Disclosures: Cohen reports consulting fees from AbbVie, Pfizer, Eli Lilly & Co. and Gilead Sciences. Winthrop reports consulting fees from Pfizer, AbbVie, UCB, Eli Lilly & Co., Galapagos, GlaxoSmithKline, Roche, Gilead Sciences, Bristol Myers Squibb, Regeneron, Sanofi, AstraZeneca and Novartis, as well as research grants from Bristol Myers Squibb and Pfizer.
February 16, 2021
10 min read

'We need to see the data': FDA's Xeljanz safety warning stirs concerns for JAK inhibitors

Disclosures: Cohen reports consulting fees from AbbVie, Pfizer, Eli Lilly & Co. and Gilead Sciences. Winthrop reports consulting fees from Pfizer, AbbVie, UCB, Eli Lilly & Co., Galapagos, GlaxoSmithKline, Roche, Gilead Sciences, Bristol Myers Squibb, Regeneron, Sanofi, AstraZeneca and Novartis, as well as research grants from Bristol Myers Squibb and Pfizer.
You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact customerservice@slackinc.com.

For Pfizer and its JAK inhibitor, Xeljanz, February rang in with the sound of alarm bells, as the FDA issued its third warning about the drug in 2 years.

In this latest alert regarding tofacitinib, the FDA warned providers of an increased risk for serious heart-related issues and cancer among older patients, compared with TNF inhibitors.

“I do think these results are serious,” Stanley Cohen, MD, told Healio Rheumatology. “I think it’s something we have to look at critically, and again, with just the topline — what can you say? We need to look at all the data and better understand the data.” Source: Adobe Stock

However, unlike prior FDA warnings about the JAK inhibitor, which were limited to the higher 10 mg dosage only approved for ulcerative colitis, this newest announcement raises red flags for both the 10 mg and 5 mg doses, casting a shadow over all three of the drug’s indications, including rheumatoid arthritis and psoriatic arthritis. What’s more, while previous alerts have warned about the risk for pulmonary embolism and even death, the latest interim data come with the novel finding that tofacitinib may increase the risk for malignancy.

All these alerts and warnings have been based on interim and preliminary results from ORAL Surveillance, a post-marketing safety clinical trial ordered by the FDA when it first approved tofacitinib for rheumatoid arthritis in 2012. However, despite this trial wrapping in July 2020, its full results, beyond the co-primary endpoints — including secondary endpoints such as pulmonary embolism and mortality as well as efficacy data — remain unavailable. Only the topline results have been released, and only through alerts and press releases from the FDA and Pfizer.

This has left many experts anxious to dig into the full data, which could hold drastic consequences not only for Xeljanz, but also for JAK inhibitors as a class. Until then, they said, it’s difficult to forecast what impact these results will have.

Kevin Winthrop

“It’s hard to live by release alone,” Kevin Winthrop, MD, MPH, director of the Center for Infectious Disease Studies at Oregon Health & Science University, told Healio Rheumatology. “To be honest, it’s hard to know what to think. We need to see the data in the analysis. I will be interested to see the malignancy data, and how those malignancies were distributed in time. Which dose groups? When? Were there any trends in terms of types of malignancy? How did they cluster in terms of type and subtype? A lot of that will be interesting to look at, and will be necessary to look at.”

Prior History

When the FDA first approved tofacitinib for RA, the agency required a clinical trial to examine the risk for heart-related events, cancer and opportunistic infections at both the 5 mg and 10 mg doses, compared with TNF inhibitors. Participants were required to be at least 50 years old and have at least one cardiovascular risk factor.

A total of 4,369 enrolled participants were randomized to receive either 5 mg of tofacitinib twice daily, 10 mg of tofacitinib twice daily, 40 mg of subcutaneous adalimumab (Humira, AbbVie) every other week, or 50 mg of subcutaneous etanercept (Enbrel, Amgen).

“During the trial, the Data and Safety Monitoring Board found an imbalance and the presence of pulmonary embolism, and a trend for deep vein thrombosis,” Stanley Cohen, MD, clinical professor in the department of internal medicine at the University of Texas Southwestern Medical School, and medical codirector of the Metroplex Clinical Research Center, told Healio Rheumatology. “Certainly the 10 mg dose for PEs was statistically different from the Enbrel or Humira wing, and the DVTs were not statistically different but numerically different.”

Stanley Cohen, MD
Stanley Cohen

These early findings led the FDA to issue a safety alert in February 2019, stating the 10 mg twice daily dose was associated with an increased risk for blood clots in the lungs and death. As a result, Pfizer transitioned patients who were receiving the higher dose of tofacitinib to the 5 mg arm. Meanwhile, the FDA in July 2019 tagged the drug’s 10 mg dose with a Boxed Warning for patients with ulcerative colitis.

Pfizer, via a press release, announced the co-primary endpoint results of the completed study in January. Those endpoints — noninferiority with tofacitinib compared with TNF inhibitors related to major adverse cardiovascular events and malignancies excluding non-melanoma skin cancer — were not met, the company said. In addition, based on the prespecified secondary comparisons, there was no evidence of a difference in the primary endpoints between the two tofacitinib treatment groups.

According to the Pfizer release, the primary analyses included 135 participants with major adverse cardiovascular events and 164 subjects with malignancies. For tofacitinib, the most frequently reported cardiac event was myocardial infarction, while the most frequently reported malignancy was lung cancer.

“In the combined 5 and 10 mg arm, compared to Humira and Enbrel, tofacitinib was not not-inferior — this is an issue of concern,” Cohen said. “We would have expected it to be in the same range. Now, until we have a deeper dive on the data, and understand which patients were impacted, what were the risk factors, how this impacts the future use of JAK inhibitors is not clear.”

Where’s the Data?

Although officials from Pfizer declined an interview for this story, a spokesperson for the company provided a statement pointing out that — in contrast to previous tofacitinib studies — participants in ORAL Surveillance were required to be at least 50 years of age and have at least one additional cardiovascular risk factor.

“We continue to analyze the study data, including secondary endpoints, and believe additional analysis may further clarify the benefit and risk profile of tofacitinib to help inform medical decision-making and patient care,” the statement read, in part. “As these additional analyses become available, we will work with the FDA and other regulatory agencies to review the results. Patient safety is of the upmost importance to Pfizer and the company continually monitors the safety of its medicines.”

However, it remains unclear when the full results will be available. Meanwhile, there is no shortage of questions regarding the released preliminary results. Chief among these uncertainties, according to Winthrop, is how participants were analyzed in the transition from the 10 mg arm to the 5 mg group.

“I think one big question will be, how was that transition handled?” Winthrop said. “They still report data in the 5 and 10 mg buckets, and then a collapsed tofacitinib group. Presumably all the people in the 10 mg arm moved to 5 mg a year and a half ago, so I don’t really know what their rules were for attributing events to those dosing groups. If you were in the 10 mg arm, and then you moved to 5 mg, and then you got cancer 3 months after switching to 5 mg, did that cancer get attributed to you being on 5 mg or 10 mg? We need to look at all that to better understand the 5 and 10 mg dosing groups.”

There is also the matter of the discrepancy between the ORAL Surveillance topline results and what has been seen in real-world data, he added.

According to Winthrop, other studies have yet to find higher rates of either major cardiovascular events or malignancy with tofacitinib, compared with other treatments. He specifically pointed to one study published in 2020 by Philip Mease, MD, of the Swedish Medical Center and the University of Washington, and colleagues in the Annals of the Rheumatic Diseases, which found that the incidence rates of DVT, PE and arterial thromboembolism in the tofacitinib RA, psoriasis and PsA programs were similar across tofacitinib doses, and generally consistent with observational data and published rates of other treatments.

“When you look at the real-world data, we have not seen those signals,” Winthrop said. “The rates for MACE and malignancy are no different in tofacitinib than in anything else. Even looking at people with cardiovascular risk factors, we have not seen a difference in MACE between people on tofacitinib and people on TNF blockers. Phil Mease published an analysis a few months ago looking at this using Corrona data; they looked at TNF starters and they looked at tofacitinib starters, as well as specifically people who had cardiovascular risk factors, and they didn’t see any increase in risk. The rates were the same.”

“So, we haven’t seen this in the real world, and certainly not in other trial data,” he added. “So, this is really a novel finding.”

Cohen, meanwhile, compared ORAL Surveillance to the ENTRACTE study, which was published in 2019 in Arthritis & Rheumatology by Jon Giles, MD, MPH, of Columbia University College of Physicians & Surgeons, and colleagues. That study compared MACE risk among patients with RA treated with tocilizumab (Actemra, Genentech) or etanercept.

“What’s a little interesting, if you compare it to the ENTRACTE study, the incident rate for MACE was a little lower for the TNF inhibitors [in ORAL Surveillance] than it was in the ENTRACTE study,” Cohen said. “If the numbers for Enbrel and Humira in ORAL Surveillance had been the same as the ENTRACTE study’s Enbrel finding, then tofacitinib would have achieved noninferiority.”

“So, it’s just too early to say,” Cohen added. “Although, if I had to guesstimate, I think these new findings will continue to suggest that biologics should remain first-line after [conventional synthetic] DMARD failure, and that JAK inhibitors should be avoided, unless there are no alternatives, in patients who have these risk factors.”

Concerns for Rheumatologists

Despite the dearth of data, both Cohen and Winthrop agreed the currently available results are nonetheless concerning.

For Cohen, the fact that ORAL Surveillance was a head-to-head trial suggests there is “certainly a signal” for major cardiovascular events and cancer with tofacitinib.

“I do think these results are serious,” Cohen said. “I think it’s something we have to look at critically, and again, with just the topline — what can you say? We need to look at all the data and better understand the data. I am comforted that the absolute risk is low and similar to what we have seen, but there is a differential here. I think this would suggest that biologics should remain a first-line or triple therapy after a csDMARD failure, before JAK inhibitors in these patients.”

“Now, if I have a 40-year-old patient who is otherwise healthy, and they want to take an oral versus an injectable, I think a JAK inhibitor would be fine,” Cohen said. “If I have a 75-year-old with COPD, hypertension and a previous stent? I don’t know if this would be my first choice.”

Winthrop, too, is willing to give significant weight to Pfizer’s results coming from a randomized controlled trial. And although he has many questions about the findings — particularly regarding the malignancy data — he admitted that they nonetheless raise the possibility that “something real is going on here.”

“I think rheumatologists should be concerned,” Winthrop said. “As someone who cares about safety, I am concerned.”

However, he added that these results, as they stand now, have not altered his perception of tofacitinib, and its safety profile, in any drastic way.

“I think all of this — the malignancy and MACE data — they’re concerning enough that steering people who have multiple risk factors for those things away from JAK inhibitors might be reasonable, particularly VTE,” Winthrop said. “I think if you have a history of DVT or PE, I would not be excited about starting a JAK inhibitor in such a patient. That’s how I’ve felt for a couple years now based on the data to date, so nothing has changed for me there.”

‘A Cloud over JAK Inhibitors’

Tofacitinib is one of three JAK inhibitors approved for RA by the FDA; the other two being baricitinib (Olumiant, Eli Lilly & Co.) and upadacitinib (Rinvoq, AbbVie). Although the FDA’s past warnings regarding Xeljanz have not affected the drug’s popularity — its prescription volume rose 12% over the past year, and it is prescribed to more than 208,000 adults worldwide — it is possible this latest alert could benefit its competitors in a tightening U.S. market.

On the other hand, it’s also possible that tofacitinib’s cardiovascular and cancer concerns may call into question whether all JAK inhibitors share these risks.

According to Cohen, Pfizer’s results with Xeljanz indicate that other companies that produce JAK inhibitors should likely go back and examine their clinical trial programs, selecting out those patients with the same risk factors to see if there was any imbalance in MACE or malignancies.

“Baricitinib has a study that is ongoing, which is also looking at outcomes such as DVT and VTE in a somewhat enriched population of patients with cardiovascular disease,” Cohen said. “We also have baricitinib data that suggest an increased risk for VTE in single controlled trials, so I do think there’s a cloud over the JAK inhibitors at this point, as far as the safety of patients with multiple comorbidities and risk for cardiovascular events.

“I think that this safety concern, combined with the VTE/PE issue, will probably relegate JAK inhibitors to second-line for patients who are at risk for cardiovascular events,” he added.

In lieu of available data, the real impact will come when the relevant regulatory authorities have their say, Cohen said.

He also predicted more Boxed Warnings on the horizon.

“Right now, we are waiting for the FDA and the European Medicine Agency, and other regulatory agencies to review this data, and they are going to guide us,” he said. “There will probably be future black box warnings, so I think it’s going to impact JAK utilization.”

For Winthrop, the possibility of all JAK inhibitors being painted with these risks remains an open question.

“We don’t really know if all JAK drugs increase the risk or if really any of them do, because we just haven’t had the studies yet to prove it,” he said. “Certainly, that interim analysis and [ORAL Surveillance] suggested that this is a real potential adverse effect for JAKs and that it might be a JAK class issue.”

In addition, although there are strong safety similarities between all three JAK inhibitors approved by the FDA for RA, Winthrop noted that it remains uncertain whether “all is equal when it comes to some of these outcomes.”

“Whether VTE is a class issue or just the issue of one drug, I don’t know yet,” he said. “I just think that we need more study.”

And until more data becomes available, the question of just how concerned rheumatologists should be regarding Xeljanz and other JAK inhibitors will remain unanswered.

“It’s ‘Houston, we have a problem,’ I can tell you that,” said Cohen. “We just don’t know how significant a problem it is yet.”

For more information:

Stanley Cohen, MD, can be reached at 8144 Walnut Hill Ln., Suite 800, Dallas, TX 75231; email: arthdoc@aol.com.

Kevin L. Winthrop, MD, can be reached at 270 Southwest Pavilion Loop OHSU Physicians Pavilion, Suite 320, Portland, OR 97239.