Pulse cyclophosphamide therapy in chronic inflammatory demyelinating polyneuropathy

Chronic inflammatory demyelinating polyneuropathy (CIDP) is considered to be immune-mediated. The peripheral nerves of CIDP patients contain inflammatory mononuclear infiltrates, patchy demyelination, and immunoglobulin deposition. In addition, there is an association of CIDP with certain human leukocyte antigen (HLA) types, suggesting that these patients are genetically predisposed to developing a chronic immune response to peripheral nerve antigens. CIDP is associated with significant morbidity and mortality; up to 25% of these patients, if untreated, become wheelchair bound or bedridden and 10% die of complications.¹

The best treatment for this disease is not defined but controlled studies suggest that immunomodulatory therapy promotes clinical recovery. Corticosteroids² and plasma exchange³ have been shown to have beneficial effects on some manifestations of CIDP. Intravenous gammaglobulin (IVγ) has been shown to be superior to placebo⁴ but another prospective double-blind placebo-controlled study showed improvement in only 27% of CIDP patients.⁵ Although the ease of administration and relative lack of side effects make IVγ an attractive form of therapy for patients who respond, the transient nature of this response, reflecting the half-life of the circulating protein, requires retreatment every 3 to 8 weeks, contributing to expense and patient inconvenience.

More definitive immunosuppressant therapies in combination with corticosteroids or plasma exchange have been attempted with azathioprine⁶ and cyclosporin A (Sandimmune, East Hanover, NJ).⁷ In a study by Dyck et al., azathioprine (2 mg/kg) did not enhance the clinical response over corticosteroids alone.⁶ Treatment of CIDP patients with cyclosporin A was also temporarily effective; however, the requirement for ongoing treatment and high occurrence of side effects (including potential for renal toxicity) has limited its use.⁷ An open, prospective, uncontrolled study concluded that short-term treatment with oral pulse cyclophosphamide and prednisone may have a long-term favorable effect in patients with inflammatory demyelinating polyneuropathy associated with monoclonal gammopathy of undetermined significance (MGUS).⁸ Prolonged oral cyclophosphamide therapy is complicated by poor patient compliance, bone marrow toxicity, bladder irritation, and tumorigenicity potential.⁹ In this article, we present our experience with pulse intravenous cyclophosphamide (IVCY) in 15 patients with CIDP.

Patients. Fifteen patients (11 women and 4 men) with CIDP were treated with IVCY over an 11-year period between 1985 and 1996. They ranged in age from 26 to 64 years (mean, 45 years) at onset of CIDP. All patients had previous therapy with various combinations of IVγ, plasma exchange, and prednisone. Ten patients had 1 to 2 courses of plasma exchange in various combinations with prednisone, azathioprine, or IVγ. Eleven patients received a 5-day course of IVγ before pulse IVCY therapy. The average time from initial treatment until IVCY was 15.6 months (range, 1 to 42) for plasma exchange and 6.2 months for IVγ (range, 1 to 42). Patients were followed up for a minimum of 6 and a maximum of 104 months (mean, 36.3). Patients in this study fulfilled the established criteria for CIDP. They had the clinical features of a polyneuropathy with progressive weakness for 2 to 264 months (mean, 37). Our patients' mean median motor nerve conduction velocity was 34 m/second. Conduction block of the median nerve occurred in five patients and prolonged median distal latency in six. CSF protein ranged from 52 to 591 (mg%; mean, 215). Exclusion criteria included diabetes mellitus, vasculitis, HIV, multifocal motor neuropathy, paraproteinemia (anti-myelin-associated glycoprotein and MGUS), pure sensory neuropathies, and immune complex disorders such as mononeuritis multiplex.

Methods. Clinical assessment. The disability of the patients was assessed using the modified Rankin scale.¹⁰ The modified Rankin scale was chosen to assess disability and functional status:

• 0 = asymptomatic

• 1 = nondisabling symptoms that do not interfere with daily activities

• 2 = slight disability, unable to carry out all activities but still able to look after themselves

• 3 = moderate disability, requiring assistance with some activities but able to walk without assistance

• 4 = moderately severe disability, unable to walk without assistance and unable to attend to own bodily needs without assistance

• 5 = severe disability, totally dependent, requiring constant nursing care

Improvement was defined as at least one point decrease on the Rankin scale, and the time to peak sustained improvement after first infusion was noted. Remission was defined as achieving a level 0 to 1 on the modified Rankin scale for 6 months or more off any immune-modulating treatment.

Treatment protocol. Infusion protocols were adapted from the lupus nephritis cyclophosphamide trial.⁹ The patients were given an IV pulse of cyclophosphamide, 1 g/m² in 250 mL of fluid, initially over 1.5 to 2 hours. They were prehydrated for 24 hours with at least 1,500 mL and premedicated with IV dexamethasone (20 mg) and metaclopramide, 0.2 to 0.5 mg/kg. Lorazepam (1 to 2 mg) or diphenhydramine (50 mg) was given for sedation as needed. IV and oral hydration was continued for another 24 hours after infusion. They were given monthly pulse treatments for a maximum of 6 months unless they showed sustained improvement over three courses of pulse cyclophosphamide therapy. Leukocyte count was monitored weekly. The intention was to effect a drop of the leuckocyte count to 1,500 to 2,000 at 2 weeks after infusion. the dose of IVCY was increased by 25% on subsequent pulses in those patients with no significant leukocyte count drop. High-dose oral prednisone (1 to 2 mg/kg) followed by a taper over 6 months on average was used empirically with IVCY in 10 of 15 patients based on the aforementioned protocol. This therapy was not used in the remaining five patients, however, partly because of side effects but predominantly because it was found to be unnecessary.

Results. Patients who improved. Eleven patients experienced complete remission and one patient improved his functional scale from 4 to 2 (table). Twelve of 15 patients returned to routine work. All patients who improved had some response to previous other immunotherapy. These patients had an average disease duration of 9.4 months before treatment. Ten of the 12 respondents had the disease for less than 14 months. All patients improved by at least 1 point in an average of 3.3 months (range, 1 to 7). The average number of months to peak sustained improvement (persisting for more than 6 months) was 8.5 (range, 1 to 21), with average Rankin scale changing from 3.8 to 0.9. One patient improved five grades, three patients improved four grades, two patients improved three grades, four patients improved two grades, and two patients improved one grade. Improvement was achieved with an average of 4.3 monthly treatments. The two groups with and without concurrent prednisone use achieved the same average Rankin scores of 1.2 (with prednisone) and 1 (without). Eight of the 12 responding patients required no subsequent therapy at the time of last follow-up. Four patients required subsequent therapy for recurrent symptoms. Two of these four patients received repeat cyclophosphamide treatment for a Rankin scale of 2 at 4 months and 3 at 44 months after first IVCY course. Two patients received IVγ for minor paresthesia but did not require additional IVγ or cyclophosphamide therapy.

Patients with no improvement. Three patients showed no improvement (average Rankin scale pre- and post-treatment, 3.4 to 3.2) with an average of 5.4 treatments, and one patient (Patient 14; see table) worsened on two separate occasions within 10 to 14 days after IVCY and responded to prednisone and IVγ treatment. Each of the three patients had a prior illness duration of 1.7, 7, and 22 years. Two of the three had failed to respond to plasma exchange and IVγ. Patient 14 (see table) had evidence of CNS demyelination and experienced worsening of tremor and ataxia after pulse IVCY.

Side effects of IV cyclophosphamide. Nausea was experienced by four patients, headache/lightheadedness by one, and rash by another. Moderate alopecia was noted in two patients. The average reduction in hematocrit was 5% from baseline to nadir. Two patients had an increase in hematocrit. Drug-induced leukopenia developed in all patients, with an average reduction of 5.8 k/µL overall. Those patients who had been taking azathioprine had an average reduction in leukocytes of 7.83 k/µL but leukocyte count returned to normal 2 to 3 weeks after last IVCY. No patients developed hematuria, prolonged bone marrow depression, or malignancy during the reported period.

Discussion. In this case series using pulse IVCY, clinical remission was induced in 11 of 15 patients with CIDP that has persisted for up to 9 years. Improvement of up to 5 points on the Rankin scale effectively allowed 14 of 15 patients to resume normal daily activities, and 10 of 15 returned to their former occupations. Analysis of electrodiagnostic studies did not reveal any pattern predictive of prognosis, particularly in compound muscle action potential amplitudes, presence of conduction block, or both. Older age at onset or degree of drug-induced leukopenia also did not correlate with response. Time to first IVCY treatment was noticeably linked to outcome, however, with more effective response in patients with disease duration of less than 10 months. The number of treatments was determined by our decision to discontinue therapy if patients had sustained improvement after three successive treatments. Short-term complications such as leukopenia and hair loss were well-tolerated. Hair loss was more obvious in women, although our population in this limited study had 2.75 times as many women as men.

The nonrespondent group tended to have longer disease duration before therapy (10.2 years average) and received more treatments (5.4 versus 4.3). The single patient who worsened with pulse IVCY had the longest clinical course before therapy (22 years) and had significant CNS demyelination on MRI. The worsening associated with treatment may have reflected an adverse effect on a B-cell population, producing a protective anti-idiotype or direct cytotoxic effect on glial cells.

Cyclophosphamide is a potent suppressor of B-cell activity and antibody formation. Its effect on T-cells is complex and dose-related. It reportedly augments the immune response at low doses by possibly inhibiting suppression but causes T-cell lymphopenia and suppression of effector function at higher doses. Similar to our experience, IVCY in patients with systemic lupus erythematosus has been demonstrated to be well-tolerated and associated with improved long-term outcome.⁹ Concurrent use of prednisone may permit patients to tolerate larger doses of cyclophosphamide. In our series, 10 patients initially received prednisone in association with IVCY. The IVCY therapy and improvement allowed the corticosteroid dose to be tapered in 3 to 6 months. Five patients of this series had no concurrent steroids. As mentioned, the two treatment groups with and without concurrent prednisone achieved similar Rankin scores (1.2 and 1), although the numbers did not allow statistical comparison.

The immunologic effects of monthly administration of IVCY in severe systemic lupus erythematosus was investigated, with data revealing a progressive decline in the absolute number of T-cells (T3), both T₄ and T₈ subsets, but no global loss of T-cell function. The number of B-cells declined during treatment but returned to near baseline at follow-up.⁹ Specific antibody and T-cell response were not followed up in our patients. We believe, however, that the IVCY modulates the chronic immune response to peripheral nerve antigens in CIDP patients by decreasing antibody production through its effects on B-cells and regulatory T-cells.

Prednisone, IVγ, and plasma exchange treatment of CIDP patients are reported to have temporal benefit. The study using prednisone alone did not provide an answer to the value of oral steroids for longer than 3 months.² The beneficial effects of IVγ lasted 6.4 weeks in one double-blind placebo-controlled crossover study.⁴ In our experience, plasma exchange benefited patients only 1 to 2 months. The side effects, expense, and necessity for frequent repeated applications usually make these other immunotherapies impractical over the long term. In this study, selected patients with CIDP had sustained functional improvement after immunosuppressant therapy with IVCY. The improvement continued after completion of treatment and persisted for up to 9 years, with minimal side effects and a limited need for retreatment in some patients. Patients with the best outcome had disease duration of less than 10 months and had previously shown a positive clinical response to some other type of immunotherapy, such as IVγ. The long-term effects of IVCY administration on subsequent development of malignancies and on reproductive potential are unknown, however. Thus, its use should be only after adequate patient education and careful follow-up until long-term controlled studies demonstrating its superiority to other modalities and lack of serious side effects have been completed.