A three-year, multi-parametric MRI study in patients at presentation with CIS.
ABSTRACT To define the extent of overall brain damage in patients with clinically isolated syndromes (CIS) suggestive of multiple sclerosis (MS) and to identify non-conventional magnetic resonance (MR) metrics predictive of evolution to definite MS.
Brain conventional and magnetization transfer (MT) MRI scans were obtained from 208 CIS patients and 55 matched healthy controls, recruited in four centres. Patients were assessed clinically at the time of MRI acquisition and after a median period of 3.1 years from disease onset. The following measures were derived: T2, T1 and gadolinium (Gd)- enhancing lesion volumes (LV), normalized brain volume (NBV), MTR histogram-derived quantities of the normal-appearing white matter (NAWM) and grey matter (GM).
During the follow-up, 43 % of the patients converted to definite MS. At baseline, a significant inter-centre heterogeneity was detected for T2 LV (p = 0.003), T1 LV (p = 0.006), NBV (p < 0.001) and MTR histogram-derived metrics (p < 0.001). Pooled average MTR values differed between CIS patients and controls for NAWM (p = 0.003) and GM (p = 0.01). Gdactivity and positivity of International Panel (IP) criteria for disease dissemination in space (DIS), but not NAWM and GM MTR and NBV, were associated with evolution to definite MS. The final multivariable model retained only MRI IP criteria for DIS (p = 0.05; HR = 1.66, 95 % CI = 1.00-2.77) as an independent predictor of evolution to definite MS.
Although irreversible tissue injury is present from the earliest clinical stages of MS, macroscopic focal lesions but not "diffuse" brain damage measured by MTR are associated to an increased risk of subsequent development of definite MS in CIS patients.
- [Show abstract] [Hide abstract]
ABSTRACT: We summarize MRI measures currently available to assess treatment efficacy and safety in multiple sclerosis (MS) clinical trials and discuss novel metrics that could enter the clinical arena in the near future. In relapsing remitting MS, MRI measures of disease activity (new T2 and gadolinium-enhancing lesions) provide a good surrogacy of treatment effect on relapse rate and disability progression; however, their value in progressive MS remains elusive. For the progressive disease forms, these measures need to be combined with quantities assessing the extent of irreversible tissue loss, which have already been introduced in some clinical trials (e.g., evolution of active lesions into permanent black holes and brain atrophy). Novel measures (e.g., quantification of gray matter and spinal cord atrophy) have demonstrated a great value in explaining patients' clinical outcome, but still need to be fully validated. Despite showing promise, evaluations of cortical lesions, of microscopic tissue abnormalities, and of functional cortical reorganization are still some way off for monitoring of treatment effects. Trial outcomes in MS should include measures of inflammation and neurodegeneration, which should be combined according to the disease clinical phenotype, phase of the study, and the supposed mechanism of action of the drug tested.Current opinion in neurology 06/2014; 27(3):290-9. · 5.73 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: Background Gray matter (GM) and white matter (WM) pathology has an important role in disease progression of multiple sclerosis (MS). Objectives To investigate the association between the development of GM and WM pathology and clinical disease progression in patients with clinically isolated syndrome (CIS). Methods This prospective, observational, 48-month follow-up study examined 210 CIS patients treated with 30 µg of intramuscular interferon beta-1a once a week. MRI and clinical assessments were performed at baseline, 6, 12, 24, 36 and 48 months. Associations between clinical worsening [24-weeks sustained disability progression (SDP) and occurrence of a second clinical attack] and longitudinal changes in lesion accumulation and brain atrophy progression were investigated by a mixed-effect model analysis after correction for multiple comparisons. Results SDP was observed in 32 (15.2%) CIS patients, while 146 (69.5%) were stable and 32 (15.2%) showed sustained disability improvement. 112 CIS patients (53.3%) developed clinically definite MS (CDMS). CIS patients who developed SDP showed increased lateral ventricle volume (p < .001), decreased GM (p = .011) and cortical (p = .001) volumes compared to patients who remained stable or improved in disability. Converters to CDMS showed an increased rate of accumulation of number of new/enlarging T2 lesions (p < .001), decreased whole brain (p = .007) and increased lateral ventricle (p = .025) volumes. Conclusions Development of GM pathology and LVV enlargement are associated with SDP. Conversion to CDMS in patients with CIS over 48 months is dependent on the accumulation of new lesions, LVV enlargement and whole brain atrophy progression.NeuroImage: Clinical. 01/2014;
- [Show abstract] [Hide abstract]
ABSTRACT: We aim to determine if machine learning techniques, such as support vector machines (SVMs), can predict the occurrence of a second clinical attack, which leads to the diagnosis of clinically-definite Multiple Sclerosis (CDMS) in patients with a clinically isolated syndrome (CIS), on the basis of single patient's lesion features and clinical/demographic characteristics. Seventy-four patients at onset of CIS were scanned and clinically reviewed after one and three years. CDMS was used as the gold standard against which SVM classification accuracy was tested. Radiological features related to lesional characteristics on conventional MRI were defined a priori and used in combination with clinical/demographic features in an SVM. Forward recursive feature elimination with 100 bootstraps and a leave-one-out cross-validation was used to find the most predictive feature combinations. 30 % and 44 % of patients developed CDMS within one and three years, respectively. The SVMs correctly predicted the presence (or the absence) of CDMS in 71.4 % of patients (sensitivity/specificity: 77 %/66 %) at 1 year, and in 68 % (60 %/76 %) at 3 years on average over all bootstraps. Combinations of features consistently gave a higher accuracy in predicting outcome than any single feature. Machine-learning-based classifications can be used to provide an "individualised" prediction of conversion to MS from subjects' baseline scans and clinical characteristics, with potential to be incorporated into routine clinical practice.NeuroImage. Clinical. 01/2015; 7:281-7.