Might Misdiagnosis of Multiple Sclerosis Compromise Outcomes of Clinical Trials?

Background: Multiple sclerosis (MS) is a neurological disorder characterized by inflammatory demyelination and abnormal neurological function. MRI is the image technique of choice, and provides an essential component of modern diagnosis of MS (the McDonald criteria). These guidelines have been periodically revised.

Objectives:  This study was designed to categorize and quantify lesion patterns in patients enrolled into a phase III clinical drug trial for MS. We suspected that patients with atypical lesion patterns upon entry into the trial might have different on-study outcomes than those with more typical MRI findings. 

Methods: MRI scans from 1008 randomized patients were reviewed and processed in the UT MRI Analysis Center. Lesions were categorized and counted, and any “special features” were noted. “Typical” vs. “atypical” lesion patterns were identified. Scans were categorized by 2005 and 2010 McDonald criteria. Multiple on-study MRI and behavior comparisons were considered, including binary gadolinium (Gd) enhancement, burden of disease (BOD), clinical activity-free status (CAFS), combined unique activity (CUA), disease activity-free status (DAFS), protocol-defined exacerbations (PDE), and clinical progression (PROG). 

Results: Significant differences were found when comparing 2005 and 2010 McDonald criteria patients with binary Gd enhancement, BOD, and CUA. 2005 criteria patients also had a significant difference with DAFS. Significant differences appeared when comparing typical vs. atypical pattern presentation with 2005 and 2010 McDonald criteria, binary Gd enhancement, BOD, CUA, and DAFS. Differences were not found when PDE and PROG were considered.

Conclusions: Having Gd enhancements and a higher BOD at baseline corresponded with meeting both 2005 and 2010 McDonald criteria more often. Thus, more stringent entry criteria may assure more on-study events, a finding consistent with analysis of an earlier independent trial¹. This was also evident when evaluating comparisons between the 2005 and 2010 McDonald criteria. However, having more MRI features at study entry did not correspond with differences in on-study clinical events. In addition, having more MRI features considered as typical of MS at study entry corresponded with more on-study activity in the face of partially effective treatments. It did not correspond with differences in on-study clinically defined events. The presence of typical MRI lesion patterns met both 2005 and 2010 McDonald criteria more often, had Gd enhancements, and a higher BOD at baseline. These results suggest that although patients with atypical lesion patterns sometimes showed baseline or on-study MRI activity, they did not show any difference in clinical disease progression.

References: ¹Barkhof F, et al. (2003) Validation of diagnostic MRI criteria for multiple sclerosis and response to interferon beta-1a. Ann Neurol 53:718.