Background: Dual anti-HER2 antibodies pertuzumab (P) and trastuzumab (T) in combination with taxane (D), followed by PT maintenance, is the standard first line treatment for HER2 positive advanced breast cancer (HER2+ ABC). Treatment associated cardiotoxicity necessitates regular cardiac function surveillance, which is a burden particularly for treatment long-responders. Data for cardiac safety of prolonged P+T exposure is scarce. We investigate the real-world impact on cardiac function in long-responders to treatment with dual anti-HER2 antibodies. Methods: We identified consecutive patients with HER2+ ABC who received the CLEOPATRA regimen (PT-D) between Jan 2014 and Dec 2020 from an institutional cancer registry. All patients had pre-treatment multiple-gated acquisition (MUGA) scan or echocardiogram, and subsequently at 3-monthly intervals until end of treatment to monitor left ventricular ejection fraction (LVEF). Patients on treatment for ≥36 months were considered long-responders. The Wilcoxon signed-rank test was used to assess any significant difference in LVEF at various landmark time-points in comparison to their pretreatment baseline. Results: 101 women with HER2+ ABC were eligible for analysis. Median age at treatment was 62 (IQR, 56.0-69.0). The median duration of treatment was 17.3 months (IQR, 9.0-31.3). 22.8% of patients were long-responders, who received a median of 67 cycles of treatment (IQR, 58-88). Compared to baseline, median LVEF was significantly decreased at 6m (median, 66% vs 69%, p=0.02), however there were no significant differences for any of the subsequent time-points up to 84 m. All of the larger LVEF drop (≥10% from baseline) occurred by the first 24 months, representing 4.7% of the overall measurements. Risk factors present for patients experienced treatment suspension (n=3) included previous exposures to anthracycline and left sided radiotherapy. Conclusions: In patients with HER2+ ABC who were long-responders to first-line PT-D, prolonged exposure to dual anti-HER2 antibodies was not associated with significant cardiotoxicity. It is safe to de-escalate the cardiac surveillance for this population.