Background/Purpose:

In the last few year the introduction of biological agents has radically changed  the clinical outcome of patients with Rheumatoid Arthritis (RA). However, no single drug is able to control all patients with RA and it is known that each drug may be poorly effective in a sizable proportion of the treated patients. For these reasons the early identification of clinical responder patients would be a crucial advantage for both a clinical and socioeconomic point of view. Recently, mathematic algorithms, based on classical clinical parameters,  have been proposed  to predict the clinical response to Anti TNF and DMARDs. In the present study we have applied the mathematic algorithm  proposed to predict early response to anti TNF on our cohort of patients treated with antiTNF agents.

Methods: We collected the data of 96 patients followed in our unit and treated with antiTNF (Adalimumab,  Etanercept , Infliximab, Golimumab and Certolizumab  Pegol) from Jan 2010 to Jan 2013, with a follow up of at least 12 months. The mathematic algorithm, utilizing the following parameters: Tender Joint, Swollen Joint, Illness activity VAS by Physician and patient, Pain VAS, ESR and CRP; was applied to calculate the putative responders after one month of treatment and this value was compared with the DAS 28 at one month and after  one year (yr). The patients were classified as good responders if they had a delta DAS28>1.2. In table 1 we summarized  the main epidemiological and  clinical data of patients under investigation.

Results:

The clinical response at 1 yr was very significant for all kind of treatment.  After 1 month of therapy a delta DAS 28>2.6 was recorded in 50% of all treated patients, while at one yr a delta DAS28>1.2 was found in percentage variable between 86% and 91.6% (table1). In contrast, the mathematical model allows to predict 100% of the final responders for patients treated by Infliximab, Golimumab and Certolizumab  Pegol, 93% for patients treated with Adalimumab and 90% with Etanercept, 5 false negative were registered for Etanercept and 3 false negative for Adalimumab.

Conclusion: These data indicate that in a routine clinical practice the application of a simple mathematical model is capable, at one month, to predict a good response in the majority of patients. Prospective studies are underway.

Table 1
	Certolizumab Pegol	Adalimumab	Infliximab	Etanercept	Golimumab
Number of  patients	12	25	13	42	4
Age	52.63±15.58	54.76 ± 15.57	58.00 ± 10.30	59.73 ± 15.32	51.21±13.60
Sex	10 ♀ 2 ♂	21 ♀ 4 ♂	10 ♀ 3 ♂	35 ♀ 7 ♂	2 ♀ 2 ♂
DAS28 at baseline	5.77 ±0.90	5.60 ± 0.89	6.46 ± 1.13	6.24±0.78	5.27 ± 0.61
DAS28 at final visit	1.99 ± 0.43	2.80 ± 1.23	2.75 ± 0.74	3.14 ± 0.96	3.93 ± 0.84
% of Response Delta DAS28>1.2	91.6	86.0	91.9	88.1	89.4

---

« Back to 2013 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/mathematical-model-to-predict-the-early-responders-in-a-monocentric-cohort-of-patients-with-rheumatoid-arthritis-treated-by-anti-tnf-alpha/