BACKGROUND: HER2-positive breast cancer (BC) requires anti-HER2 therapy. We aimed to determine whether the expression of the HER2/centromeric probe for chromosome 17 (CEP17) ratio was associated with prognosis in patients with HER2-positive non-metastatic BC. METHODS: 267 HER2-positive BC were enrolled between January 2010 and December 2011. Stabilized inverse probability treatment weighting (sIPTW) was used to balance baseline characteristics. Real-world disease-free survival (DFS) and overall survival (OS) was analyzed. RESULTS: The median follow-up time was 10.3 years (interquartile range: 9.4-10.8 years). HER2/CEP17 ratio of >
7.0 was defined as the HER2 ultra-positive group
a HER2/CEP17 ratio of ≤ 7.0 was defined as the HER2 normal-positive group. After sIPTW adjustment, no differences were observed in DFS and OS when anti-HER2 therapy was unknown, and similarly in the patients who were recorded as not receive trastuzumab (all p >
0.05). Interestingly, HER2 ultra-positive group had a worse DFS than the normal-positive group (hazard ratio [HR] = 2.72, p = 0.02), but there was no difference in OS (p = 0.30) in patients did receive trastuzumab. The multivariate Cox models also showed that the HER2 ultra-positive had worse DFS than HER2 normal-positive patients (HR = 3.71
p <
0.01). CONCLUSION: For non-metastatic HER2-positive BC with or without trastuzumab treatment, the HER2/CEP17 ratio did not predict DFS and OS. However, our study supported that HER2 ultra-positive group had a worse DFS than the normal-positive group among non-metastatic HER2-positive BC patients receiving trastuzumab
therefore, this could be a potential predictor of DFS in these patients.