This Is MS Multiple Sclerosis Knowledge & Support Community

I just read better2gether's post in the Tysabri forum about the Sentinal results (Avonex/Tysabri combo). I've always wondered...
There were 1171 Avonex treated patients spread out over 123 trial sites. If it was a double blind study, does that mean there were double that number of actual participants? Why do they spread it out over such a large number of sites? Wouldn't 50 or so sites provide a representative sample and greatly reduce costs?

Denver,

DenverCO wrote:I just read better2gether's post in the Tysabri forum about the Sentinal results (Avonex/Tysabri combo). I've always wondered...
There were 1171 Avonex treated patients spread out over 123 trial sites. If it was a double blind study, does that mean there were double that number of actual participants? Why do they spread it out over such a large number of sites? Wouldn't 50 or so sites provide a representative sample and greatly reduce costs?

"Double blind" in trials refers to the fact that both the doctors and patients do not know who is taking the drug or getting the placebo. Only when the trial ends do they break the code and know who was actually on the drug. It has no bearing on the number of patients in the trials.

Spreading trials out over many sites usually gives a better opportunity for the many MS clinics to participate as well as get a good cross section of patients in various parts of the country. I would imagine that having too many patients in one center might also make it difficult to obtain the numerous MRIs that are required in these trials.

Harry

HarryZ wrote: "Double blind" in trials refers to the fact that both the doctors and patients do not know who is taking the drug or getting the placebo. Only when the trial ends do they break the code and know who was actually on the drug. It has no bearing on the number of patients in the trials.

Harry, I know what a double blind study is, but can we assume that there is one person receiving the placebo for every person receiving the actual drug? If that is the case then when they report that a study revealed that 500 patients receiving a new drug experienced certain results, can we assume that there were 1000 participants (500 who got the drug and 500 who got the placebo)?

I'm not asking for any particular reason, just curious.

Denver,

Harry, I know what a double blind study is, but can we assume that there is one person receiving the placebo for every person receiving the actual drug? If that is the case then when they report that a study revealed that 500 patients receiving a new drug experienced certain results, can we assume that there were 1000 participants (500 who got the drug and 500 who got the placebo)?

I'm not asking for any particular reason, just curious.

It all depends on the design of the study. Some studies have 50% on the drug and 50% on placebo (I believe the Tysabri Phase III trials did this) and others have different numbers. For instance, when my wife took part in a trial a number of years ago, they had one patient on the high dose of the drug, one on the medium dose, one on the low dose and one on the placebo. So you had a one in four chance of being on the placebo.

If you are able to get your hands on the actual study or a detailed abstract, the breakdown will be given very clearly.

The design of a study takes a lot of work and there are many variables that have to be carefully outlined. This design can effect the outcome so that's why it is so important to read the studies to determine just what happened. It is quite an art.

Harry

Here's the answer from Biogen's press release...


SENTINEL is a two-year, randomized, multi-center, placebo-controlled, double-blind study of 1,171 AVONEX-treated patients in 123 clinical trial sites worldwide. In the trial, AVONEX-treated patients who continued to experience disease activity were randomized to add TYSABRI (n=589) or placebo (n=582) to their standard regimen.