediated pathogenesis. In CIDP, the drug proved to be efficacious in single cases,5–10 whereas in a prospective pilot trial (two cases), it afforded no reduction in IVIg dosage.11

We present 13 patients with CIDP who were treated with rituximab in eight Italian centres after unsatisfactory responses to standard therapies.


We reviewed nationally registered12 CIDP patients diagnosed in accordance with the EFNS/PNS criteria.13 Two of these cases had been previously reported.9 10 Table 1 summarises the demographic and clinical characteristics of the selected 13 patients (eight men, five women; age range, 30–72 years, mean age, 55 years; mean neuropathy duration, 7 years). Routine laboratory tests, including glycaemia, serum thyroid hormones, B12 vitamin, tumour markers and hepatitis C serology, excluded other causes of neuropathy. Anti-MAG and anti-sulfatide antibody tests were negative. Clinical impairment and disability before rituximab were highly variable: mean value of 48 (range, 36–56) on the Medical Research Council (MRC) sum score14 and of 5 (range, 3–8) on the Inflammatory Neuropathy Cause and Treatment (INCAT) disability score.15 Disease course was relapsing–remitting in nine patients and chronic-progressive in four. In eight cases, CIDP co-occurred with haematological diseases. Rituximab was proposed for the following reasons: seven patients were refractory to conventional immunosuppressive therapies, and quality of life was compromised in six patients by frequent recourse to PE/IVIg cycles (mean between-cycle interval, 44 days; range, 14–120); additionally, haematological diseases worsened in four cases. All patients gave written informed consent to rituximab treatment.

Table 1

Demographic and clinical characteristics of the patients with CIDP

Rituximab dosage was 375 mg m−2 IV, weekly for four consecutive weeks, for 12 patients; one patient, who had severe Waldenstrom macroglobulinemia, received 1000 mg IV every 6 months for 2 years (four infusions).

Neurological assessments were performed at baseline and every month for the first 6 months, then every 3 months (MRC sum score for muscle strength, and INCAT scales for arm and leg disability), with ≥1 year follow-up. CD19+ B cells, serum immunoglobulin levels and platelet counts were checked regularly.

Before- and after-treatment electrophysiological data were available for six patients. Our analysis was based on ulnar nerve motor conduction velocity (MCV) because peroneal nerve MCV was not recordable in three patients. We considered ≥10% changes in MCV values as indicators of therapeutic efficacy.

Responders were defined as those patients who improved by at least 2 points on each of the two clinical scales, or who maintained the degrees of improvement obtained with IVIg/PE cycles, without further cycles of IVIg/PE over the follow-up.

Our review of the literature included patients with pure CIDP and excluded CIDP associated with anti-MAG/sulfatide antibodies.


Table 1 shows the clinical data. Nine of 13 (69%) patients responded to rituximab. Clinical improvements occurred at a median time of 2.0 months after the end of rituximab cycles and lasted for up to 1 year. Among the seven patients who were refractory to conventional therapies, one worsened under rituximab, whereas the other six improved by at least 4 points in the MRC score, and by 2 points in the INCAT scales, with a reduction of disability in quality-of-life-related daily activities, such as deambulation, handling knife and fork, washing hair and doing–undoing zips. Of the six patients for whom rituximab replaced IVIg/PE cycles, three were responders: two patients ceased IVIg/PE and maintained the improvement achieved with these treatments before rituximab (benefits started at a median time of 2.0 months after rituximab, and lasted for ≥1 year); one patient stopped IVIg and clinically improved.10 The three non-responders recorded worsened clinical symptoms after a 25% reduction in IVIg dosage (table 1). No responders required any retreatment throughout the follow-up period. We tried to correlate the response to rituximab with clinical variables, namely, disease duration, course of disease and association with haematological diseases. Mean disease durations in responders (4.3 years) were significantly shorter than in non-responders (12.7 years; p=0.025, Mann–Whitney U test). Clinical improvement and course of the disease did not correlate. Seven of the nine responders and one of the four non-responders had haematological diseases, but the between-group difference was not significant (p=0.12, Fisher exact test).

Blood CD19+ B cells were undetectable in all 13 patients at month 1 post-rituximab cycles; they reappeared at month 6 and returned to pre-treatment values at months 9–12. Serum IgM levels were approximately 50% lower at 1–3 months post-treatment, and were still lower at month 12 than at baseline in patients with IgM monoclonal gammopathy of undetermined significance (MGUS) or with Waldenstrom macroglobulinemia. In contrast, serum IgA levels were unchanged in the IgA monoclonal gammopathy of undetermined significance patient. Platelet counts improved in the patient with idiopathic thrombocytopenic purpura.10 Rituximab therapy respectively circumvented autologous peripheral blood stem cell transplantation and splenectomy in Waldenstrom macroglobulinemia and idiopathic thrombocytopenic purpura patients.

Electrophysiological data showed ≥10% ulnar nerve MCV improvement in five responders and worsening in one non-responder (data not shown).

No major adverse effects were recorded. During the first rituximab infusion, one patient complained of flu-like symptoms, and another showed a mild skin allergy, which responded to corticosteroids.


B-cell-depleting antibody rituximab has current FDA approval for the treatment of non-Hodgkin's lymphoma and rheumatoid arthritis, but its off-label use extends to several autoimmune diseases. Because rituximab could target the immunological mechanisms deemed to be involved in CIDP pathogenesis, the drug qualifies for CIDP patients who do not respond to standard therapies. The few relevant reports show controversial results in terms of drug efficacy. One study on two CIDP cases reports no benefit in the primary end point, namely, a 25% reduction in IVIg dosage 1 year after rituximab therapy versus IVIg dosage in the year before rituximab.11 In contrast, four case reports describe clinical improvement in four rituximab-treated CIDP patients (table 2).5–8 Specifically, rituximab induced the complete disappearance of a gastric lymphoma and long-lasting improvements in neurological symptoms in a CIDP case,5 the remission of both neurological symptoms and haematological abnormalities for 17 months in a patient with CIDP and Evans syndrome,6 and the stabilisation of CIDP course for 11 months in a patient with diabetes mellitus.7 Combined with chemotherapy, rituximab proved successful in a case of CIDP associated with non-Hodgkin's lymphoma.8

Table 2

Results of published studies on rituximab therapy in patients with CIDP

To date, our cohort is the largest to be used to evaluate rituximab in CIDP. Our data show that rituximab induced a sustained remission of neurological symptoms in about 70% of patients with CIDP. Although we do not advocate a syncretic approach, we note that this percentage is similar to that obtainable by pooling the six CIDP cases previously reported in the literature. Comparably with what was reported for anti-MAG neuropathy,16 rituximab efficacy lasted 1 year, or longer, both in our series and in those previously reported in the literature. As a result, patients either improved clinically or were spared periodic IVIg/PE cycles. Regarding the responders, 3 of the 6 previously reported cases, and 7 of our 13 patients, had associated haematological diseases, and rituximab was effective for both pathologies. In all the given series, the drug was less effective for idiopathic CIDP than for CIDP and haematological diseases. These findings suggest that, when associated with haematological diseases, CIDP may have different pathogenic mechanisms from those underlying isolated CIDP. Although CIDP is considered an autoimmune disease, the target antigen and the precise roles of humoral and cell-mediated immunity remain unknown.17 Rituximab's greater efficacy in CIDP associated with haematological diseases could imply that B cells, whether as antibody, or cytokine-producing cells, play a predominant role in the pathogenesis of these CIDP types. Responses to rituximab in demyelinating lymphoma-associated neuropathies18 could support this view.

As a whole, the percentages of responders reported in the literature and in our series are similar to those reported for anti-MAG polyneuropathy,3 4 but the intervals between rituximab cessation and clinical improvement differed: 2–3 months in CIDP versus 6–8 months in anti-MAG polyneuropathy.3 4 It is likely that the longer intervals in anti-MAG polyneuropathy depend on the fact that rituximab does not target plasma cells, namely, the cells directly involved in disease pathogenesis, and that benefits thus occur only after a sustained depletion of their precursors, the CD20+ B cells.

Both for our patients with CIDP and for those previously reported, mean disease duration in responders was shorter than in non-responders. The phenomenon might derive from cumulative axonal damage in patients with greater disease durations, and from the possible downregulation of CD20 molecule as a result of aggressive and prolonged pre-rituximab immunosuppressive therapies; such downregulation could deny targets and hence efficacy to rituximab.

In conclusion, although obtained in a relatively small number of patients, our data suggest that rituximab therapy for refractory CIDP may be promising and free from major adverse events, particularly if the disease has short duration and is associated to haematological diseases that are potentially responsive to the drug.



  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.