COX-2 inhibition in osteoarthritis:effects on cartilage

Abstract

The topic of this thesis was to provide more insight in the direct effects of one of the selective COX-2 inhibitors, celecoxib on articular cartilage. Issues of major relevance to clinical practice since it is essential that compounds used to treat osteoarthritis do not impair the ability of chondrocytes to repair the already damaged extracellular matrix.

First we study the in vitro effect of celecoxib on cartilage under normal, inflammatory conditions and on osteoarthritic cartilage. It was found that celecoxib had no direct effects on healthy human cartilage. Under inflammatory conditions celecoxib reversed the disturbed proteoglycan turnover of the cartilage. Also osteoarthritic cartilage was positively influenced by addition of celecoxib. There were even clues of cartilage repair. These results are in contrast with the effects of some of the non-selective NSAIDs. Furthermore in vitro experiments revealed involvement of prostaglandin-E2, the main product of COX-2, in enhanced cartilage proteoglycan release but not synthesis. Nitric oxide elevation seems to be involved in inhibition of proteoglycan synthesis, independent of prostaglandin-E2 and with that probably less sensitive to regulation by (selective) COX-2 inhibitors. NSAIDs inhibit both COX-1 and COX-2. This appears to be associated with the well-known gastrointestinal adverse events: the more COX-1-selective, drugs appear to have the tendency to cause more gastrointestinal damage. Maybe also the direct effects on cartilage by NSAIDs are dependent on their COX-2 (un)selectivity. It could well be that the adverse effects on cartilage of some of the conventional NSAIDs results from inhibition of COX-1. For this reason, we evaluated the in vitro effect of some frequently used NSAIDs on human osteoarthritic articular cartilage. It was concluded that NSAIDs with low COX-2/COX-1 selectivity show adverse direct effects on osteoarthritic cartilage (which was confirmed by use of an experimental selective COX-1 inhibitor), whereas high COX-2/COX-1 selective NSAIDs did not show such effects and even might induce cartilage reparative effects.

Although in vitro systems have the advantage that the direct effects on osteoarthritic cartilage can be studied, without interference of synovial inflammation, extrapolation to the in vivo situation has its limitations. Therefore we used an animal model of osteoarthritis, the canine "groove" model. This model is based on surgically applied damage of the articular cartilage followed by transient forced loading of the affected joint and shows characteristics mimicking human osteoarthritis. To establish whether the observed characteristics of degeneration in this model represents the surgically applied damage or whether they are the results of progressive features of osteoarthritis we evaluated this "groove" model shortly after surgery, as described in chapter 6. In chapter 7 we actually evaluated whether the beneficial in vitro effects of celecoxib could also be demonstrated in vivo. Dogs received daily placebo, or celecoxib orally. Induction of osteoarthritis using the groove model resulted in macroscopic, histological and biochemical damage of cartilage. However, none of the parameters was significantly changed upon celecoxib treatment. Additional experiments showed that celecoxib had reached the joint and that canine cartilage is sensitive to celecoxib. It could be that celecoxib had been beneficial to degenerated cartilage in vivo but that these effects had been counteracted by increased loading of the affected joint because of the well-known analgesic effects of celecoxib and the important role of joint loading in progression of joint damage in this model. However, this needs to be established.

As evaluation of human cartilage in a clinical trial is difficult, we used an approach in which the advantages of in vivo (clinical) treatment were combined with detailed biochemical evaluation of the cartilage as performed in vitro. In chapter 8 we studied the effects of selective COX-2 inhibition after short-term treatment of patients with severe knee osteoarthritis. Patients were treated 4 weeks prior to a planned knee replacement surgery with celecoxib, naproxen, indomethacin, or received no treatment. Using this approach we demonstrated an in vivo generated chondro-reparative effect of celecoxib (the most outspoken) and naproxen, in contrast to indomethacin, in osteoarthritic cartilage.