Mixed cryoglobulinemic disease was initially described by Meltzer and Franklin (1) and by Brouet and colleagues (2) in the late 1960s and early 1970s, when they characterized in detail the serum cryoglobulins in patients with cryoglobulinemia of various causes. Among this group were patients with type II and type III serum cryoglobulins, which contained a mixture of IgM and IgG rheumatoid factor (RF), and a clinical picture of purpura, skin necrosis, Raynaud's phenomenon, arthralgia, asthenia, glomerulonephritis, chronic hepatitis, and neuropathy. While many of these cases were associated with hematologic or immune-mediated disease, cases of mixed cryoglobulinemia without apparent cause were termed essential mixed cryoglobulinemia. For the next 2 decades, patients with essential mixed cryoglobulinemia were treated empirically with corticosteroids, plasmapheresis, and cytotoxic drugs according to the severity of organ system involvement, without realizing the potential negative impact of these interventions on the chronic liver disease.
The landscape changed radically in 1989 with the discovery of the hepatitis C virus (HCV) (3) and the subsequent breakthrough by Ferri and colleagues (4) showing that a high proportion of patients with mixed cryoglobulinemia and chronic liver disease had serologic evidence of HCV infection. It became apparent shortly thereafter that antiviral treatment of HCV-related mixed cryoglobulinemic vasculitis with interferon-α (IFNα) transiently lowered the viral load of HCV RNA in about 50–60% of patients, leading to improved clinical and laboratory manifestations of the vasculitic syndrome (5–7). However, soon after stopping IFNα therapy, the vasculitic manifestations reoccurred with the reappearance of circulating HCV RNA.
Over the next decade, substantial progress was made in the treatment of chronic HCV infection. Combination therapy with IFNα and ribavirin was shown to be more efficacious than IFNα monotherapy (8). PEGylated IFNα replaced the unmodified IFNα molecule because of its longer half-life and greater efficacy in eradicating the virus. A prospective study showed that HCV-related cryoglobulinemic vasculitis could be successfully treated with a 6-month course of PEGylated IFNα plus ribavirin in 62.5% of cases, with a sustained virologic response in 58.3% and clearance of cryoglobulins in 45.7% (9). However, these outcomes, while an advance over the previous era, left many nonresponders in a tenuous position without alternative therapy. Moreover, in 10–20% of patients, PEGylated IFNα and ribavirin therapy was poorly tolerated because of side effects, and other patients were not suitable candidates for this antiviral regimen because of contraindications, such as severe psychiatric disease. In addition, the predominant variant in developed countries, HCV genotype 1, had a lower likelihood of responding to antiviral therapy than HCV genotypes 2 and 3.
Since the cold-dependent insolubility of the type II and type III cryoglobulins is dependent on IgM-RF, it may be hypothesized that eliminating the expanded B cell clone or clones producing the pathogenic RF will abrogate cryoglobulin formation, inhibit immune complex–mediated inflammation, and ameliorate the signs and symptoms of vasculitis. Rituximab, a chimeric anti-CD20 monoclonal antibody, depletes more than 95% of circulating B cells and has been shown to lower serum levels of IgM-RF by 30–60% in patients with rheumatoid arthritis (10). Rituximab therefore seems fit for the task of interrupting this chain of events initiated by a virally triggered B cell clonal expansion and culminating in a chronic inflammatory response. In this issue of Arthritis & Rheumatism, the findings of two randomized controlled trials evaluating the clinical efficacy and safety of rituximab therapy for cryoglobulinemic vasculitis are reported (11, 12). Rituximab therapy was shown in both studies to be vastly superior to conventional immunosuppressive drugs for treating cryoglobulinemic vasculitis, providing further evidence of its clinical efficacy and safety in this setting. These results, while compelling, also leave us with a host of tantalizing questions about the immunologic mechanisms at play in HCV-related cryoglobulinemic vasculitis and the place of rituximab among the inventory of possible therapies for this disease.
These two studies approached the question about the efficacy and safety of rituximab therapy for cryoglobulinemic vasculitis somewhat differently. The open-label, randomized controlled trial described by De Vita and coworkers from Italy (11) compared the use of rituximab (1 gm intravenously on days 0 and 14) with conventional therapy (corticosteroids, plasmapheresis, azathioprine, or cyclophosphamide) in 57 patients with severe manifestations of cryoglobulinemic vasculitis. Evidence of HCV infection was not required for entry, but 53 of the 57 patients that were enrolled in this study had positive findings on tests for anti-HCV antibodies or serum HCV RNA. The trial was stopped early, after an interim analysis revealed a large treatment difference between the two groups. By intent-to-treat analysis, the proportion of patients who continued to take their initial therapy at 12 months was significantly higher in the rituximab arm compared with the conventional therapy arm (64.3% versus 3.5%; P < 0.0001). It was not possible to directly compare the clinical response rates in the two arms due to the substantial proportion of early dropouts from the conventional therapy arm because of treatment failure. Only 13.8% of the patients in the conventional therapy arm continued their initially assigned therapy beyond 3 months.
Caution must be exercised in interpreting these results because of the open-label study design, which may have biased against the survival of conventional therapy. Patients in whom treatment failed, as determined by the investigators, had the opportunity to crossover to the rituximab arm, a potentially appealing choice for both the investigators and patients in this study. As expected, rituximab therapy produced the desired effect on disease mechanisms, lowering serum RF activity and increasing C4 levels. However, further conclusions about the effects of rituximab on these mechanisms were limited, owing to the absence of any information about virologic responses and other immunologic parameters, such as serum cryoglobulin levels.
Sneller and colleagues from the National Institutes of Health (NIH) (12) also report in this issue the results of their open-label, randomized controlled trial of rituximab versus conventional therapy in 24 patients with HCV-related cryoglobulinemic vasculitis. The patients in this study were randomly allocated to receive treatment with 4 weekly infusions of rituximab 375 mg/m2 or standard therapy (12). Standard therapy was considered to be maintenance or intensification of conventional immunosuppressive therapy. Unlike the study by De Vita et al (11), the patients receiving rituximab in the NIH trial were allowed to continue taking their background immunosuppressive therapy. The NIH investigators also took a different tack in their study design by choosing clinical remission at 6 months as their primary end point, rather than survival of therapy. Clinical remission was defined as a Birmingham Vasculitis Activity Score of 0, indicating no new or worsening disease activity within the previous month or persistently active disease (13). The study population was roughly comparable to that of the other trial and included patients with HCV-related active vasculitis in whom previous treatment with antiviral therapy had failed because of lack of efficacy or intolerance. Almost 8 years were needed to enroll the 24 patients in this trial, attesting to the lower prevalence of HCV-associated cryoglobulinemic vasculitis in the US compared with southern Europe.
Consistent with the results reported by De Vita et al (11), the difference in clinical efficacy was shown to be strikingly in favor of rituximab over conventional immunosuppressive therapy, with an 83.3% response rate at 6 months compared with 8.3% (P < 0.001). Since most of the patients in this trial continued to take their assigned treatment through 6 months, it was possible to directly compare the clinical response rates of the two treatments, whereas in the other trial, survival of therapy was used as a surrogate end point. Serum cryoglobulin levels declined in the rituximab group, along with an increase in total hemolytic complement. However, relapses were not closely tied to a rise in serum cryoglobulin levels, a finding which is at variance with other studies showing a correlation between clinical relapse and increase in serum cryoprecipitates after successful treatment with rituximab and antiviral agents (14–16).
The concerns about the safety of rituximab therapy for cryoglobulinemic vasculitis center mainly on the risk of exacerbating the HCV infection and worsening the underlying liver disease. No alarm bells were sounded in either study, although this question was not addressed in any depth. De Vita et al (11) did not evaluate HCV viral load in their trial. However, reassuringly, no mention was made of worsening liver disease. Sneller and colleagues (12) found the HCV viral load to be unchanged following rituximab therapy and noted only mild elevations in serum transaminase levels, which were no different between the two treatment groups.
Most previous studies about the safety of a single course of rituximab therapy for this indication do not point to any major liver toxicity issues. For example, in an earlier uncontrolled study, rituximab therapy was evaluated in 19 patients with HCV-related cryoglobulinemic vasculitis and advanced liver disease (17). Without the cover of antiviral agents, the use of rituximab was associated with improvement in the clinical manifestations of systemic vasculitis and at least some of the complications of cirrhotic liver disease (e.g., ascites, serum albumin concentrations) despite transient increases in the HCV viral load. Other studies have raised the specter of serum sickness (18). Although serum sickness was not observed in either of the two clinical trials reported herein, it has been previously described following the use of rituximab therapy for cryoglobulinemic vasculitis. Intriguingly, rituximab may form a complex with IgMκ mixed cryoglobulin, resulting in enhanced cryoprecipitation and severe systemic reactions, including serum sickness (19). Patients with the highest cryoglobulin levels and the lowest C4 levels seem to be particularly prone to this adverse event.
HCV-related mixed cryoglobulinemic disease is considered to be a “benign” lymphoproliferative B cell disease, which provides a solid scientific ground for tackling it with a B cell–depleting therapeutic strategy. It also has malignant potential and may evolve into a full-fledged lymphoma. HCV-associated cryoglobulinemic vasculitis has been associated with an increased risk of non-Hodgkin's B cell lymphoma, most commonly a low-grade marginal-zone lymphoma, such as splenic lymphoma with villous lymphocytes and mucosal-associated tissue lymphomas, but also lymphoplasmacytic lymphoma and diffuse large B cell lymphoma (20). Thus, it serves as a good model for a viral-induced malignancy. Indeed, effective antiviral therapy has led in some cases to the regression of HCV-associated low-grade marginal-zone lymphoma (21, 22).
Patients with mixed cryoglobulinemic disease, as well as patients with chronic HCV, harbor B cell clonal expansions in the liver, bone marrow, and peripheral blood, exhibiting a highly diverse set of antigen receptors, which are somatically hypermutated and show features of an antigen-driven response (23). The aberrantly expanded B cell clones are IgM+CD27+ and contain an overrepresentation of VH1–69 and Vκ3–20 variable genes that encode RFs of the Wa cross-reactive idiotype (24). In a small study, the majority of cases of type II mixed cryoglobulinemia were found to be associated with monoclonal RFs bearing the Wa cross-reactive idiotype (25). This clonal expansion of B cells likely results from chronic antigenic stimulation and not from direct infection or transformation. It has been shown that the HCV envelope protein E2 binds human CD81 on naive B cells, resulting in cellular activation (26). CD81 forms a costimulatory complex with CD19, CD21, and CD225 that reduces the threshold for B cell activation by bridging antigen and CD21-mediated complement recognition.
Rearrangement of the bcl-2 gene via a t(14:18) translocation may also be important in the mechanisms underlying B cell expansion during HCV infection. This translocation event, a frequent genetic aberration in lymphoma, has been shown to occur with higher frequency in patients with HCV-associated mixed cryoglobulinemia than in those with chronic HCV infection but without mixed cryoglobulinemia (75.7% versus 37.6%) (27). The t(14:18) translocation may disappear with effective antiviral therapy and therefore appears to be driven by HCV infection. Based on these observations, it seems likely that HCV-containing immune complexes are stimulating the clonal expansion of RF-positive B cells in this disease. However, at some stage in the disease process, B cell expansion may become independent of HCV stimulation, since some patients with cryoglobulinemic vasculitis may experience a disease relapse despite clearance of their HCV infection (14).
How should rituximab be incorporated into the management of HCV-related cryoglobulinemic vasculitis? Most would agree that the cornerstone of treatment is antiviral therapy: PEGylated IFNα and ribavirin for 48 weeks (for HCV genotypes 1, 4, 5, and 6) or 24 weeks (for HCV genotypes 2 and 3) in the absence of contraindications, such as severe psychiatric disease or cytopenias. Using this approach, a sustained virologic response is obtainable in 40–50% of patients with genotype 1 and ∼80% or more of those with genotypes 2 or 3 (28). Antiviral therapy may suffice for the treatment of mild-to-moderate cryoglobulinemic disease, combined with low doses of prednisone to control the signs and symptoms of vasculitis until a virologic and clinical response are realized.
Based on the results of the clinical trials discussed herein, rituximab may be an attractive option for managing severe cryoglobulinemic disease in patients ineligible for antiviral therapy and failing treatment with conventional immunosuppressive agents. One additional advantage of using rituximab for the treatment of this disease may be a reduced exposure to corticosteroids. The responders to rituximab therapy from both the NIH and Italian studies appeared to receive lower total doses of prednisone than did those allocated to conventional immunosuppressive therapy. Although not directly addressed by these two clinical studies, a case can also be made for considering rituximab as first-line therapy for the treatment of organ- or life-threatening disease, in contrast to embarking on treatment with standard immunosuppressive drugs such as azathioprine or cyclophosphamide.
Should rituximab be used in combination with antiviral therapy? To address this question, Dammacco and coworkers (15) performed a randomized controlled trial involving 37 subjects with HCV-associated cryoglobulinemic vasculitis in which background PEGylated IFNα and ribavirin therapy was used, with or without rituximab at a dosage of 375 mg/m2 weekly for 4 weeks, followed by two infusions of 375 mg/m2 at months 5 and 10. At month 12, they found that significantly more patients in the rituximab group than the no rituximab group had achieved a complete response (54.5% versus 33.3%; P < 0.05). It is worth noting that the complete clinical responders in this study showed complete disappearance of the cryoprecipitates, a complete virologic response, and loss of serum RF activity. Similar outcomes were obtained in a prospective cohort study of patients with HCV-related cryoglobulinemic vasculitis receiving a combination of antiviral agents and rituximab (29). Thus, the combination of rituximab plus antiviral agents may be emerging as the best choice for treating patients with severe disease manifestations of cryoglobulinemic vasculitis, provided no new safety concerns are raised using this approach.
The recent development of two direct-acting antiviral agents, boceprevir and telaprevir, has changed the playing field for treatment of genotype 1 HCV infection. Both of these drugs inhibit the HCV nonstructural protein 3-4A serine protease, and when used in combination with PEGylated IFNα and ribavirin (triple therapy), increase the sustained virologic response rates in genotype 1 HCV infection to more than 70% (30). For treatment of genotype 1 HCV infection, the recommendations now call for the use of boceprevir or telaprevir in combination with PEGylated IFNα and ribavirin (triple therapy) for both treatment-naive and treatment-experienced patients (30). There are as yet no published studies examining the role of triple therapy in HCV-related cryoglobulinemic vasculitis, but it seems likely that this more effective antiviral regimen should improve outcomes to the extent that they are influenced by achieving a sustained virologic response. With these new advances, further studies of treatment with combined antiviral and immunomodulatory agents are likely on the horizon. These will not only improve the virologic and immunologic responses in HCV-related cryoglobulinemic vasculitis, but also dramatically increase the cure rate for this disease.