Head and neck squamous cell carcinoma (HNSCC) was traditionally associated with smoking and alcohol use; however, human papillomavirus (HPV) infection has recently been implicated as a novel risk factor for oropharyngeal carcinoma (OPC) (1). HPV positivity (+) appears to be associated with improved loco-regional control (LRC) but not necessarily improved distant control (2). There may exist a subgroup of patients with LRC that may benefit by use of less aggressive therapy, such as radiation treatment deintensification (2). To perform personalized medicine, it is a prerequisite to better characterize HPV+ OPC. DW-MRI has been utilized in tumor characterization and evaluation of treatment response in HNSCC (3, 4). In this study we aim to characterize HPV+ OPC patients using intravoxel incoherent motion (IVIM) diffusion weighted (DW)-MRI performed before and during chemo-radiation therapy.IVIM DW-MRI data acquisition: Our institutional review board approved this prospective study. We obtained written informed consent from 15 HPV+ OPC patients with neck nodal metastasis prior to the MRI. A total of 60 MRI studies (1 pre- and 3 intra-treatment (TX)) were performed and the patients were treated with chemo- and radiation therapy (dose 70Gy). MRI protocol consisted of multi-planar T₁/T₂ weighted (T_1w/T_2w) imaging followed by IVIM DW-MRI on a 3.0T scanner (Ingenia, Philips Healthcare, Netherlands) using a neurovascular phased-array coil. The IVIM DW-MRI images were acquired using a single shot spin echo planar imaging (SS-EPI) sequence with TR=4000 ms, TE=57 ms, FOV=20-24 cm, matrix=128×128, slices=8-10, slice thickness=5mm, NEX=2 and b=0,20,50,80,200,300,500,800,1500,2000 s/mm2. IVIM DW-MRI data analysis: The data were fitted using (a) mono-exponential model to calculate the apparent diffusion coefficient (ADC) and (b) bi-exponential model (IVIM DW-MRI) to estimate the true diffusion coefficient (D), perfusion fraction (f) and pseudo diffusion coefficient (D*) (5, 6). Regions of Interest (ROIs) were delineated on the metastatic neck node by radiation oncologists on IVIM-DWI image (b = 0 s/mm2) and T2w images using the Eclipse treatment planning system (Varian Medical Systems) and were imported into in house MATLAB software (Math Works, Natick, MA) for image analysis. The total tumor volume was calculated from T2w images. Statistical analysis: A consensus clustering algorithm (7) based on hierarchical clustering and Pearson correlation distance was performed using the pre-TX and intra-TX weekly IVIM DW-MRI metric values of D, f and D*, total tumor volume and delivered radiation dose to assess whether there is a presence of HPV+ subtypes. Analysis was performed using non parametric Wilcoxon-test. Results are presented as mean±SD and a p value <0.05 was chosen for statistical significance. All data analyses were performed using the Stata (Stata Corp.) and R (https://www.r-project.org/) software.Figure 1 A shows the heatmap of consensus clustering matrix for clusters 1 and 2 in HPV+ OPC patients. The heatmap indicates the extent to which two samples are clustered, representing a high consensus with dark blue color (cluster 1) and a low consensus with light blue color (cluster 2). Figure 1B shows the box-and-whisker plot comparing the D metric values between pre- and intra-TX (week 3) [D3w-0w] for HPV+ OPC patients in cluster 1 and cluster 2. The cluster 1 DD3w-0w value was significantly different (p=0.007) from the cluster 2 and the relative changes in metric values for cluster 1 and cluster 2 were 22% and 62%, respectively. The mean pre-TX total tumor volume was significantly (p=0.02) different between the 2 clusters. Figure 2 shows pre-TX MRI data from two representative HPV+ OPC patients [(51 yr, male (top row), from cluster 2) and 45 yr, male (bottom row), from cluster 1)]. The histograms display the distribution of pre-TX ADC (Fig.1. d and j) and D (Fig.1. f and i) in the two neck nodal metastases. Table 1 shows the IVIM DW-MRI metric values and total tumor volumes for pre-TX and intra-TX (week 3).The weekly IVIM DW-MRI metric values, total tumor volume and delivered radiation dose data were able to characterize HPV+ OPC patients into 2 clusters. The results indicated that HPV+ OPC patients in cluster 2 may benefit patients with the use of less aggressive therapy, such as radiation treatment deintensification.These novel findings need to be validated in a larger HPV+ OPC patient population to enable personalized medicine in radiation therapy.