Advertisement

Impact of Methotrexate Discontinuation, Interruption, or Persistence in US Patients with Rheumatoid Arthritis Initiating Tofacitinib + Oral Methotrexate Combination

Open AccessPublished:June 03, 2022DOI:https://doi.org/10.1016/j.clinthera.2022.05.002

      Abstract

      Purpose

      Using data from real-world practice, this analysis compared outcomes in patients with rheumatoid arthritis (RA) initiating treatment with an oral Janus kinase inhibitor, tofacitinib, in combination with persistent, discontinued, or interrupted treatment with oral methotrexate (MTX).

      Methods

      This retrospective claims analysis (MarketScan® databases) included data from US patients with RA and at least one prescription claim for tofacitinib, dated between January 1, 2013, and April 30, 2017. Eligible patients were continuously enrolled for ≥12 months before and after treatment initiation, and initiated tofacitinib in combination with oral MTX, with at least two prescription claims for each. Patients were grouped according to treatment pattern (MTX-Persistent, MTX-Discontinued, or MTX-Interrupted). Tofacitinib treatment persistence, adherence, and effectiveness, as well as all-cause and RA-related health care costs, were assessed.

      Findings

      A total of 671 patients were eligible for inclusion; 504 (75.1%) were MTX-Persistent; 131 (19.5%), MTX-Discontinued; and 36 (5.4%), MTX-Interrupted. Rates of tofacitinib treatment persistence, adherence, and effectiveness at 12 months were similar between the MTX-Persistent and MTX-Discontinued cohorts. The percentage of patients switched from tofacitinib to another advanced disease-modifying antirheumatic drug within 12 months of tofacitinib initiation was greater in the MTX-Persistent cohort compared with that in the MTX-Discontinued cohort. RA-related health care costs at 12 months post-initiation were significantly greater in the MTX-Persistent cohort compared with those in the MTX-Discontinued cohort.

      Implications

      The findings from this analysis of real-world data indicate that patients who initiate tofacitinib in combination with oral MTX may discontinue MTX and still experience outcomes similar to those in patients who persist with MTX, with lesser RA-related health care costs. These results support those from a previous clinical study on methotrexate withdrawal in patients with RA (NCT02831855).

      Key words

      Introduction

      Clinical guidelines acknowledge disease-modifying antirheumatic drugs (DMARDs) as the standard treatment in patients with rheumatoid arthritis (RA): a conventional synthetic (cs)-DMARD, preferably methotrexate (MTX), in the first instance, followed by the addition of a biologic (b)-DMARD or a targeted synthetic (ts)-DMARD in cases of inadequate response.
      • Smolen JS
      • Landewé RBM
      • Bijlsma JWJ
      • et al.
      EULAR recommendations for the management of rheumatoid arthritis with synthetic and biological disease-modifying antirheumatic drugs: 2019 update.
      ,
      • Singh JA
      • Saag KG
      • Bridges Jr, SL
      • et al.
      2015 American College of Rheumatology guideline for the treatment of rheumatoid arthritis.
      In patients with low disease activity, it is recommended that treatment with a csDMARD, as well as a bDMARD and tsDMARD, is continued
      • Singh JA
      • Saag KG
      • Bridges Jr, SL
      • et al.
      2015 American College of Rheumatology guideline for the treatment of rheumatoid arthritis.
      ; however, to reduce treatment burden, the bDMARD or tsDMARD may be tapered in patients achieving stable remission, particularly when used in combination with a csDMARD.
      • Smolen JS
      • Landewé RBM
      • Bijlsma JWJ
      • et al.
      EULAR recommendations for the management of rheumatoid arthritis with synthetic and biological disease-modifying antirheumatic drugs: 2019 update.
      ,
      • Curtis JR
      • Emery P
      • Karis E
      • et al.
      Study of Etanercept And Methotrexate in Combination or as Monotherapy in Subjects with Rheumatoid Arthritis.
      Additionally, the dose of csDMARD may be reduced in patients who achieve persistent remission on a csDMARD alone, or who maintain persistent remission following tapering and subsequent withdrawal of a concurrent bDMARD or tsDMARD.
      • Smolen JS
      • Landewé RBM
      • Bijlsma JWJ
      • et al.
      EULAR recommendations for the management of rheumatoid arthritis with synthetic and biological disease-modifying antirheumatic drugs: 2019 update.
      Tofacitinib is an oral Janus kinase inhibitor used for the treatment of patients with moderately to severely active RA and an inadequate response or intolerance to MTX. Tofacitinib can be used as monotherapy or in combination with MTX or other nonbiologic DMARDs. The efficacy and tolerability of tofacitinib 5 and 10 mg b.i.d. administered as monotherapy or in combination with a csDMARD, mainly MTX, in patients with moderately to severely active RA, have been reported in Phase III
      • Burmester GR
      • Blanco R
      • Charles-Schoeman C
      • et al.
      Tofacitinib (CP-690,550) in combination with methotrexate in patients with active rheumatoid arthritis with an inadequate response to tumour necrosis factor inhibitors: a randomised phase 3 trial.
      • Fleischmann R
      • Kremer J
      • Cush J
      • et al.
      Placebo-controlled trial of tofacitinib monotherapy in rheumatoid arthritis.
      • Lee EB
      • Fleischmann R
      • Hall S
      • et al.
      Tofacitinib versus methotrexate in rheumatoid arthritis.
      • Kremer J
      • Li Z-G
      • Hall S
      • et al.
      Tofacitinib in combination with nonbiologic disease-modifying antirheumatic drugs in patients with active rheumatoid arthritis: a randomized trial.
      • van der Heijde D
      • Tanaka Y
      • Fleischmann R
      • et al.
      Tofacitinib (CP-690,550) in patients with rheumatoid arthritis receiving methotrexate: twelve-month data from a twenty-four-month phase III randomized radiographic study.
      • van Vollenhoven RF
      • Fleischmann R
      • Cohen S
      • et al.
      Tofacitinib or adalimumab versus placebo in rheumatoid arthritis.
      • van der Heijde D
      • Strand V
      • Tanaka Y
      • et al.
      Tofacitinib in combination with methotrexate in patients with rheumatoid arthritis: clinical efficacy, radiographic, and safety outcomes from a twenty-four-month, phase III study.
      and Phase IIIb/IV
      • Fleischmann R
      • Mysler E
      • Hall S
      • et al.
      Efficacy and safety of tofacitinib monotherapy, tofacitinib with methotrexate, and adalimumab with methotrexate in patients with rheumatoid arthritis (ORAL Strategy): a phase 3b/4, double-blind, head-to-head, randomised controlled trial.
      ,
      • Ytterberg SR
      • Bhatt DL
      • Mikuls TR
      • et al.
      Cardiovascular and cancer risk with tofacitinib in rheumatoid arthritis.
      studies with up to 72 months' follow-up, and in long-term extension studies of up to 114 months' observation.
      • Wollenhaupt J
      • Silverfield J
      • Lee EB
      • et al.
      Safety and efficacy of tofacitinib, an oral Janus kinase inhibitor, for the treatment of rheumatoid arthritis in open-label, longterm extension studies.
      • Yamanaka H
      • Tanaka Y
      • Takeuchi T
      • et al.
      Tofacitinib, an oral Janus kinase inhibitor, as monotherapy or with background methotrexate, in Japanese patients with rheumatoid arthritis: an open-label, long-term extension study.
      • Wollenhaupt J
      • Lee EB
      • Curtis JR
      • et al.
      Safety and efficacy of tofacitinib for up to 9.5 years in the treatment of rheumatoid arthritis: final results of a global, open-label, long-term extension study.
      Tofacitinib monotherapy has been associated with reduced signs and symptoms of RA when administered as first-line therapy in MTX-naïve patients,
      • Lee EB
      • Fleischmann R
      • Hall S
      • et al.
      Tofacitinib versus methotrexate in rheumatoid arthritis.
      and in patients with inadequate response to prior csDMARDs or bDMARDs.
      • Fleischmann R
      • Kremer J
      • Cush J
      • et al.
      Placebo-controlled trial of tofacitinib monotherapy in rheumatoid arthritis.
      ,
      • Fleischmann R
      • Mysler E
      • Hall S
      • et al.
      Efficacy and safety of tofacitinib monotherapy, tofacitinib with methotrexate, and adalimumab with methotrexate in patients with rheumatoid arthritis (ORAL Strategy): a phase 3b/4, double-blind, head-to-head, randomised controlled trial.
      Although combination therapy with tofacitinib + MTX has been reported to be effective in patients with RA, it is important to consider multifactorial problems with medication adherence.
      • Brown MT
      • Bussell J
      • Dutta S
      • Davis K
      • Strong S
      • Mathew S.
      Medication adherence: truth and consequences.
      • Gast A
      • Mathes T.
      Medication adherence influencing factors-an (updated) overview of systematic reviews.
      • Barton A
      • Jani M
      • Bundy C
      • et al.
      Translating research into clinical practice: quality improvement to halve non-adherence to methotrexate.
      Adherence to MTX over time may be affected by many factors, such as a lack of response or tolerability, with elevated liver enzymes and gastrointestinal issues among notable adverse reactions.
      • Tilling L
      • Townsend S
      • David J.
      Methotrexate and hepatic toxicity in rheumatoid arthritis and psoriatic arthritis.
      • Curtis JR
      • Xie F
      • Mackey D
      • et al.
      Patient's experience with subcutaneous and oral methotrexate for the treatment of rheumatoid arthritis.
      • Hope HF
      • Hyrich KL
      • Anderson J
      • et al.
      The predictors of and reasons for non-adherence in an observational cohort of patients with rheumatoid arthritis commencing methotrexate.
      Indeed, about 30% of patients with RA discontinued MTX (when used in combination with other csDMARDs) within 2 years, with nearly half citing tolerability issues.
      • Curtis JR
      • Xie F
      • Mackey D
      • et al.
      Patient's experience with subcutaneous and oral methotrexate for the treatment of rheumatoid arthritis.
      ,
      • Müller S
      • Wilke T
      • Fuchs A
      • et al.
      Non-persistence and non-adherence to MTX therapy in patients with rheumatoid arthritis: a retrospective cohort study based on German RA patients.
      Similarly, over one third of patients with RA taking MTX in combination with the bDMARD tocilizumab discontinued treatment or underwent an MTX dose decrease by 6 months (31%) and 12 months (38%) after tocilizumab initiation.
      • Pappas DA
      • Blachley T
      • Zlotnick S
      • Best J
      • Emeanuru K
      • Kremer JM.
      Methotrexate discontinuation and dose decreases after therapy with tocilizumab: results from the Corrona rheumatoid arthritis registry.
      Nonadherence to MTX can lead to poor clinical outcomes, such as greater disease activity and more affected joints.
      • Curtis JR
      • Bykerk VP
      • Aassi M
      • Schiff M.
      Adherence and persistence with methotrexate in rheumatoid arthritis: a systematic review.
      In addition, concurrent MTX use has been associated with greater persistence on bDMARDs
      • Zhang J
      • Xie F
      • Delzell E
      • et al.
      Impact of biologic agents with and without concomitant methotrexate and at reduced doses in older rheumatoid arthritis patients.
      ; therefore, it is important to optimize the treatment regimen in patients with RA by understanding these issues and providing alternative options. One area to consider is the withdrawal or tapering of concurrent csDMARDs that may be associated with long-term persistence or adherence challenges.
      In an analysis of data from two pooled, open-label, long-term extension studies, in some patients in whom concurrent treatment with MTX or glucocorticoids (GCs) was discontinued, a clinical response to tofacitinib was maintained.
      • Fleischmann R
      • Wollenhaupt J
      • Cohen S
      • et al.
      Effect of discontinuation or initiation of methotrexate or glucocorticoids on tofacitinib efficacy in patients with rheumatoid arthritis: a post hoc analysis.
      Recently, in the Phase IIIb/IV ORAL Shift study (Methotrexate Withdrawal in Patients with Rheumatoid Arthritis Who Achieve Low Disease Activity with Tofacitinib Modified-Release 11 Mg Once Daily Plus Methotrexate: A Randomised, Phase 3b/4, Non-Inferiority Trial),
      • Cohen SB
      • Pope J
      • Haraoui B
      • et al.
      Methotrexate withdrawal in patients with rheumatoid arthritis who achieve low disease activity with tofacitinib modified-release 11 mg once daily plus methotrexate (ORAL Shift): a randomised, phase 3b/4, non-inferiority trial.
      some patients with RA who achieved low disease activity with tofacitinib + MTX could withdraw from treatment with MTX without unexpected tolerability issues or significant worsening of disease activity. Using data from a claims database, the present analysis compared the characteristics, treatment patterns, and health care costs incurred by patients initiating tofacitinib treatment in combination with oral MTX, and who, after a period of treatment with tofacitinib + MTX, either were changed from their MTX treatment (discontinued or interrupted) or persisted with MTX.

      Materials and Methods

      Study Design and Patients

      This retrospective cohort study included data from the MarketScan® Commercial and Medicare Supplemental databases (Truven Health Analytics™; now IBM MarketScan®) on US patients with RA between January 2012 and April 2018. The databases encompass enrollment data from large-scale employers and health plans that provide commercial health care coverage for employees, dependents, and retirees, including health care claims information across the continuum of care (eg, inpatient, outpatient). The data were statistically deidentified, in compliance with the Health Insurance Portability and Accountability Act of 1996, and the research protocol was deemed exempt from institutional review board approval.
      For the analysis, the date of the first health care insurance prescription claim in patients newly initiating treatment with tofacitinib was defined as the index date. Patients with data eligible for inclusion were ≥18 years of age at index and had a diagnosis of RA (International Classification of Diseases, Ninth Revision—Clinical Modification [ICD-9-CM] code for RA, 714.0x–714.4x and 714.81; ICD-10, M05.* and M06.0*–M06.3* or M06.8*–M06.9*) on or within 12 months pre-index. Eligible patients had continuous medical and pharmacy enrollment for ≥12 months pre- and post-index, and initiated immediate-release tofacitinib 5 mg b.i.d. or modified-release (MR) tofacitinib 11 mg once daily in combination with oral MTX (no dose restrictions), with at least two prescription claims for each.
      With regard to tofacitinib prescription claims (at least two claims), at least one claim had to be dated between January 1, 2013, and April 30, 2017, with the second claim occurring ≤60 days after the end of the days supplied from the first (index) claim. With regard to oral MTX prescription claims (at least two claims), the first claim had to be dated ≤90 days before or after the index date. If the first MTX claim was dated ≤90 days before the index date, then the patient was not permitted a gap of >60 days between the end of the days supplied and the start of subsequent qualifying MTX claim(s), until a claim occurred on or after the index date. If the first MTX claim was on or ≤90 days after the index date, then the patient must have had a second qualifying MTX claim, with a gap of ≤60 days between the end of the days supplied from the first MTX claim and the start of the subsequent (second) claim.
      At the end of 12 months, eligible patients were assigned to one of three mutually exclusive cohorts, based on their 12-month post-index MTX persistence status, as follows: (1) patients who had a gap of ≤90 days (between the end of the days supplied from the prior MTX claim and the start of the subsequent claim) between all MTX claims until the end of 12 months post-index were classified as MTX-Persistent, and patients with a gap of ≤90 days between the end of the days supplied and the end of 12 months post-index were also considered persistent; (2) patients who had a gap of >90 days between the end of the days supplied from the prior MTX claim and the end of 12 months post-index were classified as MTX-Discontinued (these patients were permitted to have multiple persistent claims, but then have a longer gap and no new claims for MTX during the 12 months post-index); (3) patients who did not meet the criteria for MTX-Discontinued, but had a gap of >90 days (between the end of the days supplied from the prior MTX claim and the start of the subsequent claim) between any pair of consecutive MTX claims through the end of 12 months post-index were classified as MTX-Interrupted. The threshold of 90 days for the gap between MTX claims was chosen because it was believed to be an accurate claims-based proxy for discontinuation, given that patients may have had a 28-, 30-, or 84-day supply of MTX.
      Patients were excluded from this analysis if they had a claim indicating a non-RA diagnosis for which a bDMARD may have been prescribed during the 12-month pre-index period (including on the index date). These diagnoses included ankylosing spondylitis, Crohn's disease, psoriasis, psoriatic arthritis, ulcerative colitis, and juvenile RA. Patients were also excluded if they had a claim indicating at least one bDMARD on index, had received tofacitinib treatment at any time pre-index, had at least one claim for injectable MTX in the 12-month pre- and post-index periods, and/or was enrolled in a capitated health care plan during the 12-month pre- and post-index periods.

      End Points

      Patients' demographic and disease characteristics at baseline, including bDMARD and comorbidity histories, were captured as of the index date and/or at least 12 months pre-index. End points at 12 months post-index included tofacitinib treatment patterns, switching patterns, and health care costs in each MTX cohort.
      Tofacitinib treatment patterns included assessments of persistence, adherence, and effectiveness. Tofacitinib persistence was defined as a ≤90-day gap in treatment and was based on the days' supply and fill dates on the claims for tofacitinib. Tofacitinib adherence was defined in one instance as the mean proportion of days covered (PDC), calculated as the total days supplied divided by the number of days from the index date during follow-up until day 360 post-index. Tofacitinib effectiveness was estimated based on a validated composite measure that included six criteria (ie, the percentage of patients achieving all six criteria): (1) adherence (per the conventional threshold of ≥0.80 PDC

      PQA Adherence Measures [Pharmacy Quality Alliance website]. 2021. Available at: https://www.pqaalliance.org/adherence-measures. Last accessed May 2022.

      ), (2) lack of tofacitinib dosage escalation (eg, from 5 to 10 mg b.i.d.), (3) lack of a switch to a bDMARD, (4) lack of addition of a nonbiologic DMARD, (5) lack of increased oral GC use, and (6) the receipt of not more than one GC injection.
      • Curtis JR
      • Baddley JW
      • Yang S
      • et al.
      Derivation and preliminary validation of an administrative claims-based algorithm for the effectiveness of medications for rheumatoid arthritis.
      ,
      • Oladapo A
      • Barner JC
      • Lawson KA
      • et al.
      Medication effectiveness with the use of tumor necrosis factor inhibitors among Texas Medicaid patients diagnosed with rheumatoid arthritis.
      The probability of patients being switched from tofacitinib to a bDMARD within 12 months post-index was estimated. Tofacitinib treatment persistence, adherence, effectiveness, and switching patterns were assessed overall and stratified by prior bDMARD use (0, 1, and 2+ prior bDMARDs).
      Health care costs (mean and median) were assessed based on all-cause costs and RA-related costs (in patients with a diagnosis of RA noted during the visit) in the 12-month pre- and post-index periods. All-cause costs included pharmacy costs (prescriptions only) and medical costs (consisting of inpatient costs [including admissions via the emergency department (ED)] and outpatient costs [including outpatient ED visits], and including treatment-administration costs [eg, IV]).
      RA-related costs included pharmacy costs (prescriptions, treatment administration, costs by specific therapy [tofacitinib and MTX], and Healthcare Common Procedure Coding System for bDMARDs and nonbiologic DMARDs) and medical costs (consisting of inpatient costs [including admissions via the ED] and outpatient costs [including outpatient ED visits], and excluding RA-related treatment administration).

      Statistical Analyses

      Patient demographic and disease characteristics at baseline, including bDMARD and comorbidity histories, were summarized descriptively. Two sample tests (t test, χ2 test) were applied separately to the MTX-Persistent versus the combined MTX-Discontinued and MTX-Interrupted cohorts, and pairwise testing among the three cohorts, with no adjustment for multiple comparisons or for imbalances in baseline covariates. Sensitivity analyses of alternative gaps in tofacitinib and MTX persistence (>60 and >120 days) and adherence thresholds (>0.70 and >0.90) were also performed. Kaplan-Meier analysis of the cumulative probability of a switch from tofacitinib to a bDMARD, stratified by bDMARD history, was conducted.
      Generalized linear modeling was performed to evaluate the impact of baseline variables on month 12 post-index tofacitinib persistence, tofacitinib effectiveness (logistic regression models), and total all-cause and RA-related health care costs (γ model). The following baseline variables were adjusted for in the model: cohort (MTX-Persistent, MTX-Discontinued, MTX-Interrupted), index year (2013–2017), age, sex, US region (Northeast, North Central, South, West, unknown), health plan (employer or health plan), insurance type (commercial or Medicare), pre-index bDMARD use (0, 1, 2+), pre-index Quan-Charlson comorbidity score, pre-index comorbidities (cardiovascular disease, diabetes, sleep disorders, depression, anxiety), pre-index oral corticosteroid use, pre-index claims-based index for RA severity (CIRAS) score, and days from earliest claim (RA diagnosis). In the health care cost model, corresponding 12-month pre-index costs were also included. Odds ratios (logistic regression) or relative differences (γ regression), with corresponding 95% CIs, and P values (χ2) were calculated.

      Results

      Study Population

      There were 671 patients who initiated tofacitinib treatment between January 2013 and April 2017, and were eligible for inclusion (Figure 1): 504 (75.1%) were MTX-Persistent; 131 (19.5%), MTX-Discontinued; and 36 (5.4%), MTX-Interrupted. The median (interquartile range) times from the index date to the fill date of the second MTX prescription were similar across all cohorts: MTX-Persistent, 27.0 (13.0–57.0) days; MTX-Discontinued, 27.0 (15.0–52.0) days; and MTX-Interrupted, 32.0 (22.5–46.0) days. Due to the small sample size of the MTX-Interrupted cohort, this analysis was primarily focused on the findings from the MTX-Persistent and MTX-Discontinued cohorts.
      Figure 1
      Figure 1Patient disposition. Index was defined as the first tofacitinib claim. aIf the first methotrexate (MTX) prescription claim was dated as ≤90 days before the index date, then the patient was not permitted a gap of >60 days between the end of the days supplied and the start of the subsequent qualifying MTX claim(s), until a claim occurred on or after the index date. If the first MTX claim was dated as ≤90 days after the index date, then the patient must have had a second qualifying MTX claim, with a gap of ≤60 days between the end of the days supplied from the first MTX claim and the start of the subsequent (second) claim. bExclusion diagnoses: ankylosing spondylitis, Crohn's disease, psoriasis, psoriatic arthritis, ulcerative colitis, juvenile rheumatoid arthritis (RA). bDMARD = biologic disease-modifying antirheumatic drug; MTX-Discontinued = patients had a >90-day gap between end of last MTX supply and end of 12 months post-index; MTX-Interrupted = patients had a >90-day gap between MTX claims until end of 12 months post-index; MTX-Persistent = patients had a ≤90-day gap between MTX claims until end of 12 months post-index.
      Demographic and clinical characteristics at baseline were similar between the MTX-Persistent and MTX-Discontinued cohorts (Table 1). The most common comorbid conditions were cardiovascular disease (MTX-Persistent, 26.4%; MTX-Discontinued, 28.2%) and diabetes (MTX-Persistent, 17.0%; MTX-Discontinued, 18.3%). In general, just over half of the MTX-Persistent cohort received an initial 30 (rounded) days' supply of MTX, and just under half received an initial 90 (rounded) days' supply; in the MTX-Discontinued cohort, over 60% received an initial 30 days' supply of MTX, and over 30% received an initial 90 days' supply. A similar pattern was seen with the subsequent qualifying MTX supply. A total of 6 patients in the MTX-Persistent cohort had ≤120 days between the end of the subsequent qualifying MTX supply and the end of 12 months post-index, and could not have been classified as nonpersistent per the cohort definitions used in this analysis (data not shown).
      Table 1Baseline demographic and clinical characteristics and out-of-pocket health care costs at index. Data are given as number (%) of patients unless otherwise noted.
      CharacteristicMTX-Persistent (n = 504)MTX-Discontinued (n = 131)MTX-Interrupted (n = 36)
      Age, mean (SD), y56.1 (10.6)55.4 (11.1)58.8 (10.2)
      Female401 (79.6)107 (81.7)29 (80.6)
      Region
       Northeast104 (20.6)27 (20.6)6 (16.7)
       North Central120 (23.8)31 (23.7)8 (22.2)
       South225 (44.6)56 (42.8)19 (52.8)
       West53 (10.5)14 (10.7)2 (5.6)
       Unknown2 (0.4)3 (2.3)1 (2.8)
      Health plan
       Employer424 (84.1)111 (84.7)29 (80.6)
       Health plan80 (15.9)20 (15.3)7 (19.4)
      No. of prior bDMARDs
      No intra- or intergroup statistical testing was conducted on number of prior bDMARDs.
       0131 (26.0)31 (23.7)12 (33.3)
       1171 (33.9)58 (44.3)11 (30.6)
       2+202 (40.1)42 (32.1)13 (36.1)
      Initial MTX supply (1st claim)
      Rounded days' supply from typical 28-, 30-, and 84-day prescriptions.
       30 d272 (54.0)84 (64.1)18 (50.0)
       60 d9 (1.8)4 (3.1)2 (5.6)
       90 d223 (44.3)43 (32.8)16 (44.4)
      Subsequent qualifying MTX supply
      Rounded days' supply from typical 28-, 30-, and 84-day prescriptions.
       30 d262 (52.0)80 (61.1)18 (50.0)
       60 d7 (1.4)2 (1.5)1 (2.8)
       90 d235 (46.6)49 (37.4)17 (47.2)
      Quan-Charlson comorbidity score, mean (SD) [min–max]1.7 (1.1) [1.0–9.0]1.8 (1.3) [1.0–8.0]1.8 (1.5) [1.0–8.0]
      Comorbid conditions (>10% of patients)
       Cardiovascular disease133 (26.4)37 (28.2)7 (19.4)
       Diabetes85 (17.0)24 (18.3)5 (13.9)
       Sleep disorder68 (13.5)18 (13.7)5 (13.9)
       Depression61 (12.1)19 (14.5)1 (2.8)
       Anxiety55 (10.9)14 (10.7)3 (8.3)
      Rheumatologist visit
       90 d pre-index351 (69.6)89 (67.9)25 (69.4)
       12 mo pre-index382 (75.8)100 (76.3)27 (75.0)
      Oral corticosteroid use368 (73.0)100 (76.3)29 (80.6)
       Daily dose, mean (SD), mg20.3 (51.8)20.2 (37.5)17.3 (30.8)
      Out-of-pocket health care costs,
      Twelve months pre-index.
      mean (SD), USD
      All-cause2466.95 (1915.74)
      P < 0.05 for MTX-Persistent versus MTX-Discontinued.
      ,
      P < 0.05 for MTX-Persistent versus MTX‑Interrupted.
      3138.88 (2582.45)
      P < 0.001 versus MTX-interrupted.
      1997.58 (1217.06)
       Pharmacy
      Prescriptions.
      1013.92 (1096.41)1286.99 (1621.72)
      P < 0.05 for MTX-Discontinued versus MTX-Interrupted.
      821.74 (800.38)
       Inpatient
      Admissions via the emergency department (ED) and treatment administration.
      105.34 (518.25)168.54 (666.46)
      P < 0.05 for MTX-Discontinued versus MTX-Interrupted.
      38.11 (159.59)
       Outpatient
      Outpatient ED visits and treatment administration.
      1347.70 (1404.58)
      P < 0.05 for MTX-Persistent versus MTX-Discontinued.
      1683.35 (1801.17)
      P < 0.05 for MTX-Discontinued versus MTX-Interrupted.
      1137.73 (800.86)
      RA related
      In patients with a diagnosis of RA noted during the visit.
      1222.76 (1469.51)1485.44 (1754.52)
      P < 0.05 for MTX-Discontinued versus MTX-Interrupted.
      1016.40 (1011.88)
       Pharmacy
      Prescriptions, treatment administration, and Healthcare Common Procedure Coding System for bDMARDs and nonbiologic DMARDs.
      755.90 (1300.17)998.00 (1666.58)
      P < 0.05 for MTX-Discontinued versus MTX-Interrupted.
      583.46 (796.80)
        Administration220.06 (913.52)⁎⁎216.74 (741.47)71.24 (240.50)
        Treatment535.84 (976.95)781.26 (1544.55)512.22 (791.15)
       Inpatient
      Admissions via the ED.
      68.43 (434.67)76.15 (361.79)20.03 (120.16)
       Outpatient
      Outpatient ED visits.
      398.43 (484.67)411.28 (495.06)412.92 (478.22)
      Demographic and clinical characteristics, including biologic disease-modifying antirheumatic drug (bDMARD) history and comorbidities, were measured as of the index date and/or at least 12 months pre-index.
      MTX = methotrexate; MTX-Discontinued = patients had a >90-day gap between end of last MTX supply and end of 12 months post-index; MTX-Persistent = patients had a ≤90-day gap between MTX claims until end of 12 months post-index; RA = rheumatoid arthritis.
      a No intra- or intergroup statistical testing was conducted on number of prior bDMARDs.
      b Rounded days' supply from typical 28-, 30-, and 84-day prescriptions.
      c Twelve months pre-index.
      d Prescriptions.
      e Admissions via the emergency department (ED) and treatment administration.
      f Outpatient ED visits and treatment administration.
      g In patients with a diagnosis of RA noted during the visit.
      h Prescriptions, treatment administration, and Healthcare Common Procedure Coding System for bDMARDs and nonbiologic DMARDs.
      i Admissions via the ED.
      j Outpatient ED visits.
      low asterisk P < 0.05 for MTX-Persistent versus MTX-Discontinued.
      low asterisklow asterisk P < 0.05 for MTX-Persistent versus MTX‑Interrupted.
      low asterisklow asterisklow asterisk P < 0.05 for MTX-Discontinued versus MTX-Interrupted.
      P < 0.001 versus MTX-interrupted.
      Despite having similar clinical characteristics, the MTX-Discontinued cohort had significantly greater all-cause out-of-pocket health care costs during the 12 months pre-index versus the MTX-Persistent cohort (P = 0.006); this difference was attributable mainly to the significant difference in outpatient costs (P = 0.049). RA-related out-of-pocket health care costs during the 12 months pre-index were similar between the MTX-Discontinued and MTX-Persistent cohorts. In the MTX-Interrupted cohort, all-cause out-of-pocket costs were significantly less, compared with those in the other two cohorts (P = 0.038 vs MTX-Persistent, P = 0.0003 vs MTX-Discontinued), and RA-related out-of-pocket costs were significantly less compared with those in the MTX-Discontinued cohort (P = 0.042), during the 12 months pre-index (Table 1).

      Tofacitinib Treatment Patterns

      Tofacitinib treatment persistence values at 12 months post-index were similar (overlapping CIs) between the MTX-Persistent and MTX-Discontinued cohorts (Figure 2A). Treatment adherence values, as per the mean (SD) PDC, in the MTX-Persistent and MTX-Discontinued cohorts were 0.74 (0.26) and 0.73 (0.25), respectively. Although overall tofacitinib medication effectiveness (all criteria met) values were similar (overlapping CIs) between the MTX-Persistent and MTX-Discontinued cohorts (Figure 2B), an incrementally small, but significantly greater, percentage of patients in the MTX-Persistent cohort met the medication-effectiveness criterion of no addition of nonbiologic DMARD, compared with the MTX-Discontinued cohort (95.4% vs 88.6%; P = 0.003). Values of treatment adherence, as measured by the percentage of patients achieving a mean PDC of ≥0.80 (Figure 2B) at 12 months, were similar (overlapping CIs) between the MTX-Persistent and MTX-Discontinued cohorts.
      Figure 2
      Figure 2Tofacitinib treatment persistence (A) and effectiveness (B) in the methotrexate (MTX)-Persistent and MTX-Discontinued cohorts. aEffectiveness was defined as meeting all of the following six criteria: (1) adherence (defined as ≥0.80 percentage of days covered), (2) absence of dose escalation, (3) no switching of switch to a biologic disease-modifying antirheumatic drug (bDMARD), (4) no addition of a nonbiologic DMARD, (5) no increase in oral glucocorticoids (GC) use, and (6) not more than one GC injection.
      • Curtis JR
      • Baddley JW
      • Yang S
      • et al.
      Derivation and preliminary validation of an administrative claims-based algorithm for the effectiveness of medications for rheumatoid arthritis.
      ,
      • Oladapo A
      • Barner JC
      • Lawson KA
      • et al.
      Medication effectiveness with the use of tumor necrosis factor inhibitors among Texas Medicaid patients diagnosed with rheumatoid arthritis.
      MTX-Discontinued = patients had a >90-day gap between end of last MTX supply and end of 12 months post-index; MTX-Persistent = patients had a ≤90-day gap between MTX claims until end of 12 months post-index. *P = 0.003 for MTX-Persistent versus MTX-Discontinued.
      Tofacitinib persistence and adherence (mean PDC) values were numerically less (partially overlapping CIs) in the MTX-Interrupted cohort (58.3% and 0.66, respectively) compared with those in the MTX-Persistent and MTX-Discontinued cohorts. In the MTX-Interrupted cohort, the percentage of tofacitinib-adherent patients (mean PDC ≥0.80) was significantly less (38.9%, P = 0.033) compared with that in the MTX-Persistent cohort, and medication effectiveness was significantly less (19.4%) compared with both other cohorts (P = 0.016 vs MTX-Persistent, P = 0.024 vs MTX-Discontinued).
      On stratification by prior bDMARD use, the patterns of tofacitinib treatment persistence, adherence, and effectiveness were generally similar (overlapping CIs) within the MTX-Persistent and MTX-Discontinued cohorts (see Supplemental Table I). Within the MTX-Discontinued cohort, persistence was numerically greatest with 1 prior bDMARD and numerically least with 2+ prior bDMARDs (nonoverlapping CIs).
      Sensitivity analysis of tofacitinib treatment persistence in patients with >60- and >120-day gaps in MTX claims showed no significant difference between the MTX-Persistent and MTX-Discontinued cohorts (data not shown). Additionally, sensitivity analysis of adherence using thresholds of 0.70 and 0.90 yielded no significant differences between the MTX-Persistent and MTX-Discontinued groups (data not shown).
      Logistic regression analysis of tofacitinib treatment persistence, adjusted for baseline variables, demonstrated no significant differences between cohorts, whereas tofacitinib effectiveness (all criteria met) was significantly less in the MTX-Interrupted cohort than in the MTX-Persistent and MTX-Discontinued cohorts (both, P = 0.02) (Table 2).
      Table 2Logistic regression analysis of tofacitinib treatment persistence and effectiveness,
      All criteria met: adherence (achieved ≥0.80 proportion of days covered), absence of dose escalation, no switching to a biologic disease-modifying antirheumatic drug (bDMARD), no addition of a nonbiologic DMARD, no increased oral glucocorticoid (GC) use, and not more than one GC injection.28,29
      adjusted for baseline variables.
      Baseline variables adjusted for include cohort, index year, age, sex, region, health plan, insurance type, prior bDMARDs, pre-index Quan-Charlson comorbidity score, pre-index comorbidity, pre-index oral corticosteroid use, pre-index claims-based index for rheumatoid arthritis (RA) severity (CIRAS) score, and days from earliest claim (RA diagnosis).
      OR95% CIP
      Persistence
       MTX-Persistent vs MTX-Discontinued0.870.56–1.370.56
       MTX-Persistent vs MTX-Interrupted1.780.86–3.670.12
       MTX-Discontinued vs MTX-Interrupted2.040.91–4.550.08
      Effectiveness
      All criteria met: adherence (achieved ≥0.80 proportion of days covered), absence of dose escalation, no switching to a biologic disease-modifying antirheumatic drug (bDMARD), no addition of a nonbiologic DMARD, no increased oral glucocorticoid (GC) use, and not more than one GC injection.28,29
       MTX-Persistent vs MTX-Discontinued0.900.59–1.370.62
       MTX-Persistent vs MTX-Interrupted2.801.17–6.740.02
       MTX-Discontinued vs MTX-Interrupted3.121.22–7.940.02
      MTX = methotrexate; MTX-Discontinued = patients had a >90-day gap between end of last MTX supply and end of 12 months post-index; MTX-Interrupted = patients had a >90-day gap between MTX claims until end of 12 months post-index; MTX-Persistent = patients had a ≤90-day gap between MTX claims until end of 12 months post-index; OR = odds ratio.
      a All criteria met: adherence (achieved ≥0.80 proportion of days covered), absence of dose escalation, no switching to a biologic disease-modifying antirheumatic drug (bDMARD), no addition of a nonbiologic DMARD, no increased oral glucocorticoid (GC) use, and not more than one GC injection.28,29
      b Baseline variables adjusted for include cohort, index year, age, sex, region, health plan, insurance type, prior bDMARDs, pre-index Quan-Charlson comorbidity score, pre-index comorbidity, pre-index oral corticosteroid use, pre-index claims-based index for rheumatoid arthritis (RA) severity (CIRAS) score, and days from earliest claim (RA diagnosis).
      Overall, 20.4% of patients in the MTX-Persistent cohort and 13.7% of patients in the MTX-Discontinued cohort switched from tofacitinib treatment to a bDMARD within 12 months (Figure 3). The most common first bDMARDs that patients switched to were tocilizumab (MTX-Persistent, 6.2%; MTX-Discontinued, 3.8% [of all patients in the cohort]) and abatacept (MTX-Persistent, 3.4%; MTX-Discontinued, 3.8%) (Figure 3). On Kaplan-Meier analysis, the cumulative probability of patients switching from tofacitinib to a bDMARD, stratified by prior bDMARD use, is shown in Supplemental Figure 1.
      Figure 3
      Figure 3Percentages of patients switching from tofacitinib to a biologic disease-modifying antirheumatic drug (bDMARD) in the methotrexate (MTX)-Persistent and MTX-Discontinued cohorts. The overall patterns of switching in the MTX-Persistent versus MTX-Discontinued cohorts were not significantly different. MTX-Discontinued = patients had a >90-day gap between end of last MTX supply and end of 12 months post-index; MTX-Persistent = patients had a ≤90-day gap between MTX claims until end of 12 months post-index.

      Health Care Costs

      The total all-cause and RA-related health care costs (paid by insurance, not out-of-pocket) during the 12 months pre-index were not significantly different between the MTX-Persistent and MTX-Discontinued cohorts. Total all-cause health care costs in the 12-month post-index period, as well as the changes in all-cause costs from 12 months pre- to post-index, were not significantly different between the MTX-Persistent and MTX-Discontinued cohorts (Table 3). However, the all-cause pharmacy costs in the 12-month post-index period were significantly greater in the MTX-Persistent cohort compared with those in the MTX-Discontinued cohort (P < 0.05).
      Table 3Health care costs of patients in the MTX-Persistent and MTX-Discontinued cohorts at 12 months pre- and post-index. Data are given as mean [median
      A median value of 0.00 means that ≥50% of patients had zero costs.
      ] (SD) costs in year-2017 USD.
      MTX-PersistentMTX-Discontinued
      12 mo Pre-index (n = 504)12 mo Post-index (n = 504)Unadjusted Cost Difference
      Calculated as post-index costs – pre-index costs.
      12 mo Pre-index (n = 131)12 mo Post-index (n = 131)Unadjusted Cost Difference
      Calculated as post-index costs – pre-index costs.
      All-cause costs
       Total health care37,421.02

      [33,468.29]

      (29,870.94)
      52,086.86

      [46,266.40]

      (28,903.60)
      14,665.84

      [13,852.38]

      (35,243.18)
      36,652.15

      [31,889.15]

      (34,454.51)
      50,765.82

      [40,505.75]

      (44,926.25)
      14,113.67

      [8855.69]

      (45,964.55)
       Pharmacy
      Prescriptions.
      19,620.82

      [14,591.71]

      (18,607.36)
      37,190.81

      [36,635.79]

      (16,110.08)
      P < 0.05
      17,569.99

      [17,595.06]

      (20,434.87)
      17,276.20

      [13,363.05]

      (16,486.39)
      32,956.31

      [32,503.07]

      (12,406.40)
      15,680.11

      [15,846.10]

      (18,244.55)
       Medical
        Inpatient
      Admissions via the emergency department (ED) and treatment administration.
      3302.20

      [0.00]

      (12,658.57)
      4228.97

      [0.00]

      (17,412.22)
      926.76

      [0.00]

      (21,060.12)
      2759.01

      [0.00]

      (9404.05)
      6841.30

      [0.00]

      (39,714.20)
      4082.29

      [0.00]

      (41,007.35)
        Outpatient
      Outpatient ED visits and treatment administration.
      14,498.00

      [6150.72] (21,960.50)
      10,667.08

      [4914.74]

      (14,245.91)
      -3830.92

      [-703.98]

      (23,667.10)
      16,616.94

      [6569.41]

      (30,790.32)
      10,968.21

      [5440.81]

      (15,801.94)
      -5648.73

      [-606.88]

      (24,709.11)
      RA-related
      For patients with a diagnosis of rheumatoid arthritis (RA) noted during the visit.
      costs
       Total health care25,644.20

      [24,764.41]

      (22,979.81)
      39,896.25

      [37,706.83]

      (19,911.85)
      P < 0.0001 versus MTX-Discontinued.
      14,252.04

      [14,240.46]

      (28,353.63)
      22,722.45

      [16,826.75]

      (23,728.58)
      33,427.21

      [32,615.07]

      (13,056.61)
      10,704.76

      [10,390.82]

      (21,853.68)
        Pharmacy
      Prescriptions, treatment administration, and Healthcare Common Procedure Coding System for biologic disease-modifying antirheumatic drugs (bDMARDs) and nonbiologic DMARDs.
      20,191.65

      [19,727.14]

      (18,604.64)
      34,415.87

      [34,333.46]

      (12,208.09)
      P < 0.001
      14,224.22

      [13,651.03]

      (19,690.63)
      19,099.79

      [14,876.99]

      (22,274.56)
      30,404.30

      [30,181.94]

      (10,981.34)
      11,304.51

      [10,534.20]

      (22,585.37)
        Administration4456.71

      [0.00]

      (13,261.80)
      1395.18

      [0.00]

      (5942.77)
      -3061.53

      [0.00]

      (14,561.17)
      4932.23

      [0.00]

      (19,080.17)
      941.20

      [0.00]

      (4109.58)
      -3990.82

      [0.00]

      (19,224.65)
        Treatment15,734.95

      [10,913.82]

      (16,938.16)
      33,020.69

      [33,505.66]

      (13,007.18)
      P < 0.05
      17,285.74

      [17,395.91]

      (19,108.37)
      14,167.57

      [8446.89]

      (15,957.22)
      29,462.90

      [29,978.65]

      (11,318.56)
      15,295.33

      [17,075.52]

      (17,259.38)
        Tofacitinib treatmentN/A29,473.73

      [30,273.38]

      (13,435.37)
      29,473.73

      [30,273.38]

      (13,435.37)
      N/A27,256.82

      [28,331.08]

      (11,749.40)
      27,256.82

      [28,331.08]

      (11,749.40)
        MTX treatment544.03

      [517.98]

      (360.82)
      569.02

      [539.49]

      (329.67)
      P < 0.0001 versus MTX-Discontinued.
      24.99

      [-7.72]

      (377.40)
      P < 0.0001 versus MTX-Discontinued.
      486.36

      [444.69]

      (348.07)
      228.41

      [192.86]

      (187.20)
      -257.95

      [-243.56]

      (278.22)
       Medical
        Inpatient
      Admissions via the ED.
      2059.48

      [0.00]

      (9322.87)
      3247.59

      [0.00]

      (15,829.46)
      P < 0.05
      1188.11

      [0.00]

      (18,096.41)
      1177.45

      [0.00]

      (5228.05)
      1019.74

      [0.00]

      (4591.74)
      -157.72

      [0.00]

      (6886.15)
        Outpatient
      Outpatient ED visits.
      3393.07

      [1183.37]

      (12,168.13)
      2232.78

      [1007.01]

      (4271.10)
      -1160.29

      [-173.52]

      (12,377.31)
      2445.20

      [1261.85]

      (5956.38)
      2003.17

      [852.37]

      (3609.27)
      -442.03

      [-144.03]

      (6881.15)
      MTX = methotrexate; MTX-Discontinued = patients had a >90-day gap between end of last MTX supply and end of 12 months post-index; MTX-Persistent = patients had a ≤90-day gap between MTX claims until end of 12 months post-index; N/A = not applicable.
      a A median value of 0.00 means that ≥50% of patients had zero costs.
      b Calculated as post-index costs – pre-index costs.
      c Prescriptions.
      d Admissions via the emergency department (ED) and treatment administration.
      e Outpatient ED visits and treatment administration.
      f For patients with a diagnosis of rheumatoid arthritis (RA) noted during the visit.
      g Prescriptions, treatment administration, and Healthcare Common Procedure Coding System for biologic disease-modifying antirheumatic drugs (bDMARDs) and nonbiologic DMARDs.
      h Admissions via the ED.
      i Outpatient ED visits.
      low asterisk P < 0.05
      low asterisklow asterisk P < 0.001
      low asterisklow asterisklow asterisk P < 0.0001 versus MTX-Discontinued.
      During the 12 months post-index, total RA-related health care costs were significantly less in the MTX-Discontinued cohort compared with those in the MTX-Persistent cohort (difference, 6469 USD, P < 0.0001) (Table 3). Unadjusted cost difference between 12 months pre- and post-index were numerically less in the MTX-Discontinued cohort compared with that in the MTX-Persistent cohort (10,704.76 vs 14,252.04 USD, respectively). Total pharmacy and RA-related inpatient admission costs were significantly less in the MTX-Discontinued cohort compared with those in the MTX-Persistent cohort (P = 0.0007 and P = 0.006, respectively), whereas outpatient costs were generally similar. Tofacitinib treatment costs were similar between the MTX-Persistent and MTX-Discontinued cohorts. MTX treatment costs during the 12 months post-index and the change from 12 months pre- to post-index were significantly greater in the MTX-Persistent cohort compared with those in the MTX-Discontinued cohort (both, P < 0.0001). The change from 12 months pre- to post-index in total RA-related health care costs was numerically, but not significantly, less in the MTX-Discontinued cohort compared with those in the MTX-Persistent cohort.
      The total RA-related costs 12 months post-index were numerically greater in the MTX-Interrupted cohort (43,320 USD) compared with the MTX-Persistent and MTX-Discontinued cohorts. The RA-related pharmacy costs in the 12-month post-index period were significantly less in the MTX-Interrupted cohort (27,766 USD) compared with those in the MTX-Persistent cohort (P = 0.019).
      On γ regression analysis of the 12-month post-index total all-cause health care costs, adjusted for baseline variables, the differences between cohorts were not significant (Table 4). On the same analysis of total RA-related health care costs, the differences between the MTX-Persistent and MTX-Discontinued cohorts, as well as between the MTX-Discontinued and MTX-Interrupted cohorts, were significant (P 0.0001 and P = 0.018, respectively) (Table 4).
      Table 4γ Regression analysis of 12 months post-index total all-cause and total RA-related health care costs adjusted for baseline variables.
      Baseline variables adjusted for include cohort, index year, age, sex, region, health plan, insurance type, prior biologic disease-modifying antirheumatic drugs (bDMARDs), pre-index Quan-Charlson Comorbidity score, pre-index comorbidity, pre-index oral corticosteroid use, pre-index claims-based index for rheumatoid arthritis (RA) severity (CIRAS) score, days from earliest claim (RA diagnosis), and corresponding 12-months pre-index costs.
      ParameterRD95% CIP
      Total all-cause health care costs
       MTX-Persistent vs MTX-Discontinued1.030.94–1.130.51
       MTX-Persistent vs MTX-Interrupted1.090.93–1.270.29
       MTX-Discontinued vs MTX-Interrupted1.060.89–1.250.53
      Total RA-related health care costs
       MTX-Persistent vs MTX-Discontinued1.181.09–1.29≤0.0001
       MTX-Persistent vs MTX-Interrupted0.970.83–1.130.70
       MTX-Discontinued vs MTX-Interrupted0.820.70–0.970.018
      MTX = methotrexate; MTX-Discontinued = patients had a >90-day gap between end of last MTX supply and end of 12 months post-index; MTX-Interrupted = patients had a >90-day gap between MTX claims until end of 12 months post-index; MTX-Persistent = patients had a ≤90-day gap between MTX claims until end of 12 months post-index; RD = relative difference.
      a Baseline variables adjusted for include cohort, index year, age, sex, region, health plan, insurance type, prior biologic disease-modifying antirheumatic drugs (bDMARDs), pre-index Quan-Charlson Comorbidity score, pre-index comorbidity, pre-index oral corticosteroid use, pre-index claims-based index for rheumatoid arthritis (RA) severity (CIRAS) score, days from earliest claim (RA diagnosis), and corresponding 12-months pre-index costs.

      Discussion

      This analysis of real-world data in a US health care claims database aimed to assess the characteristics and treatment patterns of patients diagnosed with RA who initiated treatment with tofacitinib in combination with MTX, and subsequently either persisted with or discontinued MTX. In both unadjusted and adjusted analyses, tofacitinib treatment persistence, adherence, and effectiveness were similar between the MTX-Persistent and MTX-Discontinued cohorts in the total patient populations, and were generally similar in the subpopulations stratified by bDMARD history. On unadjusted analysis, the percentage of patients who met the medication-effectiveness criterion of no addition of a nonbiologic DMARD was significantly greater in the MTX-Persistent cohort compared with that in the MTX-Discontinued cohort; such a result could be expected in patients who were persistently using MTX and also receiving tofacitinib. Furthermore, less than one fourth of patients switched from tofacitinib to a bDMARD across both cohorts, although the percentage of patients who made this switch was numerically greater in the MTX-Persistent cohort compared with that in the MTX-Discontinued cohort. In the subgroup of patients with no prior bDMARD experience versus experience with 1 or 2+ bDMARDs, for probabilities of switching from tofacitinib of ≥0.50, the time to switch (days since index, up to 360 days) was numerically greater in the MTX-Persistent cohort. Conversely, for lesser probability of switch, patients with prior bDMARD experience generally took longer to switch than did patients with no prior bDMARD experience. These results suggest that patients who discontinued from MTX treatment experienced a clinical response similar to those who persisted on MTX, and that discontinuing MTX may have decreased the need or motivation to switch to a bDMARD.
      This analysis also examined the financial implications of tofacitinib treatment in patients with RA who either persisted with or discontinued from concurrent MTX. Differences in all-cause or RA-related health care costs prior to the index date were not statistically significant between the MTX-Persistent and MTX-Discontinued cohorts. However, on unadjusted analysis, at 12 months post-index, overall RA-related costs, as well as pharmacy and inpatient costs, were significantly less in the MTX-Discontinued cohort compared with those in the MTX-Persistent cohort (differences in costs, 6469, 4012, and 2228 USD, respectively). Differences in RA-related costs between these cohorts were also significant in adjusted analyses. The trend in the MTX-Persistent group of a switch to bDMARDs, which would be a rational approach in clinical practice, may account for some of the observed differences in treatment costs in the 12-month post-index period, although further analyses are warranted. Prior analyses comparing tofacitinib and abatacept in patients with RA and 1 prior bDMARD have indicated significantly lesser total RA-related costs with tofacitinib.
      • Harnett J
      • Gerber R
      • Gruben D
      • Koenig AS
      • Chen C.
      Evaluation of real-world experience with tofacitinib compared with adalimumab, etanercept, and abatacept in RA patients with 1 previous biologic DMARD: data from a U.S. administrative claims database.
      These data are suggestive of a significant financial burden associated with MTX and/or the bDMARDs to which patients switched, compared with tofacitinib.
      In an effort to determine the characteristics of patients who discontinued and subsequently restarted MTX, the MTX-Interrupted cohort was identified. However, the limited sample size of the group categorized as MTX-Interrupted shifted the focus to the MTX-Discontinued and MTX-Persistent comparator cohorts. We did consider combining the MTX-Interrupted cohort with the MTX-Discontinued cohort, given that all patients experienced a significant gap in MTX persistence, and that the demographic and clinical characteristics at baseline were generally similar; however, there were some important differences in findings that warrant further exploration in a larger sample of patients. These differences included less tofacitinib treatment persistence, adherence, and effectiveness, as well as lesser pre-index pharmacy costs and greater total costs. These data suggest that in this small population, medication adherence may be driven by a lack of effectiveness, multimorbidity, costly intercurrent illness, or financial issues that may influence access to care. Further research is required to examine this cohort of MTX-Interrupted patients.
      The findings from this study are consistent with those from another real-world, retrospective claims analysis of data, in which differences in tofacitinib adherence or persistence, or all-cause/RA-related total costs, were not significant in patients with RA who initiated tofacitinib as monotherapy versus combination therapy.
      • Harnett J
      • Curtis JR
      • Gerber R
      • Gruben D
      • Koenig A.
      Initial experience with tofacitinib in clinical practice: treatment patterns and costs of tofacitinib administered as monotherapy or in combination with conventional synthetic DMARDs in 2 US health care claims databases.
      In addition, the findings lend real-world context to the previously mentioned ORAL Shift study.
      • Cohen SB
      • Pope J
      • Haraoui B
      • et al.
      Methotrexate withdrawal in patients with rheumatoid arthritis who achieve low disease activity with tofacitinib modified-release 11 mg once daily plus methotrexate (ORAL Shift): a randomised, phase 3b/4, non-inferiority trial.
      That study assessed the sustained efficacy and tolerability of modified-release (MR) tofacitinib 11 mg once daily in patients with RA who achieved low disease activity with tofacitinib MR 11 mg once daily in combination with background MTX and underwent immediate MTX withdrawal.
      • Cohen SB
      • Pope J
      • Haraoui B
      • et al.
      Methotrexate withdrawal in patients with rheumatoid arthritis who achieve low disease activity with tofacitinib modified-release 11 mg once daily plus methotrexate (ORAL Shift): a randomised, phase 3b/4, non-inferiority trial.
      The 24-week, double-blind phase of ORAL Shift demonstrated non-inferiority of monotherapy with tofacitinib MR 11 mg once daily compared with tofacitinib MR 11 mg once daily + MTX, with respect to change in disease activity (primary efficacy end point). Other studies have also examined the effects of MTX withdrawal. In the COMP-ACT study (Sustained Response Following Discontinuation of Methotrexate in Patients with Rheumatoid Arthritis Treated with Subcutaneous Tocilizumab: Results from a Randomized Controlled Trial),
      • Kremer JM
      • Rigby W
      • Singer NG
      • et al.
      Sustained response following discontinuation of methotrexate in patients with rheumatoid arthritis treated with subcutaneous tocilizumab: results from a randomized controlled trial (COMP-ACT).
      patients who achieved low disease activity with tocilizumab in combination with MTX were able to discontinue MTX without significant worsening of disease activity for at least 16 weeks. Conversely, in the CAMEO study (The Canadian Methotrexate and Etanercept Outcome Study: A Randomised Trial of Discontinuing Versus Continuing Methotrexate After 6 Months of Etanercept and Methotrexate Therapy in Rheumatoid Arthritis),
      • Pope JE
      • Haraoui B
      • Thorne JC
      • Vieira A
      • Poulin-Costello M
      • Keystone EC.
      The Canadian methotrexate and etanercept outcome study: a randomised trial of discontinuing versus continuing methotrexate after 6 months of etanercept and methotrexate therapy in rheumatoid arthritis.
      patients who discontinued MTX after a 6-month period of combination treatment with etanercept + MTX showed inferior outcomes, in terms of improvement in disease activity, compared with patients who continued to receive combination treatment. However, in that study, patients were not required to have achieved low disease activity, and MTX withdrawal was not blinded. Furthermore, the recent SEAM-RA study (Etanercept or Methotrexate Withdrawal in Rheumatoid Arthritis Patients in Sustained Remission)
      • Curtis JR
      • Emery P
      • Karis E
      • et al.
      Study of Etanercept And Methotrexate in Combination or as Monotherapy in Subjects with Rheumatoid Arthritis.
      assessed the withdrawal of MTX or etanercept in patients with RA who had achieved sustained remission while receiving etanercept in combination with MTX. More patients who continued to receive etanercept maintained remission compared with patients receiving MTX monotherapy.
      The limitations of this study are consistent with those of claims data in general, and include restrictions on the types of data available and the robustness of statistical analyses. Notably, it was not possible to confirm with certainty whether tofacitinib was newly prescribed, given that this analysis was based only on the first claim captured in the database used. The definition of the MTX-Persistent cohort did not delineate cases in which prescriptions were filled but not used by the patient, or cases in which prescriptions were filled early or overlapped. Generally, a similar percentage of patients in each cohort had 30-/60-/90-day (rounded) initial and subsequent qualifying MTX supplies. Definitive reasons for MTX withdrawal, interruption, or discontinuation, and RA-specific disease activity measures, were not captured. As such, it is plausible that patients in the MTX-Persistent group had greater disease activity, or less MTX-related toxicity, than did patients in the MTX-Discontinued group. Cost data were considered to be related to RA solely on the basis of mention of RA diagnosis in the claim record. Statistical tests were limited to analyses of nonrandomized groups from observational data, with no multivariate adjustment or other analytic methods applied to control for imbalances in baseline covariates (outside of the generalized linear modeling analyses reported). A limitation of claims data is the omission of missing data; however, to mitigate this, strict requirements (eg, 12 months pre- and post-index data) were applied to the data from patients who were included in the analysis. Due to the strict inclusion and exclusion criteria of this analysis (at least two claims for both tofacitinib and MTX within a short period), a large number of patients were excluded. The use of subcutaneous MTX was also excluded. As previously discussed, analysis and interpretation of data from the MTX-Interrupted group were limited by the small sample size, and further analysis with a larger sample size is needed to confirm the robustness of these findings. A prospective study powered for propensity score matching of patients would provide such a robust assessment.
      These real-world data are consistent with the findings of the ORAL Shift clinical study, in which patients who achieved low disease activity while receiving tofacitinib + MTX were able to discontinue MTX without worsening of disease activity. Together, these data can inform treatment decisions in patients with inadequate response and/or intolerability to MTX when received in combination with tofacitinib.

      Conclusions

      The findings from this analysis of US health care claims data suggest that patients who initiate tofacitinib in combination with oral MTX are able to discontinue MTX without a negative impact on tofacitinib treatment persistence, adherence, or effectiveness, compared with those who persist with MTX. Patients who discontinue MTX may have lesser RA-related health care costs, compared with patients who continue MTX. These findings reiterate that there are opportunities for personal tailoring of care to the specific needs and tolerances of individual patients with RA.

      Declaration of Competing Interest

      Prof. Cohen has received research grants, consulting fees, or other remuneration from Pfizer Inc. Dr. Haraoui has received research grants from AbbVie; consultant's fees or other remuneration from AbbVie, Amgen, Eli Lilly, Merck, Pfizer Inc, and UCB; and has participated in speakers' bureaus for Amgen, Pfizer Inc, and UCB. Dr. Curtis has received research grants, consultant's fees, or other remuneration from AbbVie, Amgen, Bristol-Myers Squibb, CorEvitas LLC (formerly Corrona LLC), Eli Lilly, Janssen, Myriad, Pfizer Inc, Radius, Roche, and UCB. Drs. Woolcott and Gruben are employees of, and hold stock options in, Pfizer Inc. Drs. Smith and Murray were employees of Pfizer Inc at the time of this analysis. The authors have indicated that they have no other conflicts of interest with regard to the content of this article.

      Acknowledgments

      This research and its publication were supported by Pfizer Inc.
      Medical writing support, with the guidance of the authors, was provided by Eric Comeau, PhD, CMC Connect, McCann Health Medical Communications, and was funded by Pfizer Inc in accordance with the Good Publication Practice (GPP3) guideline (Ann Intern Med. 2015;163:461–464).
      Drs. Woolcott and Murray were involved in the study conception and design. Drs. Smith and Gruben were involved in the acquisition, analysis, and interpretation of the data. All of the authors, including those employed by Pfizer Inc, contributed to the interpretation of data, were involved in drafting the manuscript or revising it critically for important intellectual content, and approved the final version to be published. The views and opinions expressed within this manuscript are those of all authors and do not necessarily represent those of the funding organization.

      Appendix

      Figure S1
      Figure S1Probability of patients switching from tofacitinib after receiving (A) 0, (B) 1, or (C) 2+ prior bDMARDsa
      aEvent = patient switched from tofacitinib and switch day is <360 days. Censor = patient had not switched or switch day is ≥360 days. Abbreviations: bDMARD = biologic disease-modifying antirheumatic drug; MTX = methotrexate; MTX-Discontinued = patients had a >90-day gap between end of last MTX supply and end of 12 months post-index; MTX-Interrupted = patients had a >90-day gap between MTX claims until end of 12 months post-index; MTX-Persistent = patients had a ≤90‑day gap between MTX claims until end of 12 months post-index.
      Table SITofacitinib treatment persistence, adherence (by mean PDC and proportion achieving mean PDC ≥80%), and effectiveness, stratified by prior bDMARD use
      MTX-PersistentMTX-Discontinued
      Persistence
      n/N% (95% CI)n/N% (95% CI)
      0 prior bDMARDs88/13167.2 (59.1–75.2)23/3174.2 (58.8–89.6)
      1 prior bDMARD129/17175.4 (69.0–81.9)48/5882.8 (73.0–92.5)
      2+ prior bDMARDs129/20263.9 (57.2–70.5)24/4257.1 (42.2–72.1)
      Adherence (mean PDC)
      MeanSDMeanSD
      0 prior bDMARDs0.730.250.740.27
      1 prior bDMARD0.760.250.760.22
      2+ prior bDMARDs0.730.260.670.28
      Effectiveness
      All criteria met: adherence (achieved ≥80% PDC), absence of dose escalation, no switching to a bDMARD, no addition of a non-biologic DMARD, no increased oral GC use, and ≤1 GC injection.1, 2 Abbreviations: bDMARD = biologic disease-modifying antirheumatic drug; CI = confidence interval; GC = glucocorticoids; MTX = methotrexate; MTX-Discontinued = patients had a >90-day gap between end of last MTX supply and end of 12 months post-index; MTX-Persistent = patients had a ≤90-day gap between MTX claims until end of 12 months post-index; PDC = proportion of days covered; SD = standard deviation.
      n/N% (95% CI)n/N% (95% CI)
      0 prior bDMARDs48/13136.6 (28.4–44.9)15/3148.4 (30.8–66.0)
      1 prior bDMARD72/17142.1 (34.7–49.5)22/5837.9 (25.4–50.4)
      2+ prior bDMARDs80/20239.6 (32.9–46.4)15/4235.7 (21.2–50.2)
      Adherence (PDC ≥0.8)
      n/N% (95% CI)n/N% (95% CI)
      0 prior bDMARDs69/13152.7 (44.1–61.2)18/3158.1 (40.7–75.4)
      1 prior bDMARD104/17160.8 (53.5–68.1)34/5858.6 (46.0 –71.3)
      2+ prior bDMARDs115/20256.9 (50.1–63.8)20/4247.6 (32.5–62.7)
      a All criteria met: adherence (achieved ≥80% PDC), absence of dose escalation, no switching to a bDMARD, no addition of a non-biologic DMARD, no increased oral GC use, and ≤1 GC injection.1, 2 Abbreviations: bDMARD = biologic disease-modifying antirheumatic drug; CI = confidence interval; GC = glucocorticoids; MTX = methotrexate; MTX-Discontinued = patients had a >90-day gap between end of last MTX supply and end of 12 months post-index; MTX-Persistent = patients had a ≤90-day gap between MTX claims until end of 12 months post-index; PDC = proportion of days covered; SD = standard deviation.

      References

        • Smolen JS
        • Landewé RBM
        • Bijlsma JWJ
        • et al.
        EULAR recommendations for the management of rheumatoid arthritis with synthetic and biological disease-modifying antirheumatic drugs: 2019 update.
        Ann Rheum Dis. 2020; 79: 685-699
        • Singh JA
        • Saag KG
        • Bridges Jr, SL
        • et al.
        2015 American College of Rheumatology guideline for the treatment of rheumatoid arthritis.
        Arthritis Rheumatol. 2016; 68: 1-26
        • Curtis JR
        • Emery P
        • Karis E
        • et al.
        Study of Etanercept And Methotrexate in Combination or as Monotherapy in Subjects with Rheumatoid Arthritis.
        Arthritis Rheumatol. 2021; 73: 759-768
        • Burmester GR
        • Blanco R
        • Charles-Schoeman C
        • et al.
        Tofacitinib (CP-690,550) in combination with methotrexate in patients with active rheumatoid arthritis with an inadequate response to tumour necrosis factor inhibitors: a randomised phase 3 trial.
        Lancet. 2013; 381: 451-460
        • Fleischmann R
        • Kremer J
        • Cush J
        • et al.
        Placebo-controlled trial of tofacitinib monotherapy in rheumatoid arthritis.
        N Engl J Med. 2012; 367: 495-507
        • Lee EB
        • Fleischmann R
        • Hall S
        • et al.
        Tofacitinib versus methotrexate in rheumatoid arthritis.
        N Engl J Med. 2014; 370: 2377-2386
        • Kremer J
        • Li Z-G
        • Hall S
        • et al.
        Tofacitinib in combination with nonbiologic disease-modifying antirheumatic drugs in patients with active rheumatoid arthritis: a randomized trial.
        Ann Intern Med. 2013; 159: 253-261
        • van der Heijde D
        • Tanaka Y
        • Fleischmann R
        • et al.
        Tofacitinib (CP-690,550) in patients with rheumatoid arthritis receiving methotrexate: twelve-month data from a twenty-four-month phase III randomized radiographic study.
        Arthritis Rheum. 2013; 65: 559-570
        • van Vollenhoven RF
        • Fleischmann R
        • Cohen S
        • et al.
        Tofacitinib or adalimumab versus placebo in rheumatoid arthritis.
        N Engl J Med. 2012; 367: 508-519
        • van der Heijde D
        • Strand V
        • Tanaka Y
        • et al.
        Tofacitinib in combination with methotrexate in patients with rheumatoid arthritis: clinical efficacy, radiographic, and safety outcomes from a twenty-four-month, phase III study.
        Arthritis Rheumatol. 2019; 71: 878-891
        • Fleischmann R
        • Mysler E
        • Hall S
        • et al.
        Efficacy and safety of tofacitinib monotherapy, tofacitinib with methotrexate, and adalimumab with methotrexate in patients with rheumatoid arthritis (ORAL Strategy): a phase 3b/4, double-blind, head-to-head, randomised controlled trial.
        Lancet. 2017; 390: 457-468
        • Ytterberg SR
        • Bhatt DL
        • Mikuls TR
        • et al.
        Cardiovascular and cancer risk with tofacitinib in rheumatoid arthritis.
        N Engl J Med. 2022; 386: 316-326
        • Wollenhaupt J
        • Silverfield J
        • Lee EB
        • et al.
        Safety and efficacy of tofacitinib, an oral Janus kinase inhibitor, for the treatment of rheumatoid arthritis in open-label, longterm extension studies.
        J Rheumatol. 2014; 41: 837-852
        • Yamanaka H
        • Tanaka Y
        • Takeuchi T
        • et al.
        Tofacitinib, an oral Janus kinase inhibitor, as monotherapy or with background methotrexate, in Japanese patients with rheumatoid arthritis: an open-label, long-term extension study.
        Arthritis Res Ther. 2016; 18: 34
        • Wollenhaupt J
        • Lee EB
        • Curtis JR
        • et al.
        Safety and efficacy of tofacitinib for up to 9.5 years in the treatment of rheumatoid arthritis: final results of a global, open-label, long-term extension study.
        Arthritis Res Ther. 2019; 21: 89
        • Brown MT
        • Bussell J
        • Dutta S
        • Davis K
        • Strong S
        • Mathew S.
        Medication adherence: truth and consequences.
        Am J Med Sci. 2016; 351: 387-399
        • Gast A
        • Mathes T.
        Medication adherence influencing factors-an (updated) overview of systematic reviews.
        Syst Rev. 2019; 8: 112
        • Barton A
        • Jani M
        • Bundy C
        • et al.
        Translating research into clinical practice: quality improvement to halve non-adherence to methotrexate.
        Rheumatology (Oxford). 2021; 60: 125-131
        • Tilling L
        • Townsend S
        • David J.
        Methotrexate and hepatic toxicity in rheumatoid arthritis and psoriatic arthritis.
        Clin Drug Investig. 2006; 26: 55-62
        • Curtis JR
        • Xie F
        • Mackey D
        • et al.
        Patient's experience with subcutaneous and oral methotrexate for the treatment of rheumatoid arthritis.
        BMC Musculoskelet Disord. 2016; 17: 405
        • Hope HF
        • Hyrich KL
        • Anderson J
        • et al.
        The predictors of and reasons for non-adherence in an observational cohort of patients with rheumatoid arthritis commencing methotrexate.
        Rheumatology (Oxford). 2020; 59: 213-223
        • Müller S
        • Wilke T
        • Fuchs A
        • et al.
        Non-persistence and non-adherence to MTX therapy in patients with rheumatoid arthritis: a retrospective cohort study based on German RA patients.
        Patient Prefer Adherence. 2017; 11: 1253-1264
        • Pappas DA
        • Blachley T
        • Zlotnick S
        • Best J
        • Emeanuru K
        • Kremer JM.
        Methotrexate discontinuation and dose decreases after therapy with tocilizumab: results from the Corrona rheumatoid arthritis registry.
        Rheumatol Ther. 2020; 7: 357-369
        • Curtis JR
        • Bykerk VP
        • Aassi M
        • Schiff M.
        Adherence and persistence with methotrexate in rheumatoid arthritis: a systematic review.
        J Rheumatol. 2016; 43: 1997-2009
        • Zhang J
        • Xie F
        • Delzell E
        • et al.
        Impact of biologic agents with and without concomitant methotrexate and at reduced doses in older rheumatoid arthritis patients.
        Arthritis Care Res (Hoboken). 2015; 67: 624-632
        • Fleischmann R
        • Wollenhaupt J
        • Cohen S
        • et al.
        Effect of discontinuation or initiation of methotrexate or glucocorticoids on tofacitinib efficacy in patients with rheumatoid arthritis: a post hoc analysis.
        Rheumatol Ther. 2018; 5: 203-214
        • Cohen SB
        • Pope J
        • Haraoui B
        • et al.
        Methotrexate withdrawal in patients with rheumatoid arthritis who achieve low disease activity with tofacitinib modified-release 11 mg once daily plus methotrexate (ORAL Shift): a randomised, phase 3b/4, non-inferiority trial.
        Lancet Rheumatol. 2019; 1: E23-E34
      1. PQA Adherence Measures [Pharmacy Quality Alliance website]. 2021. Available at: https://www.pqaalliance.org/adherence-measures. Last accessed May 2022.

        • Curtis JR
        • Baddley JW
        • Yang S
        • et al.
        Derivation and preliminary validation of an administrative claims-based algorithm for the effectiveness of medications for rheumatoid arthritis.
        Arthritis Res Ther. 2011; 13: R155
        • Oladapo A
        • Barner JC
        • Lawson KA
        • et al.
        Medication effectiveness with the use of tumor necrosis factor inhibitors among Texas Medicaid patients diagnosed with rheumatoid arthritis.
        J Manag Care Spec Pharm. 2014; 20: 657-667
        • Harnett J
        • Gerber R
        • Gruben D
        • Koenig AS
        • Chen C.
        Evaluation of real-world experience with tofacitinib compared with adalimumab, etanercept, and abatacept in RA patients with 1 previous biologic DMARD: data from a U.S. administrative claims database.
        J Manag Care Spec Pharm. 2016; 22: 1457-1471
        • Harnett J
        • Curtis JR
        • Gerber R
        • Gruben D
        • Koenig A.
        Initial experience with tofacitinib in clinical practice: treatment patterns and costs of tofacitinib administered as monotherapy or in combination with conventional synthetic DMARDs in 2 US health care claims databases.
        Clin Ther. 2016; 38: 1451-1463
        • Kremer JM
        • Rigby W
        • Singer NG
        • et al.
        Sustained response following discontinuation of methotrexate in patients with rheumatoid arthritis treated with subcutaneous tocilizumab: results from a randomized controlled trial (COMP-ACT).
        Arthritis Rheumatol. 2018; 70: 1200-1208
        • Pope JE
        • Haraoui B
        • Thorne JC
        • Vieira A
        • Poulin-Costello M
        • Keystone EC.
        The Canadian methotrexate and etanercept outcome study: a randomised trial of discontinuing versus continuing methotrexate after 6 months of etanercept and methotrexate therapy in rheumatoid arthritis.
        Ann Rheum Dis. 2014; 73: 2144-2151