Background/Purpose:

While many clinical trials provide direct comparisons between biologic disease modifying antirheumatic drugs (bDMARD) and nonbiologic DMARD (nDMARD), there is a need for additional evidence on the effectiveness of these therapies in routine clinical practice. We evaluated changes in disease activity measures associated with bDMARD therapy, in a large cohort of patients with RA, under conditions of routine clinical practice.

Methods:

The OM1 platform collects, links, and leverages, structured and unstructured data from electronic medical records (EMR) and other sources in an ongoing and continuously updating manner. The OM1 RA Cohort includes data on >75,000 patients treated by rheumatologists. This analysis included patients who were treated with nDMARD between January 2013 and April 2017, had not received prior treatment with bDMARD, and either added or switched to another nDMARD or initiated bDMARD during the observation period (date of change in therapy is the index date). Established American College of Rheumatology cutpoints for standard disease activity measures (RAPID-3, CDAI, DAS28) were used to define remission. Advanced natural language processing was used to impute missing disease activity categories. Drug eras were defined using Observational Medical Outcomes Partnership (OMOP) definitions. Survival analyses were conducted to evaluate time to initial remission and confirmed remission defined as 2 consecutive scores denoting remission. To reduce the impact of subsequent treatment changes, data were censored at 12 months. Patients who switched or added nDMARD but subsequently initiated bDMARDs within 6 months after the index date were excluded in sensitivity analyses. To reduce the bias that more frequent disease activity measures may be associated with shorter time to remission, we matched the two groups on average number of disease activity measures per patient.

Results: The analysis cohort included 4,957 patients who met study inclusion criteria, none of whom were in remission at index date; 1,334 added or switched to another nDMARD and 3,623 added or switched to a bDMARD. There were an average of 4.2 disease activity measures per patient and a total of 20,605 disease activity measures during the 12 month study period. A larger proportion of patients in the bDMARD group achieved initial remission (18% versus 16%, p<0.05) and confirmed remission (13% versus 11%, p<0.05) compared to the nDMARD group. Time to remission was significantly shorter in the bDMARD group (mean±SD=5.2±3.4 months) compared to the nDMARD group (5.7±3.2 months, p<0.05). These results were unchanged in the sensitivity analysis.

Conclusion: Disease activity improved with changes in DMARD therapy; however, the addition of bDMARDs were associated with significantly shorter time to remission. This study uses novel data collection techniques to replicate findings from prior observational studies in a much larger and contemporaneous cohort of patients under conditions of routine clinical practice.

ACR Meeting Abstracts - https://acrabstracts.org/abstract/initiation-of-biologic-disease-modifying-antirheumatic-drug-therapy-and-associated-changes-in-disease-activity-measures-in-routine-clinical-practice-findings-from-a-large-contemporaneous-real-world-c/