Doxycycline slows x-ray progression in knee OA, but what is the significance of this “exciting finding”?

The trial showing that the tetracycline antibiotic doxycycline slows down the radiographic progression of osteoarthritis (OA) in established knee OA, which has led to suggestions that the drug may be a disease-modifying agent for osteoarthritis (DMAOA), has been published in the July 2005 issue of Arthritis & Rheumatism. The study, involving 431 patients and headed by Dr Kenneth Brandt (Indiana University School of Medicine, Indianapolis), was first presented at the 2003 meeting of the American College of Rheumatology and was reported by rheumawire in detail at that time.

“This is an exciting finding,” says an accompanying commentary from Dr Paul Dieppe (University of Bristol, UK). However, it adds, “Many concerns remain about the meaning, significance, and implications of this work.” Although the effect on radiographic progression of established knee OA was significant, there was little effect on symptoms such as pain, leading Dieppe to comment, “I myself would not want to take a drug that may improve my joint space width [JSW] if it were not going to help my pain or function.” “In my view, we should not be chasing drugs as a means of modifying the outcome of OA,” Dieppe continues. “Rather, we should be looking at simple ways to achieve the benefits that accompany mechanical interventions such as joint distraction and osteotomy. . . . While I applaud the superb work of Brandt and his colleagues, I worry that it might, paradoxically, increase the trend toward too much ‘medicalization’ of people with OA as well as the rush to prescribe drugs for them that might affect radiographic findings but with unknown and potentially serious toxicity.” Significant effect on x-ray progression The trial was conducted in a group of obese women (aged 45-64 years) who had x-ray evidence of OA in one knee but not the other when they were assessed on a standing anteroposterior (AP) radiograph. An earlier study in a similar group of patients had suggested that about half of these women would develop OA in the contralateral knee within 24 months. Hence, when the trial began, Brandt et al hoped that they would be able to investigate the effect of doxycycline on established OA in the index knee and also study the effect of the drug on the development of OA in the contralateral knee. However, as they explain in the discussion, during the course of the study they realized that while the contralateral knee was radiographically normal in the conventional standing AP view, x-rays taking at other angles (in the lateral semiflexed AP view and/or the patellofemoral view) showed evidence of OA. “Hence, this study did not assess the effect of doxycycline on incident OA in the contralateral knee, but rather on the progression of rather mild OA in that joint, which was not apparent in the baseline standing AP view,” they write. The treatment group took oral doxycycline 100 mg daily for 30 months. The drug showed a statistically significant effect on radiographic progression of OA in the index knee: at 16 months, x-rays showed that the mean loss of joint space width was 40% less in the group taking the drug than the group taking placebo, and after 30 months, it was 33% less. However, there was no difference between the drug and placebo groups in the effect seen on the contralateral knee (the relatively disease-free, or very mildly diseased, knee). Also, there were no significant differences between the groups in the severity of pain reported, although the researchers note that mean pain scores at baseline were low and remained low throughout the trial, “suggesting the presence of a floor effect.” There was one significant effect on one pain measurethe doxycycline group showed a reduction in the frequency of follow-up visits at which subjects reported a 20% or greater increase in pain in the index knee relative to the level of pain they had at their previous visit. While describing these results as “promising,” the commentary asks whether such a small statistically significant effect is of any clinical significance. The women participating in this trial had relatively mild knee OA and were not receiving extensive treatment for their condition, Dieppe points out. While this situation reduces the scope for demonstrating improvement, it also “raises the question of whether small change in someone with early disease has much meaning in the life of that individual.” He also wonders about how generalizable the findings may be, pointing out that OA may be rather different in men than in women and in early and late stages and is affected by different factors at other sites in the body. Adverse events that occurred more frequently in the group taking doxycycline than placebo were “well recognized side effects” of the drug and included monilial vaginitis, sun sensitivity, and nonspecific gastrointestinal symptoms, the researchers note. Patients in the treatment group reported fewer urinary-tract infections, and there was a trend toward fewer respiratory-tract infections. The antibiotic was investigated in OA because of its effect of inhibiting metalloproteinases, which may prevent cartilage damage, but it also affects bones and other tissues and enzyme systems, Dieppe comments. This should “raise concerns about its long-term use in older people with chronic disease, most of whom will have comorbidities,” he cautions. Results are “confusing”Approached for comment, leading osteoarthritis researcher Dr David Felson (Boston University, MA), who acts as an editorial consultant to www.jointandbone.org, describes the results of the doxycycline trials as “confusing.” He points out: “People in this trial had two knees affected by OA. For one knee, doxycycline protected against joint-space loss (probably cartilage loss). For the other knee, it did not (not even close). Further, there was no effect on knee pain. I am not certain what to do about these findings in terms of patient treatment. Given the absence of any effect on symptoms, I don’t believe that patients with knee OA will or should be willing to take this treatment long term.””The absence of consistent results for both knees makes me skeptical the drug had any effect,” Felson tells rheumawire, although he adds that doxycycline does have favorable effects on OA in animal models of disease. In their discussion of the results, Brandt et al comment on the lack of effect in the contralateral knee. Although the joint-space narrowing (JSN) was identical in the treatment and placebo groups at 16 months, the rate was numerically greater in the doxycycline group at 30 months, although the difference was not statistically significant. “The lack of evidence that doxycycline slowed the rate of JSN in the contralateral knee may have been due to the fact that some matrix degradation pathways are more important in the later stages of OA than in the earlier stages. Doxycycline may have interfered with processes driving cartilage breakdown in the index knee that were not (yet) operative in the contralateral knee,” they write. Asked whether these latest findings have practical applications as yet, Felson says: “Here’s what I’m doing for those with knee OA, I am not using doxycycline. In my view, this is a mechanically based disease and, in most cases, a drug with a questionable chondroprotective effect will not help. However, for those with generalized OA (including hand OA and other joints), in which there are no effective drugs, I am inclined to suggest doxycycline to patients, mostly because it’s safe and may have a protective effect on cartilage loss. In my view, persons with generalized OA more often have a primary OA process that may involve a predilection for cartilage loss even in a noninjurious mechanical environment. These types of patients may be more likely to respond.” (Source: Brandt KD, Mazzuca SA, Katz BP, et al. Effects of doxycycline on progression of osteoarthritis. Arthritis Rheum 2005; 52:2015-2025; Dieppe P. Disease modification in osteoarthritis: are drugs the answer? Arthritis Rheum 2005; 52:2015-2025: Rheumawire Joint and Bone: July 2005.)