ACR 2024: What Exactly do JAKs and TYKs Do? Save

Let me start by confessing something embarrassing: the immune system is just too complex for me to understand. I suspect the onslaught of new drugs with new mechanisms at ACR 2024 may have left many of you feeling similarly. This may be particularly true for the TYK/JAK/STAT signaling pathway, where each agent has a slightly-different combination of binding affinity to various JAK and TYK signaling dimers. 

In this article I want to share some diagrams from ACR 2024 that may help, though admittedly at the risk of grossly oversimplifying the immune system.

The first diagram comes from abstract 0662, which shared subgroup analyses from the deucravacitinib-in-SLE PAISLEY trial. This diagram paired signaling dimers (ie TYKE2/JAK1) with relevant downstream signaling ligands (ie Type 1 and III IFN for the TYK2/JAK1 pair). Any analysis of this type should come with strong caveats. Many of our drugs are only partially inhibitory, none of these pathways are wholly selective, and the downstream effects on target cells overlap substantially. At the same time, I do think this is an appealing conceptual framework for approaching the question, “What should I expect from this novel molecule?”

Figure 1: ACR Convergence 2024 Abstract 0662 

To make this thought experiment more practical, let’s consider a novel molecule! At ACR 2024 we had a handful of abstracts related to a highly selective TYK2 inhibitor whose name will be new to many rheumatologists: zasocitinib. One of their basic science-y abstracts (0877) reported on the in vitro plasma inhibition of zasocitinib, which was remarkably selective for TYK2 as opposed to JAK 1/3. I find this helpful – we can actually ask what it means to selectively inhibit TYK2 and – by extension – Types I/III interferon and IL12/23. Bringing the presented abstracts together, it looks like has a good effect on skin disease (Abstr 1131) and encouraging (though not earth shattering) improvements in joints (Abstr 2584). Additional studies are planned in inflammatory bowel disease.

Figure 2: ACR Convergence 2024 Abstract 0877

Can we also apply this to safety signals? In the wake of the ORAL SURVEILLANCE study and subsequent black box warning, rheumatologists have been understandably concerned about MACE/VTE/malignancy risk for JAKs. A very nice review that was recently published in ARD (see chart below) reviewed the selectivity of JAKs and is worth a read if you are interested, but my take home is that it’s plausible that more “selective” JAKs will not have this risk. An integrated analysis of the upadacitinib trials and multiple abstracts at ACR 2024 have been encouraging (Abstr 1392, Abstr 2577, among others), but the overlap in between tofacitinib and Upadacitinib feels too great for me to be fully reassured.

What to make of all this? 

Targeting JAK-STAT has been productive for rheumatologists, and I am excited to see the emergence of additional agents with less JAK selectivity and more TYK selectivity. These drugs will affect different combinations of downstream signaling ligands, which may offer opportunities to achieve a better risk and benefit profile for various autoimmune diseases.