Others have observed the marked differences in clinicopathol
Others have observed the marked differences in clinicopathological characteristics between breast cancer subtypes by IHC study. However, they reported that triple negative and HER2 patients are at high risk of recurrences after adjusting for age, stage and histological grade. Our data support this observation only at the 5-year follow-up, but not at the 10-year follow-up. In our study, the poorest survival is observed in luminal B-like subtype (Fig. 3).
The factors associated with poor prognosis in each subtype are different. In luminal-like patients the risk factors associated with breast cancer recurrence are age ≤40, nuclear grade III, and tumor size >1.5 cm. The hazard ratios for each factor are 2.2, 2.7, and 1.8, respectively (Table 3). The hazard ratio for tumor size >1.5 cm is more significant than using tumor 1 cm as the cutoff value (data is not shown here). There are no risk factors identified in HER2-positive and triple negative patients. These observations indicate that the prognoses in each subtype should be evaluated separately. Xue et al in southern China have reported similar findings, indicating that the prognostic significance of clinicopathological factors may differ among subtypes.
For the triple negative subtype, patients with tumor size ≦1 cm or nuclear I–II patients had excellent outcomes (recurrent rate 1/43 and 1/36, respectively), however, the patient number is too small for delivery of a conclusion. ln the literature, triple negative breast cancer is regarded as an aggressive disease that affects the young patient population. Our patients in this study are mainly over 50 years of age (Table 1). Ethnic differences may exist between Asian and Caucasian patients. The patient number in this study is not high enough to detect risk factors of recurrence. Tumor infiltrating prostanoid receptors are prognostic in triple negative subtype according to a recent study. Further study and investigation may use this new factor to identify high-risk patients.
The factors associated with poor prognosis in unclassified subtype were age ≤40, tumor size >1.5 cm, and no adjuvant hormonal therapy (Table 3). These observations are understandable because this group of patients was mainly HR positive (84%) and HER2 indeterminate (Table 1). The unclassified subtype has the poorest 10-year RFS (86.2%) among the three major subtypes. This subtype is similar to luminal B-like subtype, of which the 10-year RFS was only 81.7% (Fig. 2). Luminal B breast cancer has lower expression of hormone receptors, higher expression of proliferation markers, and higher tumor grade than luminal A. It also exhibits worse prognosis and has a distinct profile of response to hormone therapy and chemotherapy.
Extended hormonal therapy has become the standard of care for HR positive breast cancer patients. We have observed substantial late recurrences in luminal-like subtype, but none in the HER2 subtype (52% HR positive) (Fig. 2). Factors examined to date are related to early recurrence, while those related to late recurrence remain unclear. Progesterone receptor positivity and lymph node metastases significantly correlated with late recurrence. Using molecular markers, Breast Cancer Index, which consisted of two independent biomarkers, HOXB13:IL17BR (H:I) and the 5-gene molecular grade index, was the only significant prognostic test for risk of late distant recurrence. Our study observed no late recurrence after the initial treatment in 141 HER2 (+)/HR (+) patients (Table 1). Extended hormonal therapy for this subtype is worthy of further investigation.
The major limitation of our observations are that they take place in a hospital-based patient population of the Asian ethnicity, where premenopausal women are predominant. Whether there is a presence of racial difference is unknown. Validation by other institutions, especially in western countries, is necessary.
Conflict of interest