Molecular Subtypes and Prognostic Factors among Premenopausal and Postmenopausal Thai Women with Invasive Breast Cancer: 15 Years Follow-up Data

Background and Purpose: This study focused on molecular subtypes and prognostic factors for survival of pre- and post-menopausal breast cancer patients. Methods: A retrospective cohort study was performed on 523 patients with invasive carcinoma of the breast treated at Ubon Ratchathani Cancer Hospital, Thailand from 2002 to 2016. Patient characteristics were collected based on a systematic chart audit from medical records. Prognostic factors were performed by observe survival analysis. A Cox regression model was used to calculate hazard ratios of death, taking into account the age and menopause status, molecular subtype, stage of disease, histological grade, lymphatic and vascular invasion, resection margin, hormone receptor expression, and treatment modality. Results: The median time from the diagnosis of invasive breast cancer to the last follow-up or death was 10.2 [95% CI = 9.28-11.95] years in premenopausal women, and 7.4 [95% CI = 6.48-8.44] years in postmenopausal cases. The overall survival estimates at 5 and 10 years for younger woman of 71.2% and 51.8% respectively, appeared slightly better than the 68.3% and 40.9% for postmenopausal women [HRadj = 1.27, 95% CI =0.99-1.63]. In the multivariate analysis, 3 prognostic indicators significantly predicted a worse overall survival in premenopausal patients, triple negative subtype [HRadj = 6.03, 95% CI = 1.94-18.74], HER2-enriched status [HRadj = 4.11, 95% CI = 1.59-10.65] and stage III [HRadj = 2.73, 95% CI = 1.10-6.79]. Statistically significant increased risk of death in postmenopausal patients was noted for only chemotherapy after mastectomy [HRadj = 8.76, 95% CI = 2.88-26.61], and for a Luminal B status [HRadj = 3.55, 95% CI = 1.47-8.53]. Conclusion: Postmenopausal women with invasive breast cancer experience a significantly shorter survival than do their premenopausal counterparts. The predictors of worse overall survival were molecular subtype, stage of disease and type of treatment administered.


Introduction
The most common type of cancer and the most common cause of cancer-related death in women is breast cancer. Invasive breast cancer is the most common type of malignancy in women worldwide (Wu et al., 2016) and also in Thailand, with estimated incidence rate of and distant recurrence, and subsequent mortality (Brenner et al., 2016). Comorbidities are associated with decreased survival in older women with BC (Dialla et al., 2012).
Included in prognostic factors from previous studies were age, positive nodes, large tumor size (Plichta et al., 2016;Fayer et al., 2016), estrogen receptor negativity and tumor stage at presentation (Hartley et al., 2006), molecular subtype expression, especially triple-negative status (Koca et al., 2014 ;Keegan et al., 2013), HER2-enrichment (Kongsiang et al., 1014), and lymphovascular invasion (Mohammed et al., 2013). One of the biggest challenges in today's breast cancer care is to adjust adjuvant treatment, according to both tumor and patient characteristics, for optimal treatment outcomes. Patients survival remains a significant and challenging medical problem (Dialla et al, 2012;Koca et al, 2014).
Despite improved knowledge of prognostic factors, there are no data about the compared difference expression molecular subtype and prognostic factors of overall survival among premenopausal and postmenopausal women with BC in a whole population.
The aim of this study was to investigate differences in molecular subtype and other prognostic factors in premenopausal and postmenopausal women with invasive breast cancer. The hope was that the findings might help in making recommendations for future improvements in early diagnosis and appropriate treatment.

Patients and methods Patient Sample selection and Follow-up
The process for selection of eligible case is illustrated in Figure 1. Of the 1,949 initially considered, 1,077 were excluded because they were not treated in our cancer hospital Of the remaining 869,346 were excluded because they had non-invasive carcinoma (183 patients), experienced menopause before 51 years of age (29 patients), were aged more than 70 years (55 patients), had undergone TAH surgery with BSO (13 patients), were stage IV (25 patients), or had incomplete information (41 patients). The resulting sample cohort (523 patients) consisted of 274 premenopausal and 249 postmenopausal cases.
Patient follow-up commenced at the date of primary invasive breast cancer diagnosis and start of treatment at Ubon Ratchathani Cancer Hospital, Thailand between January 1, 2002 and December 31, 2011. It ended at the date of death, or end of the study (December 31, 2016). Retrospective review of medical records identified characteristics of the patients at presentation, including age, menopause status, molecular subtype -1) luminal A (estrogen-receptor-positive and/or progesterone-receptor-positive [ER+ and/or PR+] and negative for human epidermal growth factor receptor 2 [HER2-]), 2) luminal B (ER+ and/or PR+/ HER2+), 3) HER2-enriched (ER-and PR-/HER2+) and 4) triple-negative (ER-and PR-/HER2-), stage at diagnosis, histological type and grade, histological grading, lymphatic and vascular invasion, resection margin, hormone receptor status, treatment modality and time of the year.

Survival analyses
The outcome of interest was death due to breast cancer, and analyses of long-term breast cancer-specific survival (15 year) by prognostic factors, for comparison with previous studies. Survival was expressed as the number of years from the date of primary diagnosis to the occurrence of an event, and Kaplan-Meier analyses were performed by menopause status and prognostic factors. Significance was assessed using the log-rank test. A p-value <0.05 was considered statistically significant.
Multivariable analysis by molecular subtype was performed using Cox proportional hazard modeling adjusting for classical patient and tumor characteristics (stage of disease, histological grade, presence of lymphatic and vascular invasion, negative resection margin, hormone receptor status, and treatment modality). All factors achieving significance in the univariate analysis were integrated as covariates in the multivariate models. Independent prognostic factors were selected in a stepwise (forward conditional) fashion with use of the maximum likelihood ratio, a p-value of 0.05 or more being adopted as the criterion for exclusion from the final model. All analyses were conducted using STATA version 10.0 (StataCorp LP, 2007).

Ethics
The study was conducted in accordance with the current revision of the Declaration of Helsinki of the World Medical Assembly, and in conformity with regulations concerning clinical trials under the ICH Good Clinical Practice Guideline. The study protocol and all relevant amendments were reviewed and approved by the Khon Kaen University Ethics Committee for human research (No. HE602003) and approved by the Ubon Ratchathani Cancer Hospital Ethics Committee (No.EC001/2017).

Discussion
The results of the present study showed that risk of death overall in younger women (27 to 50 years of age: premenopausal) was marginally greater than in postmenopausal women (51 to 70 years of age) (not statistically significant) after adjustment for all known prognostic factors. Bharat et al., (2009) previously reported that younger women were more likely to die from breast cancer than older women. Age and menopausal status have no significant effect on overall survival of breast cancer patients in Pakistan (Mahmood et al., 2015).The reasons for the improved outcomes observed in our study are likely multifactorial. Such as better staging also allowed us to exclude patients with metastatic disease at diagnosis who may have been included in prior series of young breast cancer patients, the benefit of digital mammography for early screening as well as the increasing use of breast MRI in their preoperative workup (data not shown).
Several studies published before 2010 found young age at diagnosis to be an independent predictor of poor prognosis (Bharat et al., 2009 ;Gnerlich et al., 2009). After that, outcomes for women with breast cancer at age < 40 years have improved (Plichta et al., 2016). For example, Beadle et al., (2009) found OS to be 64.6 % at 10 years, Anders al., (2009) found overall mortality (RR = 1.50, P <.04) . Plichta et al., (2016) found OS to be 86.5 % at 10 years. In our study, OS was 51.8 % at 10 years. Which, suggesting that recent improvements in diagnosis and treatment have translated into improved outcomes for young women. And suggesting that BC survival in the developed countries has continually improved and has largely been accredited to the use of mammography, advanced surgical procedures, and adjuvant therapies. These technological and diagnostic advances, suggesting important opportunities for early detection and treatment and be the cause of increased opportunities for improved survival.
Regarding molecular subtypes on overall survival, we found that survival was highest for patients with luminal-A cancers, followed by those with luminal B cancers. This confirms the better outcomes in hormone-receptor-positive cancers reported in other jurisdictions (Bauer et al., 2007;Yang et al., 2007;Kennecke et al., 2010;Voduc et al., 2010;Fallahpour et al., 2017).
Younger age with molecular subtype expression has been shown in several studies to be an independent predictor of adverse outcome. Keegan et al., (2013) found young women have poorer survival after breast cancer than do older women 40 to 64 years, adolescent and young adult 15 to 39 years. Women diagnosed with HR-/HER + and triple-negative breast cancer experienced a 1.6-fold and 2.7-fold increased risk of death, respectively, from all causes [HR-/HER + hazard ratio:  (Chuangsuwanich et al., 2014). HER-2 and basal-like subtypes were likely to have high tumor grade, HER-2 subtype larger, had higher number of nodal involvement (Chuthapisith et al., 2012).
Our findings provide evidence to suggest that an interaction between premenopausal status breast cancer and molecular subtype suggests high risk triple-negative breast cancers, which are less common, have substantially worse prognosis than Intermediate risk HER2-enriched (ER-and PR-/HER2+) Intermediate risk Luminal B (ER+ and/or PR+/HER2+) and low risk Luminal A[ER+ and/or PR+] breast cancer. For postmenopausal status, high risk in Luminal B (ER+ and/or PR+/HER2+).
The strength of this study is that the analyses were done in a heterogeneous and exhaustive group of patients, using data from systematic chart audit of demographic features and medical record from Ubon Ratchathani Cancer Hospital. Therefore, the results could be considered representative of BC patients living in the department during this period.
Limitation of the study for 523 patients included in this study were those for whom a complete set of data was available. The exclusion of those with incomplete data is potential source of bias. However, the characteristics of those excluded due to incomplete data were similar to and not statistically different from those with complete data.
In conclusions, postmenopausal women of invasive breast cancer experience a significantly shorter survival than those with premenopausal women. Despite the fact that prognostic factors in premenopausal BC was increased risk of death higher than postmenopausal BC. Moreover the results of this study showed that there are numerous predictors of a worse OS, and that the most important were molecular subtype, stage of disease and type of treatment administered.

Conflict of interest
The authors disclosed no potential conflicts of interest related to the current study.