Current Series: Multiple Sclerosis: A Disease of an Immune System Gone Awry

Fred D. Lublin, MD: In testing, clinical trials are how you’d bring a drug to market. The standard is showing efficacy and safety in well-designed clinical trials. For the past 25 years, we’ve gotten better and better at designing these trials. Our outcome measures have stayed the same, or very similar, which is that we look to reduce annualized relapse rate.

We do that because when you’re doing a clinical trial, especially with a disease as variable as multiple sclerosis [MS], you need to pick an outcome measure that’s important—and relapse rates are important—and that’s going to occur frequently enough so that you could measure a difference between the treated group and the control group. Annualized relapse rate has served us very well in that regard.

A more important but less useful measure in clinical trials is reduction in worsening of disability. It’s a very important measure, and some of our studies have shown an effect in that as well. At least 1 or 2 of them have had it as a primary outcome measure. But in early MS, these changes in our disability markers don’t occur that often, so you can miss a good therapeutic effect just because of the natural history of disease not providing you with a sufficient number of events to be able to measure a difference. That’s why annualized relapse rate has been No 1, because usually that will work.

As a secondary outcome, we use changes on the MRI [magnetic resonance imaging]. MRI has revolutionized our ability to bring drugs to market because there’s something called a phase 2 trial, which is a proof-of-principle design that you do before you get to the pivotal studies—what are called phase 3 studies—that you would take to a regulator like the FDA, Health Canada, or EMA [European Medicines Agency]. Before you would do those pivotal studies, you’d do the proof of principle, which tells you something about the safety and efficacy and dosing information.

Well, it turned out from studies initially done at the NIH [National Institutes of Health] that if you look at frequent MRIs and reducing MRI activity with any of the immunomodulating agents we have, and you succeed in doing that, then you will almost surely succeed with a clinical outcome like annualized relapse rate in the pivotal trials. Since it requires an easier and shorter time frame to do those phase 2 trials, they became a very good investment. A sponsor would have a good molecule. They’d look at that phase 2 trial and see it has succeeded. They say, “Well, we’re going to have a drug that we can get to market.” You have to do the studies, but it hasn’t failed.

MRI has been very important not only in that phase II setting but also when looking at the consistency of phase III trials. That has been another feature of all our pivotal clinical trials. You lay out the metrics—annualized relapse rate first, then worsening in disability, and then MRI, T2 lesion load, gadolinium-enhancing lesions, brain volumes, and such. With the exception of brain volume, which is a little confusing, all those other metrics have always gone in the right direction. Even if they weren’t statistically significant, they still supported the primary action. They’ve been very useful in metrics. Brain volume has been a little more confusing. In some studies, it’s followed the therapeutic effect; in other studies, it isn’t, so I think it’s a more complicated metric.