Background/Purpose:

In an effort to provide a comprehensive listing of clinical trials Congress initiated the creation of the ClinicalTrials.gov (CT.gov) registry in 1997. In 2007 the FDA Amendment Act mandated registration of most non-phase I interventional drug, biologic, and device trials. Until recently, there has been no systematic analysis of the clinical trial enterprise, either broadly or for rheumatologic diseases. As part of the Clinical Trials Transformation Initiative we analyzed the CT.gov database to describe the current state of clinical trials in rheumatology and compare these findings to other specialties.

Methods:

A dataset of 96,346 studies was downloaded from CT.gov on 9/27/2010 and was restricted to 40,970 interventional studies registered between 10/2007 and 9/2010. Clinical specialists annotated medical subject heading terms and common disease terms for relevance to rheumatology. An initial dataset identified studies with ≥ 1 disease relevant term. Manual review of individual studies, and additional keyword searching yielded the final rheumatology study dataset (R).  Studies were further divided into disease and sponsorship subcategories and comparisons were made between these subcategories, and with non-rheumatology (NR) studies.  

Results:

1,622 rheumatology trials were identified, representing 4% of the CT.gov dataset. Disease groups were classified as: osteoarthritis (24.4%), rheumatoid arthritis (19.4%), lupus/connective tissue disease (8.6%), spondyloarthritis (4.1%), scleroderma (2.3%), gout (2.2%), vasculitis (0.7%), and all others (38.2%). Median (quartiles) enrollment for R trials was 80 (35, 200) compared to (64 (30,180)) for NR studies. The primary purpose was treatment in the majority of studies (86.8%). Prevention (5.0%) and diagnostic (2.3%) were less represented than in NR studies (11.1% and 4.0% respectively). 59.4% were industry-sponsored compared to 45.4% of NR studies. Biologics were used in 23.2% of R studies. There was double-blind masking in 49.0% and randomization in 75.2%, versus 32.4% and 68.6% for NR studies, respectively. 33.1% reported using a data monitoring committee (DMC) which was less than that among NR studies (40.9%). Industry sponsored studies were less likely to utilize a DMC compared to non-industry (26.1% v. 39.1%) and more likely to utilize a biologic and focus on treatment (compared to prevention and diagnosis). Only 3.6% of the rheumatology studies had posted results to the CT.gov website by date of download, compared to 2.2% among NR studies.

Conclusion:

The rheumatology clinical trial enterprise represents a small number of trials in the CT.gov database between 10/2007 and 9/2010. Rheumatologic diseases with high prevalence and substantial unmet need are under represented relative to less common rheumatologic diseases. Half of clinical trials had enrollment of 80 participants or less. Double-blind masking and use of a DMC were employed in a minority of studies and only a small fraction of rheumatology studies have reported results. Our results indicate that there is an opportunity to improve the quality of clinical trials in rheumatology and initiate a policy discussion on whether resources are optimally focused on the highest priorities.

ACR Meeting Abstracts - https://acrabstracts.org/abstract/status-of-the-rheumatology-clinical-trials-portfolio-data-from-clinicaltrials-gov/