Association of human epidermal growth factor receptor 2 expression with 21-gene recurrence score and survival among patients with estrogen receptor–positive breast cancer.

Abstract

563 Background: Trastuzumab-deruxtecan was recently shown to improve survival for metastatic human epidermal growth factor receptor 2 (HER2)-low breast cancer. HER2-negative breast cancer has been redefined as a group of heterogeneous entities. However, HER2-low status is not considered as a prognostic factor among patients with nonmetastatic breast cancer, and the role of HER2 expression levels in the setting of 21-gene recurrence score (RS) remains unclear. We performed an observational cohort study using a national oncology database to evaluate the association of HER2 expression levels with RS and survival among patients with estrogen receptor (ER)-positive breast cancer. Methods: The National Cancer Database was queried for women diagnosed between 2006 and 2018 with ER-positive, pT1-3N0-1aM0 breast cancer who received surgery followed by adjuvant endocrine therapy with or without chemotherapy. HER2 expression levels were stratified by 3 groups: HER2-0 (immunohistochemistry [IHC] 0), HER2-1 (IHC +1), and HER2-2 (IHC +2 and in situ hybridization [ISH] negative). Logistic and Cox multivariable analysis (MVA) were used to identify factors associated with high RS (26 or higher) and overall survival (OS), respectively. Interaction term was analyzed to evaluate any heterogeneous associations in survival outcomes between HER2 levels and RS. Subgroup analysis was performed by repeating Cox MVA stratified by RS. Holm-Bonferroni correction was used to adjust for multiple comparisons, with p value less than 0.05 considered as statistically significant. Results: Among 108,973 women, 31,738 (29.1%), 52,839 (48.5%), and 24,396 (22.4%) patients had HER2-0, HER2-1, and HER2-2, respectively. Logistic MVA showed that, compared to HER2-0, HER2-2 was associated with higher RS (adjusted odds ratio [aOR] 1.09, 95% confidence interval 1.04-1.15, p < 0.001), but HER2-1 was not (aOR 0.96, 95% CI 0.91-1.00, p = 0.05). Cox MVA showed that both HER2-1 (adjusted hazards ratio [aHR] 1.03, 95% CI 0.97-1.10, p = 0.34) and HER2-2 (aHR 1.06, 95% CI 0.98-1.14, p = 0.14) were not associated with OS. There was an interaction between HER2 expression levels and RS (interaction p = 0.03). Among those with RS < 26, compared to HER2-0, HER2-2 was associated with worse OS (aHR 1.10, 95% CI 1.01-1.20, p = 0.02), but HER2-1 was not (aHR 1.04, 95% CI 0.97-1.12, p = 0.24). Among those with RS 26 or higher, both HER2-1 (aHR 1.00, 95% CI 0.88-1.13, p = 0.98) and HER2-2 (aHR 0.94, 95% CI 0.82-1.09, p = 0.41) were not associated with OS. Conclusions: To our knowledge, this is the largest study using nationwide oncology database to suggest that, when compared to HER2-0, HER2-2 was associated with higher RS and poor survival among patients with RS < 26. Further studies would be warranted to evaluate the role of HER2 expression levels for risk stratification to tailor interventions.