Review of siltuximab in the treatment of multicentric Castleman’s disease

Abstract

Castleman’s disease (CD) is a rare lymphoproliferative disorder that has multiple histologic patterns, as well as two distinct clinical forms: unicentric or multicentric. Multicentric Castleman’s disease (MCD) may have mild symptoms in some cases, but in others it can progress to severe pancytopenia, life-threatening infection, secondary malignancy, multiorgan failure, or death. Recent research has determined that the etiology of the disease signs and symptoms is related to elevated cytokines, including interleukin 6 (IL-6). Siltuximab is a monoclonal antibody that targets IL-6 and is currently the only US Food and Drug Administration approved therapy for idiopathic MCD. Clinical data have demonstrated significant efficacy and tolerance of siltuximab in patients with idiopathic CD.

Introduction

Castleman’s disease (CD) is a rare lymphoproliferative disorder that is estimated to have an annual incidence of approximately 6500 cases in the United States [Munshi et al. 2015]. It was initially described by Benjamin Castleman in 1954 as the histopathological findings of angiofollicular lymph node hyperplasia [Castleman and Towne, 1985]. It is now known that CD has multiple different histologic patterns, as well as two distinct clinical forms: unicentric and multicentric. Unicentric CD is typically localized, without any systemic symptoms, and is most commonly cured with resection of the affected lymph node and no systemic therapy is required. Multicentric Castleman’s disease (MCD), which may account for approximately 1550 new diagnoses per year in the United States, typically involves more than one lymph node area and generally cannot be treated with surgical approaches [Munshi et al. 2015]. Pathologically CD has a plasmacytic, mixed cellularity or hyaline vascular histology. Unicentric CD is usually hyaline vascular variety while the plasmacytic variant presents as a multicentric disease. MCD is often associated with human herpes virus 8 (HHV-8), which is more frequently seen in the presence of a human immunodeficiency virus (HIV) infection, although a significant proportion of patients have no known viral etiology.

The clinical spectrum of MCD has significant variation. Constitutional symptoms, such as fever, fatigue, night sweats, anorexia, and weight loss, are common. Physical exam findings such as organomegaly, diffuse lymphadenopathy, effusions, and edema are frequently encountered. Other significant clinical manifestations include anemia, thrombocythemia, leukocytosis, hypoalbuminemia, polyclonal hypergammaglobulinemia, and increases in inflammatory markers such as C reactive protein (CRP), erythrocyte sedimentation rate, ferritin, fibrinogen, and interleukin 6 (IL-6) [Fajgenbaum et al. 2014]. Although disease signs and symptoms can be mild, the disease can progress to severe pancytopenia, life-threatening infection, secondary malignancy, multiorgan failure, or death. The 5-year overall survival is 65% [Dispenzieri et al. 2012], with a 3-year disease-free survival reported at 46% [Talat and Schulte, 2011]. Due to significant constitutional symptoms and the variable prognosis of MCD, systemic therapy is frequently indicated. Traditionally the treatment for MCD has included chemotherapy, immunomodulators, or steroids, in addition to antivirals in patients with HIV or HHV-8 infection. These treatments are known to be effective and may transiently improve symptoms, but have limited efficacy. In recent years the anti-CD20 monoclonal antibody rituximab has been investigated as part of the treatment for MCD. As knowledge of CD pathophysiology has grown, and cytokines have been discovered as the driving force for this disease, the use of monoclonal antibodies has expanded to include specific cytokine targets.

Interleukin 6 as a treatment target

Patients with MCD are known to have increased levels of multiple inflammatory markers, such as IL-6, IL-1, IL-2, and tumor necrosis factor α (TNFα), although the etiology of this increase in cytokines is not entirely clear. IL-6 is one of many proinflammatory cytokines that are overexpressed and it is considered the most important disease mediator in MCD. It binds to a membrane receptor or a soluble receptor and leads to gp130 dimerization, phosphorylation, and activation of receptor-associated kinases [Rossi et al. 2015]. It is produced by T cells, B cells, monocytes, fibroblasts, and endothelial cells [El-Osta and Kurzrock, 2011]. It has been reported that it is produced in the germinal centers of hyperplastic lymph nodes of CD, especially in patients with plasmacytic variation, although this has not been replicated in subsequent studies [Yoshizaki et al. 1989]. IL-6 levels typically correlate with the clinical course and therefore are thought to be the etiology of constitutional symptoms [Yoshizaki et al. 1989]. IL-6 levels can vary substantially between individual patients, but lower levels are typically seen in patients with hyaline vascular subtype [Deisseroth et al. 2015]. In patients with HIV or HHV-8 infection the disease is thought to be driven by a viral homolog of IL-6 (vIL-6). This occurs in contrast, or in addition, to the endogenous IL-6 produced in patients without a viral infection [El-Osta and Kurzrock, 2011]. IL-6 and vIL-6 have multiple roles and are known to stimulate B-cell proliferation through autocrine and paracrine signaling, induce production of vascular endothelial growth factor (VEGF), support hematopoiesis, decrease the synthesis of albumin, stimulate plasma cell differentiation or growth, and cause immune system dysregulation. IL-6 has been identified as a target for CD therapy because it plays a significant role in disease pathology, biochemical alteration, and symptomatology.

Anti-IL-6 monoclonal murine antibodies were initially found to have some efficacy in plasma cell disorders, but the use of these agents was limited due to neutralizing antimurine antibodies produced in humans [Klein et al. 1991]. The first use of an anti-IL6 monoclonal antibody in MCD was reported by Beck and colleagues. In this report, a patient with MCD and an elevated serum IL-6 concentration was treated with BE-8, a prolonged monoclonal antibody that targets IL-6. The treatment led to quick resolution of symptoms and signs of the disease, as well as resolution of most of the abnormal laboratory findings. Subsequently, the patient’s mesenteric mass was resected, leading to a sustained remission of the disease [Beck et al. 1994].

Additional research led to the development of the first humanized IL-6 receptor antibody, tocilizumab. Tocilizumab binds directly to the IL-6 receptor and inhibits binding of IL-6. In 2000 the initial data for tocilizumab were published [Nishimoto et al. 2000]. In this observational study, seven patients with multicentric plasmacytic or mixed type CD were treated with increasing doses of tocilizumab (1, 10, 50, and 100 mg per patient) until the maximum tolerated dose of 100 mg twice weekly was established. Upon treatment with the monoclonal antibody, the patients had an objective decrease in lymphadenopathy, as well as improvement in CRP, fibrinogen, anemia, thrombocytosis, hemoglobin, and albumin. Their constitutional symptoms also improved. These data established the foundation for using IL-6 as a target in this rare disorder. Tocilizumab was then investigated in a multicenter prospective trial in 28 patients with MCD, demonstrating that patients had an improvement in lymphadenopathy, hemoglobin, and albumin, as well as an improvement in constitutional symptoms [Nishimoto et al. 2005]. Tocilizumab was granted approval for treatment of MCD in Japan, but is not approved for this indication in the United States. Tocilizumab has also been used in the treatment of many other inflammatory conditions that are known to have hypercytokinemia, such as rheumatoid arthritis, juvenile idiopathic arthritis, Crohn’s disease [Venkiteshwaran, 2009] and cytokine release syndrome secondary to immunotherapy [Lee et al. 2014].

Siltuximab

More recently, siltuximab, a monoclonal antibody with high affinity and specificity for IL-6, was developed and approved in the United States and the European Union for the treatment of idiopathic MCD [Deisseroth et al. 2015]. This approval was based on the results of a placebo-controlled trial performed in HHV-8 and HIV negative patients [Van Rhee et al. 2014]. Here we review the pertinent clinical data that have been established during its development and early use.

Pharmacodynamic properties of siltuximab

Siltuximab is a chimeric, human-murine immunoglobulin monoclonal antibody produced in Chinese hamster ovary cells that binds to and neutralizes human IL-6 directly, thereby decreasing levels of unbound IL-6 and preventing binding to its receptor [FDA, 2014; Deisseroth et al. 2015]. When IL-6 binds to the soluble or membrane-bound receptor, a complex forms that binds gp130 and results in activation of the Janus kinase/signal transducer and activator of transcription (JAK/STAT) signaling pathway [Liu et al. 2014]. The activation of this pathway is inhibited by the use of siltuximab in MCD. Siltuximab is not used in virally associated MCD due to lack of efficacy in these patients, as vIL-6 does not bind to siltuximab and therefore remains free to bind the human IL-6 receptor. vIL-6 can also independently bind gp130 and activate the JAK/STAT pathway [Li et al. 2001], even in the absence of a bound IL-6 receptor, which leads to continued pathway activation despite the presence of siltuximab.

Although siltuximab targets IL-6, the drug efficacy cannot be measured by checking serum IL-6, as the neutralized IL-6 with bound antibody interferes with the commercially available IL-6 test results [FDA, 2014]. These tests also do not account for IL-6 presence within a tumor microenvironment or the IL-6 dependence of a tumor [Kurzrock et al. 2013], which likely explains why clinical data do not demonstrate a correlation between initial IL-6 or CRP levels and patient symptoms [Casper et al. 2015]. IL-6 is the primary driver of hepatic CRP production [Castell et al. 1990]. Although IL-6 cannot be used to monitor response to treatment, CRP can be used as a surrogate marker for serum IL-6 inhibition, as it has been demonstrated to decrease with siltuximab dosing. CRP reductions are reported as early as cycle 1 day 8. Maximum suppression of CRP was seen at a siltuximab dose equivalent to 12 mg/kg administered every 3 weeks [Kurzrock et al. 2013; Casper et al. 2015].

IL-6 is also known to induce hepcidin production, therefore these levels are also affected by siltuximab administration, with a median of 47% decrease in hepcidin levels at cycle 1 day 8 of siltuximab [Casper et al. 2015]. As siltuximab doses are increased, clinical data have demonstrated a decrease in hepcidin and resulting increases in hemoglobin, although there are likely additional IL-6 independent pathways that contribute to hemoglobin improvement [Kurzrock et al. 2013; Casper et al. 2015].

Drug–drug interaction studies have not been performed with siltuximab, but it should be remembered that siltuximab may improve or restore cytochrome P450 (CYP450) activity. If CYP450 activity was previously suppressed by elevated cytokines, then cytokine reduction may lead to increased metabolism of drugs that use the CYP450 pathway, such as warfarin. If possible, those drug levels should be checked and the dose adjusted as needed at the initiation, titration, or cessation of siltuximab. Practitioners should also use caution with drugs metabolized through the cytochrome P3A4 pathway as there may be potential effects on this pathway as well [FDA, 2014].

Pharmacokinetics of siltuximab

Siltuximab is approved in the United States to be administered at 11 mg/kg in a 1 h intravenous infusion every 3 weeks. The dosing interval may be extended to every 6 weeks in some cases, as demonstrated in the extension trial [Van Rhee et al. 2015]. Steady state of the drug is reached by the sixth dose and accumulates at approximately 1.7 fold that which is achieved after a single dose. The serum mean siltuximab concentration is 332 μg/ml at steady state, with a serum mean predose trough of 84 μg/ml. The central volume of distribution in a 75 kg man is 4.5 liters. Data indicate that siltuximab primarily distributes in the intravascular space with very little extravascular distribution [EMA, 2015].

The mean terminal half life of siltuximab is 20.6 days (range 14.2–29.7 days) after the initial dose.

Clearance of siltuximab is 0.23 liter/day via first-order elimination. Body weight has proven to be the only variable identified to affect clearance, although the drug has not been tested in end-stage renal disease or severe hepatic impairment. Age and sex have not been shown to affect pharmacokinetics.

Blood tests should be performed before each dose of siltuximab for the first year of treatment to ensure appropriate absolute neutrophil count, platelets and hemoglobin levels prior to treatment. After that time these labs can be checked prior to every third cycle.

Patients with renal dysfunction (creatinine clearance ⩾15 ml/min) do not require any dose adjustment compared with patients with normal renal function (creatinine clearance ⩾90 ml/min). The drug has not been tested in those with creatinine clearance less than 15 ml/min. Mild to moderate hepatic impairment also requires no dose adjustment, but those with severe hepatic dysfunction were not included in clinical trials [FDA, 2014; EMA, 2015].

This drug has a category C rating and, when possible, should be avoided in pregnant or nursing patients. This drug has not yet been tested in the pediatric population and, although it has been given in the geriatric population, there are not adequate data to rule out any differences in efficacy in older patients.

Clinical data for siltuximab

The initial activity of siltuximab was reported by Van Rhee and colleagues in patients with symptomatic, multicentric or unresectable, unicentric Castleman’s disease (Table 1) [Van Rhee et al. 2010]. This was an open-label, phase I dose escalation trial with seven cohorts. The 23 patients in the trial were treated at 1-, 2- or 3-week intervals. Clinical benefit or response to treatment was based on improvements in hemoglobin, fatigue, anorexia, fever, night sweats, weight loss, and lymph node size. Based on these criteria, 18 out of 23 patients (78%) had a clinical benefit, with 12 patients (52%) having an objective tumor response. All 11 patients treated with the highest dose of 12 mg/kg had a clinical benefit response and eight of these patients had an objective tumor response. The response duration ranged from 44 to over 889 days, with one patient maintaining a complete response for more than 318 days. On average, patients achieved a decrease in serologic markers of inflammation within 3 weeks. The median time to a clinical partial response was 6 months. No dose-limiting toxicity was found and no treatment-related deaths were reported. Three patients (13%) had mild infusion reactions, which did not result in cessation of siltuximab. The most common adverse events were nausea (39%), upper respiratory tract infection (35%), vomiting (30%), hypertriglyceridemia (30%), arthralgias (30%), and headache (30%). Three patients developed a grade 3 or higher adverse event reasonably related to siltuximab, but no serious adverse events were attributed to siltuximab. This trial was the first to demonstrate that siltuximab is an effective treatment for CD, with a favorable side-effect profile.

Table 1.

Siltuximab clinical trials.

Study	Histologic subtype	Clinical subtype	Number of patients	Response	Duration of response
Van Rhee et al. [2010]	10 (44%) hyaline vascular, 12 (25%) plasmacytic, 1 (4%) mixed	22 (96%) multicentric, 1 (4%) unicentric (unresectable)	23 (single-arm siltuximab)	18 (85%) with clinical benefit	Partial response range from 44 to >889 days, complete response more than 318 days
Van Rhee et al. [2014]	18 (34%) hyaline vascular, 13 (25%) plasmacytic, 22 (42%) mixed	53 (100%) multicentric	79 total: 26 placebo, 53 siltuximab	18 (34%) with durable tumor and symptomatic response: 1 complete and 17 partial	Median duration of response 383 days

Note: all patients included in the trials were human herpes virus 8 and HIV negative. Clinical benefit response was defined as improvement from baseline (with no worsening in other measures) in at least one of the following: ⩾2 g/dl increase in hemoglobin; ⩾1 grade decrease in fatigue; ➮1 grade decrease in anorexia; ➮2°C decrease in fever or return to 37°C or improvement in night sweats; ➮5% increase in weight; or ➮25% decrease bidimensionally in size of largest lymph node. Durable tumor and symptomatic response was defined by Cheson criteria [Cheson et al. 2007], which was adjusted to include assessment of cutaneous lesions.

A subsequent randomized, double-blind, placebo-controlled trial compared the use of siltuximab with best supportive care in 79 patients with symptomatic MCD (Table 1) [Van Rhee et al. 2014]. Patients were randomly assigned (2:1) to siltuximab and best supportive care versus placebo and best supportive care. Siltuximab was given at a dose of 11 mg/kg by intravenous infusion every 3 weeks. The Cheson criteria (adjusted to include assessment of skin lesions) were used to assess patient response to treatment [Cheson et al. 2007]. The median patient age was 48 years with 58% of patients previously treated with at least one therapy. Of the 53 patients receiving siltuximab, durable tumor and symptomatic responses, as determined by independent review, were seen in 18 patients (34%), compared with no patients in the placebo arm (p = 0.0012). One patient (2%) had a complete response and 17 patients (32%) had a partial response. Eight of the thirteen patients (62%) with plasmacytic subtype, 10 of the 22 patients (45%) with mixed subtype, and none of those with the hyaline vascular subtype had a durable and symptomatic response to therapy. The median duration of response was 383 days (range 232–676). Time to durable symptomatic response was 170 days in the siltuximab arm. Median duration of treatment could not be assessed at the time of study publication, as only 1 of 18 responding patients had progressed on therapy. Response rates did not correlate with baseline IL-6 or CRP levels despite the significant decline seen in CRP in patients treated with siltuximab. The incidence of grade 3 or higher adverse events was similar in both groups, with three patients (6%) with a serious adverse event potentially related to siltuximab. The most common side effects in the siltuximab arm compared with the placebo arm were pruritus (42%), increased weight (21%), rash (34%), localized edema (32%), and upper respiratory infection (36%). With the data from this single trial, siltuximab was approved by the Food and Drug Administration (FDA) in April 2014 in the United States for treatment of idiopathic MCD.

The maximum duration of treatment with siltuximab has not yet been established. An extension trial from the initial phase I trial included 19 patients and showed that siltuximab could be well tolerated and efficacious even with continued use for a median of 5.1 years (range 3.4–7.2) [Van Rhee et al. 2015]. Eleven patients were continued on the original treatment regimen with 11 mg/kg administered every 3 weeks. Eight patients were transitioned to every 6 week dosing without compromising the disease response. All patients maintained disease control during this time, with no reported deaths. Grade 3 and higher adverse events that occurred in more than one patient included hypertension (n = 3), nausea (n = 2), cellulitis (n = 2), and fatigue (n = 2). Three grade 3 and higher adverse events were potentially related to siltuximab and included leukopenia, lymphopenia, and polycythemia (n = 1 each).

Adverse reactions with siltuximab

Adverse reactions, as reported in the clinical trials above, were typically minimal. In the phase I trial, there were three adverse events (13% of participants) of grade 3 and higher that were reasonably related to siltuximab. Two patients (9%) discontinued the treatment due to adverse events (grade 4 abdominal pain and diagnosis of non-Hodgkin’s lymphoma), which may not have been related to treatment. No patient deaths were attributed to therapy. A similar trend was seen in the phase II trial, which reported three patients (6%) with serious adverse events reasonably related to siltuximab, including lower respiratory tract infection, anaphylactic reaction, and sepsis. Again, no treatment-related deaths were reported. A total of seven patients from the two trials had a mild infusion reaction, with one patient reported to have a grade 3 anaphylactic reaction. The most common adverse events seen in over 30% of patients in these trials were pruritus, rash, weight gain, upper respiratory infection, localized edema, and nausea. Many of these adverse events were also demonstrated at a similar rate in the patients receiving only placebo and supportive care. There is no evidence of cumulative toxicity in patients with prolonged exposure to the medication and no increase in serious adverse events has been reported with continued use [Kurzrock et al. 2013; Van Rhee et al. 2015].

Dosage recommendations

The FDA approval for this medication is for a dose of 11 mg/kg every 3 weeks, although the extension trial shows continued benefit with a 6-week dosing interval. The dose of 11 mg/kg, although different from that in the original trial, is considered to be an equivalent dose due to the approximately 9% difference in the absorption constant used to calculate the drug in the previously mentioned phase I clinical trial [Kurzrock et al. 2013; Markham and Patel, 2014]. The literature supports prolonged use of siltuximab, as needed for treatment of clinical signs or symptoms, with high likelihood of maintaining disease response and low risk of adverse events associated with treatment.

Future direction

Although currently approved for HHV-8 negative MCD, it will be interesting to evaluate the role of siltuximab in other settings. For example, in unresectable unicentric CD, one can contemplate tumor cytoreduction by siltuximab therapy followed by resection. Additional investigation is also warranted to identify effective therapy in those 50% of the patients in whom siltuximab is not effective. Areas of investigation would include combination therapies with other agents. This may include combination with other antibodies such as rituximab, or those targeting other cytokines with upregulated activity in CD, including IL-1, VEGF, and TNFα [El-Osta et al. 2010].

Conclusion

Therapeutic options for CD have been limited in the past and research of newer therapies has been hindered by the rarity of the disorder and the wide array of histologic and clinical presentations. IL-6 has become a promising therapeutic target for the treatment of MCD and antibodies targeting this pathway have provided significant improvement in clinical manifestations and constitutional symptoms, as well as objective tumor response in MCD. Siltuximab, the only FDA-approved medication for the treatment of MCD, provides long-term efficacy for many patients with minimal adverse effects. The approval of siltuximab is a promising step forward in the development of potential therapies for this rare disease.