Safety of JAK Inhibitors: A Closer Look at Big Pharmacovigilance Data

Infections, Blood Clots, Cancer and Gastrointestinal Perforation Identified as Increased Risks with JAK Inhibitors

JAK inhibitors are increasingly studied for use in various types of hair loss. We are just a few months away from a JAK inhibitor being approved for alopecia areata. More will likely follow, and more will follow after that.

JAK inhibitors are already approved and widely used for treating a variety of inflammatory medical conditions including psoriasisis, rheumatoid arthritis, inflammatory bowel disease, and others. In the hair clinic, we use JAK inhibitors off label for treating alopecia areata, as well as lichen planopilaris, frontal fibrosing alopecia and folliculitis decalvans.

How Safe are JAK Inhibitors for Hair Loss Patients?

A key questions I grapple with every single day is: How safe are the JAK inhibitors for my hair loss patients? The JAK inhibitors like tofacitinib and baricitinib seem pretty safe but there are some major limitations to being able to answer this question fully. Anyone who says “These drugs are completely safe!” has just not been following the published literature in the medical journals. Certainly, I feel they have pretty good safety and I do use them in my clinics nearly every day.

But there are several challenges that arise when a patient or the patient’s family asks me: “how safe are these drugs for hair loss patients?”

1) Most hair loss studies involving JAK inhibitors are small.

Let’s face it, most JAK inhibitor studies are pretty small. That’s okay. That’s just a reality of how these studies are done. The largest hair loss study to date has about 500 patients to 600 patients in each group. The BRAVE AA1 has about 600 patients and BRAVE AA-2 has about 500 patients. Just a few hundred patients get the drug and the remainder get placebo.

Most of the prior tofacitinib retrospective reviews have a few dozen or about 100 patients.

Studies are small in the hair world.

Let’s face it, JAK inhbitors will be approved by various health regulatory bodies based not on studies of 1 million patients or even a few thousand patients but rather based on studies of a few hundred patients. That’s okay. That’s just a reality of how studies are done and the number of patients that are needed to render fairly solid conclusions about short term safety and efficacy.

2) Most studies are done in patients without hair loss problems.

Most of my knowledge about what to tell patients about JAK inhibitors, what tests to order before starting JAK inhibitors and what tests to order after starting JAK inhibitors come from careful examination of studies done in the rheumatology field - mainly in patients using tofacitinib for rheumatoid arthritis. We do the exact same thing when we start baricitinib or ruxolitinib for hair loss - but the largest experience worldwide is with tofacitinib.

It’s really senseless right now for anyone to prescribe JAK inhibitors for any patient unless one feels they have pretty solid knowledge about tofacitinib. We base so much on what we do now on tofacitinib data.

If you are a hair loss specialist, one needs to understand tofacitinib literature really well if you’re going to dive into the world of JAK inhibitors to try to better help your patients.

On Nov 6 2012, the FDA approved tofacitinib for rheumatoid arthritis. So the world has a solid 10 years of experience with this medication - for rheumatoid arthritis.

3) Most studies have limited follow up.

We don’t have 70 years of data behind us about the use of JAK inhibitors. Do we really need 70 years of data? Well, yes, of course, that would be the ideal situation. If we want to be 100 % certain about long term side effects we need long term data. That’s pretty simple.

If we want to be somewhat certain about long term safety in human beings we need reasonably long term studies in human beings

if we want to be somewhat certainly about the long term side effects in human beings with hair loss issues, then we need reasonably long term studies done specifically in human beings with hair loss problems.

We are 10 years out from the date that tofacitinib started to be used in rheumatoid arthritis. The recent New England Journal study summarizing the ORAL SURVEILLANCE STUDY raised issues about whether JAK inhibitors might be associated with increased cancer risks and heart disease risks. It seems that it might increase the risk of heart disease and cancer - if one is talking specifically about a patient with rheumatoid arthritis 50 years of age and older who failed methotrexate in the past. Experts question whether there are really similar risks in younger patients who don’t have rheumatoid arthritis.

If you are looking for long term studies of JAK inhibitors in hair loss patients, there are none. The wonderfully designed BRAVE AA1 and BRAVE AA2 trials go to 36 weeks. The trials are ongoing to look at data at year 2, 3 and 4. But one needs to keep in mind that we absolutely have zero good long term data of JAK inhibitor use in hair loss patients.

For this reason, I’m always on the watch for studies addressing safety. Of course, I am interested in clinical trial data. What seems to happening to patients in clinical trials who use these medications?

But I’m also interested in real world data. What seems to happening to patients outside of clinical trials who use these medications?

Hoisnard L et al study of 2022 is one of the studies in the second type of studies - studies with real world data.

Hoisnard L et al. 2022

Hoisnard and colleagues set out to assess safety of three JAK inhibitors: ruxolitinib, tofacitinib and baricitinib. The study was of retrospective observational study and pharmacovigilance data were extracted from the World Health Organization database. The World Health Organization (WHO) international pharmacovigilance database, VigiBase, contains more than 24 million individual case safety reports (ICSRs). To identify if there was any sort of safety concern, the authors used a technique known as disproportionality analysis.

Of the over 24 million case safety reports in VigiBase, the authors found 126,815 individual case safety reports involving JAK inhibitors. This comprised a total of 376,487 adverse events. Not surprisingly, tofacitinib had the highest number of reports given that it has been on the market the longest. In 16.3% of the ICSRs for ruxolitinib, 9.6% for tofacitinib and 12.9% for baricitinib, the side effects experienced by the patient caused or prolonged hospitalization. In 14.0% of the ICSRs for ruxolitinib, 1.9% for tofacitinib and 1.4% for baricitinib, the adverse events caused death.

More than three quarters of the safety reports for tofacitinib and baricitinib involved women. Rheumatoid arthritis was the underlying diagnosis in 55 % of tofacitinib reports and rheumatoid arthritis was the underlying diagnosis is 79.7 % of baricitinib case reports, The majority of case reports in patients using ruxolitinib were in those with an underlying diagnosis of myelofibrosis or polycythemia.

Medical Conditions Increased in JAK Inhibitor Users

The authors identified several groups of adverse events for which there appears to be increased frequency of reporting compared to the full database:

“infections and infestations”

musculoskeletal and connective tissue disorders”

“investigations”

“neoplasms benign, malignant and unspecified”

“blood and lymphatic system”

“respiratory, thoracic and mediastinal disorders”

INFECTIONS

The main significant increased reporting of adverse events for viral infections were herpes infections (herpes simplex and zoster) and influenza viral infections. Interestingly, over-reported herpes viral infections were ranked from the highest for baricitinib, then tofacitinib, then ruxolitinib, whereas over-reported influenza viral infections were ranked from the highest for tofacitinib, then baricitinib, then ruxolitinib.

A variety of other types of infections were reported with greater frequency compared to other medications in the data base. For example, the authors found that pneumocystis infections (Pneumocystis jirovecii), cryptococcal infections and coccidioides infections had significantly higher reporting. Other types of infections were increased as well.

NEOPLASMS

There were several types of malignant neoplasms for which adverse event reporting was significantly increased:

“hematopoietic neoplasms (excluding leukaemias and lymphomas)”

“skin neoplasms malignant and unspecified” (IC025 2.4),

“leukaemias”

“soft tissue neoplasms benign”

“respiratory and mediastinal neoplasms malignant”

There were no differences in the overall over reporting of neoplasms according to dose of either baricitinib or tofacitinib

EMBOLISM AND THROMBOSIS

The authors identified an overreporting of blood clots in JAK inhibitor users. The greatest overreporting of embolism and thrombosis was for baricitinib, then ruxolitinib, then tofacitinib.

Interestingly, there were no differences in an over reporting of embolism and thrombosis events associated to the dose of either baricitinib or tofacitinib.

GI PERFORATION

Of the 3 JAK inhibitors, only tofacitinib had a significant increase of adverse event reporting for gastrointestinal type perforation. This included various categories such as “gastrointestinal perforation”, “large intestinal perforation”, “diverticular perforation”, “intestinal perforation” and “gastric perforation”.

CARDIAC EVENTS

No major cardiovascular adverse event were associated with higher reporting for JAK inhibitors. Similarly, no cerebrovascular events were reported with JAK inhibitors. This included no increased reports of myocardial infarction and no increased reports of heart failure.

Conclusions and Comments

This is an important study for us to understand as it gives insight into real world use of JAK inhibitors. Pharmacovigilance studies are valuable in that they help identify signals that could warrant different recommendations in use or even specific limitations in use.

In this pharmacovigilance study, JAK inhibitors were most commonly associated with infectious adverse events, embolism and thrombosis, neoplasms and gastrointestinal perforation events. The authors also identified significant increase in adverse event reporting regarding musculoskeletal and connective tissue disorders. Interestingly, they found no association with major cardiovascular events.

Baracitinib could potentially be associated with a greater risk than tofacitinib for herpes infections and blood clots. All 3 JAK inhibitors were reported to be associated with increased frequency of “skin neoplasms malignant and unspecified” as well as “respiratory and mediastinal neoplasms malignant.” GI perforation risks may be greatest with tofacitinib.

It was interesting that the authors did not find a significant increase in the reporting of major cardiovascular events. The Oral Surveillance study had suggested a possible increased risk of rheumatoid arthritis patients over 50.

Clearly this is an ongoing field of evaluation and we are far from finished monitoring the safety of JAK inhibitors. It’s going to be up to both good clinical trial design, mandated long term follow up studies (extension studies) and real world reporting of side effects to get a good sense of safety of these drugs.

For now, it appears that infections, clots and cancer are very much part of the side effect profile of JAK inhibitors. Which groups these side effects affect most and which groups they affect least (or not at all) will need to be sorted out in future longitudinal studies. Proper patient selection and diligent monitoring will be needed for all patients beginning JAK inhibitors.

There are clearly limitations to mining data with this type of pharmacovigilance data base. Spontaneous reporting cannot be used to estimate prevalence or incidence of adverse events among patients exposed to drugs. Of course, individual case safety reports are not fully reliable regarding causal association because all the potential causes can’t be determined when submissions are accepted to the VIGIBASE database.

REFERENCE

Hoisnard L et al. Adverse events associated with JAK inhibitors in 126,815 reports from the WHO pharmacovigilance database. Sci Rep. 2022 May 3;12(1):7140.