3.1 The distribution of breast cancer patients based on receptor and menopausal status in the dataset
I first showed that the percentage of patients with tumors of each receptor status (estrogen receptor (ER), progesterone receptor and human epidermal growth factor receptor 2 (HER2)) is similar between premenopausal and postmenopausal breast cancer patients (Fig. 1, bottom chart; Table 1); although, sample size (n) for postmenopausal breast cancer patients is larger in the dataset (Fig. 1, top chart; Table 1). In this dataset, most patients are ER+, PR + and HER2- (n = 361) (Table 2). The number of patients with triple negative breast cancer (TNBC) is 112 (Table 1). Sample sizes of breast cancer patients with pre- and post-menopausal status are 224 and 689, respectively in this dataset.
3.2 Estrogen receptor (ER)-alpha (ESR1) mRNA expression between tumors from pre- and post-menopausal breast cancer patients based on ER, PR and HER2 status
Next, separately for each group of breast cancer patients with tumors of different single receptor status (6 groups: ER-negative, ER-positive, PR-negative, PR-positive, HER2-negative and HER2-positive), I analyzed the mRNA expression levels of ESR1 (encoding ER-alpha) comparatively between tumors from pre- and post-menopausal patients with breast cancer (Fig. 2). Here, note that, in this dataset, receptor status was identified by IHC (immunohistochemistry), and mRNA expression was quantified by RNA-Seq. I found that, in ER-positive or PR-positive or HER2-negative tumors, ESR1 transcript levels are higher in tumors from postmenopausal women than those from premenopausal women with breast cancer (Fig. 2; p values = 7.3e-10, 1.3e-10 and 0.013, respectively for ER+, PR + and HER2- breast tumors). When analyzed based on the combination of 3 receptor status, ESR1 mRNA expression was higher in tumors from postmenopausal patients compared to premenopausal patients, only in those with hormone receptor (ER and PR)-positive status (ER+, PR+, HER2- or ER+, PR+, HER2+; p values = 2.1e-09 and 0.01, respectively) (Fig. 3). However, it should be pointed that the distribution of breast cancer patients based on the combination of 3 receptor status are highly different; for instance, there are only few patients with tumors of certain combinations such as ER-, PR+, HER + or ER-, PR+, HER- (Fig. 3), making it impossible to perform any statistical analysis based on menopausal status in these groups of patients.
When I analyzed the data to identify the top genes whose transcript levels differ most significantly between tumors from breast cancer patients with pre- and post-menopausal status (not considering receptor status), ESR1 was also the top hit as also previously reported [Liao et al., 2015] (Supplementary Fig. 1, Supplementary Table 1). ESR1 expression was more than 2 fold higher in tumors from postmenopausal breast cancer patients compared to those from patients with premenopausal status (p value < 2.22e-16) (Supplementary Fig. 1, Supplementary Table 1). The other top genes whose expression differ most significantly between breast tumors from patients with pre- and post-menopausal status was DBX2, IGSF21, CLEC3A and GREB1 (Supplementary Fig. 1, Supplementary Table 1).
3.3 Estrogen receptor (ER)-beta (ESR2) mRNA expression between tumors from pre- and post-menopausal breast cancer patients based on ER, PR and HER2 status
Next, I performed the same comparisons between tumors from pre- and post-menopausal breast cancer patients for ESR2 (encoding ER-beta). In contrast to ESR1, mRNA expression of ESR2 was lower in tumors from postmenopausal women than those from premenopausal women with breast cancer, in the same groups of patients (namely those with either ER-positive or PR-positive or HER2-negative, p values = 0.0014, 0.0091 and 0.017, respectively) (Fig. 4). When analyzed based on the combination of 3 receptor status, transcript levels of ESR2 were lower in tumors from postmenopausal patients compared to premenopausal patients, only in those with ER+, PR+, HER2- or ER+, PR-, HER2- (Fig. 5, p values = 0.014 and 0.016, respectively).
3.4 Progesterone receptor (PGR) mRNA expression between tumors from pre- and post-menopausal breast cancer patients based on ER, PR and HER2 status
Later, I compared the mRNA expression of progesterone receptor (PGR) between tumors from pre- and post-menopausal breast cancer patients with tumors of different receptor status (again 6 groups: ER-negative, ER-positive, PR-negative, PR-positive, HER2-negative and HER2-positive) (Fig. 6). I found that PGR transcript levels are lower in breast tumors from postmenopausal women than those from premenopausal women only in those with ER + or PR + status (Fig. 6, p values = 0.0064 and 0.0056 for ER + and PR + tumors, respectively). When analyzed according to the combination of 3 receptor status, there was no significant change in the mRNA expression of PGR between tumors from pre- and post-menopausal women with breast cancer (Fig. 7, all p values > 0.05).
3.5 Human epidermal growth factor receptor 2 (HER2 / ERBB2) mRNA expression between tumors from pre- and post-menopausal breast cancer patients based on ER, PR and HER2 status
Lastly, I analyzed the mRNA expression of ERBB2 gene (encoding HER2) between tumors from pre- and post-menopausal breast cancer patients with tumors of different receptor status. However, in any of the 6 groups of patients, ERBB2 transcript levels did not show any statistically significant difference between tumors from pre- and post-menopausal breast cancer patients (Fig. 8). However, when analyzed according to the combination of 3 receptor status, I found that the mRNA expression of ERBB2 is lower in breast tumors from postmenopausal women than those from premenopausal women only in these patients with ER-, PR-, HER2 + status (Fig. 9, p value = 0.02).