Background/Purpose: Neurologic involvement is one of the most serious cause of long-term morbidity and mortality in Behçet’s disease (BD). To date, no controlled trials have assessed therapy for Neuro-BD and therapy is guided by expert opinion. It is usually recommended to treat active parenchymal Neuro-BD with 5-10 intravenous pulses of methylprednisolone and a course of high-dose glucocorticoids followed by azathioprine. Biologic agents such as interferon (IFN)α or TNF antagonists are used in refractory cases. However, Brazilian rheumatologists usually prescribe intravenous pulse therapy with cyclophosphamide to treat and prevent relapses of neuro-BD. Thus, the objective of this study is to compare the use of monthly intravenous pulse therapy with cyclophosphamide to azathioprine in preventing relapses of Neuro-BD.

Methods: An observational, retrospective cohort study was performed. Inclusion criteria were BD diagnosis according to the criteria of the International Study Group (ISG) or the International Criteria for Behcet’s Disease (ICBD), neurologic involvement of BD and written informed consent. Neuro-BD was treated with high-dose glucocorticoids and at physician’s discretion patients were treated either by a monthly course of intravenous cyclophosphamide pulse therapy (0.5 to 1.0 g/m²) or azathioprine 2.5 mg/kg/day. After cyclophosphamide, therapy was switched to azathioprine 2.5mg/kg/day, with total treatment duration of approximately 5 years in both groups.

Results: Twenty-five patients with neuro-BD were followed for a mean 90.4 ± 53.2 months. Parenchymal involvement was observed in 19 (76.0%) patients and 6 (24.0%) patients presented non-parenchymal neuro-BD. Cyclophosphamide was prescribed for 18 (72.0%) patients for a mean 9.0 ± 3.6 months, whereas 7 (28.0%) patients received azathioprine. Baseline features were similar between patients using cyclophosphamide and those using only azathioprine regarding the frequency of parenchymal involvement (83.3% vs. 57.1%, p = 0.298), the median daily prednisone dose at neuro-BD diagnosis [50.0mg (40.0-60.0) vs. 55.0mg (27.5-75.0), p = 0.192] and the use of intravenous pulse methylprednisolone (82.4% vs. 42.8%, p = 0.548). Relapses of neuro-BD were observed in 8 (44.4%) patients from cyclophosphamide group and in 4 (57.1%) patients from azathioprine group (p = 0.673). Time to the first relapse of neuro-BD was shorter in cyclophosphamide group compared to azathioprine group (11.5 ± 7.4 months vs. 57.0 ± 15.1 months, p < 0.0001). By the Kaplan-Meier analysis, no significant differences were observed with the use of cyclophosphamide in the prevention of relapses of neuro-BD (p = 0.767), hazard ratio = 0.818 (95% confidence interval: 0.217-3.080) (Figure 1). Only 5 (27.8%) relapsing patients from the cyclophosphamide group vs. none from azathioprine group had to switch therapy to TNFα antagonists during the follow-up (p = 0.274).

Conclusion: This pilot study shows that the use of monthly intravenous cyclophosphamide as an induction therapy for neuro-BD is not associated with a lower relapse rate of the neurologic involvement compared to azathioprine.

Figure 1

« Back to 2019 ACR/ARP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/cyclophosphamide-therapy-for-the-neurologic-involvement-of-behcets-disease-is-it-superior-to-azathioprine-in-preventing-relapses/