Minocycline in Rheumatoid Arthritis: A 48-Week, Double-Blind, Placebo-Controlled Trial

Landmark NIH-sponsored, 48-week, multi-center, double-blind, placebo-controlled clinical trial, demonstrating the safety and efficacy of minocycline as a treatment for rheumatoid arthritis.

Minocycline in Rheumatoid Arthritis (MIRA): 48-Week, Double-Blind, Placebo-Controlled TrialAnn Intern Med. 1995;122(2):81-89. DOI: 10.7326/0003-4819-122-2-199501150-00001

In a press release, by the National Institutes of Health (NIH) on January 14th, 1995, Michael D. Lockshin, M.D., acting director of the National Institute of Arthritis and Musculoskeletal and Skin Diseases (NIAMS) stated, “The MIRA Trial is an example of the way a clinical study should be done.” Further, he said, “Minocycline is another drug to add to the armamentarium of treatments for rheumatoid arthritis.” In an accompanying editorial, Harold E. Paulus, M.D., a rheumatologist at the University of California, commented that the MIRA trial was, “…well designed and [well] executed,” and that minocycline, due to its modest, but significant, clinical benefit and minimal toxicity, “… compares favorably with currently used disease-modifying anti-rheumatic drugs.”


The MIRA Trial was a randomized, double-blind, placebo-controlled study in which neither physicians nor patients were aware of who received the test drug, minocycline, or the inactive placebo. In order to participate, patients were required to meet established criteria for rheumatoid arthritis and to have six or more swollen joints and nine or more tender or painful joints. If participants were taking non-steroidal anti-inflammatory drugs (NSAIDs) and/or low doses of oral steroids prior to the study, they were allowed to continue using these medications, but were taken off any disease-modifying anti-rheumatic drugs (DMARDs) a month prior to starting the trial. Patients meeting these criteria were then split into two randomized groups with similar characteristics. One group of 109 patients received 100mg of minocycline, twice a day by mouth, and the second group of 110 patients received a matching placebo (inactive capsules), twice daily.

The objectives of the MIRA Trial were to determine long-term safety of the use of minocycline and whether it was able to produce improvement in joint swelling and tenderness in rheumatoid arthritis. By the end of the 48-week trial, 54% of the patients taking minocycline and 39% of patients taking the inactive placebo experienced at least 50% improvement in the number of swollen joints. In those with joint tenderness, 56% of the minocycline group and 41% of the placebo group also experienced a 50% improvement. Although a degree of placebo-effect is expected in controlled studies, improvements in the MIRA placebo group were also attributed to the fact that these patients were encouraged to continue taking their usual NSAIDs and corticosteroids. Most notably, however, although improvements began by week 12 in both study groups, the placebo group plateaued in improvement by week 24, whereas the minocycline group continued to experience improvements in joint swelling and tenderness through to completion of the 48-week study. Additional significant improvements in laboratory markers of disease activity, such as erythrocyte sedimentation rate (ESR), hematocrit, platelet count and rheumatoid factor (RF) levels were noted in the minocycline group, in spite of comparable subjective progress assessments provided by patients and doctors in both groups. Adverse effects experienced in the minocycline group were mild and infrequent, the most common being dizziness, and only 7 patients dropped out of the study due to side-effects.

RBF Commentary

In the NIH press release, commentators remarked that the MIRA researchers were not assessing the mechanism of action of minocycline; rather its efficacy and safety for long-term use. Nevertheless, aside from its antimicrobial properties, speculative credit for the significant improvements of the patients taking minocycline in the trial was attributed to its metalloproteinase (MMP) effects, its ability to block the joint-destructive enzyme, collagenase, and its anti-inflammatory effects.

Although minocycline’s disease-modifying properties indeed bear merit, these fail to explain the remarkable results experienced by Dr. Brown’s patients to whom he would prescribe low, pulsed doses of minocycline. A typical pulsed protocol used by Brown would have been 100mg, once or twice a day, on just three days per week. Dr. Brown paid particular attention to carefully titrating the minocycline dose to patient tolerance, relative to pre-existing inflammation levels, and it was often prescribed in less than ¼ or ½ of the total weekly doses used in the MIRA Trial. This meant that Brown’s patients were receiving far less of the immune-modulatory and anti-inflammatory effects due to the minocycline, but still achieving comparable results. In fact, Dr. Brown found that patients with severe disease would have a superior response to a very low, initial pulsed dose, thereby limiting the expected paradoxical early worsening, called a Jarisch-Herxheimer flare (or, “herx,” for short).

A major limitation of clinical trials is that individualized dose titration is not feasible for data collection, because the exact same dose must be given to all patients receiving the medication, regardless of tolerance issues, in order to reduce study variables. Further, due to the duration of the MIRA Trial (48 weeks), it is unknown whether a greater proportion of patients receiving minocycline would have responded and achieved full remission, given more time.

In the most entrenched and recalcitrant cases, it can take up to thirty months from the beginning of therapy until the patient clearly turns the corner toward improvement, and the achievement of lasting remission can take several years… In shorter-term cases – and short term doesn’t necessarily mean less severe – complete remission can be achieved in less than six months. —The New Arthritis Breakthrough, by Henry Scammell, pages 269-270

Back To Questions