JAK Inhibitor Safety: What Did ORAL Surveillance Really Teach Us?

As many of us are learning, Janus kinase (JAK) inhibitors can be quite efficacious for various difficult-to-treat dermatologic disease processes but, unfortunately, also come with potential safety concerns. But where did these safety concerns originate? And what do they really mean? Can we perhaps develop a more nuanced understanding of this topic to better inform shared clinical decision-making with our patients? In this article, my goal is to promote a more contextualized dialogue on the relative safety of JAK inhibitors when used appropriately.

The ORAL Surveillance Trial
 

In September 2021, the US Food and Drug Administration (FDA) communicated its findings from the ORAL Surveillance trial of tofacitinib—the largest controlled interventional study in rheumatoid arthritis (RA) to date.^1,2 The FDA found that there was an increased risk of serious events, such as heart attack or stroke, cancer, blood clots, serious infections, and death, for the combined doses of tofacitinib compared with tumor necrosis factor (TNF) inhibitors. As such, the FDA began requiring new and updated label boxed warning language for all JAK inhibitors. Figure 1 shows the boxed warning for upadacitinib, which is FDA approved for moderate to severe atopic dermatitis (AD).³ Something similar can be seen in the labels for abrocitinib, also approved for moderate to severe AD,⁴ and baricitinib, approved for severe alopecia areata (AA).⁵ Although upadacitinib, abrocitinib, and baricitinib have not been studied in safety trials like ORAL Surveillance, they share a similar mechanism of action with tofacitinib, and so the FDA now requires similar boxed warning language for all 4 medications. Similar language can even be found in a boxed warning for topical ruxolitinib cream, which is FDA approved for mild to moderate AD and nonsegmental vitiligo.⁶ Why did the ORAL Surveillance trial of tofacitinib come about? One of the first JAK inhibitors developed, tofacitinib selectively inhibits JAK1, JAK3, and, to a lesser extent, JAK2. It is FDA approved for moderate to severe RA, ulcerative colitis, psoriatic arthritis, ankylosing spondylitis, and polyarticular course juvenile idiopathic arthritis. During its development, increases in serum lipid levels and the occurrence of cancers such as lymphoma were observed. This prompted the FDA to require a prospective, head-to-head, noninferiority safety trial comparing tofacitinib with TNF inhibitors in a high-risk RA population.²

Patients were randomly assigned in 1:1:1 ratio to receive oral tofacitinib 5 mg or 10 mg twice a day, or a subcutaneous TNF inhibitor (adalimumab in the United States and etanercept outside the United States). It is important to note that 100% of trial patients were also on background methotrexate, with a median dose of 17.3 mg/week, and 57.2% of trial patients were also on background systemic corticosteroids. The first patient was enrolled in March 2014. In February 2019, the patients receiving tofacitinib 10 mg twice a day were reduced to 5 mg twice a day after the safety monitoring board noted a higher frequency of pulmonary embolism (PE) and mortality.²

To be included in the ORAL Surveillance study population, patients needed to meet the following criteria:²

•          Active RA despite methotrexate treatment

•          Aged 50 years or older

•          At least 1 cardiovascular risk factor:

•          Current cigarette smoker

•          Hypertension

•          High-density lipoprotein cholesterol level less than 40 mg/dL

•          Diabetes mellitus

•          Family history of premature coronary heart disease

•          Extra-articular RA

•          History of coronary artery disease

Key exclusion criteria included current or previous cancer, except adequately treated nonmelanoma skin cancer (NMSC).² A similar number of patients were randomized to each of the 3 treatment arms. Patient years of exposure, as well as baseline characteristics, were also similar. Close to one-third (31.0%) of patients were older than age 65 with mean duration of their RA more than 10 years, and almost half (48.2%) had a history of smoking.²

Contextualizing the Results

 

When one looks closely at the Oral Surveillance data, it seems to me that the incremental increased risk for both primary end points (major adverse cardiovascular events [MACE] and malignancies) is less than many may realize even in this high-risk population, and is likely to be minimal, if even elevated at all, in low-risk dermatology populations when practicing good patient selection.

Major Adverse Cardiovascular Events

The incidence of MACE, defined as either a death from cardiovascular causes, a nonfatal myocardial infarction (MI), or a nonfatal stroke, was higher with the combined tofacitinib doses at 3.4% than with TNF inhibitors at 2.5%, and the statistical threshold for noninferiority was not achieved (Figure 2). MACE incidence rates were higher across all trial groups among patients aged 65 years and older.²

But can we contextualize these MACE results further? To have 1 more MACE relative to TNF inhibitors, the number of patient years of exposure needed to harm was 567 for the lower dose of tofacitinib.⁷ This means that in a trial designed to elicit safety events (ie, enriched with high-risk patients), you would need to line up 567 of these patients and give them low-dose tofacitinib for 1 year to elicit 1 more MACE compared with TNF inhibitors. Remember, these patients have RA, they are aged 50 years or older, they are on methotrexate, more than half are on chronic steroids, almost half have a smoking history, and they all have at least 1 cardiovascular risk factor. In addition, TNF inhibitors likely decrease the risk of MACE in the RA population, which may have also contributed to tofacitinib's inability to reach statistical noninferiority.⁸ Lastly, when we look at subsequent post hoc multivariate analyses, we can see that there are some independent risk factors for MACE, irrespective of whether patients were given tofacitinib or a TNF inhibitor. These include current smoking, aspirin use, being older than age 65, or male sex.⁹ 

All in all, I feel that these MACE data show 2 things: 1) serious cardiovascular events attributed to JAK inhibition are relatively rare even in high-risk populations and 2) they are likely to be very rare, if even elevated from baseline at all, in low-risk populations when practicing good patient selection.

Malignancies

 

The incidence of cancers was also higher with the combined tofacitinib doses at 4.2% than with TNF inhibitors at 2.9%, and the statistical threshold for noninferiority was again not achieved (Figure 3). The most common cancers were lung cancers and lymphomas with tofacitinib and breast cancers with TNF inhibitors. Cancer incidence rates were higher across all trial groups among patients aged 65 years and older.²

Similar to MACE, can we also perhaps contextualize these malignancy results further? The number of patient years of exposure needed to elicit 1 more malignancy event for tofacitinib compared with TNF inhibitors was 276 for tofacitinib 5 mg and 275 for tofacitinib 10 mg.7 I find it interesting that this number is statistically equal for both doses of tofacitinib even in this very large trial. What I also find interesting is that on subsequent multivariate analysis of this data, when you remove the risk factors of being aged 65 or older and past or current smoking, the hazard ratio for malignancy goes from a significant 1.55 to a nonsignificant 1.16.10 Lung cancers, in part correlated with smoking and age, seem to have driven much of the incremental increased malignancy risk with tofacitinib compared with TNF inhibitors in ORAL Surveillance.

Taking this all together, I think these data show again that it all comes down to patient selection for JAK inhibitors. When you remove older patients and smokers from the population, there does not appear to be a statistically elevated risk of cancers with the use of tofacitinib, at least in this large data set of high-risk patients. In addition, we have fortunately thus far not seen any increased risk of malignancy or clustering of malignancies by organ system with any of the JAK inhibitors approved in dermatology.

Infections

What about infections? We know that the rates of herpes zoster are higher with JAK inhibitors, likely at least in part due to inhibition of interferon signaling, which is important for viral immunosurveillance. However, the rates seen in Oral Surveillance are higher than what we are seeing with JAK inhibitors in dermatology thus far.² Why is that? Well, our JAK inhibitor patients tend to be younger, healthier individuals who are not on concomitant immunosuppressants. Serious infection risk was actually only significantly elevated for tofacitinib 10 mg twice daily.^2,11 As we all know, immunosuppression exists on a spectrum; for a younger, healthier population with lower levels of JAK inhibition, the rates of infections and serious infections is most likely to be relatively low. Regardless, I still do recommend bringing herpes zoster vaccination on board for all adults on JAK inhibitors, given that the Centers for Disease Control and Prevention now recommends vaccination for all adults aged 19 years and older who are or will be immunosuppressed.¹²

Thromboembolic Events

When we look at tofacitinib 5 mg twice daily compared with TNF inhibitors, we see statistically higher incidence rates and hazard ratios for deep vein thrombosis (DVT), PE, and venous thromboembolism (VTE).² Now these rates go up consistently for tofacitinib 10 mg twice daily, but this dosage is no longer even approved for patients with RA and not a level of JAK inhibition commonly used in dermatology.

But again, let’s contextualize these safety findings a bit further. These events are rare, particularly for the lower levels of JAK inhibition even in this high-risk population.^2,13,14 Comparing tofacitinib 5 mg twice daily with TNF inhibitors, the number needed to harm in this high-risk population was 763 and 870 for VTE and PE, respectively.¹⁴ So how high are these number needed to harm figures for a low-risk AD or AA population? Well, we do not really know because we have not run a similarly massive, controlled safety trial like ORAL Surveillance for the newer JAK inhibitors FDA approved in our patient populations, but I think we can safely assume that the numbers would be lower given these are healthier populations.  However, please do not think that the clinical trials for the JAK inhibitors FDA approved in dermatology “cherry picked” their patient populations to be “low risk.” Many of these phase 3 trials began enrolling patients before ORAL Surveillance even read out. For example, let’s look at the baseline characteristics for MEASURE UP 1 and MEASURE UP 2, which were the pivotal monotherapy phase 3 trials for upadacitinib in AD.^15-17 Approximately one-third of patients had at least 1 of the previously defined ORAL Surveillance cardiovascular risk factors. A similar number of patients had a smoking history. In addition, approximately one-fifth of patients were on oral contraceptives.

JAK Inhibitors in Dermatology

With all this talk about tofacitinib safety or lack thereof in RA, how should we think about the newer JAK inhibitors FDA approved in dermatology? Safety data to date for these JAK inhibitors suggest that they are associated with low incidence rates of malignancy, MACE, and thromboembolic events, but how low are they?  A recently published article in the Journal of Drugs in Dermatology lends support to the notion that these JAK inhibitors may be as safe, if not safer, than some of the other traditional systemic therapies that we have previously used quite often to manage inflammatory skin diseases.¹⁸ The study showed that traditional systemic therapies for AD (methotrexate, cyclosporine, systemic corticosteroids) demonstrated equal or higher incidence rates for malignancy (excluding NMSC), NMSCs, MACE, and VTE compared with abrocitinib and upadacitinib. The use of these JAK inhibitors also exhibited either comparable or lower incidence of malignancy (excluding NMSC), MACE, and VTE, but higher rates of NMSCs, in comparison with baseline rates in AD or control patients.

What other legacy treatments used in dermatology and medicine as a whole would fail the ORAL Surveillance test? We may never know, but I think it is an important and interesting question to ask. Regardless, in dermatology we are thankfully primarily using lower levels of JAK inhibition. For example, the starting dose for upadacitinib and abrocitinib is 15 mg and 100 mg, respectively; however, the FDA has approved a 45-mg induction dose for upadacitinib in ulcerative colitis. In addition, the JAK inhibitors FDA approved in dermatology have different in vitro selectivity profiles for the 4 JAK enzymes than tofacitinib, which could potentially lead to different clinical profiles.^19-21 However, the relevance of inhibition of specific JAK enzymes to clinical profiles is not currently known.^3-6

Conclusion

To summarize, oral JAK inhibitors are FDA approved for moderate to severe AD and severe AA. A topical JAK inhibitor is FDA approved for mild to moderate AD and nonsegmental vitiligo. Some of us are also using JAK inhibitors off -label for various other dermatologic diseases, including widespread vitiligo, cutaneous sarcoidosis, hidradenitis suppurativa, cutaneous graft-host disease, lichen planus, granuloma annulare, and severe cutaneous drug reactions.²²  Patients who have a higher risk of complications with JAK inhibitors include those aged 65 or older; smokers; or patients at high risk for or with a history of malignancy, VTE, or cardiovascular events. In these patients, JAK inhibitors should likely only be used when there are no other good options after extensive patient counseling on potential risks followed by shared decision-making. However, in dermatology, fortunately many patients do not fall into these categories, which makes them more favorable candidates for the use of JAK inhibitors when appropriate.

Taken all together, it is my opinion that we need to continue to get more comfortable having a more nuanced understanding of JAK inhibitor safety in our patient populations so we can have fair and balanced discussions with our patients on the risk-benefit profiles of these highly efficacious therapies. It is interesting to me that JAK inhibitors are reportedly being used more by dermatologists in Canada, Japan, and Europe than they are in the United States. It is my opinion that our ex-US dermatology colleagues have gained an understanding that these therapies are quite safe in healthy populations when used appropriately.

References

1. FDA requires warnings about increased risk of serious heart-related events, cancer, blood clots, and death for JAK inhibitors that treat certain chronic inflammatory conditions. US Food and Drug Administration. September 1, 2021. Updated December 7, 2021. Accessed February 6, 2023. https://www.fda.gov/drugs/drug-safety-and-availability/fda-requires-warnings-about-increasedrisk-serious-heart-related-events-cancer-blood-clots-and-death

2. Ytterberg S, Bhatt DL, Mikuls TR, et al. Cardiovascular and cancer risk with tofacitinib in rheumatoid arthritis. N Engl J Med. 2022;386(4):316-326. doi:10.1056/ NEJMoa21099272

3. RINVOQ full prescribing information. US Food and Drug Administration.Revised April 2022. Accessed February 16, 2023. https://www.accessdata.fda.gov/drugsatfda_docs/label/2022/211675s003lbl.pdf

4. CIBINQO full prescribing information. US Food and Drug Administration. Revised January 2022. Accessed February 16, 2023. https://www.accessdata.fda.gov/drugsatfda_docs/label/2022/213871s000lbl.pdf

5. OLUMIANT full prescribing information. US Food and Drug Administration. Revised May 2022. Accessed February 16, 2023. https://www.accessdata.fda.gov/drugsatfda_docs/label/2022/207924s006lbl.pdf

6. OPZELURA full prescribing information. US Food and Drug Administration.Revised July 2022. Accessed February 16, 2023. https://www.accessdata.fda.gov/drugsatfda_docs/label/2022/215309s001lbl.pdf

7. Ytterberg S, Bhatt D, Mikuls T, et al. Safety and effi cacy of tofacitinib vs TNF inhibitors in RA patients aged 50 years or older with one or more cardiovascular risks: results from a phase 3b/4 randomized safety trial. Arthritis Rheumatol. 2021;73(suppl 9). Accessed February 6, 2023. https://acrabstracts.org/abstract/safety-and-effi cacy-of-tofacitinib-vs-tnf-inhibitors-in-ra-patients-aged-50-yearsor-older-with-one-or-more-cardiovascular-risks-results-from-a-phase-3b-4-randomized-safety-trial

8. Nair S, Kahlon SS, Sikandar R, et al. Tumor necrosis factor-alpha inhibitors and cardiovascular risk in rheumatoid arthritis: a systematic review. Cureus. 2022;14(6):e26430. doi:10.7759/cureus.26430

9. Charles-Schoeman C, Buch M, Dougados M, et al. Risk factors for major adverse cardiovascular events in patients aged 50 years with RA and 1 additional cardiovascular risk factor: results from a phase 3b/4 randomized safety study of tofacitinib vs TNF inhibitors. Arthritis Rheumatol. 2021;73(suppl 9). Accessed February 6, 2023. https://acrabstracts.org/abstract/risk-factors-for-majoradverse-cardiovascular-events-in-patients-aged-%e2%89%a5-50-years-withra-and-%e2%89%a5-1-additional-cardiovascular-risk-factor-results-from-aphase-3b-4-randomized-safety-stud

10. Curtis J, Yamaoka K, Chen Y, et al. Malignancies in patients aged   50 years with RA and   1 additional cardiovascular risk factor: results from a phase 3b/4 randomized safety study of tofacitinib vs TNF inhibitors. Arthritis Rheumatol. 2021; 73(suppl 9). Accessed February 21, 2023. https://acrabstracts.org/abstract/malignancies-in-patients-aged-%e2%89%a5-50-years-with-ra-and-%e2%89%a5-1-additional-cardiovascular-risk-factor-results-from-a-phase-3b-4-randomizedsafety-study-of-tofacitinib-vs-tnf-inhibitors

11. Balanescu A, Citera G, Pascual-Ramos V, et al. Incidence of infections in patients aged   50 years with RA and   1 additional cardiovascular risk factor: results from a phase 3b/4 randomized safety study of tofacitinib vs TNF inhibitors. Arthritis Rheumatol. 2021;73(suppl 9). Accessed February 6, 2023. https:// acrabstracts.org/abstract/incidence-of-infections-in-patients-aged-%e2%89%a5- 50-years-with-ra-and-%e2%89%a5-1-additional-cardiovascularrisk-factor-results-from-a-phase-3b-4-randomized-safety-study-of-tofacitinibvs-tnf-inhib

12. Frequently asked questions about Shingrix. Centers for Disease Control and Prevention. January 24, 2022. Accessed February 10, 2023. https://www.cdc.gov/vaccines/vpd/shingles/hcp/shingrix/faqs.html

13. Molander V, Bower H, Frisell, T, Askling J. Risk of venous thromboembolism in rheumatoid arthritis, and its association with disease activity: a nationwide cohort study from Sweden. Ann Rheum Dis. 2021;80(2):169-175. doi:10.1136/annrheumdis-2020-218419

14. Charles-Schoeman C, Fleischmann R, Mysler E, et al. The risk of venous thromboembolic events in patients with RA aged   50 years with   1 cardiovascular risk factor: results from a phase 3b/4 randomized safety study of tofacitinib vs TNF inhibitors. Arthritis Rheumatol. 2021;73(suppl 9). Accessed February 6, 2023. https://acrabstracts.org/abstract/the-risk-of-venous-thromboembolicevents-in-patients-with-ra-aged-%e2%89%a5-50-years-with-%e2%89%a5-1-cardio¬vascular-risk-factor-results-from-a-phase-3b-4-randomized-safetystudy-of-tofacitinib-vs-tn

15. RINVOQ package insert. AbbVie Inc.; October 2022.

16. A study in healthy adults and adult subjects with rheumatoid arthritis to evaluate the safety, tolerability and pharmacokinetics after multiple doses of ABT-494. ClinicalTrials.gov. December 5, 2012. Updated November 20, 2017.Accessed February 14, 2023. https://clinicaltrials.gov/ct2/show/NCT01741493

17. Guttman-Yassky E, Teixeira HD, Simpson EL, et al. Once-daily upadacitinib versus placebo in adolescents and adults with moderate-to-severe atopic dermatitis (Measure Up 1 and Measure Up 2): results from two replicate doubleblind, randomised controlled phase 3 trials. Lancet. 2021;397(10290):2151-2168. doi:10.1016/S0140-6736(21)00588-2

18. Daniele S, Bunick C. JAK inhibitor safety compared to traditional systemic immunosuppressive therapies. J Drugs Dermatol. 2022;21(12):1298-1303. doi:10.36849/JDD.7187

19. Mohamed MEF, Beck D, Camp HS, Othman AA. Preferential inhibition of JAK1 relative to JAK3 by upadacitinib: exposure-response analyses of ex vivo data from 2 phase 1 clinical trials and comparison to tofacitinib. J Clin Pharmacol. 2020;60 (2):188-197. doi:10.1002/jcph.1513

20. Vazquez ML, Kaila N, Strohbach JW, et al. Identifi cation of N-{cis-3-[Methyl(7H-pyrrolo[2,3-d]pyrimidin-4-yl)amino]cyclobutyl}propane-1-sulfonamide (PF-04965842): a selective JAK1 clinical candidate for the treatment of autoimmune diseases. J Med Chem. 2018;61(3):1130-1152. doi:10.1021/acs.jmedchem.7b01598

21. Mayence A, Vanden Eynde JJ. Baricitinib: a 2018 novel FDA-approved small molecule inhibiting Janus kinases. Pharmaceuticals (Basel). 2019;12(1):37. doi:10.3390/ph12010037

22. Garcia-Melendo C, Cubiró X, Puig L. Janus kinase inhibitors in dermatology: part 2: applications in psoriasis, atopic dermatitis, and other dermatoses. Actas Dermosifi liogr (Engl Ed). 2021;112(7):586-600. doi:10.1016/j.a engl.2021.05.008