Cortical inflammatory lesions have been correlated with disability and cortical atrophy in multiple sclerosis. The extent to which cortical lesion load is associated with longer-term physical and cognitive disability in different multiple sclerosis phenotypes has not yet been investigated. Thus, a 5-year prospective longitudinal study was carried on in a large group of patients with multiple sclerosis. Three hundred and twelve consecutive patients suffering from multiple sclerosis (157 relapsing remitting, 35 paediatric, 45 benign, 44 primary progressive and 31 secondary progressive) were enrolled in a 5-year prospective clinical and neuroimaging study. Several magnetic resonance parameters (including cortical lesion number and volume, contrast-enhancing cortical lesions and grey matter atrophy) were analysed to find associations with clinical and cognitive outcomes. Patients with high cortical lesion load had higher Expanded Disability Status Scale increase (median = 1.5; range = 0-3) during the study than both patients with low cortical lesion load (median = 1.0; range = 1-3, P < 0.001) and without cortical lesions (median = 0.5; range = -1 to 2, P < 0.001). Compared with clinically stable patients, 101 (32.4%) patients showing clinical progression at 5 years had the highest rate of cortical lesion accumulation (P < 0.001). Stepwise regression analysis revealed significant and independent contributions from age (β = 0.55), cortical lesion volume (β = 0.58), T(2) white matter lesion volume (β = 0.34) and grey matter fraction (β = 0.42) as predictors (final model with r(2 )= 0.657, P < 0.001) of Expanded Disability Status Scale change. Disease duration (β = 0.52, P < 0.001), cortical lesion volume (β = 0.67, P < 0.001), grey matter fraction (β = 0.56, P < 0.001) and T(2) white matter lesion volume (β = 0.31, P = 0.040) at baseline were found to be independent predictors of cognitive status at the end of the study. While confirming the relevance of cortical pathology in all multiple sclerosis phenotypes, but benign, our study suggests that grey matter and white matter changes in multiple sclerosis occur, at least, partly independently, and that grey matter, more than white matter, damage is associated with physical and cognitive disability progression. Thus, the combination of grey and white matter parameters gives a more comprehensive view of multiple sclerosis pathology and allows a better understanding of the progressive phase of the disease, which, however, seems more related to cortical damage than to subcortical white matter changes.