Treatment Patterns and Costs in Biologic DMARD-Naive Patients with Rheumatoid Arthritis Initiating Etanercept or Adalimumab with or Without Methotrexate

BACKGROUND: Etanercept (ETN) and adalimumab (ADA) are tumor necrosis factor inhibitors indicated for treatment of moderate to severe rheumatoid arthritis (RA) and are used as monotherapy or in combination with conventional disease-modifying antirheumatic drugs (DMARDs) such as methotrexate (MTX). Data on treatment patterns and costs of ETN and ADA as monotherapies or in combination therapy with MTX are lacking in biologic DMARD (bDMARD)-naive patients with RA. OBJECTIVE: To evaluate treatment patterns and costs of ETN and ADA monotherapy and combination therapy in bDMARD-naive patients with RA. METHODS: Data from adult bDMARD-naive patients with RA were evaluated according to index therapy (ADA or ETN as monotherapy or combination therapy with MTX) in a retrospective cohort study using the IBM MarketScan Commercial Claims and Encounters and Medicare Supplemental Databases from January 1, 2010, to June 30, 2017. Participants were bDMARD-naive for ≥ 12 months before initial ETN or ADA pharmacy claim (index date) and had continuous enrollment for ≥ 12 months pre-index and 24 months post-index. Combination therapy cohorts had an MTX claim within 30 days of the index date. Outcomes included persistence (no treatment changes or gap [≥ 60 days]); modifications to index therapy (discontinuation or switching without prior gap, restarting as switch or restart after gap, or MTX initiation/discontinuation); and mean total bDMARD costs for 2 years post-index. RESULTS: Patients on ETN monotherapy (n = 2,064) had higher persistence (26.8% vs. 21.1%, respectively; P < 0.001) on index treatment and received treatment for a longer duration (mean 375.9 days vs. 339.7 days, respectively; P < 0.001) than those on ADA monotherapy (n = 1,528). Regimen changes were more common in patients on ADA monotherapy than patients on ETN monotherapy (38.0% vs. 33.4%, respectively; P = 0.004). More patients on ADA monotherapy added MTX than those on ETN (17.5% vs. 12.6%, respectively; P < 0.001). Overall, 790 patients receiving index monotherapy had a regimen change following a gap (≥ 60 days), with a similar proportion between cohorts. Among these patients, 13.8% restarted index therapy, and 7.9% switched from index therapy. Significantly more patients receiving ETN monotherapy restarted their index regimen after a gap than those receiving ADA monotherapy (14.9% vs. 12.2%, respectively; P = 0.023). The proportion of patients persistent on combination therapy was similar between the ETN and ADA combination therapy cohorts (21.9% vs. 22.2%, respectively; P = 0.818). Treatment pattern rates were similar regardless of index combination therapy. Overall, costs for ADA were consistently higher within the index regimen throughout the follow-up period irrespective of MTX. CONCLUSIONS: ETN monotherapy as first-line treatment was associated with higher persistence, lower rate of MTX supplementation, and lower bDMARD costs than ADA monotherapy. ETN monotherapy may represent a less costly option for achieving treatment targets in bDMARD-naive patients with RA.

METHODS: Data from adult bDMARD-naive patients with RA were evaluated according to index therapy (ADA or ETN as monotherapy or combination therapy with MTX) in a retrospective cohort study using the IBM MarketScan Commercial Claims and Encounters and Medicare Supplemental Databases from January 1, 2010, to June 30, 2017. Participants were bDMARD-naive for ≥ 12 months before initial ETN or ADA pharmacy claim (index date) and had continuous enrollment for ≥ 12 months pre-index and 24 months post-index. Combination therapy cohorts had an MTX claim within 30 days of the index date. Outcomes included persistence (no treatment changes or gap [≥ 60 days]); modifications to index therapy (discontinuation or switching without prior gap, restarting as switch or restart after gap, or MTX initiation/discontinuation); and mean total bDMARD costs for 2 years post-index.
RESULTS: Patients on ETN monotherapy (n = 2,064) had higher persistence (26.8% vs. 21.1%, respectively; P < 0.001) on index treatment and received treatment for a longer duration (mean 375.9 days vs. 339.7 days, respectively; P < 0.001) than those on ADA monotherapy (n = 1,528). Regimen changes were more common in patients on ADA monotherapy than patients on ETN monotherapy (38.0% vs. 33.4%, respectively; P = 0.004). More patients on ADA monotherapy added MTX than those on ETN (17.5% vs. 12.6%, respectively; P < 0.001). Overall, 790 patients receiving index monotherapy had a regimen change following a gap (≥ 60 days), with a similar proportion between cohorts. Among these patients, 13.8% restarted index therapy, and 7.9% switched from index therapy. Significantly more patients receiving ETN monotherapy restarted their index regimen after a gap than those receiving ADA monotherapy (14.9% vs. 12.2%, respectively; P = 0.023). The proportion of patients persistent on combination therapy was similar between the ETN and ADA combination therapy cohorts (21.9% vs. 22.2%, respectively; P = 0.818). Treatment pattern rates were similar regardless of index combination therapy. Overall, costs for ADA were consistently higher within the index regimen throughout the follow-up period irrespective of MTX.
CONCLUSIONS: ETN monotherapy as first-line treatment was associated with higher persistence, lower rate of MTX supplementation, and lower bDMARD costs than ADA monotherapy. ETN monotherapy may represent R E S E A R C H R heumatoid arthritis (RA) is a chronic, progressive, and disabling autoimmune disease affecting 0.6%-1.0% of adults in the United States. 1 When unmanaged, RA may lead to joint damage, chronic pain and stiffness, loss of function, disability, and decreased life span. 2 The past 15-20 years have seen significant advances in the treatment of RA, with treat-to-target strategies aimed at Study periods were defined as follows: Index date refers to the date of initial ETN or ADA claim. Pre-index period refers to the 12 months preceding the index date and was used to identify demographic and clinical characteristics and to establish a bDMARD-naive population. Follow-up period refers to the 24 months after the index date and was used for assessment of outcome measures.

Study Population
Inclusion Criteria. To be included in this analysis, patients were required to be aged at least 18 years on the index date and have at least 1 outpatient pharmacy claim carrying a National Drug Code number for ETN or ADA between January 1, 2011, and June 30, 2015, and at least 1 inpatient or 2 nondiagnostic outpatient claims carrying an International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) or International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) diagnosis for RA (714.x or M05-M06, respectively) during the pre-index period. Furthermore, patients were required to have continuous enrollment for the 12-month pre-index period and at least 24 months after the index date and at least 1 refill of index bDMARD after initial treatment (Appendix A, available in online article).
Exclusion Criteria. Patients were excluded from the final analyses if they had an outpatient pharmacy or medical claim for any bDMARD during the pre-index period. Patients were also excluded if they had a medical claim with a diagnosis of psoriatic arthritis, plaque psoriasis, ankylosing spondylitis, Crohn disease, ulcerative colitis, and juvenile idiopathic arthritis or diagnosis of cancer (except nonmelanoma skin cancer) during the pre-index period through the follow-up period. Adherence and dosing information cannot be assessed accurately on a medical claim; therefore, patients were also ineligible if they had an outpatient medical claim carrying a Healthcare Common Procedure Coding System code for ADA or ETN from the index date through the end of the patient's follow-up period. achieving low disease activity or remission as outlined in the 2015 guidelines set by the American College of Rheumatology (ACR). 3 Traditionally, patients with RA are treated with conventional disease-modifying antirheumatic drugs (DMARDs) such as methotrexate (MTX). However, many patients do not achieve treatment targets. The addition of a biologic DMARD (bDMARD) or a targeted synthetic DMARD to the treatment regimen is recommended for patients with RA who have moderate or high disease activity despite monotherapy with a conventional DMARD. 3,4 bDMARDs, including tumor necrosis factor inhibitors (TNFi), have transformed the course of disease for patients with moderate to severe RA. TNFis are indicated for the treatment of moderate to severe RA, 3 with adalimumab (ADA) and etanercept (ETN) being the most commonly prescribed. 5 Studies have demonstrated significant improvements in radiologic outcomes among patients prescribed combination therapy with TNFis plus MTX compared with MTX alone. 6 However, MTX augments the efficacy of TNFis to varying extents, which may affect treatment patterns and costs. After 2 years of treatment, 59% of patients on ADA plus MTX and 69% of patients on ETN plus MTX combination therapy achieved 50% improvement in ACR criteria (ACR 50 response). 7,8 In comparison, 37% and 48% of patients receiving ADA and ETN monotherapy, respectively, achieved an ACR 50 response. 7,8 Despite clinical benefit, a decline in bDMARD use among bDMARD-naive patients from 21% in 2006 to 18% in 2009 has been observed. Furthermore, 22% of new bDMARD patients switched to a different bDMARD within 2 years. 9 Poor adherence and low persistence to therapy in RA reduce treatment effectiveness and increase health care costs and resource utilization. 10 Additionally, monthly total health care costs have been shown to be higher for patients with RA who switch therapy than for those who do not. 11 Although patients receiving TNFi and MTX combination therapy have been shown to be less likely to discontinue treatment for efficacy reasons, 12 data are lacking on treatment patterns and costs of TNFi as monotherapies or in combination with MTX, especially over a longer period (24 months).
Our aim was to explore real-world treatment patterns among bDMARD-naive patients with RA, examining cost differences across TNFi monotherapy and TNFi in combination with MTX. Our study focused on bDMARD-naive patients receiving their first prescription of ADA or ETN and evaluated the treatment patterns and costs as either a monotherapy or combination therapy with MTX.

Outcomes
Demographic and clinical characteristics were determined from claims data in the pre-index period. We assessed treatment patterns within the follow-up period according to specific modifications to index therapy: (a) switch from index therapy with prior refill gap, (b) switch from index therapy without prior refill gap, (c) restart as switch after ≥ 60-day gap, (d) restart index therapy after ≥ 60-day gap, (e) MTX addition or discontinuation, and (f) discontinuation of index therapy. Medication persistence was defined as an absence of treatment modifications or gaps (≥ 60 days). Medication adherence was defined as when the proportion of days medication taken as prescribed covered > 80%. Frequency and magnitude of index treatment dose escalation was also assessed. Dose escalation was defined as a proportion of days' supply/metric quantity that is at least 30% lower than the value of the index claim. 13 Additionally, health care utilization and costs were assessed by the type of service (inpatient, outpatient, and pharmaceutical) and reported in the following categories: (a) outpatient pharmacy RA bDMARD costs, (b) other outpatient pharmacy costs (excluding RA costs), (c) outpatient medical costs, and (d) inpatient admissions costs. Costs were assessed during the follow-up period and were reported as mean costs over the follow-up period and mean per patient per month costs by treatment pattern group. Costs were based on paid amounts of adjudicated claims, including insurer and health plan payments, and patient cost sharing in the form of copayment, deductible, and coinsurance. All costs were adjusted for inflation using the medical care component of the Consumer Price Index and standardized to 2017 U.S. dollars.

Statistical Analysis
Outcomes were stratified by index therapy. Means and standard deviations were calculated for continuous variables, while counts and proportions were calculated for categorical variables. Statistically significant differences were assessed via Student's t-tests and chi-square tests for continuous and categorical variables, respectively. Statistical testing was exploratory, and comparisons were unadjusted.

Treatment Patterns
Monotherapy. For the 2 years after the initial ETN or ADA claim, significantly more patients in the ETN monotherapy cohort persisted on their index regimen through the end of the study than the ADA monotherapy cohort (26.8% vs. 21.1%, respectively; P < 0.001; Table 2). Additionally, duration on the index regimen was significantly longer in the ETN monotherapy cohort compared with the ADA monotherapy cohort (mean duration: 376 days vs. 340 days, respectively; P < 0.001). Adherence (proportion of days covered > 80%) to the index medication was similar between the ETN and ADA monotherapy cohorts (87.1% vs. 88.7%, respectively; P = 0.129). A larger proportion of patients in the ADA monotherapy cohort increased their dose by at least 30% compared with patients in the ETN cohort (11.2% vs. 1.8%. respectively; P < 0.001). A larger proportion of patients receiving ADA monotherapy (19.3%) discontinued treatment without restarting than those receiving ETN (16.9%); however, this difference was not significant (P = 0.064). A larger proportion of patients receiving index ADA monotherapy had a regimen change without a prior gap (< 60 days) than those receiving index ETN monotherapy (38.0% vs. 33.4%, respectively; P = 0.004). Among those with a regimen change without a prior gap, MTX supplementation was more frequent in the ADA monotherapy cohort than the ETN monotherapy cohort (17.5% vs. 12.6%, respectively; P < 0.001).
Overall, 790 patients receiving index monotherapy had a regimen change after a gap (≥ 60 days), with the proportion performing similarly between cohorts (Table 2). Among these patients, 494 (13.8%) restarted index therapy and 283 (7.9%) switched from index therapy. Significantly more patients receiving ETN monotherapy restarted their index regimen after a gap than those receiving ADA monotherapy (14.9% vs. 12.2%, respectively; P = 0.023).  lower bDMARD costs than ADA monotherapy. Furthermore, first-line ETN plus MTX combination therapy had similar persistence but lower bDMARD and total health care costs than ADA plus MTX combination therapy. Our findings are consistent with a previous report that showed ETN had the lowest estimated bDMARD cost per effectively treated patient in a commercially insured population of patients with RA. 14 Optimizing adherence and persistence with therapy is a major challenge among patients with RA, as inadequate adherence and nonpersistence with prescribed regimens can affect treatment effectiveness. 10 This study found that disruptions in therapy (defined as switches, discontinuations, and restarts) were common. More than 75% of patients experienced some form of disruption during the 24 months after starting a bDMARD. The persistence rate was lower than previously reported rates. [15][16][17] Low persistence rates may reflect the bDMARD-naive population assessed, an inadequate response seen in patients initiating their first TNFi, or the longer followup period assessed (24 months vs. 12 months). The overall low persistence rates highlight an unmet need in existing bDMARD treatment for RA. Only a quarter of patients were persistent with their bDMARD therapies by the end of the 24-month follow-up period, whereas almost one fifth of patients ceased their monotherapy treatment. The remaining patients restarted on the same bDMARD, switched to another agent, or discontinued index therapy. Interestingly, restarting after a gap of more than 60 days was more common than switching bDMARD, which suggests that physicians and patients were inclined to (Table 2). Among these patients, treatment pattern rates were similar regardless of index combination therapy and regimen changes after a gap were rare.

Health Care Costs
Overall, mean outpatient pharmacy RA bDMARD costs over the 24-month follow-up period were lower among the ETN cohorts ($49,969 monotherapy; $53,587 plus MTX) than the ADA cohorts ($52,849 monotherapy; $56,407 plus MTX; P < 0.001 for both comparisons; Figure 1A). Similarly, among patients who were persistent on their index regimen through the entire 24-month period, mean per patient per month costs and outpatient pharmacy costs for RA biologics during the 24-month follow-up period were generally lower among the ETN cohorts than the ADA cohorts ( Figure 1B). Total health care costs by treatment pattern in the 24-month post-index period are shown in Table 3. Costs for ADA were consistently higher than for ETN throughout the follow-up period (Appendix B, available in online article).

■■ Discussion
We examined the treatment patterns and health care costs from a large, nationwide database of medical and pharmacy claims of more than 7,500 bDMARD-naive patients with RA. This retrospective analysis of real-world medication use and costs of bDMARDs showed that first-line ETN monotherapy for patients with moderate to severe RA was associated with higher persistence, lower rate of MTX supplementation, and  Finally, pharmacologic treatments were based on filled prescriptions with the assumption that patients take the medications; however, it cannot be confirmed if treatments were taken as prescribed.
■■ Conclusions ETN monotherapy as first-line treatment was associated with higher persistence, lower rate of MTX supplementation, and lower bDMARD costs than ADA monotherapy and may represent a less costly option for achieving treatment targets in bDMARD-naive patients with RA. modify index treatment using the same TNFi rather than try a different TNFi or bDMARD. This may reflect familiarity with the index treatment and a reluctance to try something different or an interruption due to an illness or surgery as opposed to an economic decision. As could be clinically expected, a larger proportion of patients in the ADA monotherapy cohort increased their dose by ≥ 30% compared with patients in the ETN cohort. This likely reflects prescribing guidance in the respective package insert in regard to the RA indication. 18,19 Some patients with RA may benefit from an ADA dose increase (i.e., 40 mg every other week to 40 mg weekly). 18 Conversely, ETN dose increases are not recommended. 19 Patients with RA who switch bDMARD therapy have been shown to incur higher 1-year total postswitch health care costs after the switch compared with patients who were persistent on the index bDMARD ($44,244 vs. $41,901; P = 0.006). Furthermore, health care costs were lowest for patients who started on ETN, particularly those who persisted on ETN ($37,638). 20 Accordingly, we found that first-line ETN was associated with lower costs despite higher (ETN monotherapy) or similar (ETN combination therapy) persistence compared with ADA. Similarly, switching bDMARD therapy has been associated with increased health care costs and resource utilization among patients who do not switch to a more effective therapy. 10,11 Limitations This study was subject to limitations inherent in the analysis of administrative claims data. The analysis was limited to individuals with commercial health coverage or private Medicare supplemental coverage.
Results of this analysis may not be generalizable to bDMARD-naive patients with RA with other types of insurance or without health insurance coverage. The generalizability was also limited by the size of the analysis set, which was restricted to patients with 24-month continuous enrollment after the index date and no cancer diagnosis (15.2% of patients with RA treated with ETN or ADA).
The potential for misclassification of RA, bDMARD use, covariates, or study outcomes was present, as patients were identified through administrative claims data as opposed to medical records. These data are subject to data coding limitations and data entry error. Misclassification of bDMARD-naive patients with RA may occur due to the 12-month pre-index period limit. It is possible that some patients may have received bDMARD therapy sometime before the 12-month pre-index period, resulting in their misclassification as bDMARD naive.
Additionally, the requirement for 24-month continuous enrollment after the index date may have led to a survival bias. Descriptive and univariate analyses were conducted for all outcomes. Analytically, we used descriptive analyses that comprise the initial, necessary phase of data exploration and Treatment Patterns and Costs in Biologic DMARD-Naive Patients with Rheumatoid Arthritis Initiating Etanercept or Adalimumab with or Without Methotrexate