To examine the effect of allogeneic hematopoietic stem cell transplantation (HSCT) on myelin and axons in late-onset Krabbe disease 4 years post-procedure.Krabbe disease, or globoid cell leukodystrophy, is an autosomal recessive lysosomal storage disorder caused by a deficiency of galactocerebrosidase (GALC). The accumulation of psychosine results in death of oligodendrocytes and Schwann cells, both essential to myelin formation. The late-onset form of Krabbe disease is a very rare progressive neurodegenerative condition leading to death within 2-7 years of symptom onset. Traditional treatment was supportive but more recent research focuses on HSCT strategies whereby the newly derived white blood cells restore GALC levels thus halting accumulation of toxic metabolites that damage myelin producing cells. Very recently, one case study found HSCT led to a sustained normalization of peripheral GALC enzyme activity and halted symptom progression at 24 months post-HSCT¹. Conventional MRI can show brain lesions in Krabbe disease but not changes in the normal appearing white matter (NAWM). MRI techniques specific to myelin and axons can characterize pathology and provide insight into the effects of HSCT on the on-going disease processes. We used myelin water imaging (a marker for myelin²), diffusion tensor imaging (DTI, for fibre characterization and white matter integrity) and magnetic resonance spectroscopy (MRS, for quantification of metabolites including n-acetyl-aspartate (NAA), believed to be a marker for axons) to assess the efficacy of HSCT on Krabbe disease over 4 years. To our knowledge, this is the longest advanced imaging follow-up in late-onset Krabbe disease and the first report using 3.0T MRI.A 42 year old female with late-onset Krabbe disease and an age/gender-matched control underwent imaging on a 3.0T Philips Achieva MR system (baseline, year 1, 2, 3, 4; baseline was immediately prior to HSCT for the Krabbe subject). MR experiments: Conventional scans (sagittal, axial and coronal proton density, T2-weighted, FLAIR) were followed by (1) 3DT₂ Gradient Echo Spin Echo sequence (32 echoes, TR=1200ms, 10ms echo spacing, voxel size=1.0x1.0x5mm³, slices=10); (2) DTI (Single-shot EPI, TR/TE=9598/58ms, slices=71, voxel size=2.24x2.24x2.1mm³, voxel (reconstructed)=1.0x1.0x2.1mm³, b=800, 16 non-colinear gradient encoding directions; (3) MRS (single voxel (Krabbe: lesion and NAWM; control: location matched to Krabbe), PRESS, scan time=4:36min, TR/TE=3000/35ms, voxel=15x12x15mm³). Analysis: A regularized non-negative least-squares algorithm with extended phase graph algorithm for stimulated echo correction^3,4 fit T2 data. Voxel-wise myelin water fraction (MWF) was the area under the T₂ distribution from 10-40ms divided by the total area. FMRIB's diffusion toolbox (FDT) estimated all diffusion tensors and generate maps of fractional anisotropy (FA), mean diffusivity (MD), parallel diffusivity (D_para) and perpendicular diffusivity (D_perp). Segmented lesion and NAWM was applied to registered myelin water and DTI maps and a mean value of each parameter was computed. LCModel⁵ with water-scaling gave institutional concentrations in millimolar (mM) for total NAA, creatine (Cre), choline (Cho), myo-Inositol (mI) and glutamate and glutamine (Glx). Differences over time were assessed by linear regression.Krabbe lesion and NAWM showed reduced myelin water, NAA and FA, and increased MD, D_para, D_perp, mI and Glx at baseline compared to control (Figure 1). Conventional MRI showed no evidence of new disease activity 4 years post-HSCT (Figure 2) and all advanced imaging measures remained stable (non-significant slope) from baseline to year 4 in both the Krabbe subject and the control (Figure 1). Relative to control white matter, Krabbe NAWM had 28% less myelin water, 9.6% reduced FA, 17% reduced NAA, 18% increased MD, 25% increased Dpara, 10% increased Dperp, 30% increased mI and 11% increased Glx (Figure 1, Figure 3). Average choline and creatine levels were similar between Krabbe NAWM and control white matter. Krabbe lesion was more severely affected than NAWM (Figure 3), but lesions measures also remained stable over the 4 years.HSCT in late-onset Krabbe disease halts demyelination and axonal loss up to 4 years post-HSCT. A global process of NAWM damage occurs in late-onset Krabbe disease beyond the findings visible on conventional MRI, with myelin being more affected than axons. However, damage did not worsen over time, supporting HSCT as an effective treatment strategy.