Cardiotoxicity and incidence of brain metastases after adjuvant trastuzumab for early breast cancer: the dark side of the moon? A meta-analysis of the randomized trials

Abstract

Background: In five randomized clinical trials (RCTs), adjuvant trastuzumab (T) for early stage breast cancer with human epidermal growth-factor receptor-2 over-expression/gene-amplification has shown to decrease the risk of both recurrence and death. The issue regarding the long-term safety profile of such drug is still open; in particular, questions remain about long-term cardiotoxicity, and specific patterns of relapse such as brain metastases (BM). In order to quantify the magnitude of these two risks, and then balance those with the survival outcome, a literature-based meta-analysis was performed.

Methods: All phase III trials were considered eligible. A literature-based meta-analysis was accomplished, and event-based relative risk ratios with 95% confidence interval were derived. A fixed- and a random-effect model according to the inverse variance and the Mantel-Haenzel method were applied. Heterogeneity test was applied as well. Absolute differences (AD) and the Number of patients Needed to Treat or to Harm (NNT/NNH) were calculated. Safety end-points were: (1) Chronic Heart Failure (CHF) grade III-IV rate, (2) Significant reduction of left-ventricular-ejection-fraction (L-FEV) rate and (3) BM rate. In order to quantify the magnitude of the significant benefit already found in the original RCTs, Efficacy end-points were: (1) disease-free survival (DFS) and (2) overall survival (OS).

Results: Five RCTs were gathered (11,187 patients); at an average 2-years follow-up, all data was available for the safety and efficacy end-points, while three RCTs reported results for BM analysis (6,738 patients). When considering RCTs with trastuzumab administered for 1 year, a significant increased risk of grade III-IV Congestive Heart Failure (CHF) was found in the T-arm, with an AD of 1.61% (p < 0.00001), which translates into 62 treated patients required to harm one (NNH). When considering the asymptomatic L-FEV reduction, a significant increased risk of grade significant L-FEV reduction was found in the T-arm, although significantly heterogeneous, with an AD of 7.20% (p < 0.00001), which translates into 14 NNH. The incidence of BM was significantly higher in the T-arm, without significant heterogeneity, with an AD of 0.62 (p = 0.033), which translates into 161 NNH. The DFS, DDFS, and OS were significantly better in the T-arm, with an AD of 6.00, 4.80 and 1.96%, which translates into 16, 21 and 51 NNT, respectively.

Conclusions: The overall outcome results show that trastuzumab is one of the most important discoveries in oncology. Nevertheless, the biological activity of trastuzumab needs to be investigated more extensively to explore both long-term safety and specific relapse patterns.