We evaluated temporal changes in early diastolic strain rates by cardiovascular magnetic resonance (CMR) as an early detector of trastuzumab-induced ventricular dysfunction. We conducted a prospective, multi-centre, longitudinal observational study of 41 trastuzumab-treated breast cancer women who underwent serial CMR (baseline, 6, 12, and 18 months). Two blinded readers independently measured left ventricular ejection fraction (LVEF), peak systolic strain parameters (global longitudinal strain [GLS] and global circumferential strain [GCS]), and early diastolic strain rate parameters (global longitudinal diastolic strain rate [GLSR-E], global circumferential diastolic strain rate [GCSR-E], and global radial diastolic strain rate [GRSR-E]), by feature tracking (FT-CMR) using CMR42. There was a significant decline in peak systolic strain GLS and GCS at 6 months (p = 0.024 and p < 0.001, respectively) and 12 months (p = 0.002 and p < 0.001, respectively), followed by recovery at 18 months, which paralleled decline in LVEF at 6 months (p = 0.034) and 12 months (p = 0.012). Conversely, early diastolic strain rates GLSR-E and GCSR-E did not significantly change over 18 months (p > 0.10), while GRSR-E was marginally significant at 12 months (p = 0.021). There was no significant correlation between changes at 6 months in LVEF and GLSR-E or GRSR-E (p > 0.10), and a marginally significant weak correlation between LVEF and GCSR-E (p = 0.046). Among trastuzumab-treated patients without overt cardiotoxicity, there was no consistent temporal change in FT-CMR-derived diastolic strain rate parameters up to 18 months, in contrast to decline in systolic strain and LVEF. Systolic strains by FT-CMR are likely more useful than diastolic strain rates for monitoring subclinical trastuzumab-related myocardial dysfunction.ClinicalTrials.gov identifier NCT01022086.