Study reference

Study characteristics

Patient characteristics

Intervention (I)

Control (C)

Follow-up

Outcome measures and effect size

Comments

METHOTREXATE VS MYCOFPHENOLATE MOTEFIL

Rathinam, 2014

Type of study: RCT

Country:

South India

Competing interests:

Funding for this trial was provided by That Man May See Foundation at

University of California, San Francisco (UCSF), and The South Asia

Research Fund

Inclusion criteria:

16 years or older,

noninfectious intermediate, posterior or panuveitis,

in past 60 days prior to enrolment, presence of any of the following, acc. to SUN criteria, in at least 1 eye: ≥ 1+ anterior chamber cells, ≥ 1+ anterior vitreous cells, ≥ 1+ vitreous haze, active retinal or choroidal lesions,

≥ 15 mg oral prednisolone dose at enrolment,

history of corticosteroid taper failure (i.e. inability to taper to prednisolone 10 mg or less) or an obvious chronic disease necessitating corticosteroid sparing immunosuppressive treatment (i.e. Behcet disease, multifocal choroiditis and panuveitis, serpiginous choroidopathy or birdshot retinochoroidopathy) as per standardization of SUN working group guidelines.

Exclusion criteria:

any infectious cause of uveitis, chronic hypotony (IOP < 5 mmHg for > 90 days), prior use of any immunosuppressive drug for the treatment of uveitis (other than CS), periocular or intravitreale CS injection in past 90 days, fluocinolone acetonide implant surgery in either eye in past 3 years, intraocular surgery in past 30 days, VA of hand motions or worse in better eye, pregnant or currently breastfeeding, evidence of active TB, syphilis or hepatitis

Group A: N=40 eyes

oral methotrexate (MTX)

Group B: N=38 eyes

oral mycophenolate mofetil (MMF)

Treatment regimen:

Patients received antimetabolite therapy at a reduced dose for the

first 2 weeks to ensure tolerability (15 mg a week oral methotrexate

and 500 mg twice daily oral mycophenolate mofetil).

After 2 weeks, the dose was increased to a maintenance level for the remainder of the trial: 25 mg a week oral methotrexate or 1 g twice daily oral mycophenolate mofetil. Additionally, all patients taking methotrexate were given 1 mg of folic acid daily.

The protocol provided guidelines regarding the administration of systemic and topical corticosteroids. Patients were placed on a minimum of 15 mg of oral prednisone a day at enrollment and tapered according to SUN guidelines.

If patients experienced intolerable symptoms or a nonserious adverse event, they were allowed to reduce their study dose by 1 or 2 levels; the physician remained masked. The first

dose reduction was to 20 mg for methotrexate or 1 g in the morning and 500 mg in the evening for mycophenolate mofetil. Treatment could be further reduced to 15 mg for methotrexate or 500 mg twice daily for mycophenolate mofetil.

Length of follow-up:

6 months after start treatment

Lost to follow-up:

MTX: 6/41=14.6% patients (did not receive intervention, lost to follow-up, or drop out)

MMF: 7/39=17.9% patients (did not receive intervention, lost to follow-up, or drop out)

The primary outcome was analyzed with a complete case analysis using a Fisher exact test, excluding patients who dropped out or were lost to follow-up. An additional intent-to-treat analysis was conducted using regression-

based multiple imputatio

Outcome measures:

Primary outcome:

Treatment success (i.e. achieving

corticosteroid-sparing control of inflammation in both eyes at the 5- and 6-month visit) defined as:

1) ≤0.5+ anterior chamber cells, ≤0.5+ vitreous cells, ≤0.5+ vitreous haze, and no active retinal/choroidal lesions;

2) ≤ 10 mg of oral prednisone daily and ≤2 drops of prednisone acetate 1% (or equivalent) a day; and;

3) no declaration of treatment failure because of intolerability or safety concerns.

Results:

Treatment success:

MTX: 24/35=69% patients

MMF: 15/32=47% patients

p=0.09 (no difference between the groups)

Adverse events (≥ 1 nonserious AE):

MTX: 33/41=80% patients

MMF: 32/39=82% patients

There were no differences by arm in the

number of nonserious adverse events (p= 0.38) or in the number of patients who experienced ≥ 1 adverse event (p=0.99)

Randomisation by using computer-generated permuted blocks of size 4 and 6.

All personnel responsible for outcome measurements (i.e., ophthalmologists, visual acuity examiners, and optical coherence tomography operators) were masked.

Patients and study coordinators were unmasked.

The primary outcome was analyzed with a complete case analysis, excluding patients who dropped out or were lost to follow-up. An additional intent-to-treat analysis was conducted using regression-based multiple imputation (results not reported).

Unclear how many eyes of each patient were included.

Small sample size (underpowered to detect statistically significant difference).

Co-medication (1 mg of folic acid daily to MTX group and prednisolone) (indirectness of effect)

GRADE: low quality of evidence due to lack of blinding of patients and investigators, imprecision and indirectness.

Even though both medications are generic, a 4-week supply of maintenance dose methotrexate (25 mg/wk) costs $169, whereas an equivalent supply of mycophenolate mofetil (2 g/d) costs $887.

CYCLOSPORIN VS. PLACEBO OR STEROIDS

BenEzra, 1988

Type of study: RCT

Country:

Israel

Competing interests:

Not stated

Inclusion criteria:

Diagnosis of ocular Behcet's syndrome was solely based

on the findings of involvement of the triad: eyes, oral

mucous membranes, and genitalia.

Exclusion criteria:

abnormalities of kidney or liver functions or gross hematologic or immunologic impairment

Intervention: N=20 patients

Cyclosporin

Treatment regimen

Treatment started with a single daily dose of 10 mg/kg/d, which was slowly tapered to 5 mg/kg/d as soon as a control of the intraocular inflammation was observed or a serum creatinine level higher than 150% of baseline was recorded. The internist was responsible for the modulation of the drug regimen according to systemic clinical manifestations, side effects, and laboratory data.

Control: N=20 patients

Conventional therapy (corticosteroids, N=17or chlorambucil, N=3)

Treatment regimen:

started on a regimen of 1.0 to 1.5 mg prednisolone/kg/d or chlorambucil

(Leukeran), 0.1 to 0.2 mg/kg/d, and the dosage was lowered according to clinical response. The internist was responsible for the modulation of the drug regimen according to systemic clinical manifestations, side effects, and laboratory data.

Length of follow-up:

6 months after start treatment

Lost to follow-up:

Intervention: 4/10=40%

N=2 lost for follow-up, N=1 GI discomfort, N=1 unreliable patient

Control: 4/10=40%

N=3 lost for follow-up (steroids), N=1 refused to take assigned treatment (leukeran)

Four patients in control group switched to cyclosporine in first six months because of

marked worsening of their ocular condition

Outcome measures:

Visual acuity

Extraocular manifestations of Behcet

Results:

Visual acuity:

Most eyes of patients receiving CsA demonstrate a better visual acuity after treatment than when entering the study, whereas many patients receiving conventional therapy have a similar or worse vision.

After 2 years of receiving CsA most patients continue to show stable or better visual acuity than baseline, and only in two patients was unmasking of the treatment necessary because of worsening of

the intraocular inflammatory processes and/or visual acuity. When receiving conventional therapy, most patients had

to be unmasked because of marked worsening, and only one patient is still under control after a follow-up of 2 years.

Extraocular manifestations:

Conventional therapy appears more beneficial than CsA. Aphthous stomatitis is slightly

better controlled with conventional therapy, whereas genital ulcers tend to recur less with CsA therapy. Skin lesions and arthritic manifestations are definitely less problematic, with conventional therapy than with CsA.

Randomisation by using sealed envelopes.

No report of blinding.

Small sample size, lost to follow up rate high (40% in both groups) and 4 patients (40%) in control group switched to cyclosporine in first six months (imprecision).

No statistical analysis performed.

GRADE: very low quality of evidence due to lack of blinding and very serious imprecision.

De Vries, 1990

Type of study: RCT

Country:

Netherlands

Competing interests:

Unclear. Sandoz supplied cyclosporin and placebo but no mention of further role

Inclusion criteria:

active idiopathic posterior uveitis, panuveitis or intermediate uveitis, an insufficient response to systemic corticosteroids, and a best corrected visual acuity of 0 5 or less in their best eye. This last criterion did not apply for patients with Behcet's disease or sympathetic ophthalmia, since they can be well separated from the other uveitis cases and since their poor prognosis under current therapy is well documented.

Exclusion criteria:

age under 18, presumed infectious uveitis, endstage

disease with irreversible retinal damage, corneal or lens opacities preventing the evaluation of the efficacy parameters, anticipated

intraocular surgery during the study, contraindications

to immunosuppression (that is, uncontrollable infections, malignancy, or a

history of malignancy), contraindications to oral

corticosteroid therapy, concomitant therapy

with cytostatic agents or nephrotoxic medicines,

impaired kidney or liver function, hypertension,

pregnancy, malabsorption syndrome, drug or

alcohol abuse, and non-cooperation.

Intervention: N=14 patients

cyclosporin plus low dose prednisone

Control: N=13 patients

Placebo plus low dose prednisone

Treatment regimen:

At entry into the study all patients received a low dose of prednisone (0 3 mg/kg/day to a maximum of 20 mg/day). In one group this was combined with cyclosporin in a single dose of 10 mg/kg/day, while the second group received a placebo instead. After two weeks of therapy and depending on the clinical response (see below) the corticosteroid dosage was reduced by 2.5 mg

per fortnight and eventually stopped. After oral corticosteroids had been tapered off to zero,

cyclosporin (or placebo) was reduced by 1 ml (= 100 mg) per month.

The dosages were tapered off in accordance with protocol in order to compare the two treatment groups with regard to the number of months of successful therapy before the uveitis relapsed.

The dosage of corticosteroids or cyclosporin (or placebo) could be kept constant only in case of a decrease in visual acuity of 1 rank number, compared with the best visual acuity, owing to uveitis, or in case of an unchanged visual acuity in combination with an

increase in inflammatory activity score of more than 4 points.

In order to safeguard the double-masked design of the study dose reductions were also made in the placebo group. The dosage of

corticosteroids or cyclosporin (or placebo) could not be increased.

The only concomitant medication

for uveitis allowed was dexamethasone 0.1% and/or atropine 1% eye drops. Subconjunctival

or parabulbar injections of corticosteroids were not allowed. A combination therapy was given because a

double-masked comparison of cyclosporin versus prednisone was not possible owing to the different administration forms and the outward side effects of corticosteroids.

Length of follow-up:

One year after start treatment

Lost to follow-up:

Intervention: 3/14=21%

One patient moved abroad and was lost for follow-up, one did not comply with the protocol, and one developed cataract.

Control: 3/13=23%

One patient had to undergo glaucoma surgery, one developed cataract, and

one had unexplained positive cyclosporin blood levels, probably owing to an accidental change in

assigned medication.

The findings from all the patients who entered the study were analysed. To account for those

patients who dropped out during the study the

efficacy results were expressed in Kaplan-Meier estimates of the 'survival' curves for the two treatment groups

Outcome measures:

Treatment success (i.e. number of months of successful therapy before the uveitis relapsed defined as decrease in visual acuity of ≥ 2 rank numbers compared

with the best visual acuity*, due to uveitis, or (b) an unchanged visual acuity in combination with an increase in inflammatory activity score of ≥ 4points).

* in order to make comparisons between the two measurement scales the visual acuity of each eye was given a rank number. For example, visual acuities of hand movements in one eye and 20/80 in the other were given the rank numbers 2 and 8 respectively.

Safety parameters: creatinine concentrations, diastolic blood pressure

Results:

Treatment success:

The cyclosprin group fared better than the

placebo group in number of months of successful therapy. However, the positive effect of cyclosporine was not lasting after dose reduction, and the difference between the two 'survival' curves did not reach statistical significance (p=0155, log rank test).

Safety

Under treatment three patients of the cyclosporin group had an increase in

serum creatinine concentration exceeding 50% of the baseline value; one patient for three consecutive months and the others for only one month. Interestingly also one patient of the placebo group had such an increase for one month. This patient was diagnosed as having an intermediate uveitis. No patient received other drugs known to be nephrotoxic.

Only one patient of the placebo group had a reduction in creatinine clearance exceeding 25% of the baseline value for one month, while four patients in the cyclosporin group had such a reduction.

A trace of proteinuria was occasionally

found in three patients of the placebo group and four patients of the cyclosporin group.

In four cyclosporin patients the diastolic

blood pressure increased 15-20 mmHg.

Randomisation by using a randomisation list (how it was generated not reported).

One masked investigator, with no information about the

assigned treatment, evaluated the efficacy parameters

and decided about possible changes in therapy according to the protocol. A second investigator who knew the assigned treatment checked the safety parameters and recorded the

subjective side effects.

Patients were likely to be blinded to allocation (not explicitly mentioned).

Small sample size (imprecision).

Co-medication (prednisone and dexamethasone 0.1% and/or atropine 1% eye drops) (indirectness of effect)

GRADE: low quality of evidence due to lack of blinding of all investigators, imprecision and indirectness.

Nussenblatt, 1991

Type of study: RCT

Country:

USA

Competing interests:

Not stated

Inclusion criteria:

Patients with intermediate or posterior noninfectious uveitis, 10 years of age or older, capable of understanding the goals of the project; had active bilateral disease with a visual acuity of 20/40 or worse in both eyes because of intraocular inflammation (active inflammation was defined as a recent [within four months] decrease of two or

more lines of visual acuity or active retinal lesions or vitreal haze); and had no underlying chronic infectious disorder, major organ

dysfunction, or history of neoplasia.

Exclusion criteria:

Patients who anticipated ocular surgery within the first six months of the study; who had been

treated for ocular disease with systemic corticosteroids or cytotoxic agents for at least one month before entering the study; who had insulin-dependent diabetes or uncontrolled

systemic hypertension with a diastolic pressure greater than 90 mm Hg on two occasions; or a female patient was pregnant (females of childbearing age were required to use an accepted method of contraception).

Intervention: N=28 patients

Cyclosporin

Treatment regimen:

10 mg/kg of body weight/day as a starting dosage. The dosage of cyclosporine could be as high as 15 mg/kg of body weight/day, but only for a short interval. Dosage depended on the clinical status of the patient

Topical medication could be used any time during the study.

Control: N=28 patients

Prednisolone

Treatment regimen:

Patients were given a dose of prednisolone (64 mg) that was pharmacologically equivalent

to 80 mg of prednisone if they weighed 70 kg or more, or the equivalent of 60 mg of prednisone (42 mg of prednisolone) if they weighed less than 70 kg. The maximal dose of prednisolone was the prednisone equivalent of 80 mg/ day for all patients. Dosage depended on the clinical status of the patient.

Topical medication could be used any time during the study.

Length of follow-up:

Three months after start treatment

Lost to follow-up:

6/56=11%

Not reported from which groups they were.

Outcome measures:

Treatment success

Defined as: improvement in visual acuity of 15 letters (three lines) or more in at least one eye, or an improvement of at least two increments on the vitreal haze scoring scheme. Additionally, patients taking prednisolone could take no more than 20 mg/day at the three-month evaluation.

Side effects

Results:

Treatment success:

Cyclosporine: 13/28=46%

Prednisolone: 13/28=46%

Difference n.s.

17 patients receiving prednisolone

therapy as their initial course of therapy were crossed over to cyclosporine therapy, whereas 13 of the cyclosporine patients crossed over to the prednisolone regimen. In both groups 2 patients had ≥ 15 letters improvement in VA.

Side effects

Hypertension, alterations in the serum creatinine concentration and higher mean systolic and diastolic blood pressures were more often observed in cyclosporine group.

Randomisation method not described.

Patients and outcome assessors blinded.

Small sample size (imprecision).

Co-medication (topical medication) (indirectness of effect)

GRADE: low quality of evidence due to unclear randomisation method, imprecision and indirectness.

AZATHIOPRINE VS. PLACEBO

Mathews, 1969

Type of study: RCT

Country:

Australia

Competing interests:

The work was supported by a grant from

the National Health and Medical Research Council of Australia.

Inclusion criteria:

Adult patients with diagnoses of active iritis or iridocyclitis

Exclusion criteria:

Not stated

Intervention: N=8 patients

azathioprine

Control: N=8 patients

Placebo

Treatment regimen:

Patients in the intervention

group were given 100 mg azathioprine twice daily for 3 months and patients in the control group received identical inert tablets. Systemic or local corticosteroid preparations and other standard treatments were maintained (or added) as necessary for uveitis, or for complicating systemic disease or glaucoma.

Length of follow-up:

3 months after start treatment

Lost to follow-up:

None

Outcome measures:

Improvement in symptoms (visual acuity, flares, intraocular pressure, cells in anterior chamber)

Results:

Improvement in symptoms

Patient-reported: 1/8 (12.5%) control patients and 5/11 (45%) intervention patients reported improvement (p=n.s.)

Improvement in visual acuity, flare, IOP not statistically significant more frequent in intervention group.

Reduction in VOK more frequently in azathioprine group (41%) vs. placebo group (25%), but difference was not statistically significant.

Adverse effects

One patient developed a transient neutropenia while receiving azathioprine.

Randomisation method not stated.

The patient, the ophthalmologists, and the supervising physician were unaware of which

preparation was supplied.

Very small sample size (imprecision).

Three patients who completed a course of placebo, received later a course of azathioprine, the results of which were included in the analysis; thus the nineteen completed courses included eleven with azathioprine and eight with placebo (possible confounding)

Co-medication (systemic or local corticosteroid preparations and other standard treatments) (indirectness of effect)

GRADE: very low quality of evidence due to unknown randomisation method, imprecision and indirectness.

Yazici, 1990

Type of study: RCT

Country:

Turkey

Competing interests:

The work was financially supported by Wellcome Foundation

Inclusion criteria:

Group 1:

Male patients with Behçet ‘s syndrome, < 40 years, duration of disease ≤ 24 months, no history or evidence of uveitis

Group 2:

Male patients with Behçet ‘s syndrome of any age or duration of disease, and ophthalmologic evidence of uveitis

Exclusion criteria:

Irreversible bilateral eye disease, fundi not assessable, lived more than a day’s journey from the trial center, receiving immunosuppressive agents, levamisole, or penicillamine in previous 3 months or systemic corticosteroids in previous month

Intervention: azathioprine

Group 1: N=12 patients

group 2: N=25 patients

Control: placebo

Group 1: N=13 patients

Group 2: N=23 patients

Treatment regimen:

Patients received either azathioprine (2.5 mg/kg/day , rounded to the nearest 25 mg) or matching placebo tablets. Patients left trial after severe drug reactions or on ophthalmologic advice in cases of severe eye disease; treatment with an alternative immunosuppressive agent was started without breaking the trial code. Patients with less severe exacerbations of eye disease remained in the study and were given additional methylprednisolone (usually 1000 mg intravenously on three alternate days) or oral corticosteroids.

Length of follow-up:

24 months after start treatment

Lost to follow-up:

Intervention: 3/37=8.1%

Two for military service, one due to concern about possible azoospermia.

Control: 7/36=19%

Two for military service, three for unknown reasons, one death due to pulmonary vascular disease due to Behçet ‘s syndrome, one due to seeking medical care elsewhere.

Outcome measures:

Effects on eye disease (indolent anterior and posterior uveitis and retinitis that can lead to blindness)

Results:

Withdrawal

Group 2:

6/23=26% patients on placebo were withdrawn due to severe eye disease.

Development of new eye disease

Group 1: 8/13 (62%) patients on placebo and 1/12 (8.3%) patients on azathioprine developed eye disease (p<0.001).

Group 2: 5/7 (71%) patients on placebo and 0/7 patients on azathioprine with uniocular eye disease at start developed eye disease in unaffected eye (p<0.001).

Hypopyon uveitis

Group 1: no episodes of hypopyon uveitis

Group 2: 1/25 (4%) patients on azathioprine and 7/23 (30%) patients on placebo had episodes of hypopyon uveitis (p=0.018).

Visual acuity

There was a small but significant deterioration in visual acuity in the placebo group (delta visual acuity 3.06 ± 2.40; p<0.025) (change from baseline).

Adverse effects

Patients taking azathioprine had less frequent oral ulcers, genital ulcers, arthritis. Gastrointestinal side effects were more frequent in azathioprine than placebo groups (23% vs. 18%; p<0.05). No serious side effects requiring withdrawal were attributable to azathioprine.

Randomisation method not stated.

Unclear who were blinded to allocation (article only mentions that trial was double-blind)

Small sample sizes (imprecision).

Co-medication (corticosteroid treatment remained available to all patients) (indirectness of effect)

GRADE: low quality of evidence due to unknown randomisation method, unclear blinding, imprecision and indirectness.