MBDA scores unreliable for assessing RA disease activity after corticotropin injection
Click Here to Manage Email Alerts
Overall multibiomarker disease activity scores are insufficient for determining rheumatoid arthritis disease activity following repository corticotropin injection, according to data published in Arthritis Care & Research.
“Disease control was more accurately assessed with using the Clinical Disease Activity Index (CDAI) or the Disease Activity Score 28–erythrocyte sedimentation rate (DAS28‐ESR),” Roy Fleischmann, MD, MACR, co-medical director of the Metroplex Clinical Research Center, and clinical professor of medicine at the University of Texas Southwestern Medical Center, in Dallas, told Healio Rheumatology. “Patients may be in low disease activity or remission by these validated metrics, and this may not be apparent if one uses the multibiomarker disease activity scores (MBDA).”
Data derived from Fleischmann R, et al. Arthritis Care Res. 2021;doi:10.1002/acr.24583.
To compare the MBDA score’s ability to assess RA disease activity after repository corticotropin injection (RCI; Acthar Gel, Mallinckrodt Pharmaceuticals) to that of the DAS28-ESR and the CDAI, Fleischmann and colleagues conducted a multicenter, randomized, placebo‐controlled trial. Participants were adults with persistent, active RA with a DAS28-ESR of more than 3.2 despite disease-modifying antirheumatic drug and glucocorticoid use. In addition, all participants were required to be receiving glucocorticoids for at least 12 weeks and a stable dose of prednisone or equivalent for 4 weeks prior to screening.
Patients must have also received methotrexate and one biologic, targeted synthetic or conventional synthetic DMARD, or treatment with one biologic DMARD for at least 12 weeks before screening.
During the study’s open-label period, 259 participants received 80 U of RCI twice weekly for 12 weeks. Those who achieved low disease activity — defined as a DAS28-ESR of less than 3.2 — at week 12 then moved on to the double-blind period, where they were randomized to receive either 80 U of RCI or placebo twice weekly for another 12 weeks. In all, 154 participants were randomized 1:1 to the two groups, with the aim of assessing low disease activity maintenance. Low disease activity was assessed at baseline, week 12 and week 24 using the DAS28-ESR, CDAI and MBDA.
The researchers then analyzed the changes in disease activity, as well as the correlations between MBDA scores and both DAS28‐ESR and CDAI scores.
According to the researchers, changes in DAS28‐ESR and CDAI scores from baseline suggested that RCI therapy led to clinically meaningful improvements in disease activity. However, improvements in MBDA scores from baseline were below the minimally important difference threshold. For the DAS28‐ESR and CDAI scores, correlations with total MBDA and individual component scores were generally low (r 0.3), and occasionally moderate (r between 0.3 and 0.5).
“Importantly, although the MBDA did correlate with the DAS28-ESR and CDAI, the correlation was very low indicating that they do not correlate well,” Fleischmann said. “In addition, although there was a statistically significant change in the MBDA, the mean change was less than the minimally clinically important difference strongly suggesting that the statistical significance is not reflective of the clinical significance.”
According to Fleischmann, these results are similar to what has been previously reported with other molecules, including abatacept (Orencia, Bristol Myers Squibb), adalimumab (Humira, AbbVie), baricitinib (Olumiant, Eli Lilly & Co.) and tocilizumab (Actemra, Genentech) in prospective studies.
“The MBDA does not correlate well with the validated disease metrics suggested for clinical use by the American College of Rheumatology,” he said. “One might assume these results would be similar with other molecules with a similar mechanism of action, such as other TNF inhibitors, IL-6 inhibitors and JAK inhibitors, but one cannot know for sure unless these specific molecules were studied in a similar manner.”
“Until then, I would be cautious about relying on the MBDA to assess clinical disease control with these other molecules,” Fleischmann added. “The treat-to-target approach for the treatment of RA is well established. The treatment target should include a metric which examines joints and not the MBDA. Hopefully, a biomarker — or biomarkers — will be developed that can accurately reflect disease activity with high correlation with our validated metrics.”
Perspective
Back to Top
Over the past several years, the treat-to-target paradigm has emerged as a best practice approach in rheumatoid arthritis disease management, and with good reason. Low-disease activity correlates with less radiographic destruction, preservation of functional status and improved clinical outcomes. Various validated measuring instruments have emerged to objectively assess disease activity including the DAS-28, the CDAI, and the MBDA score, among others.
In their study, Fleicshman and colleagues demonstrate a discrepancy between a biochemical marker of disease activity (the MBDA) and more traditional disease activity measures (the DAS-28-ESR and the CDAI) in patients with persistently active rheumatoid arthritis treated with repository corticotropin injection (RCI) or a placebo control.
This study challenges us to question the accuracy of our measuring devices. Do they always behave competently in a broad range of clinical circumstances in a disease state as heterogenous as RA? Or are our yardsticks more like high-performance automobile tires that perform admirably on dry pavement but fail miserably when the weather turns inclement? Might there be a subset of RA patients where one validated measure of disease activity is superior to another?
Because the biology of RA is inconsistent and changes over time, is it possible a measuring device that is accurate in early disease, may become less precise in subjects with more established disease? Are all measures equally as predictive in seronegative disease as they are in seropositive disease? Does age make a difference? Gender? Disease activity at baseline? Duration of disease? Can treatments skew scores? If so, which treatments affect what scores?
If NSAIDs and analgesics influence tender joint counts and patient global assessments, but are not disease modifying, might these interventions alter outcome measures that include a patient global assessment and/or a tender joint count as a component of measure and introduce a false sense of security to the rheumatologist?
This study serves as a reminder that discordance exists among all disease activity instruments, to a greater or lesser degree, when they are compared to each other. It reminds us that we still haven’t found the perfect measuring instrument that is highly precise over a broad range of clinical circumstances. Fleischmann and colleagues conclude that the MBDA scores are not sufficiently responsive for use in assessing rheumatoid arthritis disease activity after treatment with RCI. That is one possible conclusion.
Another possible conclusion is that the MBDA is telling us something in this population that the DAS-28-ESR and the CDAI are not, specifically that RCI improves a rheumatoid arthritis patient’s sense of well-being, but is not disease modifying. We’ll never know for sure from this analysis because radiographic scores were not a captured outcome in this study.
Perspective
Back to Top
David A. McLain, MD, MACR, FACP, FRCP
According to the ACR Disease Activity (DA) recommendations, the MBDA score is among 11 DA measures that can be used in routine clinical practice; however, MBDA is not designated as a preferred measure.
I personally have used the MBDA score to judge disease activity in patients who appear to be fairly well-controlled with their present regimen, but who have a lot of pain. Sometimes, this pain is due to coexistent fibromyalgia or osteoarthritis. If the MBDA is low and there is low risk of radiographic progression, I continue with the same treatment and address the pain component.
On the other hand, with patients who have active rheumatoid arthritis who are hesitant to take biologics or JAKs (or even DMARDs), I have used the MBDA as a “sales piece”. I can show them in black and white that their disease is out of control and they have a high risk for radiographic progression.
The use of repository corticotropin injection (RCI) to treat rheumatoid arthritis is unusual, as the authors note. I may have used it once. Throughout this study, every finding of a difference between the MBDA and CDAI or DAS28-ESR was interpreted as evidence that the MBDA was inadequate.
The authors of the paper agree that MBDA is a more objective measure than CDAI or DAS28-ESR. Interestingly, and not addressed by the authors, the proportional improvements were not only much greater for CDAI than DAS28, but the CDAI reductions were large: 69% from baseline (BL) to week 12 for all 259 patients and ~85% from BL to week 12 among those with DAS28-ESR <3.2 at week 12. In addition, they were proportionately larger than the DAS28-ESR changes (44% and 56.5%, respectively).
A proportionately greater response with CDAI, compared with DAS28-ESR, might thus reflect: 1) a disproportionately large improvement in MD global and/or 2) a disproportionately small improvement in ESR, as might occur in patients with relatively low levels of inflammation. This interpretation is speculative but appears to be consistent with the data in the publication, exclusive of supplementary materials.
Either way, this disparity could be a sign that placebo effect was unusually large and/or baseline global assessments were exaggerated, as sometimes occurs. With no placebo group from baseline to week 12, we don’t know how much of the change seen with subjective measures was real but it is possible that the objective MBDA score is actually closer to actuality.
The authors wrote “…DAS28-ESR and CDAI include subjective assessment of RA disease activity. Compared with a more objective measure like MBDA, the DAS28-ESR and CDAI conceivably could be more likely to show improvements during an open-label treatment period when patients are aware of the additional treatment they are receiving”. This statement may, in fact, be a carefully worded reference to a problem in the study that amplified discordance with the MBDA score.
Collapse
Click Here to Manage Email Alerts