Background: To evaluate the clinicopathological and prognostic significance of the percentage change between maximum standardized uptake value (SUV max) at 60 min (SUV max 1) and SUV max at 120 min (SUV max 2) (ΔSUV max %) using dual time point 18 F-fluorodeoxyglucose emission tomography/computed tomography (18 F-FDG PET/CT) in breast cancer. Methods: Four hundred and sixty-four patients with primary breast cancer underwent 18 F-FDG PET/CT for preoperative staging. ΔSUV max % was defined as (SUV max 2 − SUV max 1) / SUV max 1 × 100. We explored the optimal cutoff value of SUV max parameters (SUV max 1 and ΔSUV max %) referring to the event of relapse by using receiver operator characteristic curves. The clinicopathological and prognostic significances of the SUV max 1 and ΔSUV max % were analyzed by Cox's univariate and multivariate analyses. Results: The optimal cutoff values of SUV max 1 and ΔSUV max % were 3.4 and 12.5, respectively. Relapse-free survival (RFS) curves were significantly different between high and low SUV max 1 groups (P = 0.0003) and also between high and low ΔSUV max % groups (P = 0.0151). In Cox multivariate analysis for RFS, SUV max 1 was an independent prognostic factor (P = 0.0267) but ΔSUV max % was not (P = 0.152). There was a weak correlation between SUV max 1 and ΔSUV max % (P < 0.0001, R 2 = 0.166). On combining SUV max 1 and ΔSUV max %, the subgroups of high SUV max 1 and high ΔSUV max % showed significantly worse prognosis than the other groups in terms of RFS (P = 0.0002). Conclusion: Dual time point 18 F-FDG PET/CT evaluation can be a useful method for predicting relapse in patients with breast cancer. The combination of SUV max 1 and ΔSUV max % was able to identify subgroups with worse prognosis more accurately than SUV max 1 alone.