Dr Ramiro reviews her presentation at ACR Convergence on the SPACE cohort study of the difference in structural lesions of the spine as assessed on radiographs and on MRI between patients with early axial SPA and patients with nonaxial SpA chronic back pain. 

Sofia Ramiro, MD, PhD, is a senior researcher at the Leiden University Medical Center, Leiden, the Netherlands, and a rheumatology consultant at the Zuyderland Medical Center, Heerlen, the Netherlands.

• This prospective 2-year analysis from the SPACE cohort included patients aged 16-45 years with chronic back pain (<2 years) who underwent radiography (mSASSS) and MRI (modified Canada-Denmark method) scanning.
• The cohort included 318 patients with spinal radiographs and 351 with MRI. They were grouped by definite or most likely axial SpA (axSpA) vs nonaxial SpA.
• Radiographic findings showed minimal progression in mSASSS (Δ≈0.01/2 yrs) and no significant group differences in number of syndesmophytes.
• MRI findings showed erosions, bone spurs, and ankylosis were stable over 2 years in both groups. Fat lesions significantly increased in axSpA patients (+0.5 units/2 yrs; baseline 1.1→1.6), with no increase seen in nonaxSpA patients. Fewer fat lesions were seen in women vs men.
• Minimal progression in spinal lesions show on radiographs in both axSpA and nonaxSpA chronic back pain. Radiographs of the spine do not detect early structural progression. Fat lesions may offer additional insights into disease evolution as change over time only in axSpA.

Hello everyone. My name is Sofia Ramiro. I'm a rheumatologist and researcher from Leiden University Medical Center in Leiden, Netherlands. And today I'll tell you about our study on differences in structural lesions of the spine between patients with early axial spondyloarthritis and nonaxial spondyloarthritis, chronic back pain—2-year results of the spondyloarthritis SPACE cohort. Structural damage is a very important domain for our patients and they express a lot of concerns about damage developing in them. And damage progression has been assessed in patients, but mainly with established axial SpA. We know less about patients about damage progression in the early phases of the disease, especially comparing it with patients who have nonaxial SpA chronic back pain.

And that's exactly what we have done. We have investigated the difference in structural lesions of the spine as assessed on radiographs and on MRI between patients with early axial SPA and patients with nonaxial SpA chronic back pain.

And for this we use the baseline and the 2-year follow-up of the SPACE cohorts. What's the SPACE cohorts? It's a cohort that started in Leiden and in which patients are included if they have the chronic back pain of a unknown origin for at least 3 months and less than 2 years. Patients are between 16 and 45 years of age. And we have expanded it so that several centers, as you can see them on the map now, participate in this SPACE cohorts.

After 2 years, patients are categorized into 6 diagnostic groups: definite, most likely possible SpA and nonSpA. For this specific analysis, we have only used the definite or most likely diagnosis, so we excluded a few patients that after 2 years only had a possible diagnosis and we have only used patients that had baseline and 2-year image of the spine, so either radiographs or MRIs.

Imaging was performed at baseline at 1 year and 2 years; at 1 year only MRIs and on the other time points MRI and radiographs. So we have radiographs of the spine that was scored according to the mSASSS— modified Stoke Ankylosing Spondylitis Spinal Score—by 3 readers and MRI assessed by 2 of the same three readers and we scored according to the modified Canada-Denmark method. In this here you see in detail what we have assessed on radiographs and MRIs. And I will show you, as I show you the results. We have compared groups, compared patients with axSpA and nonaxSpA and also compared whether there was a change between baseline and 2 years. And we have looked with statistical models, so we generalized estimating equations whether there was progression over time in these patients in terms of spinal damage. And this was adjusted for age at baseline, sex, and the use of NSAIDs.

And then we wanted to see whether the progression was different in patients with axSpA versus nonaxSpA. In the space cohort, 702 patients were included and as I mentioned before, we excluded patients who only had a possible diagnosis and then we had 434 visits with baseline and 2-year available of which having images we have at both time points, 318 patients could be included with radiographs of the spine, 351 with MRIs.

And here are some descriptives of those patients. Two-thirds of the patients had axial spondyloarthritis and here are the first results based on radiographs of the spine. And the way we look at this graph is that on the left we have the mSASSS, on the right we have the number of syndesmophytes. In blue we have axial SPA. In yellow we have nonaxial SPA. In light blue we have baseline. In dark blue we have 2 years. And also light yellow is a baseline and dark yellow it's 2 years. So as we can see there was very little damage of the spine after 2 years and actually comparable between axial SpA and nonaxial SpA and there was a minor increase in both groups. So no statistically significant difference between the groups in terms of what we can see on radiographs, both based on the mSASSS or only the number of syndesmophytes. As we can see on the right hand side, if we look at evolution over time, so the change over these 2 years adjusted for the variables that you can see in this slide. In this model we see that there was a progression of 0.01 ans per 2 years. In other words, no progression at all in these patients and no difference between axSpA and non axSpA.

Here we look at MRIs and first we start with the status scores. So the MRIs at baseline and at 2 years we see some bars that are very low, namely for erosions, bone spurs and ankylosis. There we see a very low frequency of these lesions in both in patients with axSpA and without axSpA. A small increase in the bone spurs between baseline and 2 years in patients with axSpA but very minor. And for the rest, no differences between groups.

We see a different situation for fat lesions where we see 1.1 fat lesions at baseline that's increased to 1.6 fat lesions in axspA and in non axspA there is a lower level and no difference between the two years. Then we see the total structural lesions, which is a graft that mirrors the one from fat lesions because the other remaining lesions were hardly prevalent.  So it's almost the same as the fat lesions.

Here we are looking at the change over the 2 years and we see a similar story. So we see that there is no change or hardly any change in erosions, in bone spurs, or ankylosis, but there is some change in fat lesions. So especially in axSpA 0.5 units in 2 years compared to absolutely no change in nonaxSpA. And again the same situation is mirrored to the total of structural lesions. When we look at progression of these lesions over time, for most of them there's hardly any progression at all. That's the same for erosions, bone spurs, and ankylosis for fat lesions. We see some changes as I showed you in the previous slide, and they are different for axial spondyloarthritis and nonaxial SPA. In axial SpA, there's a progression of 0.16 units per year and we see that females have less fat lesions compared to male.  And on the right hand side we see that in nonaxial SPA there's no change in fat lesions over time.

So in conclusion, there's a minimal progression in spinal lesions on radiographs in both axSpA and nonaxSpA chronic back pain. MRI revealed a significant increase over 2years in the number of fat lesions in the axSpA group, but not in the non axSpA group. And although radiographs of the spine don't detect early structural progression, fat lesions may offer additional insights into disease evolution as they are already prevalent and change over 2 years only in axSpA and not in nonaxSpA.

So this is work done in the SPACE cohort and we started looking at spinal structural lesions. We are now continuing to do the same in SIJ, sacroiliac joint lesions to see whether there is different change between axSpA and non xSpA. And we are continuing following up the cohort with longer follow up where we will want to see how and we will analyze how these processes evolve and whether, what are the factors that lead to the starting and the continuing evolution of structural damage in axial SPA. So this was in a nutshell, the explanation of the project.