The Glucocorticoid Low-dose Outcome in RheumatoId Arthritis Study

Rationale: Rheumatoid arthritis (RA) is a condition with high impact both on the individual and society. In the context of comparing the effectiveness of existing healthcare interventions in the elderly, RA is a condition highly relevant to the community since it has a strongly negative impact on the quality of life of the individual, is particularly frequent in the elderly, and is associated with significant costs. RA management remains challenging: there is an urgent unmet medical and societal need for improved treatment strategies that are effective, safe and affordable. Evidence based information on the glucocorticoid (GC) harm/benefit balance will have a high impact on RA treatment strategies and on treatment strategies for the many other inflammatory disorders for which GCs are considered. The Glucocorticoid Low-dose Outcome in RheumatoId Arthritis Study (GLORIA) is a 5-year project funded by the European Commission under the Horizon 2020 Program, designed to address these needs.

Objective: The primary objectives of the GLORIA project are twofold:

1. To assess the effectiveness, safety and cost-effectiveness of low-dose GC therapy (5 mg/day) compared to placebo given for two years as co-treatment for elderly RA patients (≥ 65 years) in a pragmatic randomized trial;
2. To assess study medication adherence through a medication packaging solution, and test the effectiveness of smart device technology to improve adherence.

Other objectives of the GLORIA project are: to deliver an outcome prediction model for individual patient outcome, to tailor treatment strategies for elderly RA patients with comorbidities; and deliver data to support:

1. better guidelines on RA treatment in the elderly;
2. more accurate information for elderly RA patients, their physicians and researchers;
3. improved strategies for trial design and conduct in the elderly.

Study design: The GLORIA study is a randomized, double-blind, placebo-controlled pragmatic multicenter clinical trial to assess the effectiveness and safety of a daily dose of 5 mg prednisolone or matching placebo in elderly RA patients. Patients will be randomized into two arms: the experimental arm (receiving prednisolone 5 mg/day) or the control arm (receiving placebo). The design emulates the routine care setting: eligibility criteria are very liberal, assessments and procedures are tailored to represent standard of care, and concurrent antirheumatic treatment is allowed next to the trial medication with minimal limitations. Furthermore, all patients will have an adherence monitoring device loaded into the cap of the drug bottle; adherence data will be monitored throughout the trial. In addition, to test the effect of adherence reminders, a substudy (another trial) will be nested in the main GLORIA trial.

Substudy design: The substudy is limited to patients with a smart device (smartphone, tablet, etc) who have completed at least 3 months of the main study on treatment. The experimental arm of the substudy will receive an application loaded onto their smart device that communicates with an adherence monitoring device loaded into the cap of the drug bottle and delivers reminders to improve adherence. The control arm of the substudy will not have this application and reminders. The substudy has a duration of three months.

Study population: Patients of 65 years of age and older with RA according to the 1987 or the 2010 classification criteria of the American College of Rheumatology (ACR) and the European League Against Rheumatism (EULAR), requiring antirheumatic therapy because of inadequate disease control, as evidenced by a disease activity score of 28 joints calculated with erythrocyte sedimentation rate (DAS28) ≥2.60.

Substudy population: Patients in the main study who have completed at least 3 months on treatment, in possession of and familiar with a smart device.

Intervention: In this two-armed clinical trial, patients will be randomized to either the experimental arm or the control arm. The experimental arm will receive prednisolone 5 mg/day added to existing antirheumatic treatment. The control arm will receive matching placebo added to existing antirheumatic treatment. Treatment duration is two years per patient. Subsequently, study drug is tapered in linear fashion to zero in 6 weeks by inserting increasing numbers of non-treatment days. Patients experiencing a flare at that time can restart open label prednisolone at the discretion of the treating rheumatologist.

Co-interventions: As part of standard of care all patients will receive Calcium 500 mg/Vitamin D3 800 IU. Besides the study medication, almost all treatment is allowed; both treatment for comorbidities, as well as antirheumatic treatment. This includes biologic and nonbiologic disease-modifying antirheumatic drugs (DMARDs), non-steroidal anti-inflammatory drugs (NSAIDs), short term oral or parental GCs for comorbidities and acetaminophen. However, it is advised not to start other antirheumatic therapy (DMARD, biologic) or give intra-articular or intramuscular GC injections, especially not in the first 3 months, but it is allowed if clinically judged as unavoidable. In such cases, preferred administration is at baseline.

Substudy intervention: All patients will have an adherence monitoring device loaded into the cap of the drug bottle that will be equipped with a wireless transmitter, which not only tracks adherence but can also communicate real time with a smart device of the patient (smartphone, tablet, etc.), through special software to remind patients of the time of medication. In the substudy, patients with a smart device will receive an application loaded on their smart device that communicates with the adherence monitoring device loaded into the cap of the drug bottle. The app sends a reminder message to the patient when it's time to take the medication and in case of non-adherence the app also sends an alert message to inform the patient he or she has forgotten to take the medication. Eligible patients will be randomized to either the experimental arm that receive reminders on their smart device, or to the control arm that will not receive reminders, for a period of 3 months.

Main study endpoints:

• To measure benefit, primary endpoints are a) signs and symptoms: the time-averaged mean value (estimated from linear mixed models) of the DAS28; b) damage progression: change from baseline after 2-years in total Sharp/van der Heijde damage score of hands and forefeet radiographs.
• To measure safety, the primary endpoint is the total number of patients experiencing at least one serious adverse event, or one clinical event related to the disease or its therapy.
• Other major outcomes are cost-effectiveness, cost-utility, and medication adherence.

Assessment takes place at varying intervals, and includes seven clinic visits and 3 assessments by telephone.

Nature and extent of the burden and risks associated with participation, benefit and group relatedness: GLORIA is a pragmatic trial, with measurements that are almost all part of standard of care, and therefore there are no additional risks associated with these measurements. However, the intervention in this trial is that patients are randomized to either standard of care with low-dose GCs or standard of care without low-dose GCs for a duration of two years. It is known that GCs have strong favorable effects on disease activity, of which patients randomised to the control arm (placebo) cannot benefit. However, side effects of GCs are known as well, and patients in the experimental arm (prednisone 5 mg/day) might suffer from them. Although side effects predominantly occur when GC are used in high doses for long periods of time, elderly are more likely to suffer from adverse effects of the disease and its therapy than younger patients. However, elderly are underrepresented or even excluded from many clinical trials, so it is difficult to estimate their risk: this forms the rationale for the trial. To reduce risks, participants are monitored and patients with a low probability of benefit and patients with a high probability of harm will be excluded from participation to this trial. For the substudy to measure adherence through an innovative application, no additional risks are expected: if the application appears to be successful, patients randomised to the intervention can only benefit from it since it improves their medication adherence; in case the application is not successful, for patients not randomised to the application or for patients without a smart device, no difference is expected with the 'normal' situation, in which patients do not have access to an adherence application.