The first Janus kinase (JAK) inhibitor received regulatory approval for the treatment of psoriatic arthritis (PsA) more than 5 years ago. Although there are limited comparative data between this and other JAK inhibitors approved or in development for the treatment of PsA, it is reasonable to anticipate variability in therapeutic effect and the risk of adverse events between different JAK inhibitors. So far, there have been considerable differences in the relative selectivity of each agent on the 4 JAK isoform enzymes, JAK1, JAK2, JAK3, and TYK2. This selectivity determines the downstream signal transducers and activators of transcription proteins (JAK-STAT [signal transducer and activator of transcription] pathway) that ultimately mediate both anti-inflammatory and off-target effects. In this review of JAK inhibitors in PsA, differences between JAK inhibitors will be explored for their potential impact on benefit-to-risk ratio while treating PsA.
Data from the National Psoriasis Foundation (NPF) estimates that 8 million individuals in the United States have psoriasis.1 PsA, an inflammatory spondyloarthritis associated with psoriasis, develops in about 30% of these individuals, but precise epidemiology on this subset of psoriasis patients is complicated by missed and delayed diagnoses. Of patients with psoriasis, only about 15% of patients with PsA have joint inflammation at the time or in advance of skin lesions.2 This might explain delays in diagnosis. In one study, 15% of patients treated for psoriasis were found to have concomitant but unrecognized PsA.3
PsA was first classified as a distinct pathologic condition only about 50 years ago, even though skeletal remains indicate that this disease existed in early civilizations.2 Based on consensus that PsA deserved definition as a distinct entity, the Classification Criteria for Psoriatic Arthritis (CASPAR) were published in 2006.4 By these criteria, cumulative points are allotted for clinical signs of skin, nail, and joint involvement, as well as radiographic signs in patients judged to have inflammatory disease in the joints, spine, or entheses to classify them as having PsA.
There are numerous recommendations for the treatment of PsA, including those issued by the American College of Rheumatology (ACR),5 the European Alliance of Associations for Rheumatology (EULAR),6 and the Group for Research and Assessment of Psoriasis and Psoriatic Arthritis (GRAPPA).7 Although generally compatible with the others, the GRAPPA recommendations, which are the most recent, have addressed the heterogeneity of PsA by recommending therapies for specific disease domains, such as the skin, nail, and joint manifestations.
For treatment of PsA, the available drug classes for moderate-to-severe disease include immunomodulators, such as methotrexate, biologics that inhibit cytokines, such as tumor necrosis factor (TNF) and the interleukin (IL) cytokines IL-17, 1L-23, and IL12/IL-23, phosphodiesterase-4 (PDE4) inhibitors, and JAK inhibitors. In the GRAPPA recommendations, JAK inhibitors are listed along with other targeted therapies as first-line choices for peripheral arthritis, axial disease, enthesitis, dactylitis, and plaque psoriasis.
JAK Inhibitors and PsA
There are multiple ways to classify JAK inhibitors. Tofacitinib, the first JAK inhibitor approved for PsA, is labeled a first-generation agent because it is relatively nonselective for the 4 JAK isoforms.8 Second-generation agents, such as upadacitinib, have been distinguished from tofacitinib, baricitinib, and other first-generation drugs by greater relative selectivity on the JAK1 enzyme. Other drugs in development for PsA target different JAK isoforms. Deucravacitinib, for example, which was approved for psoriasis after a favorable phase 3 trial9 and has shown promise for PsA in a phase 2 trial, is selective for the TYK2 isoform.10 A rapidly growing list of JAK inhibitors with different selectivity profiles, including dual JAK inhibitory effects, are being explored in a host of inflammatory diseases.
The relationship between selectivity on specific JAK isoforms, anti-inflammatory effects, and off-target effects is not fully understood.8 In addition, characteristics beyond JAK selectivity have potential pharmacologic importance. For example, JAK inhibitors can be classified as ATP competitive inhibitors and allosteric inhibitors, both of which are reversible binding modes.8 Within each of these subcategories, the site of kinase binding has the potential to influence clinical activity.8
JAK Inhibitors: Clinical Experience in PsA
Tofacitinib, a first-generation JAK inhibitor, initially licensed for use in the treatment of rheumatoid arthritis (RA), received regulatory approval for PsA on the basis of the OPAL Beyond trial.11 Approval of upadacitinib for PsA followed about 4 years later on the basis of the SELECT PsA-1 trial.12 The primary endpoint in both of these studies was proportion of patients with an ACR response, signifying degree of improvement from baseline, of ≥20%. For the JAK inhibitors, the ACR20 rates were about 50% and 70% in the tofacitinib and upadacitinib phase 3 trials, respectively. Other JAK inhibitors have been evaluated in PsA but none so far are approved in the United States.
Despite experimental evidence supporting the hypothesis that JAK1 selectivity is clinically relevant to the treatment of PsA and other spondyloarthritides,13 there is no level 1 evidence of an efficacy or safety advantage for second- relative to first-generation JAK inhibitors. A small number of indirect comparisons, such as one employing a network Bayesian analysis to compare these drugs for the treatment of RA,14 have supported a clinical advantage for JAK1 selectivity, but head-to-head comparisons are needed to confirm differences.
Prescribing information for both tofacitinib and upadacitinib in PsA and other indications include a black box warning for risk of serious adverse events, including major adverse cardiac events (MACE) and thromboembolism. The warning is based on the placebo-controlled ORAL trial with tofacitinib in RA.15 The study population was enhanced for risk with eligibility that required older age and the presence of cardiovascular risk factors. In this high-risk RA population, tofacitinib was associated with modest increases in serious adverse events, including MACE and thromboembolism, relative to placebo over several years of follow-up. A similar trial has not been conducted with upadacitinib or in patients with PsA.
In a phase 3 trial with the TYK2-selective deucravacitinib in psoriasis, there was no increase in the rate of MACE or thromboembolism.9 When granted regulatory approval for psoriasis, the product information did not include a black box warning, differentiating it from other currently available JAK inhibitors. It has not yet been proven whether the absence of serious adverse events in the phase 3 psoriasis and phase 2 PsA trials with deucravacitinib are related to TYK2 JAK enzyme selectivity.
Although TYK2 is closely associated with upregulation of IL-23 and other inflammatory cytokines implicated in the pathophysiology of PSA, the JAK-STAT signaling pathway is incompletely understood.8 Moreover, all of the JAK inhibitors synthesized so far have relative rather than absolute selectivity for any specific JAK isoform. This complicates the ability to attribute benefits and risks to the inhibition of any single JAK enzyme isoform and amplifies the need for comparative studies.
While other JAK inhibitors have reached late stages of development for the treatment of PsA, such as filgotinib (a JAK1 selective drug) and brepocitinib (which is selective for both JAK1 and TYK2),16,17 it is appropriate to emphasize that currently available JAK inhibitors are effective and acceptably safe for PsA. The goal of continued drug development is the potential to develop agents with even greater efficacy but with a lower risk of off-target effects. Currently, the black box warnings included in the labeling of tofacitinib and upadacitinib give pause, leading many clinicians to move to these agents after an inadequate response to biologics. Newer therapies in the JAK inhibitor class free of serious adverse effects might reverse the order, given the preference of many patients for oral agents.
The JAK inhibitor development program is rich not just for inflammatory diseases and autoimmune diseases, but for myeloproliferative diseases and neoplasms. JAK inhibitors are already identified in the GRAPPA recommendations as appropriate first-line options for most manifestations of PsA, including joint and skin involvement, but newer drugs with a more favorable JAK selectivity or other pharmacologic characteristics and decreased adverse risks might make these a more dominant treatment choice.
Relative selectivity for JAK isoforms promises therapies that are both more effective and safer for PsA as well as other inflammatory diseases. This promise is now being explored in experimental trials testing therapies with variable degrees of selectivity in the context of other characteristics, such as kinase binding, with the potential to influence clinical effects. However, the promise will not be fulfilled until large clinical trials, particularly comparative trials, can confirm the importance of JAK isoform selectivity. If specific types of selectivity prove relevant to the benefit-to-risk ratio of JAK inhibitors in PsA, it may alter the current order of treatment preferences for this disease.