Prediction of relapse after discontinuation of infliximab therapy in severe sarcoidosis

Introduction

Sarcoidosis is a systemic, granulomatous disease that has variable clinical course and can affect multiple organs. This disease can be self-limiting, but can also follow a chronic course in a subgroup of patients [1–3]. When immunosuppressive treatment is indicated, corticosteroids are the first choice drugs [4, 5]. Even though corticosteroids are generally effective, continued use may cause severe side-effects. Second-line therapy therefore usually involves agents with steroid-sparing effects, such as azathioprine or methotrexate [6–11]. However, some patients are resistant to these types of treatment or develop considerable side-effects. Therefore, biologicals have been introduced as new treatment options.

The biological anti-tumour necrosis factor (TNF) drug infliximab (Remicade; Janssen Biotech, Inc., Malvern, PA, USA) has been used widely for treatment of inflammatory diseases such as rheumatoid arthritis, Crohn’s disease and psoriasis [12–14]. Furthermore, it has been the subject of investigation in case series and reports of several manifestations of sarcoidosis. Two randomised controlled trials have investigated short-term infliximab treatment in sarcoidosis and revealed improvement in lung function after 14 weeks of infliximab treatment [15, 16]. Post hoc analysis showed positive effects of infliximab treatment on extrapulmonary symptoms [17].

Due to the fact that these studies have focused on treatment outcomes after an induction phase of a maximum of 6 months, long-term outcome after discontinuation of treatment remains largely unknown. Because the optimal treatment duration or best moment to discontinue infliximab treatment have not been studied, in clinical practice, treatment duration is based on the physicians’ opinion. However, like most biologicals, infliximab is an expensive drug, which makes excessive or redundant use not desirable in times of rising healthcare costs. Conversely, re-initiation after discontinuation might contribute to anti-infliximab antibody formation, resulting in decreased drug efficacy and sometimes allergic reactions [18].

That relapse of symptoms can occur after discontinuation of infliximab in sarcoidosis was shown in a small series of 14 patients [19]. Prediction of relapse is important to identify patients with a low risk of relapse and possibly shorten therapy duration, because this will reduce the burden of therapy for the patient (long-term safety and efficacy) and lower healthcare costs. Moreover, identifying patients with a high relapse risk after discontinuation would be useful to adjust or even prolong treatment with infliximab in the case of an objectified response.

In Crohn’s disease, leukocyte count and C-reactive protein (CRP) levels were found to be predictors of relapse after discontinuation of infliximab therapy [20]. As leukocyte count and CRP are generally within normal limits in sarcoidosis patients, other markers of disease activity in sarcoidosis were also studied: angiotensin-converting enzyme (ACE), soluble interleukin (IL)-2 receptor (sIL-2R) and uptake of ¹⁸F-fluorodeoxyglucose (FDG) by positron emission tomography (PET) [21–28].

The aim of this study was to investigate the long-term outcome of infliximab treatment in patients with severe sarcoidosis and to predict which patients will relapse after discontinuation of infliximab therapy.

Results

Study subjects

Between August 2004 and October 2010, 56 patients started infliximab therapy at St Antonius Hospital. Four patients did not complete the induction phase of 6 months; three of which developed an allergic reaction and one died due to comorbidity. Three patients were still on anti-TNF-α treatment at the moment of inclusion and two patients were lost to follow-up. A total of 47 patients were therefore included in the risk analysis (fig. 1).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1–

Flow diagram of all patients included in the study. Grey boxes indicate exclusion from risk analysis; the induction phase was set at 6 months. IFX: infliximab; TNF: tumour necrosis factor.

Demographics, clinical characteristics and disease activity parameters at baseline are summarised in table 1. The most common reason for initiating treatment was a pulmonary treatment indication (30 (64%) patients); however, of the total cohort, 41 (87%) patients suffered from extrapulmonary manifestations of sarcoidosis. Most patients (46 out of 47) were treated with prednisone and/or methotrexate before infliximab was initiated. In the four patients who did not use prednisone, the most common contraindications were obesity and diabetes. The majority of patients received additional immunosuppressive or immunomodulatory agents while being treated with infliximab to prevent formation of anti-infliximab antibodies; only three patients were not on any type of concomitant medication due to intolerance. The mean±sd treatment duration was 8.5±5.8 months. After treatment discontinuation, the mean±sd overall follow-up time was 36.6±22.6 months.

View this table:

• View inline
• View popup

Table 1– Baseline characteristics of the included patient population at initiation of infliximab therapy

Relapse

Kaplan–Meier analysis revealed a median±se time to relapse of 11.1±2.57 months after discontinuation of infliximab treatment and showed that 25% of the total cohort relapsed within 4 months (fig. 2). Out of 47 included patients, 29 (62%) experienced relapse after a mean±sd time of 7.8±7.7 months. Relapse occurred within the first 10 months in 20 out of 29 relapsing patients. Moreover, only two out of 29 relapsing patients did so after a period of 20 months. Relapse in most patients was a severe deterioration of symptoms requiring retreatment, with some patients being as ill as they were before treatment with infliximab.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2–

Kaplan–Meier analysis of time to relapse after discontinuation of infliximab therapy. The median time to relapse in the total cohort was 11 months. In total, 29 (62%) out of 47 patients experienced relapse.

The majority of patients who experienced relapse were retreated with infliximab (23 out of 29 patients). Three of those 23 patients developed an allergic response to infliximab and switched to alternative medication. Six patients were retreated with agents other than infliximab due to varying reasons, including pregnancy, previous adverse events during infliximab treatment and the degree of relapse severity. Of these six patients, one was treated with prednisone, two were treated with azathioprine and another two were treated with hydroxychloroquine. One patient was enrolled in a clinical trial for an experimental drug.

Discussion

In this study we found a high proportion of relapse (62%) after discontinuation of infliximab therapy in sarcoidosis patients, after a mean time of 8 months. These relapse episodes were characterised by recurrence of sarcoidosis symptoms with need for therapeutic intervention. Both a high mediastinal SUV_max score on FDG PET and a high serum sIL-2R concentration at the initiation of therapy were found to significantly predict the chance of relapse in sarcoidosis patients.

Disease relapse after infliximab discontinuation in sarcoidosis patients was first noted in a study of 14 patients by Panselinas et al. [19]. The relapse rate in this cohort was even higher (86%) than in our cohort. Based on the natural history of disease, patients with a longer duration of disease could be more prone to relapse after discontinuation of infliximab. However, as the time from diagnosis to infliximab initiation was longer in our cohort (6.3 versus 4.7 years), this is unlikely to be the cause of the lower relapse incidence found in our cohort. Patients in our cohort were treated for a longer period of time; in the study by Panselinas et al. [19] nine out of 14 patients had been treated with fewer than six infusions. It is not known whether a more intensive infliximab treatment regimen can prevent relapse or change the course of the disease.

No studies have focused on predicting relapse after discontinuation of infliximab therapy in sarcoidosis patients. In Crohn’s disease, leukocyte count and CRP levels were found to be predictors of relapse after discontinuation of infliximab therapy [20]. However, CRP levels and leukocyte count were not predictors of relapse in our sarcoidosis cohort, as in this disease their levels are mostly within the normal range.

In this study, mediastinal SUV_max score on FDG PET and a high serum sIL-2R concentration at the start of therapy were identified as predictors of relapse. FDG PET has previously been demonstrated to be a predictor for decline in lung function in untreated patients and is more sensitive than ⁶⁷gallium imaging in the assessment of sarcoidosis activity in mediastinal and pulmonary areas [26, 27]. In previous studies, serum sIL-2R levels at disease presentation were correlated with parenchymal infiltration in pulmonary sarcoidosis and were possible prognostic markers at diagnosis [29, 30].

In our cohort, patients with SUV_max scores ≥6.0 were found to be four times more likely to experience relapse than patients with SUV_max <6.0. In addition, we showed that patients with serum levels of sIL-2R ≥4000 pg·mL⁻¹ were more than twice as likely to relapse than those with lower serum sIL-2R levels. However, this does not mean that patients with normal sIL-2R at start of therapy will not relapse.

FDG PET results were found to be more adequate predictors of relapse than serum levels of sIL-2R, although it must be acknowledged that FDG PET should not be performed repeatedly, because of radiation exposure and associated healthcare costs. Furthermore, not all treatment centres may have access to FDG PET and qualified nuclear physicians with specific knowledge in the field of sarcoidosis. Therefore, serum levels of sIL-2R, when measured in a standardised fashion, can still serve as indicators of relapse risk.

Although in our opinion sIL-2R is a parameter suited to guide infliximab therapy, and a normal sIL-2R level might be required before infliximab therapy is tapered off, we could not prove this in our study. In our cohort, infliximab treatment was sometimes prolonged when clinically indicated. These patients could have elevated sIL-2R levels. In fact, at the moment of discontinuation, only seven patients had elevated sIL-2R levels (>4000 pg·mL⁻¹). Our study might therefore underestimate the guiding effect of sIL-2R in tapering infliximab treatment.

Besides the finding that the incidence of relapse was high in our cohort, the median time to relapse in the total cohort was 11 months. Interestingly, 93% of relapsing patients did so within the first 20 months, making relapse unlikely in patients who are still in remission after this period of time. In clinical practice, follow-up of these patients can probably be performed at longer time intervals.

A limitation of this study is the retrospective nature of the design, because nonstandardisation of follow-up necessarily occurs in a retrospective study. For instance, patients differed in terms of treatment indications and conditions, especially regarding the type and dose of concomitant and consolidation treatment used. While this may be considered a limitation, this variation does reflect clinical practice.

From this study, it appears that patients with high disease activity at the start of therapy, reflected by high sIL-2R or mediastinal SUV_max on FDG PET, have an increased chance of relapse after discontinuation of treatment. A possible explanation could be that these patients, suffering from severe active disease, need a more intensive treatment regimen, either in dose or duration. No study has focused on the optimal duration of infliximab therapy in sarcoidosis after the 6-month induction phase. It might be worthwhile to investigate whether the risk of relapse would be lower if infliximab administration was continued for a longer period or at a higher dosage schedule than the standard 5 mg·kg⁻¹ in selected patients. For now, close monitoring of patients with high SUV_max scores and serum sIL-2R levels at the start of treatment is warranted.

The high relapse rate of 55% 1 year after discontinuation of infliximab therapy found in this study underlines the importance of additional research in this field and the need for new therapeutic approaches. The relapse rate in sarcoidosis after discontinuation of infliximab appears to be higher than described in inflammatory bowel disease, with a relapse rate of 25% after 1 year for ulcerative colitis and of 39–44% for Crohn’s disease [20, 31]. Even though most sarcoidosis patients benefit from infliximab therapy initially, neither infliximab nor any of the currently available drugs can actually cure the disease.

In conclusion, this study provides evidence for a high relapse rate after discontinuation of infliximab therapy in sarcoidosis patients. For the first time it was shown that both high serum sIL-2R level and high SUV_max on FDG PET at initiation of therapy are valuable predictors of relapse after infliximab discontinuation. Close monitoring of patients in this category who discontinue treatment is therefore indicated. In addition, the results suggest that longer duration of infliximab treatment might be needed in these patients, but prospective trials are needed to prove the clinical benefit of such a practice.