Article Text
PDF
Abstract
Objectives Non-adherence to rheumatoid arthritis (RA) treatments must be identified. A methotrexate (MTX) urinary dosage (METU) was recently developed. The aim of our study was to assess adherence to MTX in RA using METU in real-life conditions and to compare it with indirect adherence measurement technics.
Methods We performed a cross-sectional study at Reims University Hospital. We included over 18-year-old patients with RA treated by MTX for more than 6 months. Patients were invited to complete demographic, clinical and psychological questionnaires and adherence measurement technics (Compliance Questionnaire of Rheumatology (CQR) and Medication Possession Ratio (MPR)). A urinary sample was collected to measure MTX and information about tolerance was evaluated through Methotrexate Intolerance Severity Score.
Results 84 patients were included, 26 using oral MTX, 58 subcutaneous (SC) MTX. Among them, 73% were female, mean age was 61.5 years, MTX mean dose was 15 mg/week and 61.9% were treated by biological DMARDs (Disease Modifying Antirheumatic Drugs). 77 patients (91.7%) were adherent to treatment according to METU, whereas MPR and CQR reported less adherence (69.5% and 61.9%, respectively). MPR and METU were not significantly different in SC MTX users (p=0.059). Non-adherent patients had a higher number of tender joints and C reactive protein value (p<0.05).
Conclusion This is the first largest study evaluating MTX adherence in patients with RA using a urinary dosage. We identified that indirect adherence measurements did not reflect real-life adherence. It would be appreciable to realise METU, in a new study, in patients with RA with unexplained response to treatment, to consider it before escalating therapeutic strategy.
- methotrexate
- rheumatoid arthritis
- therapeutics
- pharmacokinetics
- sensitivity and specificity
Data availability statement
Data are available on reasonable request.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Therapeutic adherence is a crucial concept in patient management, and its measurement is complex. Adherence varies widely in rheumatoid arthritis (RA), ranging from 30% to 99%.
The urinary methotrexate (MTX) dosage developed in 2020 can measure adherence with very good sensitivity and specificity.
WHAT THIS STUDY ADDS
This study confirms that measuring adherence is difficult and that adherence varies according to the method used.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Urinary MTX dosage can be used in everyday practice to assess non-adherence in patients with RA.
A new study should be made using MTX urinary dosage in patients with RA with chronic unexplained high activity, frequent flares or the need to increase treatments.
Introduction
Rheumatoid arthritis (RA) is a chronic inflammatory joint disease that leads to significant morbidity in the absence of treatment.1 Methotrexate (MTX) is the first-line treatment for this disease. It remains an essential tool for subsequent therapeutic line due to its impact on immunogenicity and its synergy with biologics and targeted synthetic DMARDs.2 3 Moreover, The European Alliance of Associations for Rheumatology (EULAR) has recently defined the concept of ‘difficult-to-treat RA’, which takes into consideration treatment compliance and adherence before concluding that treatments are ineffective.4 5 Indeed, therapeutic adherence is a public health concern that manifests across all chronic illnesses.6 In RA, adherence is estimated at 70%, but in some studies, it can range from 30% to 99%.7 8 Current therapies are becoming increasingly expensive, sometimes with a higher risk of infections, and adherence must be assessed before considering therapeutic escalation.1 4 9–11
Various methods exist to measure adherence, but there are no strict recommendations on which measures to use, especially as each method has its advantages and disadvantages.12 13 The optimal assessment method has not been determined.13 In a recent study, a pharmacokinetic model able to describe expected patterns of urinary MTX was created, called methotrexate urinary dosage (METU).14 This allowed us to propose a new objective test of MTX adherence, helping in routine practice to differentiate adherent and non-adherent patient, thanks to CI of described concentrations depending on MTX dosage and time post intake. METU can be used in patients with RA for tasks such as blood-level assessment of hydroxychloroquine or azathioprine, which are typically used in systemic lupus erythematosus or Crohn’s disease, respectively.15–18
The aim of our study was to measure MTX adherence in patients with RA using METU. Additionally, we compared the results between patients treated with oral (PO) and subcutaneous (SC) MTX and compared our adherence measurement method with other validated methods. Finally, we analysed factors associated with non-adherence.
Methods
Study design and population
We performed a single-centre cross-sectional epidemiological study in the rheumatology department of the University Hospital of Reims between 9 June 2021 and 3 January 2022. All included patients were required to be adults and receiving outpatient or inpatient care for RA meeting the ACR (American College of Rheumatology)/EULAR 2010 criteria. They had to be using MTX for more than 6 months, with or without additional therapies, and had to agree to participate in the study. Pregnant or breastfeeding women, minors and adults under legal protection were excluded.
Data collection
Urinary methotrexate measurement (METU)
A non-sterile urine sample of at least 3 mL was collected from each patient at the time of enrolment. Patients were informed of the aim of the study. There was no lag time between the information about the study and urine collection, as the urine sample was taken directly after the consultation with the referent rheumatologist proposing the study to the patient.
The pharmacology laboratory measured MTX concentrations using chromatography coupled with mass spectrometry. The analysis takes less than half a day. The analysis is now routine in the laboratory and is usually returned to the prescribing doctor within a week.
The presence of urinary MTX is detected with an analytical threshold of 1 nM. Patients were considered adherent if their urinary MTX concentration was within the CI of values observed in the previous study, according to the delay between the intake and the dose of MTX.14 For example, the tables of distribution of attended urinary MTX values according to the time post intake (from day 0 to 7) for SC MTX doses ranging from 7.5 to 25 mg per week are available in online supplemental material (online supplemental table 1).14
Supplemental material
METU is reliable, direct and non-invasive. It is not expensive and can stay 24 hours at room temperature before achieving pharmacology laboratory.19 Sensitivity and specificity were 100% from days 1 to 6 and over 93% on day 7 for SC administration.14 Sensitivity and specificity after day 7 were evaluated in silico using a validated model, as MTX is administered only once a week.
To explore the ability of the METU kinetic profile to identify compliance, we compared the profiles of adherent patients, with repeated doses (online supplemental sub-figures of S1A, S1B, S1C, S1D on the right), with non-adherent patients who had missed one, two, three or four doses (the patient was supposed to take his dose every week). For each dose, we compared the METU values between the two groups of patients and identified the days on which the concentration ratio (R_METU=(METU) of adherent/(METU) of non-adherent) was lowest, as evidence of loss of discriminatory ability between the two groups of patients.
Supplemental material
For all doses and days (from 10 days to 28 days post administration), the most critical days were 11, 12, 13 and 14 (online supplemental figures S2A–S8A). The ratios on the other days were very high (>>>5), enabling us to identify adherent and non-adherent patients with sufficient sensitivity and specificity (online supplemental figures S2A–S8A).
For these critical days, using ROC (Receiver Operating Characteristic) curves, we analysed the discriminatory abilities of METU. For all doses and critical days,11–14 the AUCs (Area Under the Curve) of the ROC curves were greater than 90%. Sensitivity and specificity were above 80%. As indicated by the ratios (>>>5), the AUC of the ROC curve, sensitivities and specificities for other days are all greater than 90%, whatever the dose (online supplemental figures S2–S8).
Conventional adherence measurement methods
The Compliance Questionnaire of Rheumatology (CQR) is a validated self-administered questionnaire for patients with inflammatory rheumatic diseases, consisting of 19 questions.20 Scores ranged from 0 to 100, with scores above 80/100 indicating good adherence.
The Medication Possession Ratio (MPR) represents the medication possession rate. An MPR greater than 80% is considered good adherence in the literature.21 We obtained the amount of MTX dispensed by the patient’s regular pharmacy over a retrospective 6-month period starting from the date of inclusion, 6-month being the usual time between two consultations.
Data related to adherence
Clinical and demographic data collected included gender, age, weight, height, body mass index, ethnicity, educational level, socioprofessional category, marital status, disease duration, duration of MTX treatment, dose, and the administration method, associated therapy, use of external assistance for medication administration or injection, and the number of treatments and tablets prescribe to the patient.
Comorbidities were assessed using the Charlson Comorbidity Index predicting 1-year and 10-year mortality.22 Regarding the disease, we collected the number of swollen and/or tender joints, the visual analogue scale for disease activity and the immunopositivity of RA (CCP -Cyclic Citrullinated Peptides- and/or RF -Rheumatoid Factor-). On the biological side, we gathered data on creatinine, albumin, C reactive protein (CRP), mean corpuscular volume and eGFR (Estimated Glomerular Filtration Rate) levels. Using this information, we determined RA activity at enrolment using Disease Activity Score-28 (DAS28)-CRP.
Patient beliefs about medications and their illness were assessed using the Belief about Medicines Questionnaire (BMQ).23 24 MTX tolerance was evaluated using the Methotrexate Intolerance Severity Score (MISS), where a score≥6 indicates intolerance to the treatment.25
Statistical analysis
Quantitative variables were presented as medians with IQRs, while qualitative variables were presented as frequencies and percentages. The primary objective of the study was analysed based on the percentage of adherent patients and its 95% CI (using binomial distribution). The agreement between different adherence measurement tools was assessed using the kappa coefficient. A McNemar test was conducted to compare the percentage of adherence between different tools. For the comparison of patients treated with PO or SC MTX, univariate analysis was performed using the χ2 test or Fisher’s exact test according to their validity condition. Concerning factors associated with non-adherence, univariate analysis using the χ2 test, Fisher’s exact test, Student’s t-test or Mann-Whitney test were conducted. P values less than 0.05 were considered significant. CIs were calculated at 95%. Statistical analysis was performed using SAS software V.9.4, R software V.4.1.2 (The R Foundation for Statistical Computing, http://www.r-project.org) and a non-linear mixed effects modelling approach within the Monolix software (V.2021R1; Lixoft, Antony, France, http://lixoft.com/).
Legal aspects and data handling
All patients received PO information from their referring rheumatologist about the study’s purpose and procedures. Written information was provided along with a non-opposition form. For each patient included in the study, a unique identifier was assigned, and collected data were anonymised. All data were recorded in an observation notebook and transcribed into electronic format for interpretation and analysis.
Results
Population
The study was proposed to 99 patients. Five refused to participate. 10 reported having taken their MTX treatment more than 7 days ago. Descriptive analyses were conducted on a total of 84 patients, with 26 in the PO group and 58 in the SC group (figure 1). The characteristics of the 84 patients at baseline are detailed in table 1.
Flowchart. MTX, methotrexate.
Characteristics of the population at inclusion (n=84)
The study population consisted of patients with a median age of 61.5 years, with 73.8% being female (n=62). Most patients had immunopositive RA, with 55.9% of patients in remission according to the DAS28-CRP at baseline. The median dose of MTX used was 15 mg/week (IQR: 11.2–20), most commonly in combination with a biologic therapy.
Adherence according to METU
According to METU, 91.7% (CI: 83.6% to 96.6%) of patients were defined as adherent. There was no statistically significant difference between patients treated orally (PO) or subcutaneously (SC), with 92.3% of adherent patients in the PO population (CI: 74.9 to 99.1; p=1) and 91.4% in the SC population (CI: 81.0 to 97.1; p=1) (table 2).
Comparison of different adherence measurement methods
Comparison of different adherence measurement methods
Subsequently, we compared various adherence measurement methods (table 2). There was no statistically significant difference between the use of METU or MPR to quantify adherence in patients treated with SC MTX (p=0.059).
Factors associated with non-adherence
We then compared adherent and non-adherent patients according to METU (table 3). Non-adherent patients had a higher number of swollen joints and higher CRP levels at baseline (p=0.011 and p=0.048, respectively). They also had a higher number of flares within 6 months prior to the study and a higher DAS28-CRP 6 months prior to inclusion. The BMQ-Concerns Score tended to be higher in non-adherent patients (p=0.068). There were no statistically significant differences in terms of gender, age, follow-up location or disease duration. The median MISS was 0 (0–1), with the most common side effects being nausea and fatigue.
Comparison of adherent and non-adherent patients according to METU
Discussion
According to METU method, adherence rate was 91.7%. METU was not correlated with the indirect measurement methods studied (CQR and MPR). Factors associated with non-adherence were more swollen joints and a higher CRP value at inclusion in non-adherent patients.
This is the first study to evaluate METU in current practice. METU is an innovative, simple, direct and non-invasive technique.14 This technique can be developed in all healthcare establishments and for several diseases. The practitioner can have confidence in the results before discussing adhesion with their patients.
Studies evaluating adherence in patients with RA are common and the percentage of adherent patients varies widely, depending on the assessment method.8 The lack of gold standard for adherence measurement concerning MTX makes the comparison difficult. Our adherence rate is similar to 95% (n=138) found in the only study using direct measurement in current practice (by plasma assay).26 However, blood sampling appears to be more invasive, technical and time-consuming than urinalysis. Conversely, our rate is higher than those in the literature using indirect methods.27 28 This difference may be explained by the evaluation method and the safety profile of MTX. Indeed, only 9.5% of patients were intolerant, compared with 30 to 40% in most studies.29 30 Therefore, the conclusions of our study cannot stand for all patients with RA and a new adherence study using METU 3–6 months after starting MTX should be made.
In line with the literature, these different methods were not correlated with each other.8 This can be related to several points. First, the temporality of the measurement methods is different. METU has a 7-day setback, which is equivalent to one dose, while CQR and MPR evaluate adherence over months. METU reflects short-term adhesion, does not consider the dynamic process of adhesion, which varies over time. Second, MPR, CQR and METU have different limitations. MTX adherence may be overestimated by METU due to a white coat effect.31 MPR can overestimate adherence, and it does not reflect treatment intake but delivery. It is time-consuming, making it difficult to use in common practice.12 Most studies use self-administered questionnaires as CQR.20 32
However, a recent systematic review of the literature concluded that, to date, there is no simple, reliable and valid questionnaire for assessing medication adherence.33 Some argue that Medication Event Monitoring Systems are gold standards, but because of the bias due to the patient knowing he is monitored, the possible tampering of data, and its cost, it cannot be used in clinical practice.20
The limitations of the study include the monocentric recruitment in a tertiary centre, where adherence is reputed to be higher due to frequent intervention in therapeutic education.34 Moreover, the low number of non-adherent patients resulting in a lack of power to study the factors associated with non-adherence.
More swollen joints and a higher CRP value at inclusion were associated with non-adherence. Factors associated with non-adherence to treatment are largely evaluated in the current literature, but studies are discordant.34 In our study, there was also a trend toward higher BMQ-Concerns in non-adherent patients, consistent with other studies.35 36 In contrast to previous study, age, sex, duration and dose of MTX were not associated with non-adherence.37 38 The need of an external aid for medication preparation or injection’s realisation was not associated with adherence. These data have not been studied in previous studies.
In summary, there is no gold standard for measuring compliance, and it is important to have a direct method. METU has a real potential for finding non-adherent patients. However, monitoring MTX adherence in all patients with RA treated by MTX might be bothering for patients, and more costly than useful. Therefore, METU might be restricted to patients with a chronic unexplained high activity of RA or frequent flares, or the need to increase treatments. It would be appreciable to monitor MTX adherence in a better selected population. In addition, future research can focus on other diseases using MTX or level of urinary concentration to assess efficacity.
Data availability statement
Data are available on reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants and was approved by an ethics committee(s) or institutional board(s). This study protocol was approved on 26 May 2021, following evaluation by the West III Ethics Committee for the Protection of Persons (registration number: 2021-A00650-41). Written information was provided along with a non-opposition form.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Presented at The 35th congress of The Société Française de Rhumatologie. Oral communication: Théate N, Geoffroy M, Kanagaratnam L, et al. Dosage urinaire du méthotrexate dans l’évaluation de l’observance chez les patients atteints de polyarthrite rhumatoïde. Revue du Rhumatisme. 2022;89:A95. The EULAR 2023. Scientific Abstracts. Poster View Théate N, Geoffroy M, Kanagaratnam L, et al. Pos0446 urinary methotrexate dosage as an adherence measurement in rheumatoid arthritis patients. Ann Rheum Dis. 2023;82:479.
Contributors NT, MG, LK, ZD and JHS contributed to the conception and design of the study. MG, IC, LB, AH-R and JHS furnished patients. NT collected the samples. CG and ZD contributed to the analysis of the samples. LK contributed to the analysis. NT, MG, JHS, AH-R and LB revised it critically for important intellectual content. NT is responsible for the overall content as guarantor.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.