Roy Fleischmann, MD, on the Safety Profile of Upadacitinib In Patients With RA

Dr Fleischmann reviews a post hoc analysis he and colleagues conducted of the SELECT phase 3 trial of upadacitinib in rheumatoid arthritis and the safety of JAK inhibitors among RA patients at risk for cardiovascular events.

Roy Fleischmann, MD, is a clinical professor in the Department of Internal Medicine at the University of Texas Southwestern Medical Center, chief of the Division of Rheumatology at UTSMC-University Hospitals-St. Paul, and co-medical director of Metroplex Clinical Research Center in Dallas, Texas.

TRANSCRIPT:

Welcome to this podcast from the Rheumatology & Arthritis Learning Network. I'm your host, Rebecca Mashaw. And today, I'm welcoming Dr. Roy Fleischmann, who is clinical professor of medicine at the University of Texas Southwestern Medical Center, and co-medical director of the Metroplex Clinical Research Center, in Dallas, Texas. We're going to discuss his recent research with colleagues into the safety profile of upadacitinib among patients with rheumatoid arthritis, who are also at risk of cardiovascular disease. Thanks for joining us today, Dr. Fleischmann.

Roy Fleischmann:

Thank you, and you're welcome.

Rebecca Mashaw:

What led you to investigate the risks of cardiovascular disease among patients with RA, and in the context of treating them with this JAK inhibitor, upadacitinib?

Roy Fleischmann:

As you know, ORAL surveillance, which is a postmarketing safety study of tofacitinib versus TNF inhibitors, looking at cardiovascular risk and malignancies, was a postapproval program that was mandated by the FDA. The results of that began to come out in 2019, 2020, and it was published last year.

What that showed was that there were numerical differences between tofacitinib and the TNF inhibitors with a spec cardiovascular risk—not statistical, but numerical. And that led to a change in the label from the FDA, as well as from the EMA, suggesting all JAK inhibitors, not just tofacitinib, should be used after TNF failure or intolerance. The reason for that was that although the events were numerically different and not statistically different, then why not use an advanced agent that really has efficacy, like a TNF, first, with probably a little safer safety profile than tofacitinib or the JAKs?

The EMA didn't mandate that the JAKs should be used after TNF, but they did suggest that patients who are at risk should really be carefully looked at. The EMA then asked the other manufacturers, including AbbVie, upadacitinib, to take a look at their population in a post hoc analysis, to see whether or not the similar risks were there. And indeed, that's why this analysis was done, in order to satisfy the EMA requirements for looking at the risks of cardiovascular disease and malignancy, as well as mortality and serious infections in this high risk population. So this is a post hoc analysis of the entire AbbVie program. There were various studies that were looked at, and they were the studies where upadacitinib was used as monotherapy, or in combination with conventional synthetic DMARDs. There were studies where upadacitinib was compared to adalimumab. There were studies where upadacitinib was compared to methotrexate.

So those initial studies, the phase 2, phase 3 program, all were of varying length, different patient populations. So we extracted the patients who would have met the criteria for inclusion in ORAL surveillance, as closely as we could. With this being a post hoc analysis, obviously, there were various patient characteristics that were predetermined in ORAL surveillance, that weren't predetermined in these other studies. And it turned out that over 50% of the patients... I can't remember exactly. It might've been 60% of patients, actually were the patients who did have the higher risk as associated with ORAL surveillance. And they either had prior MIs, or they had a history of cardiovascular disease, or they had elevated lipids, or they had hypertension, etc. The upadacitinib population was something around 3,000 patients, give or take. The population for the adalimumab was much smaller, because there was only one study comparing upadacitinib with adalimumab. And the adalimumab population was in the 300-400 range. And the population comparing with methotrexate was even smaller. That was just several hundred patients.

Rebecca Mashaw:

What did you find out?

Roy Fleischmann:

What we found out was actually interesting, but not unexpected, at least from my standpoint. What we found out was that this analysis actually substantiated what was seen in ORAL surveillance. So what was seen in ORAL surveillance was it wasn't just patients who were over the age of 50, patients who were smokers, patients who had cardiovascular risk. It was a very specific group, who were older, who smoked, who did have cardiovascular risk, which was a preceding MI, where it had a high risk of cardiovascular disease, with respect to VTEs. It was the patients who were overweight, a higher BMI. More importantly, a previous history of VTE, etc. It was a very specific population, as was also identified ORAL surveillance.

So in ORAL surveillance, it really wasn't everybody who was at higher risk. It was only those patients. And this study, it was found that patients treated with upadacitinib, who had the same high risk identified in ORAL surveillance, not by the inclusion criteria, but by the findings, also had a high risk. And whether it was upadacitinib, or it was adalimumab, or it was methotrexate, all of those patients were the patients who had the highest risk.

So what was really interesting to me, and a point that I've tried to make about ORAL surveillance, but also about this particular study, and I think also findings that the EMA found with filgotinib and with baricitinib, it's also the patients treated with TNF. It's also the patients treated with methotrexate. If they have a higher risk, then these are the patients at risk, no matter which mechanism you use.

Now, the difference between this study and ORAL surveillance was, you actually could compare tofacitinib versus the TNF inhibitors, because the study was powered to do that. So you can see that there were numerical differences, nonstatistical. We couldn't really do that here. So what we did find was the risk was actually similar in those. That minority of patients that were higher risk was similar with upadacitinib, methotrexate, adalimumab. But if it was correctly powered, if we had done a prospective study that was respectively powered, we might find the exact same findings as ORAL surveillance.

So the take home from this is: that if a patient has a higher risk of these events, no matter what mechanism you use, it's higher. The risk is higher than a patient who is younger, who is not a smoker, who doesn't have these risks. So in practice, what this means is that if I see a patient who's 70 years old, who's a smoker, who has cardiovascular risk, and I say, "Okay, I'm not going to use the JAK. I'm going to use the TNF," that's the wrong conclusion. Because you have to still be concerned about the patient treated with a TNF, or with methotrexate. So what the rheumatologist needs to do, or any practitioner needs to do, is try and mitigate their risk, as best as possible. Patient has a high lipids, treat it. Patient had a prior VTE, think about anticoagulation. Patient has a high BMI, try and get the patient to lose weight.

Rebecca Mashaw:

So across the board, whatever drug it is, this is a high risk group of patients to treat for rheumatoid arthritis, period?

Roy Fleischmann:

Right. But not treating actually raises the risk much higher. So patients who are inadequately treated for rheumatoid arthritis—so the active disease, if they're 35 years old, nonsmokers, don't have cardiovascular risk, but their disease is active, because disease activity really is very important in these patients—they have higher cardiovascular risk, higher VTE risk, higher SIE risk. It's a systemic inflammation, which raises those risks. So you do want to treat, and you do want to get the disease under control.

Another finding—not from this paper, because we couldn't really look at it this paper—from ORAL surveillance, is which of the patients who had the higher risk actually had the events, and it was patients with disease activity. So, I don't mean just high disease activity. I don't mean moderate disease activity. I don't mean low disease activity. I mean disease activity. So we have another publication which is in press actually from ORAL surveillance, which looks at the risk of VTEs. And it turns out, only one VTE occurred in a patient who is in remission, even though they had a higher risk for VTEs. All the other VTEs occurred with patients with disease activity, whether it be low disease activity, moderate, or high disease activity. So, it would be a mistake not to treat, because treatment does get the disease activity under control. And that really is a major, major driver of these events.

Rebecca Mashaw:

Did you arrive at any hypothesis about why patients are at higher risk for cardiovascular AEs, and also had higher rates of serious infection? Is this all about the inflammatory burden?

Roy Fleischmann:

It probably is about the inflammatory burden. We know that the risk of serious infection is higher in the untreated RA population than the treated population, as is the cardiovascular risk. So it is about the comorbidities that you see with active disease. So, your first goal should be to get the disease under control, and that does reduce the risk. Now, using advanced therapies, these drugs do affect the immune system. And you do see serious infections, to some extent, with all of these medications. And you certainly see herpes zoster occurring more frequently in patients treated with JAK inhibitors, than you do with TNF inhibitors and methotrexate. And that's because of the mechanism of action of the JAK inhibitors affecting interferon.

Rebecca Mashaw:

Would you hesitate to begin treatment with a JAK inhibitor for a patient who, let's say, didn't have any of those CV risk factors, but who had not been vaccinated against herpes zoster? Or would you go ahead, and then proceed with vaccinating the patient?

Roy Fleischmann:

You can strongly suggest to a patient that they become vaccinated. But whether they do or not, it's really up to them. And that's the question of access. So if I have a 35-year-old patient who doesn't have risk factors, other than for herpes zoster treated with upadacitinib, in this case, where the insurance may not cover the vaccination, then it becomes the question of access. But do I encourage all patients to obtain the vaccination? And the answer is yes.

Rebecca Mashaw:

The rates of nonmelanoma skin cancer and herpes zoster were higher across all the populations, including those who were not at higher risk for MACE or VTE. How significant is this risk? And do you have an idea as to why this would be the case?

Roy Fleischmann:

Well, I can speak to up upadacitinib directly. So the risk of herpes zoster is higher in a patient without the cardiovascular risk, basically because of the mechanism of action of upadacitinib, affecting interferon. Why it was higher in the methotrexate group and the adalimumab group, I'm really not that clear. The risk is higher generally in all of the programs that I've looked at, all the analysis that I've looked at, with the JAKs versus the TNFs and conventional synthetics. We've also noted the risk of nonmelanoma skin cancer, for whatever reason, is higher. The reason for that, I'm not clear about as well.

Rebecca Mashaw:

What advice would you offer your colleagues in rheumatology about prescribing upadacitinib, specifically for RA?

Roy Fleischmann:

So the question is, what would I actually do in practice, forgetting the FDA or the EMA guidance? We talked about disease activity, and we talked about getting the disease activity under control, as quickly as you can. I have been the first author in several studies which have shown that the JAK inhibitors, all the JAK inhibitors actually work quicker, and to somewhat better affect than TNF inhibitors, in patients who are methotrexate inadequate responders, and haven't been on a biologic. I would have a discussion with a patient, assuming they have access to all the drugs, and would actually make that point, and really try and decide, do they have a high cardiovascular risk? Do they have a high malignancy risk? Do they have a higher risk of VTEs? If they do, I would mitigate it. I would try to mitigate it, no matter which drug I would use. And then have a discussion with the patient, about the rapidity of the response to a JAK inhibitor versus a TNF inhibitor.

And then, it's always access, and what the PBM will allow, and the insurance company will allow, and what the country specific agencies will allow. But I wouldn't be against starting upadacitinib in a methotrexate incomplete responder. Certainly, if the patient doesn't have cardiovascular risk. And certainly, if they can mitigate the risk, for the reason that I stated, which is I think I can get the disease activity under better control quicker. And that is really a major component. In reality, in the United States, that doesn't occur very frequently. And that's because of access. The PBMs want to use a TNF.

With respect to cardiovascular events, the number needed to harm was upwards of 500. That would mean I would need to treat 500 more patients with a JAK, in that case, tofacitinib, in order to have one more event than I do with a TNF. And with respect to malignancy over the 5 years, it was about 275. I'd have to treat 275 more patients with a JAK to have one more malignancy than I would with a TNF.

So patients ask me, "What does that really mean?" And I say, "Well, so your chances of having this are 2 in 1,000, or 3 in 1,000. Or even 4 in 1,000. That's the risk." And it's important, because when all this information came out, and we had patients on the JAKs; they would question this. They would come back and say, "Wait, my risk is 2 in 1,000, 3 in 1,000, 4 in 1,000, and my disease is now under control. And now, you want me to stop a drug that has my disease under control, for this small risk, to have the risk of having more active disease, where my risk is increased. Can you do that again for me?" It turns out that none of the patients that I spoke to actually, in 2019, 2020, '21, '22, wanted to discontinue the JAK. And that's an important point.

So these risks are small. They're there. They're numerically different, but they're small. So the practitioner needs to think about the patient, think about the risk factors, mitigate them, think about the disease activity, what's more likely to get the patient's disease activity under control. And then make a decision, with the patient. Because the patient is going to take the medication, or not take the medication. I can tell a patient, "You have to take this medication." And that means nothing, because if a patient doesn't take it, what good is it? So it isn't I really have to convince the patient, but I have to give the patient enough information, and all of the information, so they're comfortable with what they decide to do.

And whatever they decide to do is fine with me. We helped develop all the TNFs. The TNFs are really good drugs. They decide they want to do a TNF, fine. And if they don't do well with the TNF, then use a JAK, fine. If they want to use the JAK first, fine. But if the disease doesn't come under control with the JAK, which happens in a proportion of patients, use a TNF, or use another mechanism of action. The prime endpoint is getting the disease under control. That should be the goal. No activity, that's the goal.

Rebecca Mashaw:

Getting the patient into remission.

Roy Fleischmann:

Getting the patient into a true remission. So I'm really looking at no tendon joints, no swollen joints, normal functioning, no radiographic progression, or as close as I possibly can. Now, how often do I hit that? Maybe 20, 30% of the time. And then, I do look for the lowest disease activity I can find.

Rebecca Mashaw:

How often do you see JAK inhibitors help you achieve that very low disease activity, or true remission?

Roy Fleischmann:

So in my practice, it is probably upwards of 80%.

Rebecca Mashaw:

Now, in addition to mitigating risk factors, as you mentioned before, what about monitoring these patients? What do you do to monitor them for signs that maybe this isn't going to be the best drug for them, not in terms of their RA, but in terms of their risk factors?

Roy Fleischmann:

Well, that's a really good question. I know what I should do. I'm old school. I was trained last century. And I actually believe a lot of what was written last century, about how to follow these drugs. So, if I have a patient on methotrexate, I actually look at their labs every month, maybe every 6 weeks. I don't do it every 3 months, or every 6 months, or once a year, as many rheumatologists do. So I can begin picking up things like hepatotoxicity or neutropenia fairly rapidly, before it becomes a major problem. I can talk to patients about open wounds, or if they have COPD, whether they're having recurrent infections. There are things you can do, along those lines. I should have the patient see a dermatologist yearly, looking for skin cancers. And sometimes, I remember to do that, and sometimes I don't.

VTE—Is the patient on hormone replacement therapy, which raises the risk? Have they had a VTE? Are they going to be doing things like taking an airplane trip for 18 hours and just sitting in a seat, without wearing a support hose? So those are the types of things that we discuss, as we talk about it.

But more importantly, I see the patient at least every 3 months, even if the patient is in remission. And the reason for that is that disease activity is important. It needs to be controlled. So there are times in which I see a patient, "How are you doing?" "I'm doing fine." And the last time I saw them, they actually had no tender joints and no swollen joints, and that's fine. And then this time I see them, and they're doing fine, and they have 3 tender joints, and they have 3 swollen joints. That's not good. So then, I would think about adjusting medications, and discussing with the patient. Or the patient may tell me, "I'm doing terribly," but I can't find anything. And it turns out to be a social issue, so you deal with a social issue. There are many aspects, but it's really seeing the patients on a regular basis. And although the practice is busy, I do make it my business to see the patient every 3 months.

Rebecca Mashaw:

Do you have any final thoughts that you'd like to share with your colleagues, on treating patients with RA with JAK inhibitors?

Roy Fleischmann:

Well, just would reemphasize what I said. Get the disease activity under control, mitigate the comorbidities as best as you can. Don't think it's just a JAK inhibitor that has these risks. It's not. We know it's a TNF. We know it's methotrexate. But first of all, treat.

Rebecca Mashaw:

Well, thank you so much for sharing your experience with us. Look forward to coming back to you at a later time, to talk about the newer studies coming out.

Fleischmann R, Curtis JR, Charles-Schoeman C, et al. Safety profile of upadacitinib in patients at risk of cardiovascular disease: integrated post hoc analysis of the SELECT phase III rheumatoid arthritis clinical programme. Ann Rheum Dis. 2023;82:1130-1141.