Is It Possible to Predict Which Patients Treated with Biologic Agents for Rheumatic Diseases Will Develop Anti-Drug Antibodies ?

Pascal Zufferey¹, Melanie Favre dit Jeanfavre², Alexandre Dumusc³, Charles Benaim⁴, Matthieu Perreau⁵ and Alexander K. So Sr.⁶, ¹Rhu /Dal .Chuv, Rheumatology, Lausanne, Switzerland, ²DAL, RHU, Lausanne, Switzerland, ³DAL, RHU/CHUV, Lausanne, Switzerland, ⁴DAL, MPR, Lausanne, Switzerland, ⁵Medecine /CHUV, Immunology, Lausanne, Switzerland, ⁶Service De Rhumatologie, CHUV, Lausanne, Switzerland

Meeting: 2016 ACR/ARHP Annual Meeting

Date of first publication: September 28, 2016

Keywords: antibodies and inflammatory arthritis, Biologic agents

Session Information

Date: Tuesday, November 15, 2016

Title: Rheumatoid Arthritis – Clinical Aspects - Poster III: Treatment – Monitoring, Outcomes, Adverse Events

Session Type: ACR Poster Session C

Session Time: 9:00AM-11:00AM

Background/Purpose: All biologic agents (bDMARDs) currently used in rheumatology can induce anti-drug antibodies (ADAB), which will influence the drug levels and the drug effectiveness. Why only certain patients develop these antibodies is not yet clear, although there is already some literature dealing with anti-drug antibodies and trough drug level in rheumatic diseases.The aim of this study was to look for predictive factors of occurrence of such antibodies.

Methods: Since March 2013, we measure ADABs and trough levels for all anti TNF agent and also for rituximab and tocilizumab. Half of all our patients under biologic treatments have been tested. The method used is a sandwich ELIS A. ADABs, trough and TNF levels can be measured simultaneously. The reproducibility and the cut-offs have been tested among patients exposed and non-exposed to the medication. Clinical predictors of ADABs development were analyzed using multi and monovaraite regression analysis :.They comprised : gender, age, duration of the disease type of disease, duration of treatment, type of treatment, co-medication, previous biologic agent. Biologic predictors were: though level, TNF level, CRP and ESR

Results: 297 patients had at least one measurement of ADAB and drug through level up to January 2016. 124 patients tested were treated with a biologic agent for rheumatoid arthritis, 116 for spondylarthritis, 30 for psoriatic arthritis, 27 for other diagnoses. In 63 out of 297 (21%) ADABs against at least one of the bDMARD agents were detected. All the patients with ADABs were exposed to the medication, except for 3 patients (specificity: 98%). In patients exposed to bDMARDs, ADABs were found respectively for infliximab in 46/106 (44%) pts, adalimumab: 10/60 (16%) pts, certolizumab: 2/4 (50%) pts, etanercept: 1/20 (5%) pts, golimumab: 4/34 (12%) pts, tocilizumab: 1/75(1%) pts, rituximab: 4/46 (8%) pts. When ADABS against several bDMARDs were tested the % of patients developing ADABs against the second bDMARD: 33% tended to be higher than against a for the previous one : 21%, but not significantly p= 0.08).

	Monovariate analysis		p	multivariate	p
	ADBA+	ADAB-		ADBA+ OR (CI)*
Clinical predictors
Age: mean (SD): years	50(13)	53(014)	0.11
Sex ( F/M/ n (%)	35/30(54%)	157/74 (69%))	0.035
Type of disease (AS/apso/RA/ others)”	26/9/18/3	87/21/105/27	0.09⁺
Duration of disease : mean (SD):years	8.(6)	11(10)	0.018
Duration of treatments: mean (SD)/months)	41(33)	28(37)	0.015
Type of treatments (mab ;antiTNF/others) n (%)	60/5 (94/6%))	137/60 (69/31%)	<0-0001	26(2.6-264)	0.005
Co-medication(Y/N/ n (%)	22/40 (35%)	78/122(38%)	0.6
Previous biologic agent(Y/N/ (%))	51/13 (79%)	149/81 (65%)	0.027	5.9(1.14-30)	0.03
Biological predictors
CRP: >5, (Y/N/ n (%)	26/44(37%)	31/152(16%)	0.0005
ESR : >20 (Y/N/ n (%),	33/54(37%)	25/162(13%)	0.0001
Trough level undetectable(Y/N (%) )	41/15 (81%)	36/162/ (18%)	<0.00001	34(7.21-160.8)	0.0001
TNF level elevated (Y/N/ (%)	30/28(52%)	51/149 (26%)	0.00006	4.2(1.1-15)	0.025

On univariate analysis, several clinical and biological factors were significantly predictive of ADABS. After multivariate analysis only two clinical factors and two laboratory parameters remained independently associated: MAB anti-TNF treatments (OR: 26), previous bDMARD (OR: 5.9), undetectable through level (OR: 34) and High TNF trough level (OR:4.2).

Conclusion: In this large real world cohort of patients with rheumatological conditions requiring bDMARD therapy, either by anti TNF or other biological agents and tested for ADAbs at different time point of their treatments, the best predictors of the presence of ABABs were:treatment by an MAB anti TNF agent, previous exposure to another biologic agent, undetectable through level of the medication and elevated TNF levels

Disclosure: P. Zufferey, None; M. Favre dit Jeanfavre, None; A. Dumusc, None; C. Benaim, None; M. Perreau, None; A. K. So Sr., None.

To cite this abstract in AMA style:

Zufferey P, Favre dit Jeanfavre M, Dumusc A, Benaim C, Perreau M, So AK Sr.. Is It Possible to Predict Which Patients Treated with Biologic Agents for Rheumatic Diseases Will Develop Anti-Drug Antibodies ? [abstract]. Arthritis Rheumatol. 2016; 68 (suppl 10). https://acrabstracts.org/abstract/is-it-possible-to-predict-which-patients-treated-with-biologic-agents-for-rheumatic-diseases-will-develop-anti-drug-antibodies/. Accessed .

« Back to 2016 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/is-it-possible-to-predict-which-patients-treated-with-biologic-agents-for-rheumatic-diseases-will-develop-anti-drug-antibodies/