Personalized Algorithm Helps Guide Rituximab Retreatment

A personalized algorithm was developed by researchers in England to recommend retreatment with rituximab (Rituxan) among patients with antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis, with the intention of preventing relapses while lowering the risks of infection.

In a multivariable analysis of 60 rituximab-treated patients with ANCA-associated vasculitis followed over a decade, predictors of a longer time to relapse were concomitant immunosuppression (HR 0.48, 95% CI 0.24-0.94, P=0.034), complete response at 6 months (HR 0.24, 95% CI 0.12-0.50, P<0.001), and naive B-cell repopulation at 6 months (HR 0.43, 95% CI 0.22-0.84, P=0.013), reported Jack Arnold, MD, of the Leeds Institute of Rheumatic and Musculoskeletal Medicine.

The observation of longer time to relapse with concomitant immunosuppression was “not terribly surprising,” but the association with naive B-cell repopulation was important, he said at the European League Against Rheumatism virtual congress.

Based on these observations, the algorithm recommended retreatment for partial responders at 6 months irrespective of the naive B-cell status. However, for those who achieved a complete response at 6 months, they recommended measurement of naive B cells and consideration of retreatment only for those who do not show naive B-cell repopulation. “Patients with complete response and naive B-cell repopulation are unlikely to benefit from a 6-month fixed retreatment strategy,” Arnold noted.

Rituximab’s clinical effects result from a depletion of B cells, which can be detected by the presence of the CD19 surface marker on these cells. Rituximab has demonstrated efficacy in ANCA-associated vasculitis and other autoimmune diseases, but many patients relapse and the necessity for retreatment has been an important concern.

“Time to relapse after rituximab in ANCA-associated vasculitis is variable and the optimal retreatment strategy is unclear,” Arnold said.

One strategy has been fixed retreatment at 6 months, which is associated with a lower rate of relapse but carries a risk of low IgG and increased infections, while a second approach has been retreatment at the time of clinical relapse, which limits the risk of low IgG but may permit severe disease flares and additional glucocorticoid exposure.

A third suggested approach has been biomarker-guided retreatment, which would aim to avoid the problems with both other strategies, but to date there has been a lack of reliable biomarkers, Arnold explained.

The role of overall CD19+ cell return and the presence of ANCA as biomarkers was explored in the MAINRITSAN2 clinical trial. “Unfortunately, this trial found no significant difference in relapse rates between biomarker-guided and fixed retreatment patterns,” he noted.

However, in a post-hoc analysis, relapse rates in patients differed according to the presence of detectable B cells. “We found that 58% of patients with no detectable B cells throughout the course of the study relapsed compared with only 8% of those who had B cells detected at least once during follow-up,” Arnold said.

Study Details

Arnold’s group conducted the current observational study with the goal of establishing a data-driven treatment algorithm.

Complete response was defined as a Birmingham Vasculitis Activity Score (BVAS) of 0. Relapse was defined as worsening of disease and an increase in BVAS of 1 or more points, while major relapse required involvement of one or more major organs, a life-threatening manifestation, or both.

In this study, repeat rituximab was given at the time of clinical relapse.

Naive and memory B cells and plasmablasts were measured using highly sensitive flow cytometry.

Patients’ mean age at the time of the first rituximab treatment cycle was 51 years, one-third were men, and the specific subset of vasculitis was granulomatosis with polyangiitis in 75%. More than half had received cyclophosphamide, half were using concomitant prednisone, and 39% were given concomitant immunosuppression within 3 months of the first cycle.

The time to retreatment was longer than 6 months in the majority of patients, with median times to retreatment for cycles 1 to 5 of 87, 71, 65, 59, and 86 weeks, respectively.

During 417 patient-years of follow-up, 137 relapses occurred in 50 patients, with 16 of the relapses in 14 patients considered major, including seven renal relapses, four neurologic relapses, three ear-nose-throat-related relapses, and two respiratory relapses.

Previous work by Arnold’s group investigated B-cell subsets in different rheumatic diseases. “We demonstrated elevated plasmablasts in untreated systemic lupus erythematosus patients and conversely, lower naive B-cell populations in ANCA-associated vasculitis,” he noted.

Moreover, naive B cells were even lower in patients with ANCA-associated vasculitis whose C-reactive protein level was above 10 mg/L, suggesting that these patients had higher disease activity, he added.

“These B-cell population changes are disease-specific signatures, and more specifically, low naive B-cell populations are associated with more active ANCA-associated vasculitis. Therefore, repopulation of naive B cells, not total CD19+ cells, is indicative of response to treatment and restoration of normal B-cell homeostasis,” Arnold explained. “Repopulation of naive B cells is not a pathogenic process.”

“Retreatment based on this algorithm would have prevented unnecessary retreatment in 27% of patients,” he observed.

He recommended that all patients with ANCA-associated vasculitis receive concomitant oral immunosuppressants, but cautioned that his team’s findings should be replicated in additional populations.