How relevant are computer models? If you haven’t got data from real people over 20 years, how effective is computer modelling?
I want to know that studies have been done on human beings that show a meaningful benefit. So, for example, if you do a study, so you come to me with MS and you say, I don’t like it and I don’t want to have to face a risk. And I say, well if I give you this drug your brain’s going to look much better in ten years’ time than it will if you don’t take the drug. I would suggest you walk away from that.
Because I care about whether you can walk, whether you are continent, whether you can go out and do a job of work and enjoy your family, or whether you can’t do those things. I actually don’t care too much what your brain scan shows.
For example, we did a study here with one of the people that did the study on Botox in migraine, which I never thought would work, but it does. And what’s the really important outcome, that the quality of life is better. Now, counting headaches, actually not so interesting, it’s the quality of life that matters. There are lots of situations in which treatments ought to work, but don’t, and so you’ve got to look at human beings and show that in real life people…
Therefore are computer models worth having?
I think it’s helpful to point a direction and to show you where you might look. I mean, for example, with the interferon data, when it first came out some people said that the apparent benefit was from poor blinding in a trial that patients knew actually what they’d had. The reason that that’s probably not correct is that the brain scans were better. So it did something, it’s not just hooey. But whether it does something meaningful, and computer modelling, some people who are against testing in animals – and we’re animals – will say you should do it all in a Petri dish on cell cultures. And yeah, you can do that, but you have to, I think, do trials on human beings and measure an outcome on human beings in order to be sure that treatment works.