Pyrotinib versus pertuzumab plus trastuzumab and taxane as first-line therapy for HER2-positive advanced breast cancer: A retrospective cohort study using propensity score overlap weighting.

Abstract

e13024 Background: Trastuzumab–pertuzumab–taxane (THP) and trastuzumab–pyrotinib–taxane (PyHT) are both accessible first-line options for HER2-positive advanced breast cancer (HER2+ ABC) in China. However, direct comparative evidence between these strategies remains limited, particularly regarding central nervous system (CNS) evolution—whether introducing a TKI-containing regimen upfront can delay the development of new brain metastases. Methods: We conducted a single-center retrospective cohort study at the First Affiliated Hospital of Xi'an Jiaotong University including patients with HER2+ ABC treated with first-line THP or PyHT (Jan 2018–Jan 2025). The primary endpoint was progression-free survival (PFS). Secondary endpoints included overall survival (OS), objective response rate (ORR), brain metastasis–free survival (BMFS), and adverse events (AEs). BMFS was evaluated among patients without baseline brain metastases and defined as the time from treatment initiation to radiologically confirmed new brain metastases (CT/MRI) or death. To emulate a randomized comparison in the overlap population without discarding patients, propensity score overlap weighting (OW) was applied. Robustness was assessed by sensitivity analyses. Results: A total of 143 patients were included (THP n = 93; PyHT n = 50) with a median follow-up of 43.0 months. Data cutoff was Nov 1, 2025. All covariates achieved adequate balance after OW (absolute standardized differences < 0.10). After OW analysis, PFS and OS did not differ significantly between groups, while PyHT was associated with a higher ORR (84.1% vs 55.6%). Among patients without baseline brain metastases (n = 125), PyHT was associated with longer BMFS (HR = 0.47；95% CI 0.23–0.96). PyHT was associated with a higher burden of grade ≥3 AEs (36.0% vs 22.6%). Grade ≥3 diarrhea occurred in 30.0% with PyHT, whereas THP was characterized mainly by hematologic toxicity. Conclusions: This long-term retrospective study suggests comparable systemic disease control (PFS/OS) between THP and PyHT. PyHT was associated with a higher response rate and a signal toward delayed CNS evolution (longer BMFS ), at the cost of increased gastrointestinal toxicity. These hypothesis-generating findings may inform individualized first-line selection and support prospective validation focusing on CNS prevention. Overlap-weighted efficacy outcomes of first-line THP vs PyHT. THP (n=93) PyHT (n=50) Effect estimate (PyHT vs THP) Median PFS (mo) 17.0 25.0 HR 0.82 (95% CI 0.55–1.24); p=0.55 Median OS (mo) 83.0 75.0 HR 0.76 (95% CI 0.37–1.55); p=0.60 ORR (%) 55.6 84.1 p=0.03 Median BMFS (mo)* 68.0 80.0 HR 0.47 (95% CI 0.23–0.96); p=0.04 *BMFS evaluated among patients without baseline brain metastases (n=125; THP n=81; PyHT n=44).