Skip to main content
10 October 2010

Minocycline likely to be highly effective in MS despite poor study design

Canadian researchers show that minocycline (a cheap oral antibiotic used for acne) added to Copaxone is more effective than Copaxone alone, but poor study design means the results just miss out on being significant

Canadian researchers have shown a large improvement in people with MS taking minocycline, a common antibiotic used for acne, in combination with Copaxone (M+Co), compared with those taking Copaxone (Co) alone. The study outcomes for the M+Co were a 63-65% reduction in MRI lesions compared to the Co group, and a greater reduction in risk of relapse (54%, 0.19 vs 0.41).

The problems with the study unfortunately mean that very few if any neurologists will prescribe minocycline for people with MS. Why is this? Firstly, minocycline should have been tested against placebo. There are still many people who opt not to take the injectable drugs, and so it would not have been unethical to have a placebo group.

But it is very important that we know if this drug actually works in MS (which appears almost certain), because it is cheap, widely available, taken by mouth, and has few significant side effects. On all these counts, it is way ahead of the ‘new generation’ of drugs for MS that are taken by mouth and are about to hit the market (cladribine, fingolimod, etc).

When this study was first set up I was very critical that the drug company, Teva, which makes Copaxone, sponsored the study, and chose to test it as M+Co versus Co alone, rather than a placebo. This was a clever strategy because it meant that even if they showed that minocycline was effective, it would only be shown to be effective in combination with Copaxone, and would not therefore challenge Copaxone’s market dominance in MS.

From preliminary studies, it is quite likely that minocycline would be shown to be more effective against placebo than Copaxone, which would have been a disaster for Teva had this been reported. Secondly, the study had too few participants to find anything but a very large effect.

So the figures above for the reductions in lesions and relapses could not be ‘proven’ because there were only 44 people with MS in the study. If you have so few participants in a study, you will not find a statistically significant effect unless the size of the effect is very large. In fact, I suspect the drug company got a surprise.

The size of the effect was indeed very large, almost large enough to reach statistical significance (p=0.06 for new lesions for example), but not quite (p<0.05 is conventionally required).

So these enormously positive and potentially highly beneficial effects of minocycline in MS unfortunately get consigned to the ‘further study is warranted’ basket, rather than being immediately adopted by doctors, and widely taken up by people with MS.

Because minocycline cannot be patented (it has been around for decades as an anti-acne drug and is therefore prescribed as a generic and very cheap), it is unlikely that any drug company will be rushing to continue these studies, and most particularly to study it against placebo.

My suspicion is that it would be shown to be very effective in MS, with the potential to take over from all the injectable medications currently used. What a great shame that industry has such control of the research agenda in MS.

Professor George Jelinek

Click here to read the full study