1999 O'Dell Study

[Guest guest]

The 1999 O'Dell Study

O'Dell, JR, sen, G, Haire, CE, Blakely, K, Palmer,

W, Wees, S, Eckhoff, PJ, Klassen, LW, Churchill, M,

Doud, D, Weaver, A, , GF, Treatment of early

seropositive Rheumatoid Arthritis with minocycline:

Four-Year followup of a double-blind,

placebo-controlled trial, Arthritis & Rheumatism,

1999, 42:8, 1691-1695.

Objective:

To compare patients treated with conventional therapy

in the early phase of Rheumatoid Arthritis and those

treated with minocycline after 4 years of followup.

Methods:

Forty-six patients with seropositive Rheumatoid

Arthritis of more than 6 weeks and less than 1 year in

duration were enrolled in a double-blind study of

minocycline (100 mg, twice daily) versus placebo.

After the blinded portion of the study (3-6 months,

depending upon response), all patients were treated

with conventional therapy. If a patient had been

receiving minocycline during the blinded portion of

the study, but had a disease flare during the open

portion, minocycline was restarted in most cases.

Twenty of the 23 original minocycline-treated patients

and 18 of the 23 original placebo-treated patients

were available for followup (mean 4 years). This

report compares the patients randomized to receive

placebo for 3 months and then conventional therapy for

the duration of 4 years versus those originally

randomized to receive minocycline.

Results of the minocycline-treated group after 4

years:

65% of the patients improved their symptoms by at

least 75%, some achieving a remission.

40% of the patients fulfilled remission criteria

without DMARDs or steroids.

50% of the patients never required any DMARD or

steroids.

20% of the patients reported skin hyperpigmentation.

15% of the patients discontinued minocycline because

of hyperpigmentation.

Introduction:

Currently, rheumatologists are emphasizing the

importance of early control of RA and studies have

shown that patients respond best when treated early

with disease-modifying therapy. In a double-blind,

controlled trial of minocycline compared with placebo

in patients with early seropositive RA, we have

previously shown that the minocycline-treated patients

were significantly better at 6 months and continued to

show excellent responses after 1 year. In this

communication, we extend those observations and report

superior (and, in some cases, dramatic) results after

a median followup of 4 years in the

minocycline-treated patients compared with controls

treated in a conventional manner.

Patients and methods:

Experimental design. We enrolled 46 patients in the

original 6-month, double-blind, controlled study.

Twenty-three of the patients were randomized to

receive minocycline (100 mg, twice daily) and 23 to

receive placebo. Three months after enrollment,

patients were evaluated; if a patient did not meet 50%

improvement criteria, he or she was withdrawn from the

blinded portion of the study. All patients remaining

in the blinded portion were again evaluated for 50%

improvement after a further 3 months of therapy. The

blinded portion of the study ended after the 6-month

evaluation and, regardless of the response to therapy,

the minocycline or placebo was stopped in all patients

at 6 months.

Once the blinded portion ended and the data were

recorded, the physician was informed of the

randomization and was then free to prescribe whatever

therapy he or she deemed most appropriate, including

DMARDs alone or in combination, prednisone and

minocycline. If the patient had been receiving

minocycline during the blinded portion of the study

and had a good response (15 patients) but had a

disease flare during the open portion (all 15

patients), minocycline was restarted in most cases.

Concurrent therapy. During the open portion of the

study, physicians could prescribe any medication,

including changing NSAIDs, starting or restarting

minocycline, using DMARDs alone or in combination,

and/or initiating steroids.

Results:

In the original protocol we randomly assigned each of

the 46 patients to 1 of the 2 treatment groups (23

patients in each). There were no significant

differences between the groups at entry. Results of

the blinded portion of the study have been published

previously; 65% of the minocycline-treated group and

13% of the placebo-treated group met 50% improvement

criteria at the end of the blinded portion of the

study.

Toxicity (pages 1692-1693). None of the

minocycline-treated patients withdrew due to toxicity

during the blinded portion of the study. One patient

in the placebo group withdrew because of a

gastrointestinal bleed. Subsequent to the blinded

phase, 3 of the minocycline-treated patients

discontinued minocycline because of hyperpigmentation

and 1 patient reported mild hyperpigmentation but

elected to continue therapy. This occured at 1, 2.5,

3, and 3.5 years of therapy. In the 3 patients who

stopped minocycline, the hyperpigmentation decreased

slowly over time. None of the patients reported

dizziness that precluded continuation of the

treatment.

Table 1 (page 1693).

Long-term results in patients with early rheumatoid

arthritis treated with

minocycline versus placebo

___________________________________________________________

No. of patients available for followup: 20 vs. 18

Years of followup, mean (range): 3.8 (1.5-6.3) vs. 4

(2.0-6.1)

No. in remission *: 8 (40%) vs. 3 (17%) **

No. in remission without DMARDs: 8 (40%) vs. 1 (6%)

***

No. with ACR 75% response ****: 13 (65%) vs. 4 (22%)

*****

No. with DMARD therapy: 10 (50%) vs. 16 (89%)

No. with prednisone therapy: 9 (45%) vs. 11 (65%)

No. with current minocycline therapy: 11 (55%) vs. 4

(24%)

___________________________________________________________

Notes:

* Remissions according to American College of

Rheumatology (ACR) criteria, but measured at only a

single time point.

** One of these 3 patients was treated with

minocycline during the open phase.

*** Minocycline is not considered a disease-modifying

antirheumatic drug (DMARD) in this analysis.

**** All patients with > 75% response, including those

in remission.

***** Two of these 4 patients were treated with

minocycline during the open phase.

___________________________________________________________

Results of long-term treatment with minocycline (page

1693). Of the 23 patients who were originally treated

with minocycline, 20 have had followup past 1 year

(median 4.25 years, mean 3.8 years), as have 18 of the

23 placebo-treated patients. The current status of

these patients is shown in Table 1. The difference

between the number of patients in the minocycline

group and the number in the placebo group whose RA was

in remission without DMARDs (minocycline not

considered a DMARD) or steroids was significant, 1 of

18 (6%) in the placebo group versus 8 of 20 (40%) in

the minocycline group, as was the number of patients

requiring DMARD therapy, 16 of 18 (89%) of the

placebo-treated patients compared with 10 of 20 (50%)

of the minocycline-treated patients. One of the 3

patients originally in the placebo group whose RA was

in remission at followup was receiving minocycline at

the time of the followup evaluation. Importantly, 50%

of the patients (10 of 20) originally treated with

minocycline never required treatment with DMARDs or

steroids, and 40% (8 of 20) fulfilled remission

criteria without DMARDs or steroids.

Time course of response to minocycline (page 1693).

Figure 1 plots the total joint counts (sum of tender

and swollen joints) versus months of minocycline

treatment for the 15 patients who were responders to

minocycline. Although significant response had been

seen by 3 months (at which time the mean total joint

count of 31.1 had decreased to 13.5), maximal response

did not occur until at least 9 months [mean total

joint count of 5 at 9 months, and 1 at 18 months].

Discussion (pages 1693-1694):

" With currently available DMARD therapy, complete

remissions of RA are disappointingly rare. This

realization has fueled a surge of interest in

alternate forms of therapy for RA, including a

significant increase in the use of combination DMARD

therapy and of minocycline. Our double-blind,

placebo-controlled study has demonstrated the benefit

of minocycline when used to treat patients with

seropositive RA within the first year of disease, and

the present report confirms that these patients

continue to do well for up to 4 years (mean followup).

We believe that several key points about our study

design are worth emphasizing: all of the patients

studied had early disease (these patients have been

shown by many to be most responsive to therapy); all

were rheumatoid factor positive (and thus we studied a

relatively homogeneous patient population and a group

of patients who were destined to have a low rate of

spontaneous remission and who could be predicted to

have ongoing, aggressive disease); and, finally, we

chose to define success as a 50% improvement in

composite criteria instead of the 20% that is often

used.

Our findings and those of other investigators suggest

that the maximum benefit of minocycline does not occur

until after 1 year of therapy. Therefore, the results

of the original study are even more remarkable. We did

not want to continue placebo treatment for more than 3

months in patients with active RA; therefore, the

double-blind portion of the trial was continued for

only 6 months, and some patients may have been dropped

from the minocycline treatment arm before they had an

opportunity to have a maximal response.

The magnitude of improvement in our

minocycline-treated patients was dramatic compared

with the modest but statistically significant benefit

in the Netherlands and Minocycline in Rheumatoid

Arthritis trials. Reconciliation of these seemingly

disparate results requires acknowledgment that our

study used an entirely different patient population.

The most significant difference was the disease

duration, which averaged 8.6 years and 13 years in

those other trials and <5 months in our trial. The

observed difference in magnitude of response may be

explained by the fact that patients with early disease

respond better to most therapies. Alternatively, there

may be a window of opportunity early in RA, in which

minocycline can produce dramatic benefit.

Additionally, we observed fewer side effects,

especially dizziness, in our trial compared with the

Netherlands trial. The reasons for this are unclear,

but the young age of our patients is one possible

explanation. Like all other treatments for RA,

minocycline may need to be continued indefinitely to

remain effective; therefore, the localized

hyperpigmentation that appears to increase with

duration of minocycline therapy is problematic.

Recently, we have switched some of our patients to

doxycycline, which is similar to minocycline in most

of its known activities, but appears to be associated

with less hyperpigmentation.

Tetracyclines, particularly minocycline and

doxycycline, are inhibitors of metalloproteinases,

including collagenase and gelatinase.

Metalloproteinases are almost certainly active in RA

joint destruction, and studies in animal models of

arthritis (both RA and osteoarthritis) have shown

benefit with minocycline or doxycycline treatment.

Modified derivatives of minocycline that retain their

ability to inhibit metalloproteinases but do not have

antibacterial effects remain effective in some of

these models. In patients with RA, minocycline or

doxycycline treatment has been shown to result in

decreased synovial collagenase production, decreased

levels of metalloproteinase breakdown products in the

urine, and decreased collagenase activity in the

saliva. In this latter open-label study, clinical

features of RA also improved significantly.

Early advocates for the use of tetracyclines in the

treatment of RA based their choice on the

antibacterial effect, believing that RA was initiated

and perpetuated by an infectious agent. Two currently

well-accepted disease-modifying drugs, gold and

sulfasalazine, were initially used for similar

reasons. Recent experiences with Lyme disease, human

immunodeficiency virus, and hepatitis C are vivid

reminders of how much we have to learn about

infectious triggers of diseases with immunologic and

rheumatic manifestations. Therefore, it is clearly

possible that an infectious agent will be shown to

play a role in the pathogenesis of RA. Recent data on

evidence of organisms demonstrated by polymerase chain

reaction in the joints of some RA patients,

differences in the bowel flora of RA patients with and

those without erosive disease, and the ability of one

of the most commonly used and effective DMARDs,

sulfasalazine, to alter bowel flora are intriguing.

In addition to their antimicrobial and

antimetalloproteinase effects, the tetracyclines have

been shown to have antiinflammatory effects,

immunomodulating effects, and the ability to inhibit

angiogenesis. With regard to the immunomodulating

effects of tetracyclines, the recent reports of

apparent drug-induced lupus in acne patients treated

with minocycline are of interest. Finally, there has

been much recent enthusiasm for, and some evidence to

support the use of, agents with activity against tumor

necrosis factor alpha (TNFa) in the treatment of RA.

Metalloproteinases are involved in the processing of

TNF and may be affected by matrix metalloproteinase

inhibitors.

Our study does not address the critically important

question of the mechanisms of action of minocycline.

Based on the observed benefit in animal models of

arthritis when tetracyclines are used, we postulate

that part of the efficacy is due to inhibition of

matrix metalloproteinases. We believe that

metalloproteinase inhibition will be a key part of

combination therapy for the future treatment of RA.

Whether antibacterial effects are important is

unclear, but we certainly cannot rule out this

possibility. Interestingly, in the majority of our

patients who had favorable responses to minocycline,

the RA flared when this treatment was stopped. Whether

this reaction favors one of the proposed mechanisms

over another is unclear.

We believe that minocycline is effective for treating

seropositive RA within the first year of disease.

Further studies are needed to define the optimal

duration of treatment, mechanism(s) of action, and to

compare minocycline with other DMARDs given alone and

in combination early in the disease. "