p-mTOR Expression in Luminal Breast Carcinoma
p-mTOR Expression in Luminal Breast Carcinoma
We were able to include in this cohort 331 tumours with data on p-mTOR expression and outcome information. A summary of the patients' characteristics is provided in Table 1. Tumours numbering 43.8% (43.8%; n=145) were considered positive for p-mTOR expression (figure 2). Information missing was considered missing completely at random due to study design and it is indicated for each variable in Table 1. With regard to the complete cohort, 80.36% of the tumours were considered luminal, and most (74.32%) were histologically classified as invasive carcinomas of no special type (IC-NST), according to the WHO Classification of Breast Tumours. 'Other Special types' category mostly comprised mucinous carcinomas (n=10), tubular carcinomas (n=7) and micropapillary carcinomas (n=4).
(Enlarge Image)
Figure 2.
Two examples of phosphorylated mammalian target of Rapamycin expression: negative expression in a high-grade carcinoma (A, 400×) and positive expression in a low-grade carcinoma (B, 200×).
We noted a significant association between p-mTOR expression and tumour size and grade, with smaller (<2 cm) and lower-grade tumours associated with p-mTOR expression (p=0.021 and p<0.001, respectively). Patients with p-mTOR-expressing tumours also had significantly longer DFS (p<0.001) and OS (p<0.001 with HRs of 0.32 (95% CI 0.18 to 0.54; p<0.001) and of 0.20 (95% CI 0.09 to 0.46; p<0.001), respectively (figure 3; Table 2). With univariable analysis, other predictors of outcome were tumour size, grade and lymph node status (all p<0.001), reflecting the consistency of this dataset. Furthermore, in a multivariable model, the association of p-mTOR expression and longer OS was independent of all the mentioned predictors, and also of molecular subtype (HR=0.026; p=0.002 for OS and HR=0.40, p=0.002 for DFS; Table 3).
(Enlarge Image)
Figure 3.
Phosphorylated mammalian target of Rapamycin (p-mTOR) expression predicts better disease-free survival (DFS) and overall survival (OS) in breast carcinoma patients. Kaplan-Meier curves for DFS (A) and OS (B) according to p-mTOR expression (n=331).
As seen in Table 1, there was a significant association between ER expression, PgR expression and p-mTOR expression (p=0.018 and p=0.021, respectively); this relationship also translated to an association with the molecular subtype of the BC, in that p-mTOR-positive tumours were associated with luminal subtype of BC (p<0.001). Because of these associations, and the finding that ER-positive tumours generally have a better prognosis than ER-negative tumours (in this cohort: OS univariable HR=0.51 for ER negative tumours; p=0.05), we were interested in whether the associations with p-mTOR expression found in the complete cohort were dependent on the molecular subtype of the BC. To this end, and considering the important biologic differences between molecular subtypes, we next conducted an analysis stratified according to the molecular subtype of BC.
When the analysis was conducted stratified on tumour molecular subtype, the clinico-pathological associations, as well as the association with outcome, differed according the tumour subtype. The p-mTOR expression was still associated with smaller (<2 cm) and lower-grade tumours in luminal BC (p=0.003 and p=0.005, respectively), and with lower-grade in TNBC (p=0.011) (see online supplementary table S2 http://jcp.bmj.com/content/67/11/961/suppl/DC1). By contrast, no associations were detected between p-mTOR expression and clinico-pathological variables for human epidermal growth factor receptor-2 overexpressing (HER-2 OE) BCs (see online supplementary table S2 http://jcp.bmj.com/content/67/11/961/suppl/DC1); and although this group showed a better survival for p-mTOR expressing tumours on the Kaplan–Meyer plots (see online supplementary figure S2 http://jcp.bmj.com/content/67/11/961/suppl/DC1), these associations failed to reach statistical significance (log-rank p=0.159 for OS and p=0.76 for DFS). Nevertheless, the limited sample size of HER-2 OE and TNBC cohorts impairs any conclusive comments regarding these two subtypes. For luminal BC, p-mTOR expressing tumours were linked to a significantly longer DFS (p<0.001) and OS (p<0.001), with HR of 0.23 (95% CI 0.12 to 0.47; p<0.001) and of 0.13 (95% CI 0.04 to 0.36; p<0.001), respectively) (figure 4).
(Enlarge Image)
Figure 4.
The phosphorylated mammalian target of Rapamycin (p-mTOR) prognostic value is dependent on tumour molecular subtype and in luminal breast carcinoma subtype predicts better disease-free survival (DFS) and overall survival (OS) in breast carcinoma patients. Kaplan–Meier curves for DFS (A) and OS (B) according to p-mTOR expression in luminal breast carcinoma patients (n=266).
Given the clinico-pathological associations in luminal BC with p-mTOR expression status (see online supplementary table S2 http://jcp.bmj.com/content/67/11/961/suppl/DC1) and other variables' outcome predictor effect in the Cox Proportional Hazards univariable analysis (data not shown), namely HER2 expression, we conducted a multivariable Cox Proportional Hazards model in which we adjusted for size, lymph node status, grade and HER2 overexpression. In this multivariable analysis, p-mTOR-expressing luminal tumours demonstrated a significantly longer DFS and OS, with HR of 0.33 (95% CI 0.16 to 0.68; p=0.003) and of 0.20 (95% CI 0.07 to 0.59; p=0.003), respectively (Table 4), independently of the other included predictors of outcome considered for this subtype.
Results
p-mTOR Expression is Associated With Smaller, Lower-grade Carcinomas and Independently With Better Outcome
We were able to include in this cohort 331 tumours with data on p-mTOR expression and outcome information. A summary of the patients' characteristics is provided in Table 1. Tumours numbering 43.8% (43.8%; n=145) were considered positive for p-mTOR expression (figure 2). Information missing was considered missing completely at random due to study design and it is indicated for each variable in Table 1. With regard to the complete cohort, 80.36% of the tumours were considered luminal, and most (74.32%) were histologically classified as invasive carcinomas of no special type (IC-NST), according to the WHO Classification of Breast Tumours. 'Other Special types' category mostly comprised mucinous carcinomas (n=10), tubular carcinomas (n=7) and micropapillary carcinomas (n=4).
(Enlarge Image)
Figure 2.
Two examples of phosphorylated mammalian target of Rapamycin expression: negative expression in a high-grade carcinoma (A, 400×) and positive expression in a low-grade carcinoma (B, 200×).
We noted a significant association between p-mTOR expression and tumour size and grade, with smaller (<2 cm) and lower-grade tumours associated with p-mTOR expression (p=0.021 and p<0.001, respectively). Patients with p-mTOR-expressing tumours also had significantly longer DFS (p<0.001) and OS (p<0.001 with HRs of 0.32 (95% CI 0.18 to 0.54; p<0.001) and of 0.20 (95% CI 0.09 to 0.46; p<0.001), respectively (figure 3; Table 2). With univariable analysis, other predictors of outcome were tumour size, grade and lymph node status (all p<0.001), reflecting the consistency of this dataset. Furthermore, in a multivariable model, the association of p-mTOR expression and longer OS was independent of all the mentioned predictors, and also of molecular subtype (HR=0.026; p=0.002 for OS and HR=0.40, p=0.002 for DFS; Table 3).
(Enlarge Image)
Figure 3.
Phosphorylated mammalian target of Rapamycin (p-mTOR) expression predicts better disease-free survival (DFS) and overall survival (OS) in breast carcinoma patients. Kaplan-Meier curves for DFS (A) and OS (B) according to p-mTOR expression (n=331).
As seen in Table 1, there was a significant association between ER expression, PgR expression and p-mTOR expression (p=0.018 and p=0.021, respectively); this relationship also translated to an association with the molecular subtype of the BC, in that p-mTOR-positive tumours were associated with luminal subtype of BC (p<0.001). Because of these associations, and the finding that ER-positive tumours generally have a better prognosis than ER-negative tumours (in this cohort: OS univariable HR=0.51 for ER negative tumours; p=0.05), we were interested in whether the associations with p-mTOR expression found in the complete cohort were dependent on the molecular subtype of the BC. To this end, and considering the important biologic differences between molecular subtypes, we next conducted an analysis stratified according to the molecular subtype of BC.
The p-mTOR Expression Clinico-pathological Associations and Outcome Depend on Tumour Subtype
When the analysis was conducted stratified on tumour molecular subtype, the clinico-pathological associations, as well as the association with outcome, differed according the tumour subtype. The p-mTOR expression was still associated with smaller (<2 cm) and lower-grade tumours in luminal BC (p=0.003 and p=0.005, respectively), and with lower-grade in TNBC (p=0.011) (see online supplementary table S2 http://jcp.bmj.com/content/67/11/961/suppl/DC1). By contrast, no associations were detected between p-mTOR expression and clinico-pathological variables for human epidermal growth factor receptor-2 overexpressing (HER-2 OE) BCs (see online supplementary table S2 http://jcp.bmj.com/content/67/11/961/suppl/DC1); and although this group showed a better survival for p-mTOR expressing tumours on the Kaplan–Meyer plots (see online supplementary figure S2 http://jcp.bmj.com/content/67/11/961/suppl/DC1), these associations failed to reach statistical significance (log-rank p=0.159 for OS and p=0.76 for DFS). Nevertheless, the limited sample size of HER-2 OE and TNBC cohorts impairs any conclusive comments regarding these two subtypes. For luminal BC, p-mTOR expressing tumours were linked to a significantly longer DFS (p<0.001) and OS (p<0.001), with HR of 0.23 (95% CI 0.12 to 0.47; p<0.001) and of 0.13 (95% CI 0.04 to 0.36; p<0.001), respectively) (figure 4).
(Enlarge Image)
Figure 4.
The phosphorylated mammalian target of Rapamycin (p-mTOR) prognostic value is dependent on tumour molecular subtype and in luminal breast carcinoma subtype predicts better disease-free survival (DFS) and overall survival (OS) in breast carcinoma patients. Kaplan–Meier curves for DFS (A) and OS (B) according to p-mTOR expression in luminal breast carcinoma patients (n=266).
Given the clinico-pathological associations in luminal BC with p-mTOR expression status (see online supplementary table S2 http://jcp.bmj.com/content/67/11/961/suppl/DC1) and other variables' outcome predictor effect in the Cox Proportional Hazards univariable analysis (data not shown), namely HER2 expression, we conducted a multivariable Cox Proportional Hazards model in which we adjusted for size, lymph node status, grade and HER2 overexpression. In this multivariable analysis, p-mTOR-expressing luminal tumours demonstrated a significantly longer DFS and OS, with HR of 0.33 (95% CI 0.16 to 0.68; p=0.003) and of 0.20 (95% CI 0.07 to 0.59; p=0.003), respectively (Table 4), independently of the other included predictors of outcome considered for this subtype.