Millie Long, MD, on Comparative Effectiveness in Ulcerative Colitis

Dr Long reviews her presentation from the AIBD regional meeting on using comparative effectiveness data to choose from among the growing number of therapies for treating ulcerative colitis.

Millie Long, MD, is a professor of medicine, vice chief of education, and director of the fellowship program in the Division of Gastroenterology and Hepatology at the University of North Carolina at Chapel Hill.

Hi, this is Millie Long from University of North Carolina at Chapel Hill, and I just gave a talk on comparative effectiveness of therapies in the treatment of ulcerative colitis, and what we're lucky to have in the field of IBD right now is truly an explosion of available treatment options. It's a wonderful time for our patient's perspective to have choice and options and certainly if someone doesn't respond to a therapy we have many other options in terms of where to go. So what it leads us with as providers is really trying to understand how these therapies stack up to each other and how do we choose the right therapy for the patient in front of us.

So first just to define comparative effectiveness. It's really trying to understand the balance of the benefits of harms of one therapy as compared to another, or one therapy as compared to a different intervention for the individual patient in front of us. And this has actually been prioritized as one of the top 100 comparative effectiveness research topics for our, for us right now, over the next decade.

These biologic agents and advanced therapies are quite expensive. It would be ideal if we could understand what to use first, what to use second, how to really optimize for the individual patient therapy that will be the best in terms of effectiveness, but also the best in terms of safety for that individual patient. So as we review the data from a comparative effectiveness standpoint, probably the highest level of data is head-to-head randomized controlled trials.

And in UC, we don't really have too many of those. So we have a lot of data that we have the UC success trial. And this was a trial that looked at infliximab alone as compared to infliximab combined with azathioprine as compared to basically azathioprine alone.

So you have each agent on its own as compared to the combined. And what this showed was that after 16 weeks, we really saw much higher rates of clinical remission associated with those that use the combined infliximab and azathioprine alone. And so what do I take from this for my practice? I really predominantly use combined therapy, at least during induction, to help to optimize the effectiveness for my patients and then potentially during maintenance, like a drop-off of thiopurine, for example, at that time period, once I've truly controlled the inflammation and gotten that patient into remission.

Another head-to-head trial is the VARSITY trial, and this was a really landmark study. It actually compared to bylaw. biologic options for the treatment of ulcerative colitis.

So individuals were randomized to either vedolizumab or to adalimumab, and then a year later outcomes were assessed. And importantly, there was a delta. Vedolizumab was about 9% more effective as compared to the adalimumab monotherapy.

So the anti-integrin was more effective than the anti-TNF. Now there's some caveats to study. You know, in my practice I don't start someone on monotherapy adalimumab and never check a level and kind of forget about it. It wasn't actively managed. We don't know the difference between those two. And there was less difference seen in the individuals who were on combined immunosuppression.

So overall, I think that it's a good study. It helps us to understand the positioning of the anti-integrin as compared to standard-dose, nonoptimized adalimumab. Unfortunately, that's pretty much it.

There aren't other head -to -head studies in UC to help us to guide this positioning. So what are we left with then? We're left with network meta-analyses. And these are very indirect measures of comparative effectiveness, meaning that what these studies do is they take each registry trial that is used for the approval of the agent where it's compared to placebo.

And they say, if you consider all these placebos the same, let's then compare the drug to the other, taking into account that difference as compared to placebo. So it's very indirect. And we all know that those placebo arms were not exactly the same.

So again, we have to recognize that this is a lower level of data. But one thing that these network med analyses have shown is that quite consistently, the JAK inhibitor classes seem to be our most effective. Whether these are TNF-naive patients, which it's not approved in right now in the US, but was studied in, or TNF-exposed populations.

So this becomes really important. These drugs are now kind of placed by the FDA after TNF exposure. And for those patients, these are very effective drugs.

And so from an effectiveness standpoint, we know we can prioritize those jacks. These network meta-analyses have also showed in that TNF exposed population. that the IL-12/23 agents, medicines like ustekinumab, are actually quite effective as well—ore so than vedolizumab, our anti-integrin therapy, which was so great upfront as compared to TNF, it really seems to do better in biologic, naive populations. So that helps us in terms of selecting drugs for each individual patient.

Finally, just kind of from a comparative safety standpoint each of these drugs has specific safety signals. We certainly have patients for whom they have underlying comorbidities that may be a contraindication to initiation of the therapy, we need to keep that in mind.

I think the biggest thing from a safety perspective is just that corticosteroids are worse than any of our agents and anything we can do to provide a steroid-sparing therapy is actually going to improve outcomes for our patients. You know, one safety signal that we discussed is that of the thrombosis risk with JAK inhibitors.

So this came about not in a UC study but actually in a rheumatoid arthritis study called the ORAL surveillance study. This was originally designed by the FDA to look at high -risk patients. They were over the age of 50, they had rheumatoid arthritis, they often were smokers, they had cardiovascular risk factors— to try to understand if there was any increased cardiovascular risk.

And what this study found is that there was increased major adverse cardiac event in the tofacitinib arm as compared to a TNF-exposed arm as a comparison. They also found a slightly increased rate of malignancy, lung cancers actually, and with the higher dose tofacitinib, an increased risk of deep venous thrombosis, so there may be a thrombosis risk. I just want to emphasize that this is a population very, very different than the population we care for with Crohn’s disease and ulcerative colitis, and that when they actually did a subanalysis of this study the patients with RA who had complications were all over the age of 65, often former smokers, or had history of atherosclerotic cardiovascular disease.

So if your patient with UC for whom you're starting a JAK inhibitor, not over the age of 65, an active smoker. or former significant smoker, and does not have a history of atherosclerotic cardiovascular disease, we have not seen any of these safety signals. So again, patient selection helps to make your therapy more effective.

So just finally, as we're wrapping up our discussion of comparative effectiveness, we have to think about different aspects for the patient in front of us. We need to think about aspects of the drug, how fast is the onset, does it need a concomitant immunomodulator. We have to think about safety aspects and the specific patient characteristics. We have to think about the patient comorbidities and how those may influence therapy selection. If I have a patient who has an underlying heart block, for example, I'm not going to put them on an S1P. If I have a patient who has advanced CHF, I'm not going to put them on a TNF. So that can help us in terms of deciding on an individual drug.

And then finally, there are disease-specific characteristics, the presence of extraintestinal manifestations, the severity of the inflammation, the distribution that can help to guide an individual therapy for a patient.

So putting it all together, we certainly need more studies, we need more comparative effectiveness head -to -head trials to help us, but right now, by using those characteristics, we can help to get the right patient very safely on the right drug. Minimize corticosteroids, which really drive most of the risk, and really allow them to attain excellent quality of life, remission, and hopefully prevent downstream complications. So with that, thank you, and thank you for joining us, and I hope to see you at a future AIBD regionals or at the AIBD national meeting in December in Orlando.