WASHINGTON – What variables predict the course of an individual patient’s rheumatoid arthritis? It’s a question two research groups are beginning to answer by mining data from a large patient database.
Using data from CORRONA (the Consortium of Rheumatology Researchers of North America), a multicenter, longitudinal, prospective database in the United States with more than 30,000 patients, George W. Reed, Ph.D., of the University of Massachusetts in Worcester, and his colleagues have constructed a Markov model as a framework for analyzing the probabilities of transition between disease states in RA.
It’s a way to "talk about the probability of what’s next [in the course of a patient’s disease], and what’s associated with what’s next," Dr. Reed reported in a poster presentation at the annual meeting of the American College of Rheumatology.
Dr. Reed and his colleagues began by defining disease activity states: Low disease activity was defined as a clinical disease activity index (CDAI) score of 10 or less, moderate disease activity as a CDAI score between 11 and 22, and severe disease activity as a CDAI score of 23 or greater. Then states were determined at each visit, and variables were measured to see if they correlated with any changes. Covariates included initiation of disease-modifying antirheumatic drugs (DMARDs), duration of RA, patient age, and insurance status.
The covariates are just examples of "some possible variables that could be plugged in" to future models, he said.
Overall, the investigators examined 160,262 visits from 24,136 RA patients in the CORRONA database who had CDAI measures at the current and the prior visit.
The data imply that RA management, in general, is improving. Transition from any disease state to a low disease state improved from 2001-2005 to 2009-2012, said Dr. Reed.
If a DMARD was not initiated at the prior visit, "a patient in a prior state of moderate disease had a relative risk ratio of 7.6 (95% confidence interval, 7.08-8.21) to still have moderate disease at the current visit." However, if a DMARD was initiated, CDAI declined, and the relative risk of remaining in a moderate disease state was cut to 4.1 (95% CI, 3.59-4.63).
A longer duration of RA affected transitions in disease states, but patient age and insurance status did not.
In a separate CORRONA-based study, patients with moderate disease were found to transition frequently in and out of severe and low disease states.
In that study, led by Sameer V. Kotak of Pfizer in New York, 60% of 4,118 RA patients who exhibited moderate disease activity at baseline had transitioned to a low CDAI by 6 months. About 31% continued to have a moderate CDAI, and 9% transitioned to a severe CDAI.
At 12 months, about 75% of the patients with a low CDAI at the 6-month benchmark remained in that category, with 20% transitioning back to a moderate CDAI and the remaining 5% transitioning to severe disease.
Similarly, 46% of patients with a moderate CDAI at baseline remained at that level at 6 months and transitioned to a low CDAI by 12 months, while 11% of these patients transitioned to a severe CDAI.
The data show transition potential and disease instability in an understated population, "even within a short follow-up window of 6 months," Mr. Kotak said.
Dr. Reed disclosed that he is supported by a research contract with CORRONA through the University of Massachusetts. Mr. Kotak, along with another coinvestigator, is an employee of Pfizer. Employment with, and other financial relationships to, additional pharmaceutical companies were disclosed by multiple investigators in the studies.