Role of interleukin-6 in a patient with tumor necrosis factor receptor–associated periodic syndrome: Assessment of outcomes following treatment with the anti–interleukin-6 receptor monoclonal antibody tocilizumab

Authors


Abstract

In this report, we describe treatment outcomes in the first case of a patient with tumor necrosis factor receptor–associated periodic syndrome (TRAPS) treated with the anti–interleukin-6 (anti–IL-6) receptor monoclonal antibody tocilizumab. Since IL-6 levels are elevated in TRAPS, we hypothesized that tocilizumab might be effective. The patient, a 52-year-old man with lifelong TRAPS in whom treatment with etanercept and anakinra had failed, was administered tocilizumab for 6 months, and the therapeutic response was assessed by measurement of monocyte CD16 expression and cytokine levels. Following treatment, the evolving acute attack was aborted and further attacks of TRAPS were prevented. The patient did not require corticosteroids and showed significant clinical improvement in scores for pain, stiffness, and well-being. Moreover, the acute-phase response diminished significantly with treatment. Monocyte CD16 expression was reduced and the numbers of circulating CD14+CD16+ and CD14++CD16− monocytes were transiently decreased. However, cytokine levels were not reduced. This case supports the notion of a prominent role for IL-6 in mediating the inflammatory attacks in TRAPS, but blockade of IL-6 did not affect the underlying pathogenesis. These preliminary findings require confirmation.

Tumor necrosis factor (TNF) receptor–associated periodic syndrome (TRAPS) is a rare autosomal, dominantly inherited autoinflammatory disease characterized by recurrent episodes of fever, myalgia, arthralgia, migrating erysipelas, and serositis. It is associated with missense mutations of the 55-kd TNF receptor superfamily 1A (TNFRSF1A) (1). The frequency and length of attacks are variable, but untreated attacks may last from days to weeks. Poorly controlled inflammation may predispose to amyloid formation with associated morbidity.

Treatment is aimed at improving quality of life, preventing acute attacks of disease, which may require the concomitant use of corticosteroids, and controlling inflammation to prevent amyloidosis. Conventional immunosuppressants, e.g., cyclosporine, methotrexate, azathioprine, and cyclophosphamide, do not help this condition. Initial studies with etanercept yielded promising results (2), but the effect of this treatment may decline over time (3). Paradoxically, infliximab and adalimumab induce attacks of inflammation (4). Anakinra may have clinical benefit in patients with TRAPS (5, 6), although our experience has been that the response may be variable and frequent side effects may occur (McDermott EM, et al: unpublished observations).

Interleukin-6 (IL-6) is a proinflammatory cytokine that is produced by a variety of cells, including T and B lymphocytes, monocytes, and fibroblasts. Tocilizumab is a humanized monoclonal antibody that binds specifically to both soluble and membrane-bound IL-6 receptors and inhibits IL-6 receptor–mediated signaling. It is licensed for use in rheumatoid arthritis and may have benefits in diseases that may be considered to be part of the autoinflammatory disease spectrum, i.e., systemic-onset juvenile idiopathic arthritis (7) and adult-onset Still's disease (8). We have previously reported that the levels of both IL-6 and IL-8 are elevated in TRAPS (9), and we therefore considered tocilizumab as a treatment option for our patient with TRAPS, whose disease had been treated unsuccessfully with etanercept and anakinra.

CASE REPORT

The patient, a 52-year-old man with a C33Y TNFRSF1A mutation, had experienced symptoms of TRAPS since age 5 years. His typical attacks of TRAPS are characterized by recurrent fevers, myalgia, rash, abdominal pains, joint pains, and lymphadenopathy. Prior to being treated with corticosteroids, his attacks lasted 8–12 weeks. He underwent abdominal surgery on 4 occasions (appendectomy, diverticulitis surgery, and 2 exploratory laparotomies). In 2004, he had glomerulonephritis that, in retrospect, may have been a manifestation of TRAPS (10). Between 2003 and 2006, the patient received etanercept for 2 periods of time, totaling 18 months. On the first occasion, treatment with etanercept was stopped because only a partial response, according to clinical and laboratory measures, was achieved, and on reintroduction, the treatment was stopped due to development of liver function abnormalities. In January 2009, he received anakinra for 4 months, which led to clinical improvement, but the response according to laboratory test results was only partial. Moreover, he developed recurrent chest infections and neutropenia, warranting discontinuation of treatment. Prior to intervention therapy with tocilizumab, he was experiencing attacks bimonthly.

Tocilizumab was obtained commercially and was approved for use by the Hospital Drugs and Therapeutics Board. Scientific assays were approved by the Nottingham (UK) National Health Service Local Research Ethics Committee and by the University of Nottingham Medical School Research Ethics Committee.

The treatment plan was to administer 8 mg/kg tocilizumab every 4 weeks, with dose adjustment according to the response and side effect profile. The pretreatment period was defined as the 6 months immediately prior to initiation of tocilizumab treatment (during which the patient received oral prednisolone, only as required). Thereafter, the period during which tocilizumab was administered, from week 1 to week 25, was defined as the treatment period.

The patient received his first infusion of 8 mg/kg tocilizumab during the early stages of an acute attack, characterized by low-grade fever, inguinal lymphadenopathy, poor appetite, eye pain, and thigh myalgia. This did not evolve into a full attack of TRAPS, and the symptoms resolved within 3 days. Thereafter, the patient developed thrombocytopenia, characterized by platelet levels of 97 × 109/liter and 92 × 109/liter at week 2 and week 4, respectively, which necessitated a delay in the second infusion, at a reduced dose, to week 5. Prior to this second infusion, the patient reported that he believed that his TRAPS symptoms may have been returning. He subsequently received 4 mg/kg tocilizumab at weeks 5, 9, 13, and 17, during which no attacks of inflammation occurred. However, 3 days prior to each infusion, the patient described prodromal symptoms (joint and muscle stiffness, myalgia, fatigue, and vomiting). The infusion at week 21 was therefore increased to 6 mg/kg, which prevented the prodromal symptoms. During this period, the platelet level remained at >100 × 109/liter. No further side effects were reported.

The patient was monitored weekly for the first 5 weeks of the tocilizumab regimen. Subsequently, he underwent clinical evaluations every 4 weeks, with the C-reactive protein (CRP) level, erythrocyte sedimentation rate (ESR), and serum amyloid A (SAA) measurements determined every 2 weeks. The CRP level, ESR, and SAA level normalized (normal laboratory values <10 mg/liter, <20 mm/hour, and <10 mg/liter, respectively) within 1 week of the first tocilizumab infusion (Figures 1A–C). This effect was maintained for 4 months, but mild elevations in the levels of CRP and SAA were noted at weeks 17 and 21, warranting an increase in the dose to 6 mg/kg at week 21. The acute-phase responses subsequently returned to normal values at week 26. The improvement in the CRP level, ESR, and SAA level during treatment with tocilizumab, as compared with the pretreatment values, was statistically significant (for CRP, P = 0.0009; for ESR, P = 0.0007; and for SAA, P = 0.006, by Mann-Whitney 2-tailed test).

Figure 1.

Consecutive blood samples from a patient with the C33Y TNFRSF1A mutation in tumor necrosis factor receptor–associated periodic syndrome were assessed for markers of the acute-phase response before (week −27 to 0) and during (weeks 1–26) treatment with tocilizumab. Arrows indicate the time points of administration of tocilizumab at week 0 (8 mg/kg), weeks 5, 9, 13, and 17 (4 mg/kg), and week 21 (6 mg/kg), which was infused after blood sampling. Changes in the level of C-reactive protein (in mg/liter) (A), erythrocyte sedimentation rate (in mm/hour) (B), and serum amyloid A concentration (in mg/liter) (C) were assessed.

The patient was advised to take oral corticosteroids intermittently as required, during both the pretreatment and treatment periods, and was instructed to document the dose. The total dose of oral prednisolone taken by the patient in the 6 months prior to treatment with tocilizumab was 1,325 mg, with a median dosage of 10 mg per week and a mean of 47 mg per week, in comparison with no oral prednisolone being taken during treatment with tocilizumab (Figure 2).

Figure 2.

The mean dosage of prednisolone taken (in mg/week, averaged over 4-week periods) before (weeks −27 to 0) and during (weeks 1–24) treatment with tocilizumab was recorded. Arrows indicate the time points of tocilizumab administration at week 0 (8 mg/kg), weeks 5, 9, 13, and 17 (4 mg/kg), and week 21 (6 mg/kg).

Daily scores for both general well-being and pain and stiffness, recorded by the patient using an established daily diary card (where a score of 100% represents maximal well-being or no pain and stiffness), were also compared. Significant improvements in the pain and stiffness score and in the well-being score were also seen with treatment (P < 0.0001 and P = 0.0001, respectively, by Mann-Whitney 2-tailed test).

We previously reported that the expression of CD16 on proinflammatory monocytes (expressed as the mean fluorescence intensity [MFI]) was significantly elevated in patients with the C33Y mutation in TRAPS, as compared to the MFI of CD16 expression on healthy control monocytes (11). Blood was collected for monocyte and cytokine assays from the patient and from a single age- and sex-matched healthy control. The MFI of CD16 expression on monocytes, numbers of CD16+CD14+ (proinflammatory) and CD14++CD16− (conventional) monocytes, and lipopolysaccharide (LPS)–induced TNFα production by monocytes were determined by flow cytometry and intracellular cytokine staining, as described previously (11). Serum cytokines were investigated using the FlowCytomix multiplex fluorescent bead immunoassay system, according to the manufacturer's instructions (Bender Medsystems).

The MFI of monocyte CD16 expression was raised in the patient at the time that treatment with tocilizumab was commenced (week 0), but fell to the same level as that in the healthy control subject by week 2, which was sustained throughout the followup. In control experiments, it was shown that incubating normal peripheral blood with tocilizumab at a concentration equivalent to that expected in the patient's circulation did not consistently reduce the monocyte CD16 MFI, indicating that CD16 blockade by tocilizumab did not account for the reduction in CD16 staining. The numbers of circulating proinflammatory CD14+CD16+ monocytes and conventional CD14++CD16− monocytes were elevated at the start of treatment, and fell to control levels by week 3, but were starting to rise again by weeks 11–13.

LPS-induced TNFα production in proinflammatory monocytes (CD14+HLA–DR++) and in conventional monocytes (CD14++) fluctuated over the treatment period, but did not show a clear trend in response to tocilizumab treatment and was constantly elevated relative to that in the healthy control subject. Similarly, the patient's serum TNFα levels fluctuated in a random manner over the period of monitoring, although, as previously reported (9), circulating TNFα levels were minimally raised in the patient relative to those in the control subject.

Very high levels of IL-1α and IL-8 were detected in the patient's serum compared to the healthy control subject's serum. These levels remained high, and there was no apparent effect from tocilizumab treatment. Serum IL-6 levels were actually elevated by administration of tocilizumab, particularly following the initial dose of 8 mg/kg. We hypothesize that this latter effect may be attributable to the mechanism of action of the treatment, in which tocilizumab competes effectively with IL-6 for binding to IL-6 receptors, thereby leading to raised circulating levels of free IL-6. Very few of either the patient or control serum samples contained detectable IL-1β or granulocyte colony-stimulating factor, and levels of the chemokines monocyte chemotactic protein 1 and RANTES were similar in the patient and control samples.

DISCUSSION

To our knowledge, this is the first case of a patient with TRAPS treated with tocilizumab. After treatment with tocilizumab, the evolving acute attack was aborted and further attacks of TRAPS were prevented, over a 6-month period. This suggests that tocilizumab may be a satisfactory treatment for both terminating acute attacks of TRAPS and prophylaxis.

During treatment with tocilizumab in this patient, no attacks occurred and no corticosteroid usage was required, which is of clinical relevance because other subjects in the Nottingham cohort of TRAPS patients have developed corticosteroid-associated morbidity, including corticosteroid dependence, cataracts, and osteoporosis. Improvements in patient-recorded symptom scores were supported by the patient's self-reports of improved quality of life and increased activity level, e.g., ease of climbing stairs and ability to do gardening, which was not previously possible. The significant improvement in the acute-phase response may have important implications for prevention of amyloidosis and is particularly encouraging to this patient's family, since 3 of his extended family members have developed this complication to date.

Thrombocytopenia is a recognized side effect of tocilizumab (for detailed discussion, see http://www.roactemra.com/portal/eipf/pb/actemra/roactemra/safety_and_tolerability). The development of thrombocytopenia in this patient delayed the second infusion to week 5, allowing further insight into the optimum dosing schedule for this treatment. At this time point, the patient experienced prodromal symptoms of TRAPS, suggesting that the 4-week dosing intervals may be appropriate, in line with the recommended protocol for treatment of rheumatoid arthritis. The 4 mg/kg dose of tocilizumab appeared sufficient for symptom control initially, but as mentioned above, the patient experienced prodromal symptoms toward the end of the fourth and fifth cycles, with mild increases in the acute-phase response prior to receiving the infusions. This did not occur after the dose was increased to 6 mg/kg. We infer that dose adjustment may need to be individualized on a per-patient basis, with the possibility of increasing the dosage over time. In addition, an acute attack may be anticipated to occur on discontinuation of tocilizumab, but this was not tested, due to ethical reasons.

We have previously reported that levels of IL-6 and IL-8, in particular, are elevated in patients with the C33Y mutation in TRAPS, and that IL-6 levels are correlated with the CRP level (9). The effects of tocilizumab observed in this patient with TRAPS indicate that IL-6 plays a prominent role in stimulating many of the clinical features of the attacks of inflammation in TRAPS. IL-6 has long been recognized as a key cytokine in stimulating the production of acute-phase proteins, including CRP and SAA, and thereby elevating the ESR. Our present results indicate that IL-6, directly or indirectly, also stimulates CD16, leading to elevated CD16 expression by proinflammatory monocytes in TRAPS, as was also observed in our previous study (11).

Blockade of IL-6 also led to reduced numbers of circulating proinflammatory and conventional monocytes, but these numbers increased following reduction of the tocilizumab dose, which mirrors the findings regarding platelet numbers in this patient. In contrast, however, TNFα production by LPS-stimulated monocytes, and levels of circulating cytokines (TNFα, IL-1α, IL-8, and IL-6) were not reduced as a consequence of tocilizumab treatment. This suggests that IL-6 blockade does not inhibit the etiologic factors that drive cytokine production. This is consistent with the evidence supporting the hypothesis that the underlying mechanism of TRAPS involves ligand-independent, proinflammatory signaling by intracellular, misfolded mutant TNFR1 (12–15).

As far as we are aware, our observation of raised IL-1α levels in TRAPS has not been reported previously. This finding was replicated in 1 of 3 other family members with TRAPS. Of relevance, these 2 patients with raised IL-1α levels had a good clinical response to anakinra, whereas the 2 family members with normal IL-1α levels did not (McDermott EM, et al: unpublished observations). The discrepancy between IL-1α levels and IL-1β levels may also have implications for therapy, since anakinra and rilonacept inhibit the action of both, whereas canakinumab is specific for IL-1β.

In conclusion, this preliminary report opens up a promising, needed treatment option for the management of TRAPS. Experience from other biologic agents in TRAPS patients, however, suggests that there is a variable response to drugs between patients (even within the same mutation and family), and therefore extension of treatment to other patients to confirm reproducibility of the findings is required. Long-term followup is also essential to ensure that the beneficial effects described do not remit over time, as has occurred with etanercept, and that side effects are documented.

AUTHOR CONTRIBUTIONS

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Vaitla had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Vaitla, Radford, McDermott, Todd, Drewe.

Acquisition of data. Vaitla, Radford, McDermott, Drewe.

Analysis and interpretation of data. Vaitla, Radford, Tighe, Powell, Todd, Drewe.

Ancillary