No evidence of disease versus residual disease in long-term responders to first-line HER2-targeted therapy for metastatic breast cancer

Background Long-term response to HER2-targeted therapies is infrequent in metastatic breast cancer (MBC). We evaluated clinical characteristics of HER2-positive MBC patients with no evidence of disease (NED) vs residual disease (RES) experiencing long-term response to first-line HER2-targeted therapy. Methods Patients receiving first-line chemotherapy-trastuzumab (CT) or taxane-trastuzumab-pertuzumab (THP) with response duration >= 2-fold higher than in phase II/III trials (CT [18.2 months]; THP [40.4 months]) were included. Clinical characteristics and radiographic review for NED or RES was evaluated by Cox-regression (hazard ratio; HR) or Kaplan-Meier (log-rank). Characteristics associated with NED were evaluated by logistic regression (Odds; OR). Results From 01/2005-01/2016, N = 103 (4.6%) patients were identified. In multivariate analyses, NED (N = 46) showed improved progression-free (PFS) and overall survival (OS) [p < 0.001] versus RES (N = 57), with high 5-year PFS/OS for NED (93.2%/97.4%) relative to RES (10.6%/61.3%). Premenopausal status (p = 0.006), de-novo metastases (p = 0.002), and no palliative radiotherapy (p = 0.01) were associated with NED. Overall, 6/7 (85.7%) patients with NED were alive and disease-free after discontinuing HER2 treatment (>= 1 year) versus 1/17 (5.9%) with RES. Conclusions Long-term responders with NED have better survival compared to RES. Premenopausal status and de novo metastatic disease are associated with NED. Prospective studies of HER2 therapy discontinuation with NED in MBC are warranted.