BAFF levels in association with response to TNFa antagonists; comment on the article by Genovese et al

Athina Pyrpasopoulou, ,
January 17, 2013

Triantafyllou Areti, Pyrpasopoulou Athina, Anyfanti Panagiota, Balaska Ekaterini, Aslanidis Spyros, Douma Stella

2nd Propedeutic Dept of Internal Medicine, Hippokration GH, Thessaloniki, Greece

Dear Editor,

We read with great interest the article by Genovese et al on the effect of the BAFF antagonist tabalumab in rheumatoid arthritis patients with inadequate response to TNFa blockade. The results of this study indicated a moderate effect of tabalumab early (week 6/ week 9) after initiation of therapy, both in terms of clinical as well as laboratory markers of disease activity, which however was not significantly sustained later on in the course of treatment (week 16). In the previous decade, biological therapies that target the B cell have become established in the treatment of rheumatoid arthritis and theories regarding a possible connection of seropositivity and an enhanced peripheral compartment of memory cells have emerged. [1] Interestingly, not all biological B-cell targeted treatments have proved to be equally effective in rheumatoid arthritis. Our group published recently the observation that TNF antagonism (infliximab) does not significantly affect the levels of BAFF in previously biologically naive rheumatoid arthritis patients, regardless of the presence or absence of seropositivity. [2] Prompted by the results of Genovese et al we attempted to reanalyze our data to investigate whether BAFF levels could represent a biomarker to predict flare of the disease and potentially response to anti-BAFF therapy. We recategorized the patients included in the study in responders and non-responders and compared the effect of antiTNF treatment on BAFF levels in each individual patient, pre and post treatment, in the two separate categories. Responders were identified as the patients whose DAS28 value decreased by at least 1.2 units post treatment with infliximab.

There wasn't any statistical significant correlation between impact on DAS28 and Baff difference in the total sample (Spearman test r=0.454), in the responders group (Spearman test r=0.391) and in the non-responders group (Spearman test r=0.541). Within the non-responders group, seronegative patients had comparable levels of BAFF pre- and post treatment, while in seropositive patients, BAFF levels appeared to decrease post TNFa antagonism, p=0.046. It needs however to be stressed out that a. the number of the patients is small for a safe conclusion to be drawn, and b. the BAFF measured in this study represents only the soluble and not the membrane bound molecule.

It would be of great interest to know whether the authors looked at BAFF levels during their study. Another issue to be addressed would be the prevalence of specific BAFF promoter polymorphisms among responders and non-responders to tabalumab, which have been associated with disease severity in the presence or absence of increased serum BAFF levels. [3,4]
The pathogenesis of rheumatoid arthritis is complex and leads to a wide diversity of disease phenotypes and the role of BAFF in the development and treatment of the disease, as well as the physiological meaning of exogenous pharmaceutical interventions have not become fully clarified.
The interplay of soluble cytokines and membrane bound receptors and its clinical outcome may only be elucidated when all involved partners have been identified and the individual intermolecular interactions have become mapped.


1.Roll P, Muhammad K, Schumann M, et al. RF positivity has substantial influence on the peripheral memory B-cell compartment and its modulation by TNF inhibition. Scand J Rheumatol. 2012 May;41(3):180-5. doi: 10.3109/03009742.2011.645056. Epub 2012 Mar 9

2. Pyrpasopoulou A, Balaska E, Triantafyllou A, et al. B-cell activating factor levels in rheumatoid arthritis patients in response to treatment with biologics. J Interferon Cytokine Res. 2012 Jul;32(7):338- 40. doi: 10.1089/jir.2011.0118. Epub 2012 Apr 17.

3. Ruyssen-Witrand A, Rouanet S, Combe B, et al. Association between -871C>T promoter polymorphism in the B-cell activating factor gene and the response to rituximab in rheumatoid arthritis patients. Rheumatology (Oxford). 2012 Dec 22. [Epub ahead of print]

4. Nossent JC, Lester S, Zahra D, et al. Polymorphism in the 5' regulatory region of the B-lymphocyte activating factor gene is associated with the Ro/La autoantibody response and serum BAFF levels in primary Sjogren's syndrome. Rheumatology (Oxford). 2008 Sep;47(9):1311-6. doi: 10.1093/rheumatology/ken246. Epub 2008 Jul 10.

Conflict of Interest:

None declared

Conflict of Interest

None declared