PARIS—When faced with patients with clinically isolated syndrome (CIS) or primary progressive multiple sclerosis (PPMS), diagnosis can be challenging. The 2010 McDonald criteria and the 2016 Magnetic Resonance Imaging in Multiple Sclerosis (MAGNIMS) criteria offer sound guidance for relapsing-remitting multiple sclerosis (MS), but for special patient populations, these criteria may fall short. At the Seventh Joint ECTRIMS–ACTRIMS Meeting, two studies looked at the suitability of diagnostic criteria for special MS populations.
Clinically Isolated Syndrome
The recently proposed MAGNIMS dissemination in space criteria include lesions in the optic nerve, cortex, and symptomatic region, in addition to an increase in the required number of periventricular lesions from one to three. Raquel Lamas Pérez, MD, and colleagues aimed to compare the diagnostic performance of the 2010 McDonald and 2016 MAGNIMS MRI criteria for dissemination in space in predicting the conversion to clinically definite MS in patients with CIS. Dr. Lamas Pérez is affiliated with the Hospital Universitario Virgen del Rocío in Sevilla, Spain.
Study inclusion criteria included CIS suggestive of CNS demyelination (since 2008), clinical assessment and baseline brain MRI within six months of CIS onset, availability of spinal cord MRI if patients presented with spinal cord syndrome, and clinical follow-up of at least 24 months.
The researchers included 161 patients with CIS (113 women) with a mean age at onset of 34. After a mean follow-up of 58 months, 102 (63.4%) patients had a diagnosis of MS according to the 2010 McDonald criteria. The overall conversion rate to clinically definite MS was 48.4%. Forty-six (45%) patients initiated a disease-modifying treatment before the second clinical event. The 2010 McDonald dissemination in space criteria were met in 100 (62.1%) and the 2016 MAGNIMS dissemination in space criteria in 95 (59%) patients with CIS. Six patients with one periventricular lesion fulfilled the 2010 McDonald criteria but not the 2016 MAGNIMS criteria. In contrast, when symptomatic infratentorial or spinal cord lesions were included, two more patients met the 2016 dissemination in space criteria than met the 2010 McDonald criteria. The sensitivity, specificity, and positive and negative predictive values of 2010 McDonald criteria were 80.7%, 55.4%, 63%, and 75.4%, respectively, and those for the 2016 MAGNIMS criteria were 75.6%, 56.6%, 62.1%, and 71.2%, respectively. Both dissemination in space criteria identified a subset of patients with CIS who were at high early risk of developing clinically definite MS (hazard ratio: 2.17 for McDonald and 2.07 for MAGNIMS).
“In our CIS patient cohort, 2016 MAGNIMS MRI criteria for dissemination in space showed lower sensitivity with similar specificity than 2010 McDonald criteria in predicting conversion to clinically definite MS, probably related to the increase in the required number of periventricular lesions,” Dr. Lamas Pérez said. “Because disease-modifying therapy can delay or prevent the conversion to clinically definite MS, the high number of patients that initiated these therapies before the second relapse would explain the intermediate specificity values obtained with both MRI criteria.”
Primary Progressive MS
The 2010 McDonald criteria for PPMS have not been fully validated, while 2016 MAGNIMS MRI guidelines have not been studied in PPMS yet.
To assess the sensitivity and specificity of 2010 McDonald and 2016 MAGNIMS criteria for patients with PPMS, Alberto Gajofatto, MD, PhD, Assistant Professor in Neurology in the Department of Neurological and Movement Sciences at the University of Verona in Italy, and colleagues applied the criteria to two retrospective cohorts.
Patients who were seen at the University of California San Francisco and Verona University MS Centers for suspected PPMS were retrospectively identified from existing databases between November 2015 and October 2016. Data were obtained from review of patient charts with adequate documentation of clinical, MRI, and CSF status to determine the fulfillment of 2010 McDonald criteria for PPMS and 2016 MAGNIMS guidelines for dissemination in space at first visit at study centers. PPMS diagnosis was confirmed at last available visit using stringent criteria (ie, dissemination in space according to 2005 McDonald criteria, dissemination in time according to 2001 McDonald criteria, and exclusion of a better explanation).
Dr. Gajofatto and colleagues included 108 patients with a mean follow-up duration of 10.1 years. The 2010 McDonald criteria had a sensitivity for PPMS of 92.1% and a specificity of 57.9%. The highest combined values of sensitivity and specificity (91.8% and 72.2%) were achieved by combining 2016 MAGNIMS dissemination in space criteria and the presence of oligoclonal bands or increased IgG index in the CSF.
“Our findings suggest that 2010 McDonald criteria for PPMS diagnosis have high sensitivity, while specificity appears to be modest,” Dr. Gajofatto said. “The substitution of 2016 MAGNIMS criteria for dissemination in space plus the incorporation of CSF status increased specificity without compromising sensitivity.”