From the Journals

Rituximab serious infection risk predicted by immunoglobulin levels



Monitoring immunoglobulin (Ig) levels at baseline and before each cycle of rituximab could reduce the risk of serious infection events (SIEs) in patients needing repeated treatment, according to research published in Arthritis & Rheumatology.

Dr Md Yuzaiful Md Yusof

Dr. Md Yuzaiful Md Yusof

In a large, single-center, longitudinal study conducted at a tertiary referral center, having low IgG (less than 6 g/L) in particular was associated with a higher rate of SIEs, compared with having normal IgG levels (6-16 g/L). Considering 103 of 700 patients who had low levels of IgG before starting treatment with rituximab for various rheumatic and musculoskeletal diseases (RMDs), there were 16.4 SIEs per 100 patient-years. In those who developed low IgG during subsequent cycles of rituximab therapy, the SIE rate was even higher, at 21.3 per 100 patient-years. By comparison, the SIE rate for those with normal IgG levels was 9.7 per 100 patient-years.

“We really have to monitor immunoglobulins at baseline and also before we re-treat the patients, because higher IgG level is protective of serious infections,” study first author Md Yuzaiful Md Yusof, MBChB, PhD, said in an interview.

Low IgG has been linked to a higher risk of SIEs in the first 12 months of rituximab therapy but, until now, there have been limited data on infection predictors during repeated cycles of treatment. While IgG is a consistent marker of SIEs associated with repeated rituximab treatment, IgM and IgA should also be monitored to give a full picture of any hyperglobulinemia that may be present.

“There is no formal guidance on how to safely monitor patients on rituximab,” observed Dr. Md Yusof, who will present these data at the 2019 European Congress of Rheumatology in Madrid. The study’s findings could help to change that, however, as they offer a practical way to help predict and thus prevent SIEs. The study’s findings not only validate previous work, he noted, but also add new insights into why some patients treated with repeat rituximab cycles but not others may experience a higher rate of such infections.

Altogether, the investigators examined data on 700 patients with RMDs treated with rituximab who were consecutively seen during 2012-2017 at Dr. Md Yusof’s institution – the Leeds (England) Institute of Rheumatic and Musculoskeletal Medicine, which is part of the University of Leeds. Their immunoglobulin levels had been measured before starting rituximab therapy and every 4-6 months after each cycle of rituximab treatment.

Patients with any RMD being treated with at least one cycle of rituximab were eligible for inclusion in the retrospective study, with the majority (72%) taking it for rheumatoid arthritis and some for systemic lupus erythematosus (13%) or antineutrophil cytoplasmic antibody (ANCA)–associated vasculitis (7%).

One of the main aims of the study was to look for predictors of SIEs during the first 12 months and during repeated cycles of rituximab. Dr. Md Yusof and his associates also looked at how secondary hypogammaglobulinemia might affect SIE rates and the humoral response to vaccination challenge and its persistence following treatment discontinuation. Their ultimate aim was to see if these findings could then be used to develop a treatment algorithm for rituximab administration in RMDs.

Over a follow-up period encompassing 2,880 patient-years of treatment, 281 SIEs were recorded in 176 patients, giving a rate of 9.8 infections per 100 patient-years. Most (61%) of these were due to lower respiratory tract infections.

The proportion of patients experiencing their first SIE increased with time: 16% within 6 weeks of starting rituximab therapy, 35% at 12 weeks, 72% at 26 weeks, 83% at 38 weeks, and 100% by 1 year of repeated treatment.

Multivariable analysis showed that the presence of several comorbidities at baseline – notably chronic obstructive pulmonary disease, diabetes, heart failure, and prior cancer – raised the risk for SIEs with repeated rituximab therapy. The biggest factor, however, was a history of SIEs – with a sixfold increased risk of further serious infection.

Higher corticosteroid dose and factors specific to rituximab – low IgG, neutropenia, high IgM, and a longer time to retreatment – were also predictive of SIEs.

“Low IgG also results in poor humoral response to vaccination,” Dr. Md Yusof said, noting that the IgG level remains below the lower limit of normal for several years after rituximab is discontinued in most patients.

In the study, 5 of 8 (64%) patients had impaired humoral response to pneumococcal and haemophilus following vaccination challenge and 4 of 11 patients had IgG normalized after switching to another biologic disease-modifying antirheumatic drug (bDMARD).

Cyclophosphamide is commonly used as a first-line agent to induce remission in patients with severe and refractory systemic lupus erythematosus and ANCA-associated vasculitis, with patients switched to rituximab at relapse. The effect of this prior treatment was examined in 20 patients in the study, with a marked decline in almost all immunoglobulin classes seen up to 18 months. Prior treatment with immunosuppressants such as intravenous cyclophosphamide could be behind progressive reductions in Ig levels seen with repeated rituximab treatment rather than entirely because of rituximab, Dr. Md Yusof said.

Dr. Md Yusof, who is a National Institute for Health Research (NIHR) Academic Clinical Lecturer at the University of Leeds, said the value of the study, compared with others, is that hospital data for all patients treated with rituximab with at least 3 months follow-up were included, making it an almost complete data set.

“By carefully reviewing records of every patient to capture all infection episodes in the largest single-center cohort study to date, our findings provide insights on predictors of SIEs as well as a foundation for safety monitoring of rituximab,” he and his coauthors wrote.

They acknowledge reporting a higher rate of SIEs than seen in registry and clinical studies with rituximab, which may reflect a “channeling bias” as the patients comprised those with multiple comorbidities including those that represent a relative contraindication for bDMARD use. That said, the findings clearly show that Ig levels should be monitored before and after each rituximab cycle, especially in those with comorbid diseases and those with low IgG levels to start with.

They conclude that an “individualized benefit-risk assessment” is needed to determine whether rituximab should be repeated in those with low IgG as this is a “consistent predictor” of SIE and may “increase infection profiles when [rituximab] is switched to different bDMARDs.”

The research was supported by Octapharma, the National Institute for Health Research (NIHR), and NIHR Leeds Biomedical Research Centre based at Leeds Teaching Hospitals NHS Trust in England. Dr. Md Yusof had no conflicts of interest. Several coauthors disclosed financial ties to multiple pharmaceutical companies, including Roche.

SOURCE: Md Yusof MY et al. Arthritis Rheumatol. 2019 May 27. doi: 10.1002/art.40937.

Next Article: