Protocol for a randomized multicenter study for isolated skin vasculitis (ARAMIS) comparing the efficacy of three drugs: azathioprine, colchicine, and dapsone

Background Skin-limited forms of vasculitis, while lacking systemic manifestations, can persist or recur indefinitely, cause pain, itch, or ulceration, and be complicated by infection or scarring. High-quality evidence on how to treat these conditions is lacking. The aim of this comparative effectiveness study is to determine the optimal management of patients with chronic skin-limited vasculitis. Methods ARAMIS is a multicenter, sequential, multiple assignment randomized trial with an enrichment design (SMARTER) aimed at comparing the efficacy of three drugs—azathioprine, colchicine, and dapsone—commonly used to treat various forms of isolated skin vasculitis. ARAMIS will enroll patients with isolated cutaneous small or medium vessel vasculitis, including cutaneous small vessel vasculitis, immunoglobulin A (IgA) vasculitis (skin-limited Henoch-Schönlein purpura), and cutaneous polyarteritis nodosa. Patients not responding to the initial assigned therapy will be re-randomized to one of the remaining two study drugs (Stage 2). Those with intolerance or contraindication to a study drug can be randomized directly into Stage 2. Target enrollment is 90 participants, recruited from international centers affiliated with the Vasculitis Clinical Research Consortium. The number of patients enrolled directly into Stage 2 of the study will be capped at 10% of the total recruitment target. The primary study endpoint is the proportion of participants from the pooled study stages with a response to therapy at month 6, according to the study definition. Discussion ARAMIS will help identify effective agents for skin-limited forms of vasculitis, an understudied group of diseases. The SMARTER design may serve as an example for future trials in rare diseases. Trial registration ClinicalTrials.gov: NCT02939573. Registered on 18 October 2016.


Background
Skin-limited subtypes of vasculitis may become chronic or recurrent in up to one third of patients and may persist for several years, with associated discomfort and psychosocial impact [1][2][3][4]. Cutaneous vasculitis can be pruritic or painful, may ulcerate and scar or become infected, and may be cosmetically or psychologically disturbing. Despite the chronic and impactful nature of these conditions, high-quality evidence for how best to treat them is lacking. One small randomized controlled trial has been performed to evaluate the efficacy of colchicine [5]. Among 20 patients, after 1 month of follow-up, no significant benefit to colchicine was found compared to placebo. Nonetheless, colchicine is routinely used to treat skinlimited vasculitis, as are azathioprine, dapsone, hydroxychloroquine, methotrexate, and other drugs, based on case series and expert opinion. Absent definitive data, treatment decisions are frequently based on physician preference rather than on patient characteristics or scientific evidence. Dosing and duration of therapies are variable. In lieu of well-defined steroid-sparing options, over-reliance on systemic glucocorticoids may lead to long-term side effects. Topical corticosteroids may relieve itch but fail to prevent new lesions [4,[6][7][8].
A well-designed comparative effectiveness study of agents frequently used for cutaneous vasculitis could contribute significantly to patient care. The ARAMIS study is a multicenter sequential multiple assignment randomized trial with enrichment (SMARTER) designed to compare the efficacy of three drugs commonly prescribed for skin-limited vasculitis: azathioprine, colchicine, and dapsone. Patients who do not respond to the first assigned study drug, or who have a contraindication or intolerance thereof, may be re-randomized to one of the remaining drugs. This approach increases the power to detect differences between the observed best treatment and the remaining drugs.

Study design
The study is an open-label, prospective, multicenter SMARTER design comparing the efficacy of azathioprine, colchicine, and dapsone for treatment of isolated skin vasculitis. Patients will be randomized (1:1:1, Stage 1) to receive one of the three study drugs, with response to treatment assessed at month 6 (primary endpoint). Patients who need to discontinue the assigned study drug because of failure, according to the study definition (Table 1), relapse, or an adverse event will be re-randomized (1:1, Stage 2) to receive one of the remaining two study drugs as part of a second 6-month study stage. Patients who have a contraindication to receiving one of the study drugs and those who have been treated previously but did not respond or who had an intolerance to one of the study drugs (per Table 1 Protocol definitions of treatment response and treatment failure Response to treatment: A complete or significant response to the study drug at month 6. Importantly, the absence of complete or significant response and/or flares of vasculitis occurring between enrollment (or day 1 of Stage 2) and the end of month 3 will not be considered failures; patients must continue taking the assigned study drug until at least the end of month 3, unless they have developed a severe adverse event due to the study drug that warrants its discontinuation. After month 3 (in Stages 1 and 2), the absence of complete or significant response to treatment will be considered a treatment failure Complete response: • No new skin lesions of vasculitis within the preceding 3 months Significant response: • Three or fewer skin lesions per flare with no more than 1 flare/month and minimally symptomatic, including no significant painful or necrotic lesions, within the preceding 3 months and • Physician global assessment reflecting minimal severity over the prior 28 days (0-2 on a 10-point scale) and • Patient global assessment reflecting an improvement compared to baseline by at least 1 point on a 10-point scale Treatment failure: A limited response or an absence of response to the study drug at month 6, or the development of a complication of vasculitis at any time after enrollment. The absence of response, a limited response, and/or flares of vasculitis occurring between enrollment (or day 1 of Stage 2) and the end of month 3 will not be considered failures unless the flares require the repeated use of prednisone for ≥ 2 courses Limited response: a reduction in the number and/or frequency of new skin lesions of vasculitis and/or in the severity of the lesions but with physician and/or patient global assessment persistently > 2 on a 10-point scale and/or frequency of flares > 1/month Absence of response: a complete absence of reduction or an increase in the number and/or frequency of new skin lesions of vasculitis and/or in the severity of the lesions ≥ 2 flares of skin vasculitis requiring prednisone use between enrollment (or day 1 of Stage 2) and month 3. One course of prednisone for a flare of skin vasculitis for a maximum of 3 weeks is allowed between enrollment or day 1 of Stage 2 and the end of month 3; prednisone is not allowed after month 3 Complication: appearance of an extracutaneous manifestation suggesting progression of the underlying vasculitis/condition or an incorrect initial diagnosis of skin-limited vasculitis (at any time between enrollment and month 6). Patients experiencing such complication(s) will have study drug discontinued, undergo a Treatment Stopping Visit, and be treated according to standard of care for their systemic vasculitis study definition) can be enrolled directly into Stage 2 of the study and randomized to receive one of the two remaining medications. A total of 90 patients are targeted for enrollment. A flow diagram of this trial is shown in Fig. 1.

Participants and recruitment
The participants will be recruited from several international centers affiliated with the Vasculitis Clinical Research Consortium (VCRC), including locations in the USA, Canada, Japan, and other countries.

Inclusion criteria
The following are the criteria for inclusion in the trial: Patients on a long-term, low, and stable dose of glucocorticoids (≤ 5 mg/day prednisone or equivalent) for other conditions (e.g., asthma) can be enrolled if the likelihood of requiring a dose increase for this other condition is low during the 6-month study period; these patients can remain on a low and stable dose during the study period. 7. Participants must be aged 18 years or older.
Eligibility criteria for Stage 2 (second randomization) of the trial are as follows: 1. Patients entered into Stage 1 of the study who subsequently discontinued the study drug during the 6 months of the first stage of the study or the follow-up period (month 6 to month 12) because of any one of the following endpoints per protocoldefined definitions: a lack of response or failure, a relapse (month 6 to month 12), or a side effect necessitating drug discontinuation 2. Patients with a history of significant intolerance, allergy, or serious adverse events to one of the study medications, deficit in glucose-6-phosphate dehydrogenase (G6PD) or thiopurine methyltransferase (TPMT) or history of hemolytic anemia, or who failed to respond (according to the study definition) to one of the study drugs prior to screening. These patients can be randomized to receive one of the other two study drugs.

Exclusion criteria
The following are exclusion criteria for the study: adverse events to any of the study medications (such patients can be enrolled directly into Stage 2 of the study and be randomized to receive one of the two other study drugs) 5. Patients who have contraindications to two or more of the study drugs or have been treated prior to enrollment with two or more of the study medications without adequate response or with adverse events requiring drug discontinuation 6. Deficiency in G6PD or history of hemolytic anemia (all patients must be tested for G6PD at the screening visit): such patients can be enrolled directly into Stage 2 and be randomized to receive one of the two other study drugs (azathioprine or colchicine) 7. Known low or absent TPMT activity (testing is not a requirement for study entry). Patients known to have low or absent TPMT can be enrolled directly into Stage 2 and be randomized to receive one of the two other study drugs (colchicine or dapsone) 8. Evidence of significant hepatic insufficiency or liver function tests > 2 times the upper limit of normal 9. Evidence of significant renal insufficiency or estimated glomerular filtration rate < 60 mL/min 10. Evidence of significant or symptomatic anemia or hemoglobin level < 10 g/dL 11. Comorbid condition that has moderate or high likelihood of requiring intermittent courses of prednisone (or equivalent) within the study period, according to the investigator (e.g., chronic obstructive pulmonary disease (COPD), unstable or severe asthma) 12. Current cancer or history of malignancy within the previous 5 years. Patients in remission of a cancer > 5 years or with non-metastatic prostate cancer or treated basal or squamous cell carcinoma of the skin can be enrolled 13. Active uncontrolled or serious infection that may compromise or contraindicate the use of any of the study medications 14. Patient unable to provide informed consent to participate in the study 15. Pregnant or lactating women.

Interventions
Eligible patients will be randomized by computer to receive one of the three oral medications under investigation for 6 months (primary endpoint). If effective, the medication will be continued, with total follow-up up to month 12. The medications are: Azathioprine 2 mg/kg/day (up to 200 mg per day) Colchicine 0.6 mg twice per day Dapsone 150 mg per day.
Azathioprine will be initiated at 1 mg/kg/day, and if no issues are noted after complete blood count (CBC) testing performed between days 7 and 15, the dose will be increased to 2 mg/kg/day, with subsequent CBC testing 7 to 15 days later, then every 2 weeks for the second and third months of treatment, then monthly or more frequently if dosage alterations occur. Colchicine and dapsone can be started at a lower dose, according to physician and/or patient preference(s), in order to improve initial tolerance to these drugs. However, the target study dose must be achieved within a maximum of 4 weeks after initiation, unless limited by a non-severe adverse event or intolerance to the medication (e.g., diarrhea with colchicine, nausea with azathioprine, or mild anemia on dapsone [drop in hemoglobin level < 2 g/dL with total hemoglobin level still > 10 g/dL]), in which case a lower dose of the study medication may be used. Patients may continue, for the duration of the study, other medications given for non-vasculitis indications.
Patients may have received systemic glucocorticoids for their skin vasculitis before enrollment. For patients on prednisone at the time of enrollment, prednisone should be stopped within a maximum of 6 weeks after initiation of the study drug, as per the predefined tapering schedule ( Table 2).
From the beginning of Stage 1 or Stage 2 until week 10 of the stage, patients who experience a flare of skin vasculitis can be given one short course of prednisone for a maximum of 19 days, according to one of two predefined regimens (Table 3), chosen based on physician preference and patient and disease characteristics.

Primary outcome
The primary study endpoint will be the proportion of participants with a response to therapy (according to the study definitions outlined in Table 1) at month 6 of the pooled study Stages 1 and 2.

Secondary outcomes
The secondary study endpoints include: 1. Proportion of patients with complete response to therapy (no new lesions within the preceding 3 months) at months 3, 6, and 12 2. Proportion of patients with significant response to therapy (three or fewer minimally symptomatic lesions per flare, with no more than one flare per month in the preceding 3 months) at months 3, 6, and 12 3. Time to achieve complete or significant response 4. Time to vasculitis flare for patients who achieved a complete or significant response before month 6 but subsequently relapsed 5. Frequency of vasculitis flares/new lesions compared to baseline 6. Physicians' global assessments of response 7. Patients' global assessments of response 8. Prednisone use during the study period 9. Skindex29 score at months 1, 3, 6, 9, and 12 (captures skin-related pain, pruritus, discomfort) [9, 10] Patients on long-term, low, and stable doses of glucocorticoids (≤ 5 mg/day prednisone or equivalent) for other conditions (e.g., asthma or adrenal insufficiency) can be enrolled if the likelihood of requiring a dose increase for the other condition is low and minimal during the 6-month study period (these patients will remain on the low and stable dose during the study period, with the option to receive one short course of prednisone at higher doses for skin vasculitis flare during the first 10 weeks of the study period

Participant timeline and monitoring
Participants will have a minimum of six study visits at the study site, including a screening visit, a baseline visit (which can be combined with the screening visit and randomization on the same day), and then visits at months 1, 3, 6, 9, and 12 if they respond to the Stage 1 study drug. If subsequently entered into Stage 2, patients will be seen at the study site following a schedule similar to that for Stage 1: day 1 of Stage 2 (baseline/randomization), then at months 1, 3, 6, 9, and 12. If entered directly in Stage 2, patients will be seen at the study site for a screening visit, a baseline visit (which can be combined with the screening visit and randomization), and then visits at months 1, 3, 6, 9, and 12.
Prior to enrollment, all patients will be screened for G6PD deficiency. TPMT testing is not routine standard of care at all centers; patients can be tested for TPMT activity before enrollment or not, according to local practice. Patients known to have low or absent TPMT can be enrolled directly into Stage 2 of the study and be randomized to receive one of the two other study drugs (colchicine or dapsone). All patients on azathioprine will start at 1 mg/ kg/day and will have close CBC monitoring, as detailed above, regardless of their TPMT testing [11,12].
Laboratory monitoring during the study will include, for azathioprine and dapsone, a minimum of CBC monitoring every 2 weeks for the first 3 months of therapy, then monthly or more frequently if dosage alterations occur. If there is any significant change in key laboratory parameters from baseline (including a reduction in estimated glomerular filtration rate to < 60 mL/min, transaminase level > 2 times the upper limit of normal, hemoglobin reduction > 2 g/dL, or hemoglobin < 10 g/ dL), the test should be repeated, and, if the change is confirmed, the medication will be discontinued with reporting of an adverse event. Patients are required to keep drug diaries throughout the study to document adherence to the course of therapy.
Additional study visits will occur in the case of a disease flare or an adverse event. The schedule for data collection within the study is outlined in Table 4.
If the patient experiences an endpoint defined as failure of treatment in Stage 2, he/she will be treated according to his/her physician's discretion, outside of the study.

Randomization method and stratification
Patients eligible for enrollment in Stage 1 will be randomized (1:1:1) to receive one of the three medications under investigation for 6 months. The endpoint is response to treatment at month 6 (Stage 1). If the patient is entered into Stage 2 (directly or because he/she discontinued the initial study drug because of a lack of response, treatment failure, or drug side effect), he/she will be randomized to receive one of the two remaining study drugs (1:1). The endpoint in Stage 2 is the response to treatment at month 6.
Randomization will be stratified at each stage according to: IgA vasculitis versus other forms of vasculitis Use of prednisone prior to and at the time of randomization versus no pre-enrollment use of prednisone.

Sample size
The statistical rationale and construct of the proposed design have been detailed elsewhere [13]. The aim of this design is, by pooling the results of the two study stages, to increase the power of the study to detect a difference between the arms. The primary hypothesis is that one of the study drugs will achieve a response in 50% of the patients at month 6 (Stages 1 and 2 combined) compared to only 25% of those receiving the best of the remaining two drugs. Under these assumptions, with an initial plan to enroll 90 participants (30 in each arm of the first study stage), the pooled analysis of the two study stages (first randomization and, for non-respondents, second randomization) will have 68% power to detect a significant difference between the study drugs (two-sided alpha = 0.10). The number of patients enrolled directly in Stage 2 of the study will be capped at 10 (10% of the total recruitment target for the study).
Because skin vasculitis is a rare disease, an interim analysis will be conducted after enrollment of 45 patients (50% of the planned accrual target) in order to determine, based on the response rate in each arm, whether an adjustment in the study design and/or enrollment target is possible and/or needed. Options might include early closure of the study for futility or deletion of one study arm for excessive rate of treatment failure (drop-the-loser design), with or without an accompanying increase in the sample size of the remaining arms.

Data management
Clinical information will be collected and stored for this study according to the study data collection schedule, Table 4. This information is collected as part of routine clinical practice and, therefore, is not specifically mandated by the protocol: Medical history data Physical examination data, including detailed descriptions of skin lesions CBC, tests of renal and liver function, urinalysis, and other laboratory tests relevant to disease assessment or adverse event monitoring Chest radiographs and other diagnostic images.
All study data will be collected via systems and online electronic case report forms created in collaboration with the Rare Diseases Clinical Research Network Data Management and Coordinating Center and will comply with all applicable guidelines regarding patient confidentiality and data integrity using encrypted communication links. Online collection forms and procedures have been designed to ensure optimal data quality for this study and incorporate reasonable checks to minimize transcription and omission errors.

Statistical analysis
Descriptive analyses of the participants and their skin vasculitis before and at the time of enrollment will be conducted.
The primary and secondary outcome endpoints will be estimated for each drug and each of the two stages of the study. The primary comparison will be the comparison of the best treatment arm in terms of Stage 1 response versus the maximum of the remaining two treatment arms. The inference for this comparison will be based on a weighted average of the Stage 1 and Stage 2 differences in response rates. An a priori weight of 0.70 to Stage 1 results has yielded the maximum power in simulations and will be used in the analysis [13].
Other continuous efficacy outcomes will also be estimated for each treatment arm in each stage. The inference for any pairwise comparison will be from a general linear model with independent factors of treatment and stage and the change from baseline as the dependent variable.
A descriptive analysis of the participants withdrawn from the study for treatment failure or any other reason will also be conducted. All patients will have follow-up visits at month 12.

Ethical considerations
The proposed research study is a randomized clinical trial to evaluate three different drugs that are all commonly used to treat skin vasculitis. The risks of participating in this study are expected to be low and will be due to possible side effects of the study drugs, which are all routinely used to treat the condition under study (skin-limited vasculitis).
Written informed consent will be obtained from all participants by the principal investigator. Participants are free to withdraw from the study at any time without compromising their relationship with their physician or their future medical care.
Registration of participants in this study will employ an interactive online database. The Data Management and Coordinating Center will use a system of coded identifiers to protect participant confidentiality and safety. Only the registering site will have access to the linkage between the local number and the personal participant ID number of the subject.
Study participants are instructed to contact their doctor immediately to report clinical deterioration between regular health checks, or at any time they want a consultation to discuss side effects or other health-related questions. Rapid access to specialist care is guaranteed for all study groups.
The trial was approved by the Vasculitis Clinical Research Consortium (VCRC) Data and Safety Monitoring Board of the National Institute of Arthritis and Musculoskeletal and Skin Diseases. The trial was registered on 18 October 2016 in the ClinicalTrials.gov database (https://clinicaltrials.gov) with identification number NCT02939573. The investigators plan to publish results of this trial, when complete, in a peer-reviewed journal.

Discussion
If successful, the ARAMIS trial will help identify effective agent(s) for treatment of primary cutaneous vasculitis, providing a major therapeutic advancement in an area of unmet need. ARAMIS may also serve as an example of feasible SMART designs for future trials of rare diseases other than skin vasculitis. The original and innovative study design, successfully employed in oncology, has not previously been utilized for the study of vasculitis or connective tissue diseases.
Chronic or recurrent isolated skin vasculitis is rare. Inclusion/exclusion criteria have been designed to mimic realworld scenarios and be minimally restrictive. ARAMIS is a multicenter, international study to be conducted through the VCRC with support from the Canadian Vasculitis research network (CanVasc).
ARAMIS relies on a SMART design adapted to a smaller patient sample size, with enrichment (SMARTER) [13]. SMARTs are most often used with large patient populations and well-established therapeutic options to determine the optimal sequences of therapies. The objectives are to determine the best treatment in the initial stage and, by pooling results of the other study stages, make inferences about the treatments in comparison to one another. Rerandomization of non-responders in the ARAMIS study will increase the power to detect differences between drugs. Despite the use of an adapted SMARTER design, the pooled analysis of the two ARAMIS study stages will only have 68% power to detect a significant difference between the study drugs. Whereas this power might be sufficient to detect a major difference, ARAMIS is a pilot study which explores a novel design and may serve as a preliminary trial before a possible larger one.
ARAMIS will be crucial to the development of a global research program on skin vasculitis, aimed at deciphering its pathogenesis and determining its optimal treatment, based on broad partnership between dermatologists, rheumatologists, and other specialists with an interest in vasculitis.