The purpose of this study was to determine if two-dimensional 2DL-COSY, an in vivo spectroscopic technique that has improved signal dispersive power⁴, can be used to identify biochemical changes in MS patients compared with healthy controls.Nineteen patients with Relapse Remitting Multiple Sclerosis (RRMS) were compared with ten age and gender matched healthy controls (mean age 42±1.8) with Expanded Disability Status Scale (EDSS) score of 1-4. All subjects were scanned using a 3 Tesla MR system (Magnetom Prisma, Siemens Healthcare) equipped with 64 Channel brain coil. 2D L-COSY MRS was acquired from the posterior cingulate gyrus (PCG) with 3x3x3 cm3 voxel employing TEinitial of 30ms, TR 1.5sec, 8 averages per increment, bandwidth 2000Hz, t1 increment of 0.8ms, vector size of 1024 points, RF offset frequency was set on 3.2ppm, and number of increments was 96. Raw spectral data was taken offline and data from different coil elements were combined and concatenated to produce a 2D time-domain array (96x1024) which was then processed and analysed with Felix 2007 software (Accelrys, San Diego, CA). The creatine methyl resonance (F2:3.02, F1:3.02ppm) was used as the internal chemical shift reference. Average peak ratios to total creatine (tCr) diagonal peak were calculated for each assigned metabolite and compared using t-test (Stata, StataCorp 2013). This study was approved by the local ethics review board and all subjects were consented in writing.L-COSY detected several metabolite resonances from PCG in healthy controls and MS group. They were fucose/threonine, glycerophosphocholine (GPC), glutamine+glutamate pool (Glx-1/Glx-2/Glx-3), and g-aminobutyric acid (GABA). The metabolites to tCr ratios with statistically significant differences between the two groups are summarised in Table 1. The t-test analysis showed metabolites to tCr ratios from MS patients were significantly different when compared to healthy controls for all cross peaks detected (P≤0.05). A typical L-COSY spectrum acquired from an RRMS patient is shown in Figure 1. Presently, the exact role of metabolites at cross peaks in MS is poorly understood. However, it can be theorised that the decrease in fucose/threonine in MS group may be due to impaired synaptic transmissions while reduction in GABA may be associated with neuronal loss⁵. The decrease in Glx pool may be attributed to neuronal metabolic dysfunction⁵. Although the clinical relevance of cross peaks remains elusive, the changes noted in this study may suggest an alteration in the neurochemical composition in the MS population. Statistically significant differences (P≤0.05) were also noted along diagonal peaks for total NAA and lipids (free lipid) which are associated with increased axonal injury and myelin breakdown as reported in other studies2-4. Our preliminary data shows L-COSY technique can be used to identify metabolites that are unresolvable by 1D 1H-MRS owing to the technical limitations. L-COSY technique has the ability to identify metabolites that may be unique to the changes in the biochemistry of MS and has potential to play an important role in biomarker discovery in MS disease.