COPENHAGEN – Temporary methotrexate in clinically suspected arthralgia delays but does not prevent clinical arthritis development; however, it does lead to sustained reduction of disease burden and MRI-detected inflammation in all at-risk groups, shows the 2-year Treat Earlier study.
“These data provide the first evidence for disease modification when intervening in ‘pre-RA [rheumatoid arthritis]’ or arthralgia,” said Doortje Krijbolder, MD, of Leiden (the Netherlands) University Medical Center, who presented her study at the annual European Congress of Rheumatology.
Dr. Doortje Krijbolder
The randomized, double-blind, study aimed to find out if giving methotrexate in the pre-arthritis phase of arthralgia (with subclinical joint inflammation) prevents the development of clinical arthritis or reduces the burden of disease.
“This is the first trial that aims to reduce the burden of disease, and as such, in pre-RA, it is important that disease modification is sustained after temporary treatment otherwise patients will start regressing with disease activity,” said Dr. Krijbolder.
She explained that methotrexate is usually initiated as first-line treatment when arthritis becomes clinically apparent with joint swelling, but “disease processes begin long before this, and only become clinically recognizable when patients develop symptoms.”
Clinically detectable arthritis development
All 236 patients included in the study had arthralgia of the small joints that, because of the character of the symptoms, was found clinically suspect for progression to RA over time.
“Importantly, these participants had not yet developed clinical arthritis that could be detected on physical joint examination, and were clinically suspected of progressing to RA. They had all undergone an MRI and subclinical joint-inflammation had been detected that was more than prevalent in symptom-free controls,” said Dr. Krijbolder.
Patients were randomized (1:1) to either a single intramuscular glucocorticoid injection (120 mg methylprednisolone [Depo Medrol]) and a 1-year course of oral methotrexate (up to 25 mg/week), or to placebo injection and placebo tablets, and were then followed for a further year without medication to see whether disease progressed.
Treatment and placebo groups were matched with an average age of 46-47 years, 62%-68% were women, they had had symptoms (joint pain) for 27-28 weeks, C-reactive protein was increased in 27%-30%, and 20%-26% were anti-citrullinated protein autoantibody (ACPA) positive.
The primary endpoint was the development of clinically detectable arthritis (fulfilling the 2010 RA-criteria or involving ≥ 2 joints) that persisted for at least 2 weeks. The main secondary endpoints were related to disease burden, including patient-reported physical functioning, along with symptoms and workability (presenteeism at work – the percentage of productivity lost caused by the joint complaints), and measured every 4 months, said Dr. Krijbolder.
She and her colleagues also followed the course of MRI-detected inflammation, which comprised the sum of tenosynovitis, synovitis, and osteitis scored with the RA-MRI Scoring (RAMRIS) method.
Analysis was carried out on an intention-to-treat basis, and two prespecified subgroup analyses were also performed to obtain a better understanding of the effect of methotrexate and glucocorticoids in participants with high risk of clinical arthritis development (positive predictive value (PPV) ≥ 70%), and in patients stratified for ACPA status.
Delays but does not prevent
There was no difference between treatment and placebo groups in the development of persistent clinical arthritis over 2 years (80% vs. 82%, hazard ratio [HR] 0.81; 95% confidence interval [CI], 0.45, 1.48]); however, in the high-risk group subanalysis, 67% in the placebo group developed persistent clinical arthritis. For the treatment group, 7%developed persistent clinical arthritis at 6 months, 27% at 12 months, 40% at 18 months, and 67% at 24 months. For the placebo group: 56% developed persistent clinical arthritis at 6 months, and 67% at 12 months, 18 months, and 24 months.
“Strikingly, in the treatment group there was a statistically significant difference between 6 and 12 months, but in year 2, this difference disappeared, suggesting a delayed arthritis development but no prevention,” said Dr. Krijbolder.
Persistent clinical arthritis became similar at 24 months, at 67% in both groups. A similar, but less pronounced, delaying effect was seen with the ACPA-positive patients, with 48% and 52% developing persistent clinical arthritis at 24 months.
In joint pain, there was a decline in the treatment group compared with the placebo group and this persisted over 2 years (–9 on scale of 0 to 100: (95%CI, –12,–4; P < .001), and a similar effect was seen in high-risk and ACPA subgroups, reported Dr. Krijbolder.
Physical functioning improved more in the treatment-group compared with the placebo group during the first months and remained better (mean between-group difference over 2 years HAQ [health assessment questionnaire] –0.1 [–0.2, –0.03; P = .004]), and morning stiffness (–12 [95%CI, –16, –8; P < .001]), and presenteeism (–8% [95%CI, –13%, –3%; P = .001]) also showed improvement over time compared with placebo.
MRI-detected joint-inflammation was also persistently improved with a mean difference over 2 years –1.4 points (95%CI, –2.0, –0.9; P < .001). “As we know that symptoms and functional impairments are associated with MRI-detected joint inflammation, it was not surprising that treatment induced a decline in joint inflammation that persisted over time, also after the stop of the treatment in the second year,” explained Dr. Krijbolder.
In the high-risk subgroup, as well as in both ACPA-positive and ACPA-negative participants, a comparable statistically significant decline in MRI-detected inflammation was found.
“It may seem counterintuitive that on the one hand we found a delaying effect of treatment on persistent clinical arthritis but no prevention, while on the other, we saw a sustained treatment effect on MRI-detected joint-inflammation and related symptoms and functional impairments,” added Dr. Krijbolder.
To help understand this, the researchers performed a post hoc analysis in high-risk participants, comparing those who did not progress with those who did progress to arthritis. “Both progressors and nonprogressors showed a sustained treatment effect for pain and MRI-detected joint-inflammation, and in nonprogressors, there was almost a complete relief of pain and they nearly returned to the normal range of MRI-detected joint-inflammation as seen in symptom-free controls,” reported Dr. Krijbolder.
“In those who progressed to arthritis, there was less pain and less MRI-detected joint inflammation in the treatment group, but also at the time when they developed clinical arthritis. So both progressors and nonprogressors benefit from treatment,” she noted.
The number of serious adverse events was the same between the groups and adverse events were as expected from methotrexate.
Dr. Krijbolder said, “the results are encouraging and will open up a new treatment landscape in pre-RA at the future, but at the moment it is too early to give recommendations for clinical practice; we definitely do not want to advocate the start of treatment in all at-risk individuals from now on.”
More research should unravel the mechanisms within the joint that contribute to the development of clinical arthritis and disease chronicity, she said. “If we understand these, we may be able to use more targeted interventions in the future and prevent the development of clinical arthritis all together.
“We are also looking forward to learning more about the long-term beneficial effects of this early treatment in our ongoing observational extension of the trial.”
Hendrik Schulze-Koops, MD, of Ludwig-Maximilians-University of Munich, who moderated the session, asked Dr. Krijbolder, “As well as giving methotrexate to the participants in the treatment group, you also gave a single glucocorticoid injection at baseline, and a placebo injection in the placebo group, can you comment on that?”
EULAR
Dr. Hendrik Schulze-Koops
Dr. Krijbolder replied that by the first study visit of 4 months, methotrexate would have already started working, and as such from her data it was not possible to distinguish what effect arose from the Depo Medrol injection, and what was from the methotrexate.
“Maybe there even is a synergetic effect, meaning that the two medications together work even better. To learn more about this would require a novel trial with a study design that would, for example, collect data quickly after the injection, for example after 2-4 weeks, because than the effect of the methotrexate would still be limited, or, of course, a novel trial with a treatment arm that only consists of a glucocorticoid injection,” she added.
No conflicts of interest were declared.