Jonathan Verma, MD, Deukwoo Kwon, PhD, Isildinha M. Reis, PhD, Jean L. Wright, MD, Cristiane Takita, MD; Sylvester Comprehensive Cancer Center, University of Miami Miller School of Medicine
Background: The use of molecular profiling in breast cancer for local risk assessment and local therapy decision is not well understood. The goal of this study is to assess locoregional outcomes associated with breast cancer subtype in stage III locally advanced breast cancer (LABC) patients after surgery, radiation therapy (RT), and modern systemic therapy.
Materials and Methods: We retrospectively collected the data from patients treated at our institution for stage III breast cancer from 1998–2008. Biologic subtype was approximated using receptor status: luminal A (estrogen receptor positive [ER+] or progesterone receptor positive [PR+], human epidermal growth factor receptor 2 negative [HER2−]), luminal B (ER+ or PR+, HER2+), basal (ER−, PR−, HER2−), and HER2 (ER−, PR−, and HER2+). Progression-free survival (PFS) and overall survival (OS) were estimated by the Kaplan-Meier method. The method of cumulative incidence allowing for competing risk was applied to the first failure event categorized as locoregional (LRR), distant metastasis (DM), simultaneous (LRR + DM within 4 months of each other), and death with no evidence of diseae (NED). Local control (LC) was defined as the complement of the combined rate of LRR and simultaneous LRR + DM failures. Log-rank test was used to compare OS and PFS by biologic subtype, and Gray’s test was used to compare incidence rates of failure events by biologic subtype. Prognostic effects relative to first failure and cause-specific death were evaluated by fitting competing risk regression models by the method of Fine and Gray.
Results: The data for 188 patients were reviewed, with a median follow-up of 74.9 months (range: 7–205 mo). The median age at diagnosis was 49 years. The subtypes included luminal A (51.6%), luminal B (10.6%), HER2 (11.7%), and basal (26.1%). Tumor grade was high in 63.8% of patients. Seventy-eight percent of patients had mastectomy, and 22% had lumpectomy. Axillary surgery was performed in all patients. Surgical margins were negative in 95.7%. Further, 93.6% of the patients received chemotherapy—adjuvantly (43.1%), neoadjuvantly (23.9%), or both (26.6%). Hormone therapy was given to 53.7% of the patients. Traztuzumab was used in 90% of patients with HER2+ disease. All patients received RT. The first failure event was LRR in 19 patients (10.1%), DM in 22 (11.7%), LRR + DM in 5 (2.7%), and deaths with NED in 9 (4.8%). The 8-year locoregional control rates were 88.4%, 90%, and 78.9% in patients with luminal A, luminal B/HER2, and basal tumors, respectively. The 8-year OS was 84%, 92.4%, and 68.8% in patients with luminal A, luminal B/HER2, and basal tumors, respectively (P = .015). In the multivariate analysis, the significant predictors for worsened locoregional recurrence were basal subtype (hazard ratio [HR] = 3.26; P = .017), tumor size > 5 cm (HR = 2.93; P = .012), and positive nodal status (HR = 5.38; P = .02). The predictors for breast cancer-specific mortality included basal subtype (HR = 6.0; P < .001), size > 5 cm (HR = 3.44; P = .008), and positive nodal status (HR = 27; P = .003).
Conclusion: Basal subtype, tumor size > 5 cm, and nodal positivity were predictive of LRR and breast cancer-specific mortality in this cohort of patients with stage III LABC treated with modern multimodality therapy. Prospective studies are needed to improve locoregional control and survival in patients with the basal subtype.
Proceedings of the 96th Annual Meeting of the American Radium Society — americanradiumsociety.org