Background/Purpose: Published data have shown that approximately one-third of patients with RA are treated with biologic (Bio) monotherapy (MT) (without concomitant DMARD) and a considerable portion of these patients are previously Bio naïve (Yazici et al. 2008). The purpose of this analysis is to determine the predictors of initiating a biologic as MT versus in combination with a DMARD (CMB) in previously Bio naïve patients in the Consortium of Rheumatology Researchers of North America (CORRONA) Registry.

Methods: Data on patients previously Bio naïve and initiating their first biologic therapy were included in this analysis. Demographic and baseline clinical characteristics were evaluated and logistic regression models were used to assess potential predictors of Bio MT initiation. The following independent variables, selected for clinical relevance and significant in the univariate model, were entered in the models simultaneously as fixed effects in various combinations: a) medical history of neutropenia, hepatic disease and malignancy, b) swollen joints counts and presence of erosions, c) whether the biologic is approved for MT in the US, and d) and if the biologic was initiated prior to 2006 which marks the increased availability of biologics approved for MT. A random effect of individual physician’s prescribing patterns was also included in the model to estimate a median odds ratio (Larsen et al., 2000).

Results: Between October 2001 and April 2012, 3,923 previously Bio naïve patients initiated biologic therapy, of which 19% initiated as MT. Baseline characteristics of patients initiating Bio MT and Bio CMB were similar: age (years; mean±SD) 57±15 vs. 58±13, female 74% vs. 76%, duration of RA (years; mean±SD) 9±9 vs. 8±9, and CDAI (mean±SD) 19±14 vs. 19±14. A significantly higher proportion of patients initiating a Bio CMB had a history of non-biologic DMARDs (86% vs. 97%, p<0.0001) and specifically, a history of methotrexate (MTX) (72% vs. 92%, p<0.0001). The most frequently reported reason for discontinuing MTX was due to toxicity (Bio MT 45% vs. Bio CMB 36%, p = 0.05); similar results were seen for history of leflunomide.

Table 1 shows the results of three different models showing various combinations of variables mentioned earlier. Biologic approved for MT use in the US and individual physician prescribing patterns significantly influenced the odds of Bio MT initiation. 

Conclusion: Biologic approved for MT and effect of individual physician prescribing patterns consistently influenced the likelihood of Bio MT use along with significant effects in some instances of less severe disease, history of conditions potentially influencing use of concomitant DMARDs and year of initiation.

Table 1. Adjusted Odds Ratiosa for Biologic Monotherapy Versus Combination in Biologic Naïve Patients
	Model 1b N=3861c	Model 2b N=2823c	Model 3b N=644c
History of Hepatic Disease	6.5  [3.20, 13.07]	7.49  [3.19, 17.58]	5.20  [0.95, 28.49]
History of Malignancy	3.79  [1.64, 8.73]	2.78  [1.02, 7.59]	1.00  [0.19, 5.40]
Swollen Joint Count	0.97  [0.95, 0.99]	0.96  [0.95, 0.98]	0.98  [0.94, 1.02]
Use of Mono Approved Biologic	1.47  [1.20, 1.81]	1.45  [1.13, 1.86]	1.93  [1.08, 3.43]
Initiated after 2006	0.83  [0.68, 1.00]	0.79  [0.63, 0.99]
Erosions		0.84  [0.68, 1.03]	0.96  [0.62, 1.49]
History of Neutropenia			4.89  [1.16, 20.59]
Random effect of individual physician’s prescribing patterns	1.89  [1.66, 2.23]	1.86  [1.61, 2.25]	1.58  [1.23, 2.72]
Notes:
a OR > 1 implies monotherapy more likely.
b Three different models with various combinations of fixed effects from independent variables described above and a random effect of individual physician’s prescribing patterns were fitted.
c Models fitted using available data among 3,923 previously Bio naïve patients initiating a biologic therapy.

---

« Back to 2012 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/predictors-of-initiating-biologic-monotherapy-in-biologic-naive-patients-with-rheumatoid-arthritis-ra-in-a-us-registry-population/