HomeThe enzymeTechnologyIngredientsStudiesAbout

A broad MMP-blocking drug brought joint side effects and no clear benefit

MMP13 biologyHuman trial2007

This 2007 trial gave 401 people with mild to moderate knee osteoarthritis PG-116800, a drug that blocks matrix metalloproteinases broadly, or placebo for 12 months. It made no significant difference to joint space or symptoms, the highest dose was stopped for side effects, and joint pain was the most common adverse effect.

Evidence typeHuman trial
Who or what was studied401 people with mild to moderate knee osteoarthritis
Dose and durationPG-116800 at 25, 50, 100 or 200 mg twice a day, or placebo, for 12 months (200 mg arm stopped early)

What they did

In a randomised, double-blind, placebo-controlled study across several centres, 401 people with mild to moderate knee osteoarthritis took placebo (80 people) or PG-116800 at 25 mg (81), 50 mg (80), 100 mg (80) or 200 mg (80) twice a day for 12 months. The main measures were narrowing of the knee joint space on microfocal X-ray and WOMAC scores for pain, stiffness and function.

What they found

The 200 mg dose was stopped during the study because of more musculoskeletal side effects. After a year, there was no statistically significant difference from placebo in knee joint space width or WOMAC scores. The most common side effect was joint pain (arthralgia), in 35%. Twenty-three per cent of evaluable patients lost at least 30% in a shoulder range-of-motion measure at follow-up, significantly more often in the two highest doses. Thirteen per cent, half of them in the 200 mg group, had hand problems such as swelling, palmar fibrosis, Dupuytren contracture or tendon thickening. Development of the drug for this use was stopped.

What it can't tell us

This tested one drug that blocks many MMPs at once, not MMP13 alone, so its failure does not show what selective MMP13 blocking would do. It tells us nothing about nutrients or plant compounds. It also used X-ray joint space, an indirect measure of cartilage.

Why it matters

This is the kind of result that pushed researchers away from broad MMP blockers and towards MMP13 alone. The authors wrote that it adds to evidence that the side effects of MMP inhibitors in general make them unsuitable for this use. Blocking many MMPs at once caused problems in shoulders, hands and tendons without protecting the joint.

The paper

  1. Krzeski P, Buckland-Wright C, Bálint G et al. Development of musculoskeletal toxicity without clear benefit after administration of PG-116800, a matrix metalloproteinase inhibitor, to patients with knee osteoarthritis: a randomized, 12-month, double-blind, placebo-controlled study. Arthritis Res Ther 2007.

Read next

Last reviewed 3 November 2026Every study cited here is indexed on PubMed. How we read the evidence