Quantitative magnetization transfer provides information complementary to grey matter atrophy in Alzheimer's disease brains Giovanni Giulietti a, ⁎, Marco Bozzali a , Viviana Figura a , Barbara Spanò a , Roberta Perri b , Camillo Marra c , Giordano Lacidogna c , Franco Giubilei d , Carlo Caltagirone b, e , Mara Cercignani a a Neuroimaging Laboratory, Santa Lucia Foundation IRCCS, via Ardeatina 306, 00179 Rome, Italy b Department of Clinical and Behavioural Neurology, Santa Lucia Foundation IRCCS, via Ardeatina 306, 00179 Rome, Italy c Institute of Neurology, Università Cattolica, L.go A. Gemelli 8, 00168 Rome, Italy d Department of Neurology, II Faculty of Medicine, Sapienza University of Rome, viale dell'Università 30, 00185 Rome, Italy e Department of Neuroscience, University of Rome Tor Vergata, via Montpellier 1, 00133 Rome, Italy abstract article info Article history: Received 28 March 2011 Revised 15 September 2011 Accepted 19 September 2011 Available online 1 October 2011 Keywords: Magnetization transfer Alzheimer's disease Two-pool model BPM Mitochondrial dysfunction Neurodegeneration Preliminary studies, based on a region-of-interest approach, suggest that quantitative magnetization transfer (qMT), an extension of magnetization transfer imaging, provides complementary information to conventional magnetic resonance imaging (MRI) in the characterisation of Alzheimer's disease (AD). The aim of this study was to extend these findings to the whole brain, using a voxel-wise approach. We recruited 19 AD patients and 11 healthy subjects (HS). All subjects had an MRI acquisition at 3.0 T including a T 1 -weighted volume, 12 MT-weighted volumes for qMT, and data for computing T 1 and B 1 maps. The T 1 -weighted volumes were processed to yield grey matter (GM) volumetric maps, while the other sequences were used to compute qMT parametric maps of the whole brain. qMT maps were warped to standard space and smoothed, and subsequently compared between groups. Of all the qMT parameters considered, only the forward exchange rate, RM 0 B , showed significant group differences. These images were therefore retained for the multimodal statistical analysis, designed to locate brain regions of RM 0 B differences between AD and HS groups, adjusting for local GM atrophy. Widespread areas of reduced RM 0 B were found in AD patients, mainly located in the hippocampus, in the temporal lobe, in the posterior cingulate and in the parietal cortex. These results indicate that, among qMT parameters, RM 0 B is the most sensitive to AD pathology. This quantity is altered in the hippocampus of patients with AD (as found by previous works) but also in other brain areas, that PET studies have highlighted as involved with both, reduced glucose metabolism and amyloid β deposition. RM 0 B might reflect, through the measurement of the efficiency of MT exchange, some information with a specific pathological counterpart. Given previous evidence of a strict relationship between RM 0 B and intracellular pH, an intriguing speculation is that our findings might reflect metabolic changes related to mitochondrial dysfunction, which has been proposed as a contributor to neurodegeneration in AD. © 2011 Elsevier Inc. All rights reserved. Introduction From a clinical point of view, Alzheimer's disease (AD) is charac- terised by an early and prominent impairment of memory functions, followed by a progressive accumulation of additional cognitive defi- cits, eventually resulting in dementia (Nelson et al., 2009). From a pathological point of view, the most prevalent AD features include the accumulation of amyloid β plaques (Markesbery et al., 2006) and neurofibrillary tangles (Braak and Braak, 1995; Gómez-Isla et al., 1996) in combination with a progressive neuronal loss, which eventually results in gross brain atrophy. In addition to the neurodegenerative processes occurring within the grey matter (GM), there is a growing body of imaging data dem- onstrating that disruption of specific white matter (WM) tracts also occurs in AD (Bozzali and Cherubini, 2007), suggesting disconnection as a strong contributor to the spread of damage in AD brains (Bozzali et al., in press). Brain volumetrics has significantly contributed to map the spread of GM atrophy over disease progression, to clarify its link with the risk for developing dementia in prodromal conditions (Bozzali et al., 2006), and to confirm how this process parallels the accumulation of cognitive disabilities (Bozzali et al., 2006; Serra et al., 2010a) and of behavioural symptoms (Serra et al., 2010b) in AD. On the other hand, diffusion techniques have clarified that microscopic damage occurs also in the WM of AD brains (Bozzali and Cherubini, 2007), highlighting its potential role in determining cognitive deficits by disconnection mechanisms (Bozzali et al., in press; Villain et al., 2008). The NeuroImage 59 (2012) 1114–1122 ⁎ Corresponding author. Fax: + 39 06 5150 1213. E-mail address: giulietti.giovanni@gmail.com (G. Giulietti). 1053-8119/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.neuroimage.2011.09.043 Contents lists available at SciVerse ScienceDirect NeuroImage journal homepage: www.elsevier.com/locate/ynimg