It’s always really hard on the last day of ACR, because it’s a half day and people leave before the end, which is a shame, because it’s really good stuff.
And it’s when one of my favourite sessions – the late-breaking abstracts – always takes place.
There were some particularly good ones this year, one with an Australian focus, which is always good to see. Always makes me a little bit proud. David Hunter presented on liposomal delivery dexamethasone intraarticular TLC599. This is a formulation of dexamethasone. It’s not the same kind of rapid release that we’re used to giving, it’s a controlled release.
The idea is that you might have a sustained delivery over the course of time.
In this phase 3 trial, the liposomal dexamethasone did quite well. It went a little bit further in terms of reduction in pain compared to just normal dexamethasone. And there was a difference between the investigational product and the placebo, which was clinically important, and that’s great.
I guess the question is, does this have a further benefit once you get the second dose? And what we saw was really even further benefit.
Even though the gains were not boundless, and it’s not disease modifying, what we do see is that we can potentially be using it in practice for knee osteoarthritis and it could be making a massive difference.
Another good thing from the late-breaking abstracts session was relating to IgG4-related disease. We hear a lot about what to do in the initial period, but not maintenance. We’ve now got a whole lot of patients who have had stable disease, and we want to know whether we can wean down their meds or not.
Quite a large Chinese study looked at withdrawing both steroids and any immunomodulator, something like mycophenolate or azathioprine mainly, versus withdrawing just the steroid and keeping the azathioprine, versus keeping them keeping some of both on board.
What we saw was that patients who stopped all of their therapy did really badly. They had a lot more flares. And that’s given me pause to consider stopping meds in these patients, because it’s hard to monitor when there’s a recurrence.
Some of the great stuff was really late in the day. There was quite a bit about precision medicine and trying to understand the causes of things like rheumatoid arthritis. One of the things that caught my eye was about faecal microbiota and trying to understand whether that could help predict patients at risk of developing rheumatoid arthritis. It’s really beautiful work (abstract 2584).
They could not see any difference in the microbiota between patients who developed full-blown rheumatoid arthritis versus those just with genetic risk, versus those who developed further auto antibodies.
And so it really tells us that it’s not just the gut that leads to people getting rheumatoid arthritis, as some people have thought in the past. Maybe it’s a small contributing factor, but it certainly isn’t the whole solution. As much as we can try and change things, like diet, potentially trying to improve our microbiota, that’s not going to be the only thing, although it may well help.
At the end of the day was really one of the most exciting parts of the whole meeting, about plasma cell-free DNA signatures and using those to separate rheumatoid arthritis from other diagnoses like osteoarthritis and inflammatory arthritis.
We don’t have a singular blood test that can diagnose rheumatoid arthritis right now. And I think it’s really frustrating for a lot of rheumatologists, when they get referrals, with people saying to them, “Well, I’ve got a positive anti-CCP, I therefore must have rheumatoid arthritis”.
In abstract 2586, Peter Taylor talked about being able to use essentially blood tests to differentiate between rheumatoid arthritis versus other diagnoses with an overall sensitivity of 90% and specificity of 96%.
But what’s really interesting is that it works for seronegative rheumatoid arthritis as well. With seronegative arthritis, there was a sensitivity of 83% and a specificity of 95%. And it was able to differentiate well against other inflammatory arthritis as well, which is really pleasing.
So I think this is potentially massive, not just in terms of diagnosis, but potentially this technology could be used with machine learning to try and identify different phenotypes of rheumatoid arthritis that we struggle to pick up now, and do it in a really scientific way, without biases, and then potentially predict what medicine a patient might respond to.
There have been studies looking at synovial tissue and using that to predict what biologic agent a patient might respond to. What they’ve done here, instead of doing a synovial biopsy, you might just be able to do a blood test, which means patients more quickly get onto a biologic that works, which means better outcomes for rheumatoid arthritis patients.
It’s all really exciting!