Objective: Presently there is no clinically feasible imaging modality that can effectively detect cortical demyelination in patients with multiple sclerosis (MS). The objective of this study is to determine if clinically feasible magnetization transfer ratio (MTR) imaging is sensitive to cortical demyelination in MS.Methods: MRI were acquired in situ on 7 recently deceased patients with MS using clinically feasible sequences at 3 T, including relatively high-resolution T1-weighted and proton density-weighted images with/without a magnetization transfer pulse for calculation of MTR. The brains were rapidly removed and placed in fixative. Multiple cortical regions from each brain were immunostained for myelin proteolipid protein and classified as mostly myelinated (MM ctx ), mostly demyelinated (MD ctx ), or intermediately demyelinated (ID ctx ). MRIs were registered with the cortical sections so that the cortex corresponding to each cortical section could be identified, along with adjacent subcortical white matter (WM). Mean cortical MTR normalized to mean WM MTR was calculated for each cortical region. Linear mixed-effects models were used to test if mean normalized cortical MTR was significantly lower in demyelinated cortex. Conclusions: This result demonstrates that clinically feasible MTR imaging is sensitive to cortical demyelination and suggests that MTR will be a useful tool to help detect MS cortical lesions in living patients with MS. Neurology â 2013;80:246-252 GLOSSARY CL 5 cortical demyelinated lesions; HSD 5 honestly significant difference; ID ctx 5 intermediately demyelinated; LSM 5 least-squares mean; MD ctx 5 mostly demyelinated; MM ctx 5 mostly myelinated; MR 5 magnetic resonance; MS 5 multiple sclerosis; MTR 5 magnetization transfer ratio; MTR ctx 5 mean cortical magnetization transfer ratio; normMTR ctx 5 normalized cortical magnetization transfer ratio; PLP 5 proteolipid protein; TE 5 echo time; TR 5 repetition time; T1W 5 T1-weighted; WM 5 white matter.Cortical demyelinated lesions (CL) have been detected in patients with multiple sclerosis (MS) for more than 5 decades.1,2 Postmortem studies have raised the possibility that CL may be more prevalent than white matter (WM) demyelination 3 and that intracortical and mixed cortical/ WM lesions together may account for .50% of all MS brain lesions. 4 Despite their detection on postmortem tissue, most CL are invisible on conventional MRI. 4,5 The double inversion recovery sequence improves detection but does not reveal most CL. 5,6 While nonquantitative imaging features low sensitivity to CL, the sensitivity of semiquantitative and quantitative MRI is unclear. Although one ex vivo imaging study performed at 1.5 T revealed that T1 and T2 relaxation rates (not magnetization transfer ratio [MTR]) are sensitive to CL, 7 a separate study performed at 9.4 T found MTR and T2 relaxation rate (not T1) to be most sensitive.