Breast cancer is the most common type of malignant tumor and the leading cause of death in women worldwide (Sung et al., 2021). The use of chemotherapy has significantly improved both mortality and morbidity outcomes in breast cancer patients. Anthracyclines are widely used in the treatment of hematological malignancies and solid tumors, with powerful antitumor effects and indispensability (Nicolazzi et al., 2018). However, cardiotoxicity is a serious side effect of these kinds of drugs, which limits their clinical application (Schwentner et al., 2016). Pegylated liposomal doxorubicin (PLD) is a new dosage form of doxorubicin confined in liposomes, which can form a stable three-dimensional structure when polyethylene glycol is grafted onto the surface (stealth liposome) (Gabizon et al., 2008). Pegylated liposomal encapsulation reduces the plasma levels of free doxorubicin and may reduce drug delivery to normal tissues, which can in turn reduce cardiotoxicity (Gil-Gil et al., 2021). Results from clinical data revealed that PLD showed a similar efficacy to doxorubicin with a lower incidence of cardiotoxicity when administered in either advanced or (neo)adjuvant stages of treatment (O’Brien et al., 2004; Liu et al., 2021).

Human epidermal growth factor receptor 2 (HER2) is overexpressed in approximately 20–25% of breast cancer cases, which is associated with poor prognosis (Lynce et al., 2017). However, remarkable progress of trastuzumab plus chemotherapy as adjuvant treatment has significantly improved the survival rates of female breast cancer patients (Slamon et al., 2011; Goldhirsch et al., 2013; Perez et al., 2014; Cameron et al., 2017). Further development in the administration of dual anti-HER2 therapy with trastuzumab and pertuzumab shows great outcomes in (neo)adjuvant as well as metastatic settings (Baselga et al., 2012; Gianni et al., 2012; von Minckwitz et al., 2017). Trastuzumab is also associated with an increased risk of cardiotoxicity, particularly when administered in combination with anthracycline-based therapy. Cardiotoxicity, in this instance, would be manifested as symptomatic congestive heart failure (CHF) or asymptomatic decline of left ventricular ejection fraction (LVEF) (Seidman et al., 2002).

Previous studies have reported that 3–7% of patients who were receiving trastuzumab exhibited cardiac dysfunction in various forms (Seidman et al., 2002). When pertuzumab was combined with trastuzumab and chemotherapy, its cardiac safety was similar to that of trastuzumab alone (Swain et al., 2013; Yu et al., 2016). Nevertheless, limited data are available on the cardiac safety of PLD-based treatment administered sequentially with trastuzumab ± pertuzumab for the adjuvant treatment of HER2-positive early breast cancer (eBC) patients. Therefore, in this study, we aimed to explore strategies to reduce the cardiotoxicity of anthracycline sequential trastuzumab treatment and to clarify the cardiac safety and efficacy of PLD sequential trastuzumab treatment in adjuvant therapy.

Over recent years, the application of anthracyclines sequentially with anti-HER2 therapy has shown to significantly improve the survival rates and prognoses of patients (Slamon et al., 2011). However, cardiotoxicity, as the major cause of breast cancer deaths, even without symptoms, may significantly limit the possibility of tumor treatment (Patnaik et al., 2011). In clinical practice, LVEF measured by echocardiography is used as an index to evaluate cardiotoxicity due to its accessibility.

PLD has been demonstrated to have equivalent efficacy but significantly less cardiotoxicity than conventional doxorubicin. In a phase III study, the cardiotoxicity rates were 3.9% and 18.8% with PLD and doxorubicin, respectively (O'Brien et al., 2004). A retrospective study compared the use of traditional anthracyclines and PLD in neoadjuvant therapy for patients with breast cancer. The study revealed higher pathologic complete response rates and lower incidences of cardiotoxicity in PLD-based cohorts (Liu et al., 2021). Therefore, it can be considered that the risk of cardiac toxicity after PLD sequential trastuzumab therapy is slightly higher than that of PLD monotherapy, but it is acceptable. Dual-target therapy with the addition of pertuzumab did not increase the incidence of cardiac-related AEs in either metastatic or neoadjuvant studies (Gianni et al., 2012; Schneeweiss et al., 2013).

PLD-C is not a standard protocol for breast cancer in adjuvant chemotherapy. However, due to the individualized treatment of cancer, some patients choose PLD as an alternative drug to anthracycline in adjuvant therapy. Through the narration of the physician in charge of these cases, the main reasons for these patients to choose PLD in our center include the following: 1) to avoid obvious hair loss caused by initial chemotherapy, 2) to avoid stronger nausea and vomiting symptoms, 3) to avoid the potential risk of developing febrile neutropenia, and 4) consideration of the potential to reduce the incidence of lifetime cardiotoxicity.

In this retrospective cohort study, PLD-C followed by anti-HER2 therapy as adjuvant therapy was effective and safe in a population of HER2-positive eBC patients. This study demonstrated that these schemes had a low risk for cardiotoxicity with slight clinical cardiotoxicity compared with treatments in previous clinical studies containing conventional anthracycline, taxanes, and trastuzumab (Table 3).

As described in Table 3, the NCCTG N9831 trial randomized patients between two arms in the adjuvant setting with AC followed by paclitaxel either with sequential or concurrent trastuzumab, and at a median follow-up of 9.2 years, the reported incidence of CHF or cardiac death (CD) was 2.8% in arm B (AC followed by paclitaxel and then trastuzumab) and 3.7% in arm C (AC followed by paclitaxel concurrent with trastuzumab) (Advani et al., 2016). In the NSABP-B31 study, at a median follow-up of 7 years, the incidence of severe CHF/CD was 3.9% in the arm with AC followed by TH. An independent retrospective review of the B-31 and N9831 trials demonstrated a 2.0% incidence of symptomatic CHF or CD in the trastuzumab-containing arm (Romond et al., 2012). In the BCIRG-006 study, at a median follow-up of 10 years, the incidence of CHF or CD was 2.0% in the arm with AC followed by docetaxel and trastuzumab (Slamon et al., 2011). While in our study, at a median follow-up of 3.5 years, the incidence of CHF or CD was 0% in the groups with PLD-C followed by TH/TPH (q3w). In the NCCTGN9831 and NSABPB-31 trials, LVEF reductions ≥15% from baseline were 14.6% (NCCTG N9831 arm B) and 18.7% (NCCTG N9831 arm C), respectively (Romond et al., 2012; Advani et al., 2016). While in our study, LVEF reduction ≥16% from baseline was 8.6%, where cardiotoxicity halved in value. Subclinical cardiotoxicity in the NSABP B-31 and BCIRG-006 studies using anthracyclines and trastuzumab was similar to that in our study (Slamon et al., 2011; Romond et al., 2012). This indicates that although PLD is used, cardiotoxicity monitoring is still necessary. In the HERA study, because 68% of patients did not use taxanes, the subclinical cardiotoxicity was relatively low (4.4%) (Cameron et al., 2017). In the NCCTGN9831 study, the 5-year DFS rates were 80.1% (arm B) and 84.5% (arm C) (Advani et al., 2016). In the BCIRG-006 study, the 5-year DFS rates were 84% (arm B:AC-TH group) and 81% (arm C:TCH group), respectively (Slamon et al., 2011). While in our study, the 5-year DFS was 96.8%, suggesting that patients had improved survival. It is worth noting that all patients completed PLD-C followed by TH or TPH every 3 weeks, which reflects the excellent tolerance; this may be related to higher DFS and efficiency.

The NCCTG N9831 trial demonstrated that age ≥60 years, registration LVEF <65%, and use of antihypertensive medications were associated with an increased risk of cardiac events, while radiation therapy, BMI, or ethnicity were not statistically significant (Advani et al., 2016). In the NSABP-B31 trial, LVEF (50–54%), age (≥60 years), and receiving antihypertensive medications and radiotherapy were also not statistically significant (Romond et al., 2012). The CANTO study indicated that obesity appeared to be associated with an important increase in risk-related cardiotoxicity in eBC patients (Kaboré and Guenancia, 2019). In our study, since the basal LVEF of the patients was not low and the incidence of cardiotoxicity was low, it was less possible to conclude that cardiotoxicity was related to BMI (<25, ≥25 kg/m²), age (<60, ≥60 years), radiotherapy, ongoing cardiovascular risk factors, or LVEF at baseline.

Our study had several limitations as follows. This study was a retrospective study, the test population was of a small sample size, and the LVEF values were not available for all patients at all corresponding observation points, leading to the possibility that the results showed a reduced incidence of cardiotoxicity. Due to the lack of necessary baseline or follow-up LVEF measurements, many patients had to be excluded, resulting in some selection bias. Therefore, as a future consideration, a prospective cohort study to further clarify the precise incidence and outcome of PLD sequential trastuzumab cardiotoxicity is worth carrying out.

PLD sequential trastuzumab treatment reduces cardiotoxicity by nearly half compared with traditional anthracyclines, which is well tolerated. Most patients can complete the established treatment plans, which may have a positive effect on survival; this merits further exploration and may become an alternative strategy for cardioprotection in patients with HER2-positive eBC.