Diffusion-Tensor cardiovascular magnetic resonance (DT-CMR) is capable of mapping myocardial fiber orientation^1,2. It has been demonstrated in myocardial infarction (MI) murine models that DT-CMR can identify the effects of stem cell therapy on myocardial fiber orientations³. The study illustrated the powerful potential of DT-CMR in identifying adverse treatment despite successful delivery of viable stem cells. However, it remains to be seen if this recent work is translatable to large animal and clinical studies. The greatest challenge for translating DT-CMR for clinical application is overcoming the sensitivity to bulk motion with gradient strengths available on clinical scanners. We propose the application of a recently developed in vivo cardiac DT-CMR technique⁴ to characterize myocardial fiber orientation before and after the novel regenerative therapy of using intramyocardial injection of exosomes from cardiosphere-derived cells (CDCs) in a MI porcine model.MI was induced in 14 Yucatan mini pigs by balloon occlusion of the mid-LAD for 2.5 hours. The MI was allowed to heal for 4 weeks for all pigs defining baseline. Group 1 (N=9) was treated with exosome proteins derived from CDCs after an additional 4 weeks of therapy. Group 2 was given a placebo injection of saline. CMR was performed at baseline and 4 weeks after treatment on a 3T Siemens Verio with the following: whole-heart 2D multi-slice (WH) morphological CMR, WH function CINE CMR, 3 short axis slice DT-CMR (M2 Diffusion Prepared bSSFP, 6 dir, b=350 s/mm^2, 8avg, free-breathing), and WH viability CMR (LGE PSIR, TI=315ms). Viability and function CMR yielded scar size (SS) and ejection fraction (EF)⁵, respectively. SS and EF were calculated semi-automatically using cvi42 software (Circle, Calgary, Canada). For in vivo DT-CMR, mean diffusivity (MD), fractional anisotropy (FA), and helix angle (HA) maps were calculated using custom software built with DIPY library in Python⁶. HA transmurality slope (HATS) was also calculated by radially sampling the transmural HA along 36 chords and fitting the slope of a linear regression between HA and transmural depth. Wilcoxon signed-rank test was performed to evaluate the difference between mean slice values (G1 N=27, G2 N=15) before and after treatment (p < 0.017 significance). Change (Δ) in MD, FA, and HATS were correlated (R^2) with ΔSS and ΔEF.For Group 1 (treated), EF, MD, FA, and HATS did not significantly change (Δ: -1±2%, -0.2±0.2 um^2/ms, 0.03±0.03, and 0.02±0.2°/%depth, respectively), while SS was significantly reduced (Δ: -4±2%, p<0.01). In contrast, Group 2 (placebo) exhibited significant (p<0.01) adverse changes with decreased EF (Δ: -6±2%), increased SS (Δ: 4±2%), increased MD (Δ: -0.4±0.3 um^2/ms), decreased FA (Δ: -0.06±0.05), and decreased HATS (-1.1±0.3 vs -0.7±0.3°/%depth). ΔMD and ΔFA weakly correlated with ΔEF (R^2: 0.2 and 0.3, respectively) and ΔSS (R^2: 0.1 and 0.2, respectively). However, ΔHATS significantly (p < 0.01) correlated highly with ΔEF (R^2: 0.8) and ΔSS (R^2: 0.7).The adverse changes seen in the placebo group is consistent with progression of chronic MI with decreased FA, increased MD, increased SS, decreased EF, and decreased HATS (towards 0). The typical chronic MI progression is not seen in the treated group and for SS it is actually reversed. This suggests that the exosomes derived from CDCs has a therapeutic effect in halting the progression of chronic MI. More interesting, DT-CMR adds a unique perspective in which the data suggests that global adverse remodeling of the myocardial architecture is preserved with treatment. The microstructural insight that DT-CMR yields demonstrates that exosome treatment may prevent permeant structural remodeling.In a MI porcine model, in vivo DT-CMR revealed that myocardial fiber orientation was preserved with CDC-derived exosome treatment and adversely changed with placebo treatment consistent with observed viability and function changes. Furthermore, changes in helix transmurality highly correlated with changes in viability and function.