Mark Freedman, MD: The next abstract I was asked to comment on was the post hoc analysis of the combined ASCLEPIOS I and II trials, which compared subcutaneous ofatumumab with the oral formulation of teriflunomide in relapsing disease. In this particular case, they were looking at NEDA-3 [no evidence of disease activity]. Just to remind you what NEDA-3 is, it’s no evidence of disease activity, which means no clinical attack, no new MRI [magnetic resonance imaging] lesion, and no particular EDS [Ehlers-Danlos syndrome] at progression, with a view to try to present ofatumumab as a superior medication to teriflunomide, which was proven in the primary analysis of these 2 drugs.
I don’t think it was particularly surprising that when you do a post hoc analysis of a study that’s already positive, the chances are you’re going to get another positive result, and that’s true for any kind of post hoc analysis.
NEDA-3 is something that people are working toward as an outcome measure that is going to be standard today but has not been accepted necessarily by authorities. Hence, the reason for putting it in a post hoc analysis. Most of these studies, to get registered with the FDA, would require a primary analysis, usually based on relapse rates, which it had already proven in both studies individually in the combined analysis.
At the end of the post hoc analysis, not surprising, ofatumumab showed significant increase in NEDA-3 over the patients who were taking just the teriflunomide.
Does this add anything to our knowledge regarding the efficacy of ofatumumab? I’m not sure. I think it’s of interest.
What does this all mean with regard to NEDA-3 as an outcome measure? Our problem is we don’t know what the natural NEDA-3 would be. We know that in some placebo-controlled studies recently that the NEDA-3 rate for even the placebo group was not insignificant, especially if not a lot of MRI studies are done. The rule of thumb is the more you look, the more you find. If you did an MRI every month, the chances of being NEDA-3 at the end of a period of time would be almost 0, since MRI picks up a lot of different new lesions. But if only 1 or 2 MRIs are done and maybe 1 or 2 clinical visits, so there’s not a lot of chance to pick up disease activity, you’re going to see that even a placebo group would probably have a substantial NEDA-3 rate.
Buried in this study is that group of people who, by natural history, are achieving NEDA. We don’t know that component in either the ofatumumab or teriflunomide group. Regardless, more of the ofatumumab patients achieved NEDA-3 at the end of the time period compared with the teriflunomide group. Keep in mind there’s probably a substantial portion of both of those who did not necessarily benefit from either treatment.
I don’t think having these outcomes necessarily adds to my confidence that ofatumumab is superior to teriflunomide. The primary outcome told us that. It’s just another way of looking at all the data compiled together. I’m sure we’re going to see other abstracts with subsequent post hoc analysis of the MRI studies and some of the other metrics that were applied in this particular trial, which were not considered the primary or secondary outcomes but the tertiary and exploratory outcomes that one likes to achieve when dealing with a large study like this. They help eventually understand the population that might best benefit from a particular therapy. Although it’s not proof that that population will benefit, it’s the kind of hypothesis-generating results that help guide us to either design new trials or to focus in and enrich a population that would be the best responder to medication X over Y.