Effect of Infliximab on Chronic Progressive Neuro-Behçet's Disease: Influence of the Timing of Introduction on the Patient Outcome

Abstract

Objectives

Chronic progressive neuro-Behcet's disease (CPNB) is characterized by progressive deterioration leading to disability. Methotrexate (MTX) has been shown to have beneficial effects on CPNB. However, while infliximab has been found to be effective for patients with inadequate responses to MTX, the appropriate timing for the introduction of infliximab remains unclear. We explored the effects of intervals before the introduction of infliximab on the functional outcome.

Methods

A retrospective analysis was performed for patients with CPNB who received infliximab and were followed up until October 2015. Functional disability was rated by the Steinbrocker functional classification as used in rheumatoid arthritis. Correlations between the outcomes and intervals before the introduction of infliximab were then analyzed by Spearman's rank correlation test.

Patients

Eleven patients with CPNB [8 men, 3 women, age 35.2±9.3 years old (mean±standard deviation)] who met the international classification criteria for Behcet's disease were included.

Results

All 11 patients had received MTX prior to infliximab. The intervals from the onset to the introduction of infliximab and the follow-up periods were 26.6±35.1 months and 65.2±43.6 months [mean±standard deviation], respectively. Among the 11 patients, 2 still showed progression after the introduction of infliximab. The functional disability grades after infliximab treatment were significantly correlated with the intervals from the onset of CPNB to the introduction of infliximab (r=0.6177, p=0.0476).

Conclusion

The results indicate that the delayed introduction of infliximab leads to irreversible functional disability in CPNB. Thus, it is recommended that infliximab be administered as soon as possible for CPNB patients with inadequate responses to MTX.

Introduction

Neurological involvement in Behçet's disease (BD) is one of the most serious complications of the disease and is called neuro-Behçet's disease (NB) (1,2). It is now understood that NB can be classified into an acute type (ANB) and a chronic progressive type (CPNB), which have different clinical characteristics (3,4). While ANB responds well to corticosteroids and rarely results in significant damage, CPNB usually presents with intractable neurobehavioral manifestations and cerebellar ataxia, which progress despite high doses of corticosteroids or conventional immunosuppressive therapy, including azathioprine or cyclophosphamide (5-7).

A particularly characteristic feature of CPNB is the persistent elevation of the cerebrospinal fluid (CSF) interleukin-6 (IL-6) value as well as progressive brainstem atrophy (5-8). More importantly, the progression rate of brainstem atrophy was shown to be closely correlated with the elevation of the CSF IL-6 value, indicating that brainstem atrophy results from persistent inflammation reflected by the elevation of CSF IL-6 (9).

Notably, methotrexate (MTX) has been shown to have beneficial effects on CPNB (6,7). An open prospective trial of six CPNB patients showed that the neuropsychological manifestations as well as the magnetic resonance imaging (MRI) findings were not worsened during treatment with MTX (6,7). Further studies of low-dose weekly MTX for up to four years confirmed its beneficial effects in CPNB (10). It should be noted, however, that there are still patients with CPNB whose progression cannot be controlled by MTX (3,10).

Of note, it has been disclosed that infliximab was effective for the treatment of recalcitrant CPNB resistant to MTX therapy (11,12). We demonstrated the efficacy of infliximab in 5 patients with CPNB refractory to MTX in a 24-week open trial (13). None of the five patients showed exacerbation of neuropsychiatric manifestations or progression of brainstem atrophy on MRI (13). Consistently, CSF IL-6 levels were dramatically decreased and stayed below 20 pg/mL before the last infusion at 14 weeks (13).

Infliximab received insurance coverage in Japan after a multicenter, prospective, open-label, single-arm phase 3 study (14). However, the appropriate timing for the introduction of infliximab remains unclear. Thus, it is possible that the later introduction of infliximab might allow disease progression, so damage might not be satisfactorily recovered despite infliximab treatment. To explore this possibility, we examined the effects of intervals before the introduction of infliximab on the outcome of patients with CPNB.

Materials and Methods

Patients

Eleven patients [8 men, 3 women, age 34.2±8.9 years old (mean±standard deviation (SD))] who satisfied the international classification criteria for BD (15) complicated with CPNB and received infliximab were included. Seven of the 11 patients were recruited from a retrospective cohort generated by the Behçet's Disease Research Committee guided by the Ministry of Health, Labour and Welfare of the Japanese Government, including BD patients who were hospitalized due to neuropsychiatric manifestations between 1988 and 2008 in the 6 institutions belonging to the Research Committee (3). Four additional patients with CPNB who developed CPNB after 2008 and whose follow-up data for at least six months were available were also included. These 11 patients were followed until October 2015.

This study was carried out in accordance with the World Medical Association Declaration of Helsinki Ethical Principles for Medical Research Involving Human Subjects and was approved by the ethical committees of the six institutions belonging to the Research Committee.

Diagnoses

The diagnosis of CPNB was based on clinical data as well as data from MRI and CSF studies. CPNB was thus defined as the insidious onset of neuropsychiatric symptoms, including dementia/neurobehavior changes, ataxia, and dysarthria, with persistent elevation of CSF IL-6 levels over 17.0 pg/mL with or without brainstem atrophy on MRI (3).

The evaluation of the patients' functional state

The functional disability of the patients was rated by the Steinbrocker functional classification as is used in rheumatoid arthritis (16). The patients were classified into four classes depending on the degree of disability: no disability (class 1), independent living with some limitations (class 2), disability that needed assistance in activities of daily living (class 3), and bedridden state (class 4). The job status of each patient was also examined.

Statistical analyses

The correlation between the patients' status and the interval to the introduction of infliximab was analyzed by Spearman's rank correlation test using the GraphPad Prism 4 software program, ver. 4.03 (GraphPad Software, San Diego, USA).

Results

Table summarizes the demographic features and the changes in the functional status of the patients before and after the introduction of infliximab. Eight and 7 patients out of 10 (one patient lacks information on smoking and HLA-51) were smokers and HLA-B51 positive, respectively, consistent with a previous report (3). However, all patients had quit smoking after the diagnosis of CPNB. Nine of the 11 patients had experienced uveitis attack, and all 11 patients had shown mucocutaneous lesions. Only one patient presented with gastrointestinal tract involvement, whereas no patients showed vascular involvement. All 11 patients had received MTX prior to infliximab and continued after the introduction of infliximab. Eight and 5 of the 11 patients were treated with low doses of prednisolone (2-10 mg/day) and colchicine, respectively. The maximum level of CSF IL-6 prior to introduction of infliximab was 1,074.0±2,543.0 pg/mL (mean±SD). The intervals from the onset to the introduction of infliximab and the follow-up periods were 26.6±35.1 months and 65.2±43.6 months (mean±SD), respectively. Overall, the functional status was not improved but rather worsened in two patients.

Table.

Demographic Features of 11 Patients with CPNB Treated with Infliximab.

N		11
Age (mean±SD)		34.2±8.9 years
Gender (men:women)		8:3
Smoker		8/10
HLA-B51		7/10
Uveitis		9/11
Gastrointestinal lesions		1/11
Vascular lesions		0/11
Maximal CSF IL-6 levels before infliximab		Range 32.0-8,620.0 pg/mL, 1,074.0±2,543.0 pg/mL (mean±SD)
Interval from the onset to the introduction of infliximab (mean±SD)		26.6±35.1 months (0-120)
Follow-up period (mean±SD)		65.2±43.6 months (10-122)
Initial state*	Class 1	6
	Class 2	3
	Class 3	2
	Class 4	0
	Brainstem atrophy on MRI	8
Final state*	Class 1	5
	Class 2	2
	Class 3	2
	Class 4	2
	Brainstem atrophy on MRI	9

* Functional status was rated by Steinbrocker functional classification: no disability (class 1), independent living with some limitations (class 2), disability that needed assistance in activities of daily living (class 3), and bedridden state (class 4)

**CSF: cerebrospinal fluid, CPNB: chronic progressive neuro-Behçet’s disease, MRI: magnetic resonance imaging, SD: standard deviation

Most patients showed brainstem atrophy on MRI. Notably, infliximab prevented the progression of brainstem atrophy on MRI in 9 patients who did not show progression of functional disability. In contrast, in two patients who did show progression of functional disability after the introduction of infliximab, brainstem atrophy developed de novo or the degree of brainstem atrophy progressed. During the follow-up period, no patients presented with an attack of uveitis, intestinal involvement, vascular involvement, or ANB that affected the functional status.

The changes in the status of each patient after the introduction of infliximab are shown in Fig. 1. Of the 11 patients, 9 did not show progression after the introduction of infliximab, whereas 2 showed progression. In one of these two patients, infliximab had been discontinued due to adverse effects (intermittent fever) and changed to adalimumab followed by the addition of azathioprine before the final follow-up evaluation. However, the efficacy of adalimumab was not appropriately evaluated by the changes in CSF IL-6 in this patient. Consistent with the results of the changes in functional status, the number of patients who received jobs was not increased after the introduction of infliximab.

Figure 1.

Changes in the status of each patient after the introduction of infliximab. The changes in functional status as well as the job status of the individual patient after the introduction of infliximab (IFX) are presented. The functional status was rated by the Steinbrocker functional classification, as follows: no disability (class 1), independent living with some limitations (class 2), disability that needed assistance in activities of daily living (class 3), and bedridden state (class 4).

Interestingly, as shown in Fig. 2, the functional disability classes of the patients after the introduction of infliximab were significantly correlated with the intervals from the onset of CPNB to the introduction of infliximab (r=0.6177, p=0.0476). A similar tendency can be seen for the functional disability classes of the patients before the introduction of infliximab, although it did not reach statistical significance (r=0.5704, p=0.0701). The results indicate that a delay in the introduction of infliximab leads to irreversible functional disability in patients with CPNB. Thus, the introduction of infliximab should be considered as soon as possible for patients with CPNB who do not adequately respond to MTX.

Figure 2.

Relationship between the interval from the onset of disease to the introduction of infliximab and functional disability. The functional status of the patients was rated by the Steinbrocker classification before and after the introduction of infliximab (IFX). The statistical analysis was conducted by Spearman's rank correlation test.

Two patients showed adverse effects of infliximab. One developed an intermittent fever, requiring a switch to adalimumab, as mentioned above. The other presented with a mild cough due to possible subclinical pneumocystis pneumonia, which was successfully managed by sulfamethoxazole and trimethoprim without discontinuation of infliximab.

Discussion

A previous retrospective study showed that MTX significantly reduced the mortality of patients with CPNB (17). However, there are still patients with CPNB who do not adequately respond to MTX (3,10). Notably, infliximab has been shown to be effective in patients who did not adequately respond to MTX (11,12,18). Accordingly, the guidelines for the management of NB recommended that infliximab be introduced immediately when there is no improvement in neurological manifestations with MTX at the highest feasible dose (19). However, the precise timing for the introduction of infliximab remains to be elucidated. In fact, it is sometimes difficult to precisely judge the progression of the manifestation during treatment with MTX due to its slow speed.

In the present study, we explored the effect of the period from the onset of CPNB to the introduction of infliximab on the functional outcomes of patients. The results demonstrated that longer intervals from the onset to the introduction of infliximab were associated with worse functional disability. These findings strongly suggest that a delay in the introduction of infliximab may lead to the irreversible functional disability of the patient.

Only a few patients showed progression after the introduction of infliximab, with most patients in a stable condition in the current study. Furthermore, longer intervals from the onset to the introduction of infliximab appeared to be associated with higher disability classes, even before the introduction of infliximab, although this finding did not reach statistical significance. It is therefore likely that longer periods of inappropriate control of disease activity, probably reflected by CSF IL-6 levels, might lead to poorer functional disability. In this regard, in the patient who showed progression, CSF IL-6 was not measured after switching from infliximab to adalimumab. Therefore, it is mandatory to check CSF IL-6 levels when the treatment regimen is changed.

Our findings also suggested that functional disability might be irreversible once it manifests in CPNB, even with the introduction of infliximab. Indeed, two patients had deteriorated functional outcomes with the development or progression of brainstem atrophy on MRI, even after infliximab introduction. In both of these patients, a higher dose of steroids (prednisolone 60 mg/day) was given for longer periods of time than that in the other 9 patients. It is therefore suggested that prolonged treatment with higher doses of steroids might be a risk factor for progression in addition to the delayed introduction of infliximab, although further studies are required to confirm this point.

Previous studies have shown that brainstem atrophy progressed most remarkably in an irreversible manner during the first two years after the onset of CPNB (9). Since disability from CPNB is irreversible in most patients, it is necessary to introduce infliximab as soon as possible when patients are judged to have a progressive course despite treatment with MTX. Thus, it is important to determine parameters for measuring progression. In this regard, it was recently reported that effective treatment of CPNB seemed to reduce the enlargement of the width of the third ventricle (WTV) on MRI in some patients (20). Thus, sequential measurement of the WTV on MRI is considered a sensitive biomarker in patients with CPNB.

CPNB is characterized by the persistent elevation of CSF IL-6 to values over 17 pg/mL (3). Furthermore, treatment to target CPNB includes the adequate control of CSF IL-6 (as low as possible below 17 pg/mL) (19). Thus, it has also been demonstrated that the progression of brainstem atrophy on MRI is closely correlated with integral values of CSF IL-6 (9). Accordingly, no progression of neurological manifestations or brainstem atrophy has been found as long as the CSF IL-6 value is maintained at low levels (13,14). However, while it is of course preferable to keep the CSF IL-6 value below 17 pg/mL, there do exist some patients with CSF IL-6 levels ＞17 pg/mL without progression. Therefore, further studies are needed to determine the critical levels of CSF IL-6 required for the prevention of progression in patients with CPNB. Accordingly, it is also important to determine the appropriate timing for the introduction of infliximab. In this regard, it would be preferable for CSF IL-6 to be measured frequently.

One limitation of the present study was that it was a retrospective cohort study. Therefore, further studies with prospective designs are required to determine the appropriate timing of the introduction of infliximab in CPNB.

Author's disclosure of potential Conflicts of Interest (COI).

Masataka Kuwana: Honoraria, Asahi Kasei, Astellas, AstraZeneca, Boehringer-Ingelheim, Chugai, Janssen, Kissei, MBL, Mochida and Ono Pharmaceuticals. Izumi Kawachi: Honoraria, Alexion, Biogen, Chugai and Novartis. Yohei Kirino: Honoraria, Amgen and Novartis; Research funding, Nippon Shinyaku. Mitsuhiro Takeno: Honoraria, Amgen, Eisai and Mitsubishi Tanabe.

Financial Support

Supported by a grant from the Behçet's Disease Research Committee, Ministry of Health, Labor and Welfare of the Japanese Government.