Introduction The assessment of hormonal receptor HR

Introduction The assessment of hormonal receptor (HR) and human epidermal growth factor receptor-2 (HER2) status by immunohistochemistry (IHC) groups breast cancer into three major subtypes: 1) luminal-like subtype (HR positive/HER2 negative); 2) HER2 subtype (HER2 positive and HR negative/positive); and 3) triple negative subtype (HR negative/HER2 negative). Luminal-like is the most common breast cancer subtype with with a superior survival rate than HER2 and triple negative subtypes; however, little is known about the failure patterns or long-term outcomes of each subtype. Unlike other breast cancer subtypes, more than half of the recurrences in luminal-like subtype occur 6 or more years after treatment. Therefore, long-term survivorship follow-up in large patient databases is required for the development of better therapeutic decisions and follow-up programs. In this study, we examined the prognostic factors surrounding the 10-year breast cancer recurrence-free survival (RFS) rate in each subtype based on the QVDOPh that each breast cancer subtype would have different clinical behavior and failure patterns.
Material and methods All treatment decisions are made based on the Breast Cancer Clinical Practice Guidelines developed in our hospital since 1993. These are revised annually in ways similar to those revisions for the guidelines developed by the National Cancer Center Network (NCCN). In principle, patients with HR positive status would have adjuvant hormonal therapy, and patients with primary tumor >1 cm and grade 2–3 would be given adjuvant chemotherapy. Patients with breast-conserving surgery would have adjuvant radiotherapy. SAS programs were written for data input, data management, data quality control, analysis, and presentation. For data quality control, our pathologists recorded pathological information on the pathology review form when the surgical specimen was available. The clinicians audited the charts to ensure that the clinical information entered into the Breast Cancer Data Base was accurate. All data entries were done twice by two independent data processors. On-line logic check was available when the data was first entered. We also regularly performed logic analysis between data forms and within of each form.
Results Of the 5508 newly diagnosed breast cancer patients treated in our hospital in the period of 1990 to 2008, 1595 (29.0%) patients with T1N0 (stage I) disease were included in this study with the last follow-up date set at August 31, 2011. The median follow-up interval was 67.8 months. Fig. 1 shows the distribution of breast cancer subtypes in stage I patients. Luminal-like subtype amounted to 51.4%, HER2 subtype 17.1%, triple negative 10.7%, and unclassified 20.8% of the patients. Table 1 shows patient characteristics by subtypes. More than 40% of the luminal-like and HER2 patients were between the ages of 41–50, while only 29.2% of the triple negative subtype, which was the age of usual diagnosis at >50 years (45.6%) (p = 0.0015). The most shared histology in this subtype was infiltrating ductal carcinoma (97.4%, p < 0.0001). Triple negative subtype was seen as more frequently associated with primary tumor size >1 cm (74.9%). HER2 and triple negative subtypes were more associated with high-grade tumors (nuclear grade and histological grade 3 than the luminal-like subtype (p < 0.0001). HER2 patients were seen most commonly with multifocal or multicentric tumors, and consequently, patients with this subtype had fewer breast conserving surgeries than other subtypes (30.5% versus 50.1–50.3%, p < 0.0001). Adjuvant chemotherapy was administered to the HER2 and triple negative groups (61.8% and 73.1%) more than the luminal-like group (46.2%, p < 0.0001).
Discussion Three major subtypes with different clinical and pathological characteristics share similar outcomes after 10-year follow-up; the differences mainly lie in age at diagnosis, tumor grading, lymphovascular invasion, and multiple foci of tumors (Table 1). The 10-year RFS rates between luminal-like, HER2 and triple negative subtypes range from 89.5% to 92.9% (Fig. 1). The observation of 10-year RFS showing no difference among subtype could be explained by the following factors: (1) patients with triple negative and HER2 subtypes had more chemotherapy than those with luminal-like subtype (Table 1); (2) 37% of patients with HR positive had incomplete adjuvant hormonal therapy, i.e. less than 4-year treatment (Table 4). It has been confirmed that a statistically significant prolongation of RFS related to longer treatment duration was observed (HR, 0.74; 95% CI, 0.59 to 0.93) in ER-positive patients. The 10-year RFS in luminal-like subtype could be improved if 10-year adjuvant hormonal therapy became the standard of care as recommended by ASCO Adjuvant Endocrine Therapy Guidelines 2014. However, when the observation period is shortened to 5 years, the luminal-like breast has the highest 5-year RFS among the three major subtypes (95.6% vs. 92.9% and 93%, p = 0.64). The RFS in luminal-like subtype after 5 years shows a decline. There is no decline of RFS after 5-year follow-up in the HER2 subtype (Fig. 1), for which late recurrence is uncommon, indicating that the benefit of extended hormonal therapy for HER2-positive patients should be re-evaluated.