Stereotactic radiosurgery (SRS) is a well-established treatment for many types of brain tumors. However, surrounding healthy tissues may also receive a significant radiation dose during SRS. This can lead to brain swelling, necrosis, and neuronal dysfunction, thus inducing delayed adverse effects such as cognitive decline and stroke-like symptoms. We propose to use diffusion MRI (dMRI) tractography and behavioral assessment tools to enhance our understanding of the neuroplasticity changes associated with brain irradiation. Thirty male Fisher rats (protocol # 363-14) were irradiated using a targeted irradiation in the primary somatosensory area (S1), hippocampus and primary motor cortex (M1) of the right hemisphere, as previously described [1]. Using the Leksell Gamma Knife Perfexion®, the mean deposited dose into the S1, M1 and hippocampus (Fig. 1) structures were respectively, 113 ± 4 Gy, 41 ± 8 Gy, 24 ± 10 Gy,. Before and at different time points after irradiation, rats were scanned with a small animal MRI scanner (Varian Inc., Palo Alto, CA) with a dedicated rat head-coil (RAPID MR International, OH), based on a single b-value dMRI acquisition to assess the integrity of neuronal interconnections. Diffusion weighted images were acquired using a multi-slice spin echo sequence with 15 non-collinear diffusion (+b0) gradient directions and b = 977 s/mm². Other imaging parameters were: TR/TEeff = 3500 ms/35 ms, FOV = 38.4×38.4 mm² covering 25 coronal slices with no gap and 0.3×0.3×0.35 mm3 resolution, and a 2 h total scan duration. For registration purposes, a spin echo T2-weighted sequence was performed with the following parameters: repetition time TR/TEeff = 3000 ms/48 ms; 8 echos; echo spacing 12 ms; field of view (FOV) = 32 × 32 mm2, 25 slices, 0.7 mm axial slice thickness. The methodology used to analyze dMRI images was, first a denoising step [2] followed by computation of fODF [3]. Tractography algorithms and visualization of streamlines were used to reveal displacements and breakdown in neuronal pathways by studying different diffusion metrics as FA, MD, AD, RD, but also advanced metrics such as fiber crossing characterization (Nufo) and maximum apparent fiber density (afd max) [4] for more appropriate fiber-density mapping [5]. All this analysis was done in Dipy [6]. Moreover, to correlate dMRI results, brain-region specific sensitivity to irradiation was determined using different behavioral tests. In addition, myelin sheath damage (Luxol Fast Blue staining), astrocytosis reaction (chicken anti-GFAP antibody staining, #AB5541, Millipore, CA, USA), and tissue inflammatory response (rabbit anti-Iba1 staining, #019-19741, Wako Chemicals, VA, USA) were characterized by immunohistochemistry.Fig. 2 shows corpus callosum fiber bundles disruption (surrounded by neocortex and hippocampus) at day 110 following irradiation, which were confirmed by histology (Fig. 3). Other regions of interest showed significant tract-density (tdi) decrease in irradiated fimbria of hippocampus (white matter), neocortex and M1 but no changes in hippocampus. The significant tdi decrease into irradiated neocortex and M1 was confirmed by afd max, which is not tracking-dependent. On the other hand, behavioral tests showed that: i) motor function (Rotarod and Actimetry), revealed that brain irradiation did not affect motor performances (M1-related) while neuronal disorganization were also observed by histology, and ii) the anxiety-like behaviors and learning/memory performances (measured by Elevated plus maze and Morris water maze assays) were significantly decreased. This result probably reflects right amygdala and hippocampus alterations, although they received respectively no and lower radiation dose than the M1. Our results revealed that sensitivity of some brain areas is not only associated to the radiation doses received. A high single dose of targeted irradiation can induce a region-specific plasticity (i.e. structural and function changes), as demonstrated by neuronal matrix remodeling using dMRI and an appropriate HARDI reconstruction, that were partially validated by behavioral assays. Brain structures are highly interconnected, thus prevention of axonal damage may be important to preserve the reserve capacity. The current interpretation of radiation-induced cognitive impairments in clinic may be overly simplistic. This study suggests that treatment planning may be improved by a better understanding of the brain response to radiation, which would take into account of its reserve capacity associated to plasticity, and also the sensitivity of white matter and not only of gray matter/functional structures.