Rheumatoid Arthritis and COVID-19: Report from the Global Rheumatology Alliance

In this podcast, Jeffrey Sparks, MD, and Zachary Wallace, MD, discuss the findings of their research into the outcomes of patients with rheumatoid arthritis taking DMARDs at baseline who contracted COVID-19.

Jeffrey Sparks, MD, is a rheumatologist and research scientist at Brigham and Women's Hospital and an assistant professor of medicine at Harvard Medical School in Boston, Massachusetts. Zachary Wallace, MD, is a rheumatologist and research scientist at Massachusetts General Hospital and an assistant professor of medicine at Harvard Medical School in Boston, Massachusetts. 

Transcript:

Rebecca Mashaw:  Hello and welcome to another podcast from the Rheumatology and Arthritis Learning Network. I'm your moderator, Rebecca Mashaw.

Today, Dr. Jeffrey Sparks of Brigham and Women's Hospital and Harvard Medical School, and Dr. Zachary Wallace, from Massachusetts General Hospital and Harvard Medical School, will discuss their research as part of the COVID-19 Global Rheumatology Alliance into how baseline use of DMARDs including TNF inhibitors, rituximab, corticosteroids, and JAK inhibitors affected outcomes among patients with rheumatoid arthritis who contracted COVID-19.

Jeffrey Sparks:  Hello, my name is Jeffrey Sparks. I'm a rheumatologist and clinical researcher of Brigham Women's Hospital and Harvard Medical School in Boston, Massachusetts.

Zachary Wallace:  I'm Zach Wallace. I'm also a rheumatologist and a clinical researcher at Massachusetts General Hospital and Harvard Medical School in Boston, Massachusetts.

Dr Sparks:  Great. We are co-first authors on a presentation at EULAR, which was also recently accepted, and also rheumatic diseases investigating baseline use of biologic and targeted synthetic DMARDs in rheumatoid arthritis for COVID severity. We're very pleased to discuss these results together.

Dr Wallace:  Jeff, we've been working on this for so long, I think we started thinking about this a while ago, just recognizing as more and more data was coming out about the impact of COVID-19 on our patients.

We wanted to try and get at the question of what was the impact of some of these biologic and targeted synthetic DMARDs that we so often use in rheumatology and were also being repurposed to treat COVID. What their impact was on patients who got COVID-19.

I don't know what you think, but I think our methodological approaches tried to address some of the potential limitations that some of the prior studies had had.

Dr Sparks:  Yeah. Just to add to that, I think this was something really early on into the Global Rheumatology Alliance, the GRA, that we were wanting to do and we hoped there weren't enough cases actually to ever do the analysis, but unfortunately, it did end up to that point. We started analyzing this probably January.

Maybe I can just summarize the sort of the punchline and then we can delve into some of the details. We wanted to only analyze rheumatoid arthritis, only analyze patients who are on advanced therapies, to try to minimize the possible impact of confounding by indication also so that patients were in a similar disease state.

Those were exposure variables, all these medications that we use in our patients, in particular for rheumatoid arthritis, and to see whether this changed the trajectory of the COVID-19 disease course.

What we found is that compared to the most common medication used— TNF inhibitors— rituximab and JAK inhibitors were associated with poor outcomes.

In the middle, I guess you could say, would be the interleukin-6 inhibitors and abatacept, which didn't seem to have a consistent association with poor COVID-19 outcomes compared to TNF inhibitors. Hopefully, that's an accurate summation, Zack.

Dr Wallace:  That's definitely a great summary of the work we did. We were lucky to be able to be a part of the rheumatology community, which came together early on in the pandemic to start the Global Rheumatology Alliance, which physicians from around the world have contributed cases.

Without all this participation, we won't have been able to do this or any of the other studies that the GRA has been able to put together. I think we should also just put a shout out there to all the people who participated and contributed cases.

Thank you for helping us do these studies. One thing that you and I get asked often is, what are the implications of this study, and what do we take from it?

Especially given the much higher odds, this is its worse outcomes among the rituximab-treated patients, and also the somewhat surprising finding among patients who use JAK inhibitors, given just what we know about the use of baricitinib in COVID? I don't know, what do you think about that?

Dr Sparks:  Maybe I'll take the rituximab part of that and I'll let you chime in on the JAK inhibitor side. You would agree, Zack, that we were probably expecting the rituximab users to have a worse course. I think really strong association was a bit surprising. It's impressive how poorly these patients did.

Another thing that comes to mind is, are these just different patients? Certainly, they could be, this is an observational study, but we tried our very hardest to try to adjust away things. We did sensitivity analyses where we eliminated patients who had ILD, or cancer, adjusted for comorbidities, and we still saw this association with rituximab users.

To me, I feel like the association is probably...Obviously, there's other studies that have shown it as well, but to me, it has a lot of face validity, it has a lot of biological plausibility. To me, it's a call to action for rituximab users.

Unfortunately, there's not an easy solution because the vaccines don't seem to be working as well for those patients. Certainly, anecdotally, I've seen patients, even fully vaccinated, with rituximab who have had poor outcomes, even deaths.

As far as what to do, I'm not exactly sure. But I feel like the finding has a lot of evidence base behind it, and it is a call to action not just for our patients but our other patients on rituximab for other indications.

Things like monoclonal antibodies, whether these patients should be social distancing, wearing masks longer, and obviously things like booster vaccines, delaying their infusions. I feel like this is a real call to action about what to do for this finding for rituximab.

Dr Wallace:  I couldn't agree more. Personally, I tend to use a lot of rituximab in my clinical practice because of the patient population that I see.

As we've learned more about the outcomes of these patients during the pandemic, I've become increasingly a little more conservative about my use of rituximab, trying to space dosing intervals out or hold on initiation, if I can, just because give these people time to get their vaccine, hopefully have a good response to the vaccine, and then start them on treatment. Or just trying to wait for the rates of COVID to go down enough that I feel pretty comfortable just in the community giving them rituximab.

I think we're lucky to live in an area where rates are dropping, and so hopefully it's becoming less of an issue, but that's not true for obviously other parts of the world where it's still a huge problem.

In terms of the JAK inhibitor observation, as I mentioned, I think we were both a little surprised about that there. There's trial data showing that baricitinib may improve outcomes in patients who have COVID-19 and are started on baricitinib to treat the hyperinflammatory response that we know leads to a lot of substantial morbidity and mortality in COVID.

This raises the question of whether or not this is also a timing question. The timing of the exposure obviously, was very different in our study. These are patients who have been using JAK inhibitors probably for many months, if not years, before they were infected with COVID-19. The effects of JAK inhibitors on the immune response and the effect that has on the host's ability to get good control of the infection immediately may be responsible for some of the associations that we observed.

Maybe when you use baricitinib or another JAK inhibitor to treat the hyperinflammatory response, you're just at a different phase of the disease. You're seeing benefits rather than some of the potential harms. One of the things that we talked about, or that you and I talked about, is that we bunched the JAK inhibitors together as a class. We know that they have different specific JAK targets, and so there may be differences across the JAK inhibitors. I think that is a future study that needs to be done to understand whether this is true for the different JAK inhibitors or worse too across the class.

Dr Sparks:  I would agree. Ours is the first to look at this question, honestly. I don't think it's necessarily in conflict with other data besides the trial data, but I definitely think this is something that needs to be replicated and looking at particular JAKs to see if those are different.

The other thing that we've been trying to rationalize the finding is that it may be a question of when this is being used, it might be that getting infected while you're on a JAK inhibitor causes the trajectory of illness to be worse. Maybe, right when the hyperinflammatory phase happens, and when there's option requirement, that might be actually be the moment clinically to use a JAK inhibitor. Similar to the steroid associations and the trial data with dexamethasone, it seems like patients with glucocorticoids do worse if they're using it at baseline, but obviously, it's a cornerstone of treatment for patients who are doing poorly with COVID-19. That does seem to be a corollary there.

Dr Wallace:  Definitely. We focused on the biologics as a class of interest here in our exposures, but folks are probably interested in looking at other DMARDs that we didn't necessarily include and also, examining this question with other populations, like patients with different types of vasculitis or patients with lupus. I think we can expect to see more data like this coming out in some of those disease subgroups as well because these are questions that are just so relevant to all of our patients with a variety of diseases.

I do think our observations are relevant to those other populations at this time. In the absence of specific data, I think what we observed in the RA patients is probably what you might see also in patients with ANCA vasculitis who are getting rituximab or lupus and other conditions.

Dr Sparks:  Wonder if you want to comment on the comparator group, TNF inhibitors. Is it that these drugs are bad or could it be that TNF inhibitors actually have maybe a beneficial effect? Just curious what your thoughts are on that.

Dr Wallace:  This is certainly a limitation of our study. Our reference group here is people who use TNF inhibitors and we don't know what their outcomes are compared to just the general population. Maybe TNF inhibitor use is beneficial. And maybe what we're seeing is while the rituximab users or the JAK inhibitor users do worse, then the TNF inhibitors, maybe compared to the general population, they're not that far off. I don't think that's probably the case. We don't necessarily expect that, but I think it's something that we need to explore further.

Something that we're working on now is trying to start to get at this question of across these different groups of drug users, what's their risk compared to the general population? One of the issues we've run into is just recognizing that these patients are using these medications for a reason, because they are sick, because they have these other diseases and trying to tease out the effect of the drug versus their condition compared to the general population. A little trickier, but certainly something we need to continue to work on.

Dr Sparks:  I mentioned that the interleukin-6 inhibitors, it was not associated, but the point estimate was actually a bit under one for that group. Obviously, there's been intense interest about repurposing those class of medications for COVID treatment. I know the recovery, was positive, but there's been trials, both positive and negative for interleukin-6 inhibitors.

It's just really interesting about how our drugs could be repurposed, and how some of them might impart a favorable course, and certainly some of them might have an unfavorable course. We haven't talked about abatacept either.

Dr Wallace:  I agree to the IL-6 observation. It's certainly interesting and it's something we tried to squeeze into the conclusion. We had so much to discuss, we couldn't really get it all in there. I think it's an important observation given what we know about some of the trials, at least looking at IL-6 inhibition, for COVID.

In terms of abatacept safety, I think the group of patients that get abatacept for RA are somewhat unique. In our approach, we tried to address some of these differences in their demographics and some of their disease-specific features. Across most of the analysis, we didn't see an association. There was one, our propensity score match analysis, where we did see a little bit of an association between abatacept and poor outcomes. That's an important observation.

We have to recognize that was sensitivity analysis, it wasn't our primary analysis. We have to take that with a grain of salt. With propensity score matching, we do lose a number of people in terms of our sample size, and so I think that's something that deserves further study and needs to be looked at in other cohorts to see if that association is true or not.

The other analyses looking at abatacept that didn't suggest that strong inverse association, and it certainly wasn't significant. For now, I'm less worried about those patients, certainly, compared to my rituximab patients, and maybe my JAK inhibitor patients. I don't know, Jeff, what do you think our next steps should be? We have so many. What do you think's our first one?

Dr Sparks:  [laughs] One thing that we haven't talked a lot about, Zach, is I think most of this was done in the prevaccine era. You wonder whether this would hold up in a vaccinated population. I'm curious what you think about how this sort of applies, at least in the US where there's a lot more vaccinations?

Dr Wallace:  I think that what we saw in our rituximab-treated patients is a reflection probably of what we're going to see in our rituximab-treated patients who get vaccinated. I'm not optimistic that they're having a really sufficient response to the vaccine that's going to protect them and think that's something that hopefully we'll have more data on soon.

Obviously, a number of our colleagues have been publishing some studies demonstrating and suggesting that rituximab users don't have a great vaccine response, but I think we also need to complement some of that translational work with clinical outcomes and looking at what happens to these people when they are infected.

So I totally agree, that's a really important next step, and probably the most important one and one that I know patients ask me about in every clinic I have—what is the efficacy of the vaccine in my case? It will be interesting once we have more data to see if there's any patients, are there alternative approaches to vaccine strategies? Whether it's booster shots, or whether we just won't necessarily be able to rely on vaccines, and maybe we need to think about passive immunity and monoclonal antibodies and things like that to protect these patients. But it's a real priority given what we observed with their really poor outcomes if they do get COVID and they're unvaccinated.

Dr Sparks:  Looking to see who's getting infected, if they're getting infected, and their outcomes after vaccination I think is going to shed a lot of light to this. I know there are some people hoping that the T-cell response is going to make up for antibody levels and understand it's really hard to measure the T cells and it's a lot easier to measure the antibodies.

Certainly, that's not the full picture. I think that will be an important question for our B cell-depleted patients who seem to have very poor antibody levels.

Again, I think, as you know, Zack, we have some anecdotal evidence, but we need to do a rigorous analysis looking at people who got COVID after vaccination and try to see if there's a signal there to try to understand what put people at risk for that. There's still more to do, unfortunately, in COVID.

Dr Wallace:  I think, to your point about the rituximab-treated patients and their outcomes, the other big knowledge gap that we have is there patients who've been on rituximab for years and have been continuously B cell-depleted for 4 or 5 years and treatment of ANCA vasculitis or other conditions.

We don't know if it's that person is the same as the one who'd gotten, let's say, 1 rituximab infusion in the last year and that's the only infusion they've ever received. I don't know how to assess the individual risk of a rituximab user based on their rituximab history if that makes sense. Hopefully, we get more data on this because it is obviously such an important area.

Dr Sparks:  We'll have to help generate that data, Zack, if you're up for it.

Dr Wallace:  Stay tuned. Jeff and I have some more data coming your way.

Dr Sparks:  We could continue talking forever, probably, Zack. We'll probably will on other Zooms, but maybe we'll have to stop.

Dr Wallace:  We'll just hop on another Zoom after this. No, this is great, though, Jeff. I always enjoy talking to you about these issues.

Dr Sparks:  It's been a great partnership, and hopefully, we'll continue.

Dr Wallace:  Definitely.

Dr Sparks:  All right. Thanks for your attention.