|Home | About | Journals | Submit | Contact Us | Français|
To report on the results of a randomized controlled trial of rituximab in hepatitis C virus (HCV)-associated mixed cryoglobulinemic vasculitis.
We conducted an open-label single center randomized controlled trial of rituximab (375 mg/m2 per week for 4 weeks) compared to best available therapy for treatment of patients with HCV-associated cryoglobulinemic vasculitis in whom antiviral therapy failed to induce remission. The primary endpoint was remission at 6 months from study entry.
A total of 24 patients were enrolled. Baseline disease activity and organ involvement were similar in the two groups. Ten patients in the rituximab group (83%) were in remission at study month 6, compared with 1 patient in the control group (8%), a result that met criterion for stopping the study (P<0.001). The median duration of remission for rituximab-treated patients reaching the primary endpoint was 7 months. No adverse effect of rituximab on HCV plasma viremia or hepatic transaminase levels was observed.
Therapy with rituximab was well tolerated and effective treatment for patients with HCV-associated cryoglobulinemic vasculitis in whom antiviral therapy fails to induce remission.
Chronic infection with hepatitis C virus (HCV) is a world wide health problem, affecting an estimated 130 million people (1). Up to 20% of individuals with chronic HCV infection will develop potentially fatal complications such as cirrhosis, liver failure, or hepatocellular carcinoma. Extra-hepatic manifestations of HCV infection are also common, occurring in up to 40% of patients and contributing to the morbidity and mortality associated with this chronic infection (2). One such extra-hepatic manifestation is a form of small vessel vasculitis associated with mixed cryoglobulinemia. HCV-associated cryoglobulinemic vasculitis is an uncommon complication of chronic HCV infection characterized by the clonal expansion of B cells that produce IgM rheumatoid factor (RF) (3, 4). The RF produced by the expanded population of B cells plays a central role in the development of vasculitis by promoting the formation of immune complexes consisting of RF, HCV, and polyclonal HCV-specific IgG. This cryoglobulin complex is deposited in blood vessel walls or glomerular capillaries triggering an inflammatory cascade that results in the syndrome of cryoglobulinemic vasculitis (5). A spectrum of disease manifestations and severity can occur in HCV-associated cryoglobulinemic vasculitis, with the primary clinical features being cutaneous vasculitis, arthralgia/arthritis, peripheral neuropathy, and membranoproliferative glomerulonephritis (6).
This lymphoproliferative disorder is driven by chronic HCV infection. Anti-viral therapy with PEGylated interferon alpha and ribavirin results in sustained remission of cryoglobulinemic vasculitis in nearly all cases where a sustained virologic response is achieved (7). However, the effectiveness of current antiviral therapy is limited by toxicity, and the failure to achieve a sustained virologic response in more than 50% of patients infected with HCV genotype 1, the most prevalent genotype in the Americas and Europe (8). For patients who do not respond to antiviral therapy, conventional immunosuppressive therapy with glucocorticoids or cytotoxic agents is ineffective at producing sustained remissions in the vast majority of cases (9–11). In addition, immunosuppressive therapy may accelerate progression of the underlying HCV liver disease. Thus, there is an important unmet need for safer and more effective treatment for patients with HCV-associated cryoglobulinemic vasculitis who do not respond to antiviral therapy.
Rituximab is a chimeric monoclonal antibody directed against CD20, which results in rapid depletion of circulating and tissue B cells. Based on this mechanism of action, rituximab has the potential to deplete the expanded population of B cells that develop in HCV-associated vasculitis thereby reducing the production of pathogenic RF and formation of the cryoglobulin immune complex. A number of published cases and uncontrolled cohort studies have reported encouraging results with the use of rituximab in mixed cryoglobulinemic vasculitis (12–17). However, these reports included some patients with non-HCV associated cryoglobulinemic vasculitis and used varying dosing regimens often in combination with other immunosuppressive or antiviral therapies. In addition, concern has been raised that treatment with rituximab may increase HCV replication (12). To address these issues, we conducted a prospective randomized controlled trial to examine the safety and efficacy of rituximab for treatment of patients with HCV-associated cryoglobulinemic vasculitis in whom prior antiviral therapy failed to induce disease remission.
The study was an open-labeled, randomized, controlled, single-center trial involving 24 patients treated at the National Institutes of Health (NIH) Clinical Center, Bethesda, MD. Inclusion in the study required the presence of active manifestations of HCV-associated cryoglobulinemic vasculitis as evidenced by one or more of the following: cutaneous vasculitis, peripheral neuropathy, or glomerulonephritis. Only patients in whom treatment with interferon alpha and ribavirin failed to induce a response or who could not tolerate this therapy were eligible for the study. Exclusion criteria included change in immunosuppressive therapy within four weeks of study entry, prior use of rituximab, diagnosis of lymphoma, severe renal insufficiency (creatinine clearance less than 30 ml/min), severe hepatic insufficiency (Childs-Pugh class B or C), co-infection with HIV or hepatitis B virus, liver transplantation, pregnancy, active systemic infection, or presence of potentially life-threatening vasculitis involving the central nervous system, heart, or gastrointestinal tract. A total of 47 patients were screened for randomization into this study. Eighteen patients did not meet one or more of the above eligibility criteria. Three eligible patients elected not to enroll in the study because of concerns about potential rituximab toxicity. One patient experienced a myocardial infarction prior to randomization and was lost to follow-up; one additional patient did not return for randomization and was lost to follow-up. The remaining 24 patients were enrolled in the study and underwent randomization.
Patients were randomly assigned in a 1:1 ratio to receive either investigational therapy (rituximab) or to continue best available therapy (control group). The NIH Clinical Center Pharmacy performed the randomizations.
The trial was sponsored by National Institute of Allergy and Infectious Disease (NIAID) and approved by the NIAID Institutional Review Board. All patients provided written informed consent. The Regulatory Compliance and Human Subjects Protection Branch, Division of Clinical Research, National Institute of Allergy and Infectious Disease monitored the study. The study drug (rituximab) was purchased by the NIH Clinical Center Pharmacy on the open market. The study was designed and carried out by the first and last author; all three authors were involved in data analysis and manuscript preparation.
Patients randomized to the investigational therapy group received rituximab 375 mg/M2 on days 1, 8, 15 and 22 beginning at the time of randomization (day 0). Patients were allowed to continue any immunosuppressive medications they were receiving at randomization, but could not increase the dose of these medications or institute new immunosuppressive therapy or plasma exchange. All patients were pre-medicated with acetaminophen (650 mg) and diphenhydramine (50 mg) prior to each rituximab infusion. Pre-medication with glucocorticoids was not given to any patient. Patients in the control group were maintained on any immunosuppressive therapies that they were receiving at the time of enrollment and were allowed to increase or initiate new immunosuppressive treatments as needed to manage worsening disease activity. After study month 6, patients randomized to the control group were offered treatment with rituximab if they continued to have active manifestations of vasculitis.
Following an initial screening visit to determine eligibility, patients returned to the NIH Clinical Center for randomization and monthly thereafter for 12 months. Patients randomized to the rituximab group were also seen weekly during the first month at the time of rituximab administration. All patients underwent comprehensive clinical and laboratory evaluations at each monthly study visit. Laboratory evaluations included routine chemistries, urinalysis, complete blood counts, cryoglobulin levels, peripheral blood flow cytometry, and HCV plasma RNA levels. Remission was defined as a Birmingham Vasculitis Activity Score (BVAS) of 0 indicating no new or worsened activity or no persistently active disease manifestations within the previous month (18, 19). Relapse was defined as the recurrence of disease activity after a period of remission as manifested by a BVAS score > 0. The primary endpoint of the study was remission at study month 6. Patients who withdrew from the study prior to month 6 were considered to have active disease regardless of their clinical status at the time of withdrawal. Secondary endpoints were duration of remission and severe adverse events.
Serial HCV RNA levels in plasma were measured by the VERSANT HCV RNA 3.0 Assay (Bayer Diagnostics, Puteaux, France).
Sample size estimation was based on the primary endpoint, i.e., remission at 6 months from study entry. The only previous standardized therapeutic study indicated a remission rate of <5% at 6 months under either no treatment or treatment with glucocorticoids (11). Assuming a remission rate of 5% in the control group and 50% in the rituximab group, a sample size of 15 patients per group was needed to attain 90% power at nominal level of two-sided alpha of 0.05. In September of 2010 an interim analysis was requested by the NIAID Institutional Review Board. Because there was no pre-specified interim analysis plan for this study, a conservative efficacy boundary requiring strong evidence for early stopping was used. A modified Haybittle -Peto (20) analysis was carried out in which the declaration of efficacy at an interim analysis required the two-tailed p-value to be .001 or less. This interim analysis, done with 12 subjects in each group showed a higher than anticipated remission rate under rituximab and indicated efficacy of rituximab with p < 0.001. The study was thus stopped with 12 patients per group for early evidence of benefit following the modified rule of Haybittle -Peto (20).
Analyses were performed on intention-to-treat basis unless stated otherwise. As primary analysis, we reported the remission rate at month 6 under each group and compare by Fisher’s exact test. Among those attaining remission under rituximab, duration of remission was summarized by Kaplan-Meier survival curve. Box plots were provided for the distribution of BVAS and cryoglobulin levels and compared by Wilcoxon rank sum test between the two groups. Median plasma HCV levels and mean peripheral blood B cell counts with 95% confidence intervals (CI) from lognormal distribution were plotted to depict the respective time trends.
Between June 2002 and April 2010, a total of 24 patients were enrolled in the study. Twelve patients randomized to the rituximab group and 12 patients to the control group. Nine of the patients in the rituximab group and 10 of the patients in the control group were infected with HCV genotype 1. The remaining patients in both groups were infected with HCV genotype 2. One patient in the rituximab group and 2 in the control group were not able to complete a full course of antiviral therapy due to toxicity. For the remaining patients, antiviral therapy failed to induce sustained virologic response. The baseline characteristics of the patients are listed in Table 1. The two groups were balanced with respect to clinical manifestations, disease activity, laboratory values, and exposure to glucocorticoids and other immunosuppressive therapies (Table 1). All 12 patients in the rituximab group and 11 patients in the control group had more than one organ system involved at study entry.
The primary endpoint of the study was the number of patients in remission at study month 6. Ten of the 12 patients in the rituximab group (83.3%; 95% CI, 51.6% to 97.9%) reached the primary endpoint as compared with 1 of 12 patients in the control group (8.3%; 95% CI, 2.0% to 38.6%), indicating significantly higher remission with rituximab treatment (p<0.001). Two patients in the rituximab group were not in remission at month 6. One patient had to discontinue rituximab after 2 infusions due to severe febrile infusion reaction and withdrew from the study at month 5. The second patient achieved remission at study month 4, but subsequently experienced a relapse of cutaneous vasculitis at study month 6.
At baseline, BVAS scores were comparable between the two groups (p =0.977), but became significantly lower in the rituximab group starting at month 4 with p values <0.02 (Figure 1A). The duration of remission for all patients in the rituximab group is illustrated in Figure 1B. Of the 10 rituximab-treated patients in remission at month 6: six remained in remission until the end of study participation. Three patients experienced a relapse after month 6 and one patient was lost to follow-up while in remission at study month 7. The median duration of remission for the 10 patients meeting the primary endpoint was 7 months (interquartile range, 4.5 to 10). The 3 patients who relapsed after month 6 were treated with a second course of rituximab on an extension phase of the study. All 3 patients achieved remission and remained in remission for more than 6 months following the second rituximab treatment.
No patient in the rituximab group increased or initiated immunosuppressive therapy following study entry. Of the 6 patients in the rituximab group who were receiving prednisone at study entry, 5 tapered and discontinued this medication during the study; 1 patient who had been on long-term glucocorticoid therapy remained on 5 mg per day of prednisone due to concerns for secondary adrenal insufficiency. Two patients in the rituximab group were being treated with plasma exchange at study entry but discontinued this treatment by study month 2. No patient in the control group discontinued prednisone treatment. One patient in the control group initiated plasma exchange at study month 5 because of worsening cutaneous vasculitis. This treatment did not result in disease remission.
Peripheral blood B-cell depletion (defined as < 0.5% CD19+ cells) occurred in the 11 rituximab treated patients who received all four infusions. Return of B cells in the peripheral blood was detected between months 4–6 in most patients with pre-treatment levels being reached by month 8–12. The 3 patients who received a second course of rituximab showed similar depletion and recovery kinetics following the second treatment (data not shown). Cryoglobulin levels were similar at baseline (p=0.684) between the two groups but became significantly lower in the rituximab group starting at month 2 (with p values <0.05; Figure 2). Return of cryoglobulin levels did not correlate with relapses. All 3 patients in the rituximab group who relapsed after month 6 had a return of cryoglobulin levels prior to the relapse as did 4 of the 6 patients who remained in remission. Total complement levels increased in the rituximab group over the course of treatment from a median of 9 CAE units (interquartile range, 9 to 55) pre-treatment, to a median of 74 CAE units (interquartile range, 36 to 129) at month 6. In contrast, total complement levels in the control group remained depressed (median month 6 value 15 CAE units, interquartile range 9 to 56). The 3 patients in the rituximab group who relapsed after month 6 had a fall in complement levels prior to relapse as did 1 of the 6 patients who remained in remission.
Rituximab treatment did not appear to effect HCV replication as measured by plasma HCV RNA levels. The median plasma HCV RNA level was higher at baseline in the rituximab group, but did not reach statistical significance (Table 1). No statistically significant difference was observed in the change in plasma HCV RNA levels from baseline between the rituximab and control group over the follow up evaluations (Figure 3).
Table 2 summarizes selected adverse events seen in the first 6 months of the study. Only one patient experienced a severe adverse event related to rituximab infusion, which was a grade 4 fever. This patient developed rigors followed by a fever to 40.5° C during his third rituximab infusion. The infusion was stopped and ibuprofen administered. The fever and rigors resolved within one hour. No patient in either group, developed serious infection, or required hospitalization for treatment of infection during the first six months of the study. One patient in the rituximab group developed probable viral bronchitis that resolved without antimicrobial therapy; 2 patients in the control group developed presumed bacterial sinusitis and were treated with oral antibiotics. Elevations in hepatic transaminases were generally mild, consistent with underlying HCV infection, and were similar between the two study groups. No patient in either group developed clinical or laboratory evidence of worsening hepatic function.
No patient in the rituximab group developed hypogammaglobulinemia while on study. Two patients in the rituximab group had significant hypogammaglobulinemia (IgG < 200 mg/dl) at study entry thought to be secondary to nephrotic syndrome. Both patients experienced a > 2 fold increase in serum IgG levels following rituximab treatment.
Of the 4 patients with glomerulonephritis in the control group, all experienced a decline in the estimated GFR over the 6 month study period. In contrast, all 4 patients in the rituximab group either maintained stable renal function or had improvement in the estimated GFR.
Nine patients randomized to the control group elected to receive rituximab treatment after month 6 on the extension phase of the study. Two of these patients withdrew from the study after completing their course of infusions and were lost to follow-up. Of the remaining 7 patients, 4 achieved remission following rituximab treatment. The median duration of remission for these 4 patients was 6 months (interquartile range, 5 to 9 months).
Rituximab has been considered as a possible therapy for HCV-associated cryoglobulinemic vasculitis based on retrospective, uncontrolled studies of cohorts that included some patients with non-HCV-associated cryoglobulinemic vasculitis. In contrast to prior reports, our study was a prospective, randomized, controlled trial that enrolled only patients with HCV-associated cryoglobulinemic vasculitis in whom antiviral therapy had failed. Furthermore, standard forms of immunosuppressive therapy had also failed to induce remission in 75% of the patients in our study. Thus, patients in this trial were highly representative of those patients with HCV-associated cryoglobulinemic vasculitis for whom effective therapy is lacking.
The trial was originally designed to enroll 30 patients but was halted at 24 when an interim analysis revealed that a sufficient difference existed between groups to warrant stopping the trial. Eighty-three percent of patients randomized to rituximab met the primary endpoint of remission at study month 6 as compared to 8% of those in the control group. The median duration of remission for the 10 rituximab-treated patients reaching the primary endpoint was 7 months. The three patients who relapsed after month 6 were all retreated with a second course of rituximab leading to sustained remission lasting more than 6 months in all three. These data indicate that therapy with rituximab is able to induce remission in a high percentage of patients, and that remission can be sustained beyond 6 months despite the presence of continuing HCV infection.
The current investigation was based upon the hypothesis that rituximab-induced depletion of the expanded population of autoreactive B cells would decrease pathogenic cryoglobulin formation and result in clinical improvement in the vasculitis. Depletion of peripheral blood B cells was observed in all patients who received four infusions of rituximab. B cell depletion was associated with a decrease in cryoglobulin levels and improvement in vasculitis activity. A return of B cells to pre-treatment levels occurred within 6–8 months, but this did not correlate with a relapse of disease activity. Although most patients remained in remission for months after the return of peripheral B cells, the persistence of low-level cryoglobulinemia and ongoing HCV infection suggests that rituximab treatment did not eradicate the pathogenic B cell population and these patients remain at risk for future relapse.
Immunosuppression is known to increase HCV viremia and accelerate the progression of chronic HCV liver disease (21–23) and a prior uncontrolled study suggested that rituximab treatment was associated with a similar risk (12). In our controlled trial, we found no evidence for worsening HCV infection. No significant change in plasma viral levels over time was detected in the rituximab group compared with the control group and there was no biochemical evidence of worsening hepatitis. Rituximab treatment was well tolerated with the frequency of adverse events not significantly different from the control group and no serious infections were seen.
A significant strength of this study is that the primary outcome measure of remission was determined prior to study initiation and was based upon the BVAS, a validated instrument previously used to assess vasculitis disease activity (18, 19). In addition, patients in this trial received standardized monthly clinical and laboratory evaluations that allowed for accurate determination of time to remission, time to relapse, and potential adverse effects of rituximab on the underlying HCV infection.
This trial does have limitations that must be weighed. Accrual into the study was slow for several reasons including the low incidence of HCV-associated vasculitis in the United States (24), restricting enrollment to patients in whom antiviral therapy had failed, and the availability of rituximab for off label use in the United States. However, the slow accrual rate should not have a major effect on study outcomes, as no new treatments for either HCV infection or HCV-associated vasculitis became available during the period in which the study was conducted. Although the study was randomized, it was not blinded. This design was chosen as the study endpoints were all based on objective measures of improvement in manifestations of vasculitis. Although BVAS has potential limitations with assignment of disease activity to subjective symptoms particularly in open-label studies, it remains the most recognized validated instrument for standardized assessment of disease activity in vasculitis clinical trials. The sample size of the study was small, due to the efficacy of rituximab and lack of efficacy of standard forms of immunosuppressive therapy in this patient population. No patients with immediately life-threatening manifestations of vasculitis involving the central nervous system, heart, or gastrointestinal tract was enrolled in our trial. Hence, the results of this trial may not be applicable to patients with these rare manifestations of HCV-associated cryoglobulinemic vasculitis. However, patients in our study did have significant disease activity as evidenced by distribution of organ involvement and their BVAS scores. Finally, our study was limited to patients with HCV-associated cryoglobulinemic vasculitis and the results cannot be extrapolated to patients with cryoglobulin disease not associated with chronic HCV infection.
It is important to emphasize that the first line treatment for HCV-associated vasculitis should be antiviral therapy. Our study only enrolled patients in whom antiviral therapy failed to induce remission either because a lack of sustained virologic response or regimen related toxicity. Limiting our study to such patients made it possible to assess the efficacy, toxicity, and effect on the underlying HCV infection of rituximab without the confounders of concomitant antiviral therapy. Recent studies suggest that combining rituximab with interferon-based antiviral therapy results in improved response rates compared to antiviral therapy alone (17, 25) a hypothesis which cannot be directly addressed by our trial.
In conclusion, our data suggests that rituximab can induce sustained remissions in patients with HCV-associated cryoglobulinemic vasculitis following failure of antiviral therapy. Rituximab treatment was well tolerated and did not appear to increase HCV replication or worsen the underlying hepatitis.
This research was funded by the Intramural Research Program of the National Institutes of Health, National Institute of Allergy and Infectious Diseases; ClinicalTrials.gov number, NCT00029107.
We thank William Sachau, Rose McConnell, and Laura Heytens for coordinating patient recruitment and study visits; Richard Kwan for coordinating medical care of the patients; the staff and patients of Outpatient Clinic 8 at the NIH Clinical Center; and Dr. Anthony S. Fauci for his work in establishing the National Institute of Allergy and Infectious Diseases Vasculitis Research Program.