Molecular Subtype May Be More Associated With Prognosis and Chemotherapy Benefit Than Tumor Size in T1N0 Breast Cancer Patients: An Analysis of 2,168 Patients for Possible De-Escalation Treatment

Abstract

Purpose: Breast cancer (BC) patients with T1N0 tumors have relatively favorable clinical outcomes. However, it remains unclear whether molecular subtypes can aide in prognostic prediction for such small, nodal-negative BC cases and guide decision-making about escalating or de-escalating treatments.

Patients and methods: T1N0 BC patients diagnosed between 2009 and 2017 were included and classified into three subgroups according to receptor status: 1) hormonal receptor (HR)+/human epidermal growth factor receptor-2 (HER2)-; 2) HER2+; and 3) triple negative (TN) (HR-/HER2-). Patients' characteristics and relapse events were reviewed. Kaplan-Meier analysis and Cox regression were used to assess the iDFS and BCSS. The effects of risk factors and adjuvant treatment benefits were evaluated by calculating hazard ratios (HRs) for invasive disease-free survival (iDFS) and breast cancer-specific survival (BCSS) with Cox proportional hazards models.

Results: In total, 2,168 patients (1,435 HR+/HER2-, 427 HER2+, 306 TN) were enrolled. The 5-year iDFS rates were 93.6, 92.7, and 90.6% for HR+/HER2-, HER2+, and TN patients, respectively (P = 0.039). Multivariate analysis demonstrated that molecular subtype (P = 0.043), but not tumor size (P = 0.805), was independently associated with iDFS in T1N0 BC. TN patients [HRs = 1.77, 95% confidence interval (CI) = 1.11-2.84, P = 0.018] had a higher recurrence risk than HR+/HER2- patients. Adjuvant chemotherapy benefit was not demonstrated in all T1N0 patients but interacted with molecular subtype status. TN (adjusted HRs = 2.31, 95% CI = 0.68-7.54) and HER2+ (adjusted HRs = 2.26, 95% CI = 0.95-5.63) patients receiving chemotherapy had superior iDFS rates. Regarding BCSS, molecular subtype tended to be related to outcome (P = 0.053) and associated with chemotherapy benefit (P = 0.005).

Conclusion: Molecular subtype was more associated with disease outcome and chemotherapy benefit than tumor size in T1N0 BC patients, indicating that it may guide possible clinical de-escalating therapy in T1N0 BC.

Keywords: breast cancer; chemotherapy benefit; de-escalating therapy; molecular subtype; prognosis.

Figure 1

Identification of the study population.

Figure 2

Kaplan-Meier analysis for invasive disease-free survival (iDFS) and breast cancer-specific survival (BCSS) according to tumor size, molecular subtype, and chemotherapy administration. (A) iDFS of T1a, T1b, and T1c tumors (P = 0.268). (B) iDFS of the HR+/HER2−, HER2+, and TN groups (P = 0.039). (C) iDFS of patients who received chemotherapy and those who did not (P = 0.681). (D) BCSS of T1a, T1b, and T1c tumors (P = 0.635). (E) BCSS of the HR+/HER2−, HER2+, and TN groups (P = 0.053). (F) BCSS of patients who received chemotherapy and those who did not (P = 0.898).

Figure 3

Exploratory analyses of invasive disease-free survival (iDFS) and breast cancer-specific survival (BCSS) rates according to patient characteristics and tumor subtype. (A) Forest plot of the hazard ratios of the iDFS rates of patients receiving chemotherapy compared with patients not receiving chemotherapy. (B) Forest plot of the hazard ratios of the BCSS rates of patients receiving chemotherapy compared with patients not receiving chemotherapy.

Figure 4

Forest plot of unadjusted and adjusted hazard ratios of invasive disease-free survival (iDFS) and breast cancer-specific survival (BCSS) rates according to molecular subtype. The unadjusted estimates were from the Cox models with only the exposures of interest. The adjusted models were estimated by incorporating factors that would influence a clinician’s decision to offer adjuvant chemotherapy: molecular subtype, age, tumor grade, and tumor size.