Arthritis & Rheumatism, Volume 63,
November 2011 Abstract Supplement

Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Chicago, Illinois November 4-9, 2011.

A Plasmablast Molecular Biomarker for Reduced Clinical Benefit From Anti-CD20 Therapy in Rheumatoid Arthritis.

Owczarczyk1,  Kasia, Lal2,  Preeti, Abbas1,  Alexander R., Wolslegel1,  Kristen, Holweg1,  Cecile TJ, Dummer2,  Wolfgang, Kelman2,  Ariella

Genentech Research and Early Development, South San Francisco, CA
Genentech, South San Francisco, CA


An important goal for personalized health care is the identification of biomarkers that predict the likelihood of treatment responses. Here, we tested the hypothesis that quantitative mRNA assays for B lineage cells in blood could serve as baseline predictors of therapeutic response to anti CD20-mediated B cell depletion therapy in subjects with rheumatoid arthritis (RA).


A reverse transcription-quantitative PCR panel comprising B cell-specific genes was developed to establish levels of B lineage cells using messenger RNA from whole blood of patients. IgJ gene expression was established as a marker of plasmablasts and plasma cells, while a B cell-specific splice variant of FcRL5 (IRTA2c) was established as a marker of mature B cells. Whole blood RNA samples from 3 large placebo-controlled trials of rituximab as well as a large placebo-controlled trial of ocrelizumab in RA patients were assessed for association of baseline B cell lineage status with clinical outcome. P values and confidence intervals for odds ratios were calculated using two-tailed Fisher's exact test.


In samples from the REFLEX trial of rituximab in anti-TNF inadequate responders (118 rituximab, 23 placebo), a 25% subgroup of treated subjects with elevated baseline mRNA levels of IgJ, a marker for antibody-secreting plasmablasts, showed reduced clinical response rates (9% ACR50 response in IgJhi, 31% ACR50 in IgJlo). There were no significant efficacy differences in the placebo arm subjects stratified by this marker. Prospective testing of the IgJ biomarker in the DANCER and SERENE rituximab clinical trial cohorts (total n=200 rituximab, 93 placebo) and the SCRIPT ocrelizumab cohort (275 ocrelizumab, 137 placebo) confirmed the utility of this marker to predict lesser clinical benefit to anti-CD20 therapy (PREPLICATION=0.006; OR = 2.4, 95% c.i. (1.2, 5.0). A combination mRNA biomarker, IgJhiFCRL5lo, showed improved test performance over IgJhi alone by removing all ACR50 responders in the REFLEX trial. Prospective testing of the IgJhiFCRL5lo profile, representing a 17% subgroup in the replication cohorts, confirmed its utility (PREPLICATION=0.008; OR = 2.7, 95% c.i. (1.3, 6.3).


This study demonstrates that baseline blood levels of molecular markers for late B lineage stage plasmablasts identify a ~20% subgroup of active RA subjects who are unlikely to gain significant clinical benefit from anti-CD20 B cell depletion therapy compared with placebo. Additional investigation into alternate dosing regimens and monitoring of plasmablast biomarkers may be warranted for these patients.

To cite this abstract, please use the following information:
Owczarczyk, Kasia, Lal, Preeti, Abbas, Alexander R., Wolslegel, Kristen, Holweg, Cecile TJ, Dummer, Wolfgang, et al; A Plasmablast Molecular Biomarker for Reduced Clinical Benefit From Anti-CD20 Therapy in Rheumatoid Arthritis. [abstract]. Arthritis Rheum 2011;63 Suppl 10 :837

Abstract Supplement

Meeting Menu