We previously observed T-bet+ lymphocytes to be associated with a good prognosis in a cohort of women with familial breast cancer. To validate this finding, we evaluated lymphocyte T-bet expression in an independent unselected prospectively accrued series of women with lymph node–negative breast carcinoma. T-bet and clinicopathologic data were available for 614 women. Hormone receptors, HER2, Ki-67, CK5, EGFR, p53, and T-bet status were determined using IHC and/or biochemical methods. Tumors were assigned to luminal A, luminal B, HER2, and basal subtypes based on the expression of IHC markers. Multiple cutpoints were examined in a univariate penalized Cox model to stratify tumors into T-bet+/high and T-bet−/low. Fisher exact test was used to analyze T-bet associations with clinicopathologic variables, IHC markers, and molecular subtype. Survival analyses were by the Cox proportional hazards model. All tests were two sided. A test with a P value < 0.05 was considered statistically significant. T-bet+/high tumor status was significantly associated with large tumor size, high grade, hormone receptor negativity, CK5, EGFR and p53 positivity, high Ki-67, and basal subtype. With a median follow-up of 96.5 months, T-bet−/low tumor status was associated with a reduced disease-free survival compared with T-bet+/high tumor status in multivariate analysis (P = 0.0027; relative risk = 5.62; 95% confidence intervals, 1.48–50.19). Despite being associated with adverse clinicopathologic characteristics, T-bet+ tumor-infiltrating lymphoid cells are associated with a favorable outcome. This supports their role in Th1-mediated antitumor activity and may provide insight for the development of new therapeutic strategies. Cancer Immunol Res; 4(1); 41–48. ©2015 AACR.

Breast cancer is a complex and heterogeneous disease with significant disparity in clinical outcomes still being seen, despite improvements in disease classification using tumor-related prognostic markers. More recently, attention has been placed on the components of the tumor microenvironment, including lymphocytic infiltration, whose interaction with the tumor can strongly influence the patient's long-term outcome. The interplay between the immune system and cancer is not straightforward: tumor cells can induce an inflammatory microenvironment that is essential for tumor growth (1); yet, components of the immune system play an essential role in both innate and adaptive immunity and are involved in tumor immune surveillance (2, 3).

Early studies that examined the relationship between lymphocytic infiltration in tumors and outcome in breast cancer showed conflicting results, with some reporting a favorable association with outcome and a dense inflammatory infiltrate (4) and others identifying an adverse association (5). More recent results from large cohorts have demonstrated a good prognosis in patients whose breast cancers showed a marked lymphocytic infiltrate as well as an association with high response rates by such tumors to neoadjuvant therapy (4, 6–12). The apparently contradictory findings may reflect the diverse functional roles of the individual components of lymphocytic subtypes (13). While recognition of the individual immune cell subsets that consistently mediate favorable effects remains elusive, the current consensus is that CD4+ Th1 and CD8+ T cells are among the players that can generate effective although potentially attenuated antitumor responses while CD4+ Th2 cells and CD4+ Tregs are among the cells that can suppress antitumor immunity and can promote tumor progression (14–16).

T-bet (T-box transcription factor 21), an immune cell–specific member of the T-box family of transcription factors, is expressed in multiple cells of the innate and adaptive immune system [including dendritic cells, natural killer (NK) cells, CD4+ and CD8+ effector cells, B cells, and a subset of regulatory T cells], and its expression is required for the survival, development, and proper functioning of immune cells (17–23). In disease states, T-bet plays an important role in infectious and inflammatory conditions as well as in tumor progression: in the absence of T-bet, susceptibility to metastases from melanoma was shown to be increased because of impaired NK-cell function and survival in vivo (24). Similarly, in a murine model of lung adenocarcinoma, T-bet deficiency led to a marked induction of tumor load (25). More recently, T-bet has been shown to be required for promoting blockade-induced CD8+ T-cell effector responses sufficient to eradicate disseminated leukemia in an animal model (26). Furthermore, high numbers of T-bet+ intratumoral lymphoid cells have been found to correlate with improved outcome in gastric cancer (27) and in high-grade cervical intraepithelial neoplasia (28).

In a cohort of women with familial breast cancer from the Ontario site of the Breast Cancer Family Registry (29), studied for expression of chemokine CXCL10 and tumoral lymphocytic infiltration, we observed that T-bet+ lymphocytes were associated with the basal molecular subtype as well as with morphologic features characteristic of such tumors, including high-grade, p53 expression, ER negativity, CK5 positivity, and EGFR positivity (30). Despite the association with an aggressive phenotype, T-bet positivity was associated with a good prognosis. To confirm this finding in an independent unselected hospital-based cohort, we assessed the presence of T-bet+ immune cells in a large cohort of axillary node negative (ANN) breast cancer patients and determined the relationship between T-bet positivity and IHC biomarkers, clinicopathologic characteristics, molecular subtypes (luminal A, luminal B, HER2, and basal), and patient outcome. We hypothesized that intratumoral T-bet+ lymphoid cells would correlate with the basal subtype and would be associated with a more favorable outcome.

Patient cohort and clinical follow-up

The patient cohort comprised a prospectively accrued consecutive series of 1,561 women with lymph node–negative, invasive breast carcinoma enrolled at eight Toronto hospitals from September 1987 to October 1996, as previously described (31, 32). This included 887 women on whom paraffin-embedded tissue blocks were available for use in the construction of tissue microarrays (TMA). The characteristics of the whole cohort and TMA cohort have been reported previously (33). Written informed consent was obtained from all study participants.

We followed women in the cohort for recurrence and death. Disease-free survival (DFS) was taken as the time between diagnosis and the confirmation of non-breast recurrence. All patients were monitored for death whether or not they experienced disease recurrence. Using clinical follow-up data, patient status on January 10, 2002, determined DFS time and censoring status. Follow-up data were monitored for an additional 6 months to confirm patient status at the termination date. Excluding the patients lost to follow-up or with distant recurrence, the minimum follow-up time was 56 months after surgery and the median follow-up time was 100 months. Approval of the study protocol was obtained from the Research Ethics Boards of Mount Sinai Hospital (#01-0313-U; Toronto, Ontario, Canada) and the University Health Network (#02-0881-C; Toronto, Ontario, Canada).

Hormone receptor status, tissue microarray construction, and IHC staining

Estrogen receptor (ER) and progesterone receptor (PgR) status were determined biochemically at the time of surgery by ligand-binding assays of frozen tissue, which was the standard approach used at the time, and by IHC detection on the TMA. Formalin-fixed paraffin-embedded tumor blocks were available for 887 patients. Areas of invasive carcinoma were selected from an H&E-stained section of each tumor, and two 0.6-mm cores of tissue were taken from the corresponding areas of the paraffin block. The selected donor cores were embedded in a recipient paraffin block, and 4-μm sections were cut and immunohistochemically stained for ER, PgR, HER2, Ki-67, CK5, EGFR, p53, and T-bet, under the conditions described in Table 1. Microwave antigen retrieval was carried out in a Micromed T/T Mega Microwave Processing Lab Station (ESBE Scientific). Sections were developed with diaminobenzidine tetrahydrochloride and counterstained in Mayer's hematoxylin.

Table 1.

Summary of antibodies and conditions of use

AntibodyCloneDilutionSourcePretreatment
ER 6F11 1/75 Vector Tris buffer (pH 9.0) 
PgR PgR 1294 1/1,000 DAKO Tris buffer (pH 9.0) 
p53 D.07 1/400 ID Lab Tris buffer (pH 9.0) 
CK5 XM26 1/400 Vector Tris buffer (pH 9.0) 
HER2 CB11 TAB250 (cocktail) 1/300 Novocastra Pepsin 10 minutes at 37°C 
  1/300 Zymed  
EGFR 31G7 1/25 Zymed Pepsin 10 minutes at 37°C 
Ki67 MIB1 1/300 DAKO Citrate buffer (pH 6.0) 
T-bet 4B10 1/100 Santa Cruz Biotechnology Citrate buffer (pH 6.0) 
AntibodyCloneDilutionSourcePretreatment
ER 6F11 1/75 Vector Tris buffer (pH 9.0) 
PgR PgR 1294 1/1,000 DAKO Tris buffer (pH 9.0) 
p53 D.07 1/400 ID Lab Tris buffer (pH 9.0) 
CK5 XM26 1/400 Vector Tris buffer (pH 9.0) 
HER2 CB11 TAB250 (cocktail) 1/300 Novocastra Pepsin 10 minutes at 37°C 
  1/300 Zymed  
EGFR 31G7 1/25 Zymed Pepsin 10 minutes at 37°C 
Ki67 MIB1 1/300 DAKO Citrate buffer (pH 6.0) 
T-bet 4B10 1/100 Santa Cruz Biotechnology Citrate buffer (pH 6.0) 

Except for Ki-67 and T-bet, each of the IHC-stained sections was scored using the Allred scoring method (34). Nuclear staining was assessed for ER, PgR, and p53. Strong complete membrane staining was assessed for HER2. Membranous and/or cytoplasmic membrane staining was scored for CK5 and EGFR. The Ki-67 labeling index was determined based on the percentage of positive tumor nuclei, regardless of intensity of staining. In all, 50 nuclei were counted and therefore tissue microarray core samples with <50 tumor cells were deemed unsatisfactory, resulting in exclusion of 220 cases. Absolute counts of T-bet+ lymphoid cells were conducted, and these were categorized as intratumoral [when within the epithelial nests or within close proximity (the distance between positive lymphocyte and tumor nest is equal to or less than the size of one tumor cell)] or peritumoral (at a distance from the epithelial nests).

The raw score data were processed using a TMA deconvoluter software program into a format suitable for statistical analysis (35). As two cores from each tumor were assessed, the larger of the two values was chosen for use in statistical analysis to minimize the effect of false negatives on the array. For ER, PgR, HER2, and p53, cutpoints to define positivity were based on previous validation studies (34, 36–39). For CK5 and EGFR, the cutpoint for positivity was arbitrarily specified as ≥4. Ki-67 was dichotomized into Ki-67 high (Ki-67 labeling index ≥ 14%) or Ki-67 low (Ki67 labeling index < 14%; ref. 40). For T-bet, following the examination of multiple cutpoints, an absolute count of 30 positive intratumoral lymphoid cells (within or within close proximity of the epithelial cell nests) was used as the cutoff for positivity (T-bet+/high). Tumors with lower levels or absence of T-bet+ intratumoral lymphoid cells were considered as T-bet−/low. Interpretable scores were obtained in 618 tumors. In four cases, clinicopathologic characteristics were unavailable, resulting in the inclusion of 614 patients in the final statistical analyses.

Definitions of intrinsic subtypes

Tumors from each group were assigned to molecular subtypes based on previous publications (41–44). Tumors that were positive for HER2 protein overexpression were assigned to the HER2 subtype. Tumors that were negative for HER2 but positive for ER were assigned to the luminal subtype. Tumors that were negative for HER2 and ER, and positive for CK5 or EGFR, were assigned to the basal subtype. The luminal subtype was subsequently subdivided into luminal A and luminal B based on PgR, p53, and Ki-67 labeling index. Tumors that had a Ki-67 labeling index ≥14% and were negative for PgR or positive for p53 were assigned to the luminal B subtype (44).

Statistical analysis

Fisher exact test was used to analyze the T-bet marker associations with clinicopathologic variables, IHC markers (markers used to define molecular subtype), and molecular subtype. Clinicopathologic variables used in analyses represent traditional and/or known prognostic factors for node-negative breast cancer and were chosen based on the literature and on previous prognostic modeling we conducted in this cohort (31–33, 44). Analyses of the association of DFS with T-bet marker status were conducted by the univariate Cox proportional hazards (PH) model (45) with Kaplan–Meier survival curves. To evaluate the additional and independent prognostic contribution of T-bet, a multivariate DFS analysis was carried out controlling for prognostic clinicopathologic factors including HER2 status by the Cox PH model. Relative risks (RR) for each factor were estimated by the HR in the Cox PH model. To deal with the low event rate in the T-bet+/high group, we applied Firth's penalized regression method for the Cox PH model with the penalized likelihood ratio (LR) test (46). All tests were two sided. The cutpoint of T-bet = 30 was chosen by examination of multiple cutpoints (10, 15, 20, 25, 30) in a univariate penalized Cox model. Adjustment of the minimum P value for multiple testing of five correlated tests (47) yielded a strict T-bet significance test criterion of P value < 0.016 for strong family-wise type I error control. A further evaluation of the association between DFS and a three-level categorization (absent, low count, high count) was performed and reported in Supplementary Tables S1 and S2, with RRs and P values for the five cutpoints examined. It is evident in the three-level KM plots (Supplementary Fig. S1A and S1B) that the absent and low count groups are indistinguishable no matter which of the cutpoints is used to define the high- versus the low-count group, and the RRs comparing absent versus high and low versus high are remarkably similar (Supplementary Tables S1 and S2). Further division of the low-count group according to present/absent did not improve discrimination further.

A P value < 0.05 was applied for tests of all other factors. Statistical analyses of associations were performed using SAS 9.1 software (SAS Institute, Inc.). Survival curves were plotted using R statistical software, version 2.15.0 (http://www.r-project.org/).

Frequency and localization of T-bet+ lymphoid cells

T-bet+ lymphoid cells were distributed among defined tumor compartments as intratumoral [when within the epithelial nests or within close proximity (the distance between positive lymphocyte and tumor nest is equal to or less than the size of one tumor cell)] or peritumoral (at a distance from the epithelial nests; Fig. 1). Intratumoral T-bet+ cells were more numerous than were peritumoral cells. The mean, median, and range of intratumoral T-bet+ lymphoid cells were 7.4, 0.0, and 0 to 220, respectively, and the mean, median, and range of peritumoral T-bet+ lymphoid cells were 2.6, 0.0, and 0 to 33, respectively. Intratumoral T-bet+/high lymphoid cells (≥30) were seen in 48 tumors, whereas 566 tumors (92.2%) had low or absent T-bet+ lymphoid cells. We performed analyses for intratumoral T-bet+ cells because this localization proved to be prognostically relevant in our previous study (30).

Figure 1.

Intratumoral T-bet+ lymphoid cells in invasive carcinoma. A, T-bet+/high (×40 magnification). B, T-bet−/low (×40 magnification).

Figure 1.

Intratumoral T-bet+ lymphoid cells in invasive carcinoma. A, T-bet+/high (×40 magnification). B, T-bet−/low (×40 magnification).

Close modal

Clinicopathologic and biologic parameters of intratumoral T-bet+ lymphoid cells

Patients with tumors showing high levels of intratumoral T-bet+ lymphoid cells (T-bet+/high) were more likely to have larger tumors (P = 0.0045), to have higher-grade tumors (P < 0.0001), and to have received chemotherapy (P < 0.0001; Table 2). These patients were also more likely to have ER-negative/equivocal tumors and PgR-negative/equivocal tumors (P < 0.0001; based on biochemical data, Table 2). Menopausal status and lymphovascular invasion did not differ significantly between the patients with T-bet+/high and T-bet−/low tumors (Table 2).

Table 2.

Association between clinical characteristics and T-bet+

T-bet−/lowT-bet+/high
(n = 566)(n = 48)
CharacteristicNumber (%)Number (%)Pa
Number of recurrences 76 (13.4) 1 (2.1)  
Menopausal status 
 Pre 199 (35.1) 22 (45.8) 0.3168 
 Peri 31 (5.5) 1 (2.1)  
 Post 336 (59.4) 25 (52.1)  
Lymphatic invasionb 
 Yes 83 (14.7) 5 (10.4) 0.5236 
 No 482 (85.3) 43 (89.6)  
Tumor size 
 ≤0.5 cm 24 (4.2) 1 (2.1) 0.0045 
 >0.5 to 1 cm 114 (20.1) 1 (2.1)  
 >1 to 2 cm 238 (42.1) 23 (47.9)  
 >2 to 5 cm 171 (30.2) 22 (45.8)  
 >5 cm 19 (3.4) 1 (2.1)  
ERc 
 Positive 361 (63.8) 13 (27.1) <0.0001 
 Negative/equivocal 120 (21.2) 34 (70.8)  
 NDd 85 (15.0) 1 (2.1)  
Progesterone receptorc 
 Positive 323 (57.1) 15 (31.3) <0.0001 
 Negative/equivocal 158 (27.9) 32 (66.7)  
 NDd 85 (15.0) 1 (2.0)  
Histologic grade 
 1 159 (28.1) 6 (12.5) <0.0001 
 2 209 (36.9) 7 (14.6)  
 3 159 (28.1) 30 (62.5)  
 NDd 39 (6.9) 5 (10.4)  
Adjuvant treatmentb 
 Hormonal 264 (46.7) 11 (22.9) <0.0001 
 Chemotherapy 85 (15.0) 22 (45.8)  
 Both 24 (4.3) 0 (0.0)  
 None 192 (34.0) 15 (31.3)  
Age, years 
 Mean 55.23 51.95 0.0656 
 SD 11.76 12.88  
 Minimum 25.49 22.43  
 Maximum 75.82 74.18  
 SEM 0.49 1.86  
T-bet−/lowT-bet+/high
(n = 566)(n = 48)
CharacteristicNumber (%)Number (%)Pa
Number of recurrences 76 (13.4) 1 (2.1)  
Menopausal status 
 Pre 199 (35.1) 22 (45.8) 0.3168 
 Peri 31 (5.5) 1 (2.1)  
 Post 336 (59.4) 25 (52.1)  
Lymphatic invasionb 
 Yes 83 (14.7) 5 (10.4) 0.5236 
 No 482 (85.3) 43 (89.6)  
Tumor size 
 ≤0.5 cm 24 (4.2) 1 (2.1) 0.0045 
 >0.5 to 1 cm 114 (20.1) 1 (2.1)  
 >1 to 2 cm 238 (42.1) 23 (47.9)  
 >2 to 5 cm 171 (30.2) 22 (45.8)  
 >5 cm 19 (3.4) 1 (2.1)  
ERc 
 Positive 361 (63.8) 13 (27.1) <0.0001 
 Negative/equivocal 120 (21.2) 34 (70.8)  
 NDd 85 (15.0) 1 (2.1)  
Progesterone receptorc 
 Positive 323 (57.1) 15 (31.3) <0.0001 
 Negative/equivocal 158 (27.9) 32 (66.7)  
 NDd 85 (15.0) 1 (2.0)  
Histologic grade 
 1 159 (28.1) 6 (12.5) <0.0001 
 2 209 (36.9) 7 (14.6)  
 3 159 (28.1) 30 (62.5)  
 NDd 39 (6.9) 5 (10.4)  
Adjuvant treatmentb 
 Hormonal 264 (46.7) 11 (22.9) <0.0001 
 Chemotherapy 85 (15.0) 22 (45.8)  
 Both 24 (4.3) 0 (0.0)  
 None 192 (34.0) 15 (31.3)  
Age, years 
 Mean 55.23 51.95 0.0656 
 SD 11.76 12.88  
 Minimum 25.49 22.43  
 Maximum 75.82 74.18  
 SEM 0.49 1.86  

aFisher exact test; ND groups were not used in testing.

bOne patient without data.

cFrom pathology reports.

dUnknown, not done, or missing.

IHC biomarkers of intratumoral T-bet+ lymphoid cells

Intratumoral T-bet+/high lymphoid cell infiltration was positively associated with CK5 expression (P < 0.0001), and negatively associated with both ER and PgR expression (P < 0.0001; Table 3). Tumors with high levels of T-bet+ cells were more likely to be positive for p53, EGFR, and Ki-67 (P < 0.0001, P < 0.0001, P < 0.0001, respectively; Table 3). HER2 status did not differ significantly between the two groups (Table 3).

Table 3.

Association between T-bet+ and IHC markers

T-bet−/lowT-bet+/high
(n = 566)a(n = 48)a
MarkeraNumber (%)Number (%)Pb
HER2 
 Negative 500 (92.1) 44 (93.6) 1.0000 
 Positive 43 (7.9) 3 (6.4)  
ER 
 Negative 129 (25.7) 30 (73.2) <0.0001 
 Positive 373 (74.3) 11 (26.8)  
PgR 
 Negative 217 (41.6) 34 (81.0) <0.0001 
 Positive 305 (58.4) 8 (19.0)  
p53 
 Negative 417 (77.5) 24 (50.0) 0.0001 
 Positive 121 (22.5) 24 (50.0)  
EGFR 
 Negative 489 (94.6) 30 (65.2) <0.0001 
 Positive 28 (5.4) 16 (34.8)  
CK5 
 Negative 459 (85.8) 21 (44.7) <0.0001 
 Positive 76 (14.2) 26 (55.3)  
Ki-67 
 <14% 181 (40.9) 2 (4.8) <0.0001 
 ≥14% 262 (59.1) 40 (95.2)  
T-bet−/lowT-bet+/high
(n = 566)a(n = 48)a
MarkeraNumber (%)Number (%)Pb
HER2 
 Negative 500 (92.1) 44 (93.6) 1.0000 
 Positive 43 (7.9) 3 (6.4)  
ER 
 Negative 129 (25.7) 30 (73.2) <0.0001 
 Positive 373 (74.3) 11 (26.8)  
PgR 
 Negative 217 (41.6) 34 (81.0) <0.0001 
 Positive 305 (58.4) 8 (19.0)  
p53 
 Negative 417 (77.5) 24 (50.0) 0.0001 
 Positive 121 (22.5) 24 (50.0)  
EGFR 
 Negative 489 (94.6) 30 (65.2) <0.0001 
 Positive 28 (5.4) 16 (34.8)  
CK5 
 Negative 459 (85.8) 21 (44.7) <0.0001 
 Positive 76 (14.2) 26 (55.3)  
Ki-67 
 <14% 181 (40.9) 2 (4.8) <0.0001 
 ≥14% 262 (59.1) 40 (95.2)  

aIHC marker data are not available for some tumors.

bBy Fisher exact test.

Molecular subtypes of intratumoral T-bet+ lymphoid cells

In the 48 tumors with high amounts of T-bet+ lymphoid cells, a molecular phenotype was assignable for 38 tumors (Table 4). Twenty-four (63.2%) of these were basal, 3 (7.9%) were HER2 overexpressing, 8 (21.0%) were luminal A, and 3 (7.9%) were luminal B. In comparison with tumors with low or absent T-bet+ intratumoral lymphoid cells, T-bet+/high tumors were more likely to belong to the basal molecular subtype (63.2% vs. 12.0%; P < 0.0001). Of basal tumors, 32% were T-bet+/high, compared with 6.5% of HER2+ tumors, 2.8% of luminal A tumors, and 5.8% of luminal B tumors.

Table 4.

Association between T-bet+ and intrinsic subgroups defined by IHC markersa

T-bet−/lowT-bet+/high
(n = 566)b(n = 48)b
SubgroupNumber (%)Number (%)Pc
Basal 51 (12.0) 24 (63.2) <0.0001 
HER2 43 (10.1) 3 (7.9)  
Luminal A 282 (66.4) 8 (21.0)  
Luminal B 49 (11.5) 3 (7.9)  
Unassigned 141 10  
T-bet−/lowT-bet+/high
(n = 566)b(n = 48)b
SubgroupNumber (%)Number (%)Pc
Basal 51 (12.0) 24 (63.2) <0.0001 
HER2 43 (10.1) 3 (7.9)  
Luminal A 282 (66.4) 8 (21.0)  
Luminal B 49 (11.5) 3 (7.9)  
Unassigned 141 10  

aTest was performed without the “Unassigned” group.

bSubgroup data are not available for some tumors due to unavailability of IHC markers data.

cBy Fisher exact test.

Prognostic relevance of intratumoral T-bet+ lymphoid cells

Univariate analysis.

When intratumoral T-bet status was considered alone throughout the entire follow-up period (minimum: 56.1 months; median: 96.5 months), intratumoral T-bet−/low was associated with reduced DFS (LR test P = 0.0133, RR = 4.72, 95% CI, 1.30–41.54; Table 5; Fig. 2).

Figure 2.

Kaplan–Meier DFS curves stratified by T-bet+ status showing a longer DFS in patients with T-bet+/high tumors.

Figure 2.

Kaplan–Meier DFS curves stratified by T-bet+ status showing a longer DFS in patients with T-bet+/high tumors.

Close modal
Table 5.

Results of DFS analysis by Cox proportional hazards model

UnivariateMultivariatea
Prognostic factorRR (95% CI)PRR (95% CI)P
T-bet 
 Low vs. high 4.72a (1.30a–41.54a0.0133b 5.62 (1.48–50.19) 0.0027b 
HER2 
 Positive vs. negative 2.30 (1.35–3.93) 0.0023 0.92 (0.40–1.86) 0.8209 
Menopausal status 
 Pre/peri vs. post 1.28 (0.88–1.87) 0.2022 0.84 (0.36–1.94) 0.6824 
ER 
 Negative/equivocal vs. ND/positive 1.26 (0.83–1.91) 0.2858 1.14 (0.62–2.032) 0.6661 
Tumor size 
 2–5 cm vs. <2 cm 2.16 (1.45–3.20) 0.0001 2.01 (1.20–3.45) 0.0102 
 >5 cm vs. <2 cm 3.46 (1.68–7.11) 0.0007 2.64 (1.00–6.02) 0.0332 
Histologic grade 
 Grade 2–3 vs. grade 1 3.48 (1.94–6.24) <0.0001 3.78 (1.67–10.44) 0.0044 
 ND vs. grade 1 3.51 (1.67–7.38) 0.0010 5.08 (1.79–15.87) 0.0033 
Lymphatic invasion 
 Present vs. absent 2.12 (1.36–3.31) 0.0010 2.38 (1.36–4.06) 0.0020 
Age at diagnosis, years 
 Linear 0.88 (0.74–1.04) 0.1317 0.94 (0.67–1.34) 0.7334 
 Quadratic 0.93 (0.81–1.06) 0.2615 0.99 (0.83–1.17) 0.9177 
Adjuvant treatment 
 Hormonal vs. none 0.55 (0.36–0.82) 0.0038 0.51 (0.29–0.86) 0.0132 
 Chemotherapy vs. none 0.94 (0.58–1.52) 0.8087 0.60 (0.29–1.17) 0.1415 
UnivariateMultivariatea
Prognostic factorRR (95% CI)PRR (95% CI)P
T-bet 
 Low vs. high 4.72a (1.30a–41.54a0.0133b 5.62 (1.48–50.19) 0.0027b 
HER2 
 Positive vs. negative 2.30 (1.35–3.93) 0.0023 0.92 (0.40–1.86) 0.8209 
Menopausal status 
 Pre/peri vs. post 1.28 (0.88–1.87) 0.2022 0.84 (0.36–1.94) 0.6824 
ER 
 Negative/equivocal vs. ND/positive 1.26 (0.83–1.91) 0.2858 1.14 (0.62–2.032) 0.6661 
Tumor size 
 2–5 cm vs. <2 cm 2.16 (1.45–3.20) 0.0001 2.01 (1.20–3.45) 0.0102 
 >5 cm vs. <2 cm 3.46 (1.68–7.11) 0.0007 2.64 (1.00–6.02) 0.0332 
Histologic grade 
 Grade 2–3 vs. grade 1 3.48 (1.94–6.24) <0.0001 3.78 (1.67–10.44) 0.0044 
 ND vs. grade 1 3.51 (1.67–7.38) 0.0010 5.08 (1.79–15.87) 0.0033 
Lymphatic invasion 
 Present vs. absent 2.12 (1.36–3.31) 0.0010 2.38 (1.36–4.06) 0.0020 
Age at diagnosis, years 
 Linear 0.88 (0.74–1.04) 0.1317 0.94 (0.67–1.34) 0.7334 
 Quadratic 0.93 (0.81–1.06) 0.2615 0.99 (0.83–1.17) 0.9177 
Adjuvant treatment 
 Hormonal vs. none 0.55 (0.36–0.82) 0.0038 0.51 (0.29–0.86) 0.0132 
 Chemotherapy vs. none 0.94 (0.58–1.52) 0.8087 0.60 (0.29–1.17) 0.1415 

aFirth penalized regression.

bFrom the likelihood ratio test; all other P values are from the Wald test.

T-bet+ lymphoid cells did not demonstrate prognostic significance when stratified according to molecular subtype or adjuvant therapy group (systemic adjuvant therapy vs. none) due to the small number of events in each subgroup.

Multivariate analysis.

The above association was retained when intratumoral T-bet status was assessed in a multivariate model for long-term follow-up that included traditional clinicopathologic factors and HER2, (LR test P = 0.0027, RR = 5.62, 95% CI, 1.48–50.19; Table 5).

In this prospectively accrued cohort of women with lymph node–negative breast cancer, we have demonstrated that intratumoral T-bet+ lymphoid cells are significantly associated with a good prognosis. This is despite being associated with adverse clinicopathologic features, including larger tumor size, higher histologic grade, hormone receptor negativity, and the basal phenotype. The data from this independent cohort of women unselected for family history add to the findings from our previous study with a cohort of women with familial breast cancer which had examined expression of the chemokine CXCL10 and tumoral lymphocytic infiltration (30). The familial cohort included tumors from patients with a strong family history of breast cancer, a significant proportion of whom carry a germline BRCA mutation. The cohort, representing incident breast cancer cases identified from population-based cancer registries, included large numbers of high-grade (54%) and basal-type (25%) tumors compared with 33% and 16%, respectively, in our current cohort. Furthermore, 46% were associated with lymph node metastases at diagnosis, in contrast with the current cohort in which patients were selected based on lymph node–negative status alone. The ANN cohort has many strengths for prognostic biomarker validation: its multi-institutional nature allows generalizability of the study population with regard to patient ethnicity, disease severity, differences in treatment, and variations in follow-up and endpoint assessments. Furthermore, the prospective accrual of patients in this cohort has the advantage of decreasing potential biases inherent in many tissue microarray studies. Thus, this current study provides important validation data for the use of T-bet as a prognostic marker in early-stage breast cancer.

Many studies have examined the prognostic effect of inflammation in breast cancer. The methods have varied considerably, with some studies assessing lymphoid infiltrates without subtype specification and other studies examining a specific subtype of the immune cell infiltrate using IHC markers. More recently, the prognostic relevance of tumor-infiltrating lymphocytes has been examined in two large prospectively accrued cohorts. In the first, the prognostic effect of tumor-infiltrating lymphocytes was assessed according to molecular subtype and type of chemotherapy in over 2,000 women with lymph node–positive breast cancer who were enrolled in the BIG 02-98 adjuvant phase III trial (12). This study found that incremental increases in lymphocytic infiltration (whether intratumoral or stromal) resulted in improved outcome in patients who had hormone receptor and HER2 disease only, regardless of chemotherapeutic regimen. Benefit (stromal lymphocytic infiltration only) was also seen in women with HER2+ disease in just one of the treatment arms (anthracycline-only). In a follow-up study from the same group (48) using tumor tissue from over 900 women enrolled in the prospective FinHER trial, a decreased distant disease recurrence rate was seen in women with increasing numbers of stromal lymphocytes in the triple-negative breast cancer group only. The strength of the effect was similar to what was seen in the earlier study.

The association of a lymphoid rich stroma with basal-like cancers is well recognized. Although this class is generally associated with a poor prognosis, outcome data suggest that it is a heterogeneous group of tumors and we and others have demonstrated that a subset of patients with basal-like breast cancer can expect to show long-term survival (33). Furthermore, medullary carcinomas cluster with the basal group, and this rare subset has been reported to be associated with a better prognosis than that of other grade 3 carcinomas (49, 50). Although one of the definitional criteria for a diagnosis of medullary carcinoma is a moderate or marked lymphoid infiltrate in the stroma between tumor nests, Rakha and colleagues (11) found no statistically significant outcome difference in patients with grade 3 medullary carcinomas and those with grade 3 ductal carcinomas with prominent inflammation, suggesting that the presence of the inflammatory infiltrate confers an improved prognosis rather than the diagnosis of medullary carcinoma per se and that this prominent inflammatory component may play an important role in determining outcome in basal-like tumors. In our study, high numbers of T-bet+ lymphoid cells were associated with the basal subtype; however, we were not able to detect a survival difference within this subgroup when analyzed separately, due to small numbers in this subanalysis. However, in a similar analysis of the basal subtype performed in our previous study (30), we found the association of higher T-bet expression with improved outcome to be of borderline significance (data not shown).

Traditionally, CD8+ cytotoxic T cells have been considered to be the key component in mounting an effective antitumor immune response and higher numbers have been associated with better patient survival (9, 51). However, CD4+ T cells have been shown to be necessary for full functioning of CD8+ cytotoxicity in vivo (52, 53). Furthermore, CD4+ T cells influence innate immunity by helping to shape the character and magnitude of the inflammatory response (54). Many breast cancer investigations before this have focused on the effects of CD4+ Tregs (55); however, increasing evidence suggests that the development of Th1 adaptive immunity is associated with improved outcome in various cancer types (56, 57), and T-bet, as the master regulator of Th1 cell differentiation, plays a pivotal role (21, 58). T-bet has been associated with better clinical outcomes in colorectal carcinoma (59) and gastric carcinoma (27); nevertheless, in breast cancer T-bet has rarely been studied. Ladoire and colleagues (60) examined T-bet expression in intratumoral lymphoid structures in women with HER2+ breast cancer who had been given prior treatment with neoadjuvant trastuzumab as well as anthracycline or taxanes. In 102 women, better recurrence-free survival was seen in those women who were treated with trastuzumab and taxanes who had T-bet+ cells in peritumoral lymphoid structures after chemotherapy. A correlation with T-bet+ cells following therapy was not seen with pathologic complete response to therapy. This study highlighted the importance, not only of the host's immune response in breast cancer, but also of the interaction between certain chemotherapeutic regimens and the immune system. The role the immune system plays in response to trastuzumab is well described: trastuzumab activates the host's immune system through antibody-dependent cellular toxicity (ADCC; ref. 61), which leads to Th1 activation and production of IFNγ that has been implicated in control of cancer growth (62). Evidence also suggests that taxanes could exert an immunostimulatory effect against breast cancer (63–66) by inducing a Th1 response. Paclitaxel has been shown to stimulate the secretion by macrophages of proinflammatory and Th1 cytokines such as IL1b or IL12 (67, 68). Carson and colleagues (65) showed that in 227 breast cancer patients, T-cell activation was significantly higher in patients receiving taxanes compared with non–taxane-containing regimens. In contrast, anthracyclines primarily activate CD8+ T cells rather than CD4+ T cells to produce IFNγ, as shown in mouse models (69).

In this study, we based analyses on intratumoral rather than peritumoral T-bet+ lymphoid cells for two reasons. First, because of our prior findings of the prognostic significance of intratumoral T-bet+ lymphoid cells in the Breast Cancer Family Registry, Ontario site (30), and second, because we used TMAs in this study. TMAs preferentially include areas rich in tumor cells, and peritumoral areas may not be represented consistently, thereby introducing a potential bias in peritumoral T-bet+ lymphoid cell counts.

In conclusion, intratumoral T-bet+ lymphoid cells in breast cancer are associated with adverse pathologic features, including the basal subtype. Nevertheless, their presence confers a favorable outcome which is independent of traditional clinicopathologic parameters and HER2 status. The potential use of T-bet as a prognostic marker in breast cancer needs to be evaluated in additional larger cohorts, particularly those rich in basal-like or HER2+ breast cancer. Understanding the mechanisms mediating these immunologic responses and biomarkers of such responses may help in better tailoring specific therapies including combinations of agents for women with breast cancer. Furthermore, modulation of T-bet expression has the potential to become a powerful therapeutic target for the treatment of cancer in the future.

No potential conflicts of interest were disclosed.

Conception and design: A.M. Mulligan, I.L. Andrulis

Development of methodology: A.M. Mulligan, I.L. Andrulis

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): A.M. Mulligan, I.L. Andrulis

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): A.M. Mulligan, D. Pinnaduwage, S.B. Bull

Writing, review, and/or revision of the manuscript: A.M. Mulligan, D. Pinnaduwage, S. Tchatchou, S.B. Bull, I.L. Andrulis

Study supervision: I.L. Andrulis

The authors thank the study participants and the Toronto Breast Cancer Group for their contributions to this work.

This work was supported in part by a grant from the Canadian Institutes of Health Research #MOP-93715 (I.L. Andrulis and S.B. Bull), the Syd Cooper Program for the Prevention of Cancer Progression and the Anne and Max Tanenbaum Chair in Molecular Medicine at Mount Sinai Hospital, and the University of Toronto (I.L. Andrulis).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
O'Callaghan
DS
,
O'Donnell
D
,
O'Connell
F
,
O'Byrne
KJ
. 
The role of inflammation in the pathogenesis of non-small cell lung cancer
.
J Thorac Oncol
2010
;
5
:
2024
36
.
2.
Finn
OJ
. 
Cancer immunology
.
N Engl J Med
2008
;
358
:
2704
15
.
3.
Nicolini
A
,
Carpi
A
,
Rossi
G
. 
Cytokines in breast cancer
.
Cytokine Growth Factor Rev
2006
;
17
:
325
37
.
4.
Lee
AH
,
Gillett
CE
,
Ryder
K
,
Fentiman
IS
,
Miles
DW
,
Millis
RR
. 
Different patterns of inflammation and prognosis in invasive carcinoma of the breast
.
Histopathology
2006
;
48
:
692
701
.
5.
Carlomagno
C
,
Perrone
F
,
Lauria
R
,
de Laurentiis
M
,
Gallo
C
,
Morabito
A
, et al
Prognostic significance of necrosis, elastosis, fibrosis and inflammatory cell reaction in operable breast cancer
.
Oncology
1995
;
52
:
272
7
.
6.
Disis
ML
. 
Immune regulation of cancer
.
J Clin Oncol
2010
;
28
:
4531
8
.
7.
Zitvogel
L
,
Kepp
O
,
Kroemer
G
. 
Immune parameters affecting the efficacy of chemotherapeutic regimens
.
Nat Rev Clin Oncol
2011
;
8
:
151
60
.
8.
Denkert
C
,
Loibl
S
,
Noske
A
,
Roller
M
,
Muller
BM
,
Komor
M
, et al
Tumor-associated lymphocytes as an independent predictor of response to neoadjuvant chemotherapy in breast cancer
.
J Clin Oncol
2010
;
28
:
105
13
.
9.
Mahmoud
SM
,
Paish
EC
,
Powe
DG
,
Macmillan
RD
,
Grainge
MJ
,
Lee
AH
, et al
Tumor-infiltrating CD8+ lymphocytes predict clinical outcome in breast cancer
.
J Clin Oncol
2011
;
29
:
1949
55
.
10.
Kohrt
HE
,
Nouri
N
,
Nowels
K
,
Johnson
D
,
Holmes
S
,
Lee
PP
. 
Profile of immune cells in axillary lymph nodes predicts disease-free survival in breast cancer
.
PLoS Med
2005
;
2
:
e284
.
11.
Rakha
EA
,
Aleskandarany
M
,
El-Sayed
ME
,
Blamey
RW
,
Elston
CW
,
Ellis
IO
, et al
The prognostic significance of inflammation and medullary histological type in invasive carcinoma of the breast
.
Eur J Cancer
2009
;
45
:
1780
7
.
12.
Loi
S
,
Sirtaine
N
,
Piette
F
,
Salgado
R
,
Viale
G
,
Van Eenoo
F
, et al
Prognostic and predictive value of tumor-infiltrating lymphocytes in a phase III randomized adjuvant breast cancer trial in node-positive breast cancer comparing the addition of docetaxel to doxorubicin with doxorubicin-based chemotherapy: BIG 02-98
.
J Clin Oncol
2013
;
31
:
860
7
.
13.
Yu
P
,
Fu
YX
. 
Tumor-infiltrating T lymphocytes: friends or foes?
Lab Invest
2006
;
86
:
231
45
.
14.
Ruffell
B
,
DeNardo
DG
,
Affara
NI
,
Coussens
LM
. 
Lymphocytes in cancer development: polarization towards pro-tumor immunity
.
Cytokine Growth Factor Rev
2010
;
21
:
3
10
.
15.
Zamarron
BF
,
Chen
W
. 
Dual roles of immune cells and their factors in cancer development and progression
.
Int J Biol Sci
2011
;
7
:
651
8
.
16.
Gobert
M
,
Treilleux
I
,
Bendriss-Vermare
N
,
Bachelot
T
,
Goddard-Leon
S
,
Arfi
V
, et al
Regulatory T cells recruited through CCL22/CCR4 are selectively activated in lymphoid infiltrates surrounding primary breast tumors and lead to an adverse clinical outcome
.
Cancer Res
2009
;
69
:
2000
9
.
17.
Lugo-Villarino
G
,
Maldonado-Lopez
R
,
Possemato
R
,
Penaranda
C
,
Glimcher
LH
. 
T-bet is required for optimal production of IFN-gamma and antigen-specific T cell activation by dendritic cells
.
Proc Natl Acad Sci U S A
2003
;
100
:
7749
54
.
18.
Spits
H
,
Artis
D
,
Colonna
M
,
Diefenbach
A
,
Di Santo
JP
,
Eberl
G
, et al
Innate lymphoid cells–a proposal for uniform nomenclature
.
Nat Rev Immunol
2013
;
13
:
145
9
.
19.
Sciume
G
,
Hirahara
K
,
Takahashi
H
,
Laurence
A
,
Villarino
AV
,
Singleton
KL
, et al
Distinct requirements for T-bet in gut innate lymphoid cells
.
J Exp Med
2012
;
209
:
2331
8
.
20.
Gordon
SM
,
Chaix
J
,
Rupp
LJ
,
Wu
J
,
Madera
S
,
Sun
JC
, et al
The transcription factors T-bet and Eomes control key checkpoints of natural killer cell maturation
.
Immunity
2012
;
36
:
55
67
.
21.
Szabo
SJ
,
Kim
ST
,
Costa
GL
,
Zhang
X
,
Fathman
CG
,
Glimcher
LH
. 
A novel transcription factor, T-bet, directs Th1 lineage commitment
.
Cell
2000
;
100
:
655
69
.
22.
Powell
N
,
Walker
AW
,
Stolarczyk
E
,
Canavan
JB
,
Gokmen
MR
,
Marks
E
, et al
The transcription factor T-bet regulates intestinal inflammation mediated by interleukin-7 receptor+ innate lymphoid cells
.
Immunity
2012
;
37
:
674
84
.
23.
Townsend
MJ
,
Weinmann
AS
,
Matsuda
JL
,
Salomon
R
,
Farnham
PJ
,
Biron
CA
, et al
T-bet regulates the terminal maturation and homeostasis of NK and Valpha14i NKT cells
.
Immunity
2004
;
20
:
477
94
.
24.
Werneck
MB
,
Lugo-Villarino
G
,
Hwang
ES
,
Cantor
H
,
Glimcher
LH
. 
T-bet plays a key role in NK-mediated control of melanoma metastatic disease
.
J Immunol
2008
;
180
:
8004
10
.
25.
Ortegel
JW
,
Staren
ED
,
Faber
LP
,
Warren
WH
,
Braun
DP
. 
Cytokine biosynthesis by tumor-infiltrating T lymphocytes from human non-small-cell lung carcinoma
.
Cancer Immunol Immunother
2000
;
48
:
627
34
.
26.
Berrien-Elliott
MM
,
Yuan
J
,
Swier
LE
,
Jackson
SR
,
Chen
CL
,
Donlin
MJ
, et al
Checkpoint blockade immunotherapy relies on T-bet but not Eomes to induce effector function in tumor infiltrating CD8+ T cells
.
Cancer Immunol Res
2015
;
3
:
116
24
.
27.
Chen
LJ
,
Zheng
X
,
Shen
YP
,
Zhu
YB
,
Li
Q
,
Chen
J
, et al
Higher numbers of T-bet(+) intratumoral lymphoid cells correlate with better survival in gastric cancer
.
Cancer Immunol Immunother
2013
;
62
:
553
61
.
28.
Origoni
M
,
Parma
M
,
Dell'Antonio
G
,
Gelardi
C
,
Stefani
C
,
Salvatore
S
, et al
Prognostic significance of immunohistochemical phenotypes in patients treated for high-grade cervical intraepithelial neoplasia
.
Biomed Res Int
2013
;
2013
:
831907
.
29.
John
EM
,
Hopper
JL
,
Beck
JC
,
Knight
JA
,
Neuhausen
SL
,
Senie
RT
, et al
The Breast Cancer Family Registry: an infrastructure for cooperative multinational, interdisciplinary and translational studies of the genetic epidemiology of breast cancer
.
Breast Cancer Res
2004
;
6
:
R375
89
.
30.
Mulligan
AM
,
Raitman
I
,
Feeley
L
,
Pinnaduwage
D
,
Nguyen
LT
,
O'Malley
FP
, et al
Tumoral lymphocytic infiltration and expression of the chemokine CXCL10 in breast cancers from the Ontario Familial Breast Cancer Registry
.
Clin Cancer Res
2013
;
19
:
336
46
.
31.
Andrulis
IL
,
Bull
SB
,
Blackstein
ME
,
Sutherland
D
,
Mak
C
,
Sidlofsky
S
, et al
neu/erbB-2 amplification identifies a poor-prognosis group of women with node-negative breast cancer. Toronto Breast Cancer Study Group
.
J Clin Oncol
1998
;
16
:
1340
9
.
32.
Bull
SB
,
Ozcelik
H
,
Pinnaduwage
D
,
Blackstein
ME
,
Sutherland
DA
,
Pritchard
KI
, et al
The combination of p53 mutation and neu/erbB-2 amplification is associated with poor survival in node-negative breast cancer
.
J Clin Oncol
2004
;
22
:
86
96
.
33.
Mulligan
AM
,
Pinnaduwage
D
,
Bull
SB
,
O'Malley
FP
,
Andrulis
IL
. 
Prognostic effect of basal-like breast cancers is time dependent: evidence from tissue microarray studies on a lymph node-negative cohort
.
Clin Cancer Res
2008
;
14
:
4168
74
.
34.
Allred
DC
,
Harvey
JM
,
Berardo
M
,
Clark
GM
. 
Prognostic and predictive factors in breast cancer by immunohistochemical analysis
.
Mod Pathol
1998
;
11
:
155
68
.
35.
Liu
CL
,
Prapong
W
,
Natkunam
Y
,
Alizadeh
A
,
Montgomery
K
,
Gilks
CB
, et al
Software tools for high-throughput analysis and archiving of immunohistochemistry staining data obtained with tissue microarrays
.
Am J Pathol
2002
;
161
:
1557
65
.
36.
Mohsin
SK
,
Weiss
H
,
Havighurst
T
,
Clark
GM
,
Berardo
M
,
Roanh le
D
, et al
Progesterone receptor by immunohistochemistry and clinical outcome in breast cancer: a validation study
.
Mod Pathol
2004
;
17
:
1545
54
.
37.
Harvey
JM
,
Clark
GM
,
Osborne
CK
,
Allred
DC
. 
Estrogen receptor status by immunohistochemistry is superior to the ligand-binding assay for predicting response to adjuvant endocrine therapy in breast cancer
.
J Clin Oncol
1999
;
17
:
1474
81
.
38.
O'Malley
FP
,
Parkes
R
,
Latta
E
,
Tjan
S
,
Zadro
T
,
Mueller
R
, et al
Comparison of HER2/neu status assessed by quantitative polymerase chain reaction and immunohistochemistry
.
Am J Clin Pathol
2001
;
115
:
504
11
.
39.
Done
SJ
,
Arneson
CR
,
Ozcelik
H
,
Redston
M
,
Andrulis
IL
. 
P53 protein accumulation in non-invasive lesions surrounding p53 mutation positive invasive breast cancers
.
Breast Cancer Res Treat
2001
;
65
:
111
8
.
40.
Cheang
MC
,
Chia
SK
,
Voduc
D
,
Gao
D
,
Leung
S
,
Snider
J
, et al
Ki67 index, HER2 status, and prognosis of patients with luminal B breast cancer
.
J Natl Cancer Inst
2009
;
101
:
736
50
.
41.
Livasy
CA
,
Karaca
G
,
Nanda
R
,
Tretiakova
MS
,
Olopade
OI
,
Moore
DT
, et al
Phenotypic evaluation of the basal-like subtype of invasive breast carcinoma
.
Mod Pathol
2006
;
19
:
264
71
.
42.
Kornegoor
R
,
Verschuur-Maes
AH
,
Buerger
H
,
Hogenes
MC
,
de Bruin
PC
,
Oudejans
JJ
, et al
Molecular subtyping of male breast cancer by immunohistochemistry
.
Mod Pathol
2012
;
25
:
398
404
.
43.
Voduc
KD
,
Cheang
MC
,
Tyldesley
S
,
Gelmon
K
,
Nielsen
TO
,
Kennecke
H
. 
Breast cancer subtypes and the risk of local and regional relapse
.
J Clin Oncol
2010
;
28
:
1684
91
.
44.
Feeley
LP
,
Mulligan
AM
,
Pinnaduwage
D
,
Bull
SB
,
Andrulis
IL
. 
Distinguishing luminal breast cancer subtypes by Ki67, progesterone receptor or TP53 status provides prognostic information
.
Mod Pathol
2014
;
27
:
554
61
.
45.
Cox
DR
. 
Regression models and life-tables
.
J R Stat Soc B
1972
;
34
:
187
220
.
46.
Heinze
G
,
Schemper
M
. 
A solution to the problem of monotone likelihood in Cox regression
.
Biometrics
2001
;
57
:
114
9
.
47.
Conneely
KN
,
Boehnke
M
. 
So many correlated tests, so little time! Rapid adjustment of P values for multiple correlated tests
.
Am J Hum Genet
2007
;
81
:
1158
68
.
48.
Loi
S
,
Michiels
S
,
Salgado
R
,
Sirtaine
N
,
Jose
V
,
Fumagalli
D
, et al
Tumor infiltrating lymphocytes are prognostic in triple negative breast cancer and predictive for trastuzumab benefit in early breast cancer: results from the FinHER trial
.
Ann Oncol
2014
;
25
:
1544
50
.
49.
Ellis
IO
,
Galea
M
,
Broughton
N
,
Locker
A
,
Blamey
RW
,
Elston
CW
. 
Pathological prognostic factors in breast cancer. II. Histological type. Relationship with survival in a large study with long-term follow-up
.
Histopathology
1992
;
20
:
479
89
.
50.
Pedersen
L
,
Zedeler
K
,
Holck
S
,
Schiodt
T
,
Mouridsen
HT
. 
Medullary carcinoma of the breast. Prevalence and prognostic importance of classical risk factors in breast cancer
.
Eur J Cancer
1995
;
31A
:
2289
95
.
51.
Ali
HR
,
Provenzano
E
,
Dawson
SJ
,
Blows
FM
,
Liu
B
,
Shah
M
, et al
Association between CD8+ T-cell infiltration and breast cancer survival in 12,439 patients
.
Ann Oncol
2014
;
25
:
1536
43
.
52.
Bos
R
,
Marquardt
KL
,
Cheung
J
,
Sherman
LA
. 
Functional differences between low- and high-affinity CD8(+) T cells in the tumor environment
.
Oncoimmunology
2012
;
1
:
1239
47
.
53.
Bos
R
,
Sherman
LA
. 
CD4+ T-cell help in the tumor milieu is required for recruitment and cytolytic function of CD8+ T lymphocytes
.
Cancer Res
2010
;
70
:
8368
77
.
54.
Gu-Trantien
C
,
Loi
S
,
Garaud
S
,
Equeter
C
,
Libin
M
,
de Wind
A
, et al
CD4(+) follicular helper T cell infiltration predicts breast cancer survival
.
J Clin Invest
2013
;
123
:
2873
92
.
55.
deLeeuw
RJ
,
Kost
SE
,
Kakal
JA
,
Nelson
BH
. 
The prognostic value of FoxP3+ tumor-infiltrating lymphocytes in cancer: a critical review of the literature
.
Clin Cancer Res
2012
;
18
:
3022
9
.
56.
Zhang
XR
,
Zhang
LY
,
Devadas
S
,
Li
L
,
Keegan
AD
,
Shi
YF
. 
Reciprocal expression of TRAIL and CD95L in Th1 and Th2 cells: role of apoptosis in T helper subset differentiation
.
Cell Death Differ
2003
;
10
:
203
10
.
57.
Roepman
P
,
Jassem
J
,
Smit
EF
,
Muley
T
,
Niklinski
J
,
van de Velde
T
, et al
An immune response enriched 72-gene prognostic profile for early-stage non-small-cell lung cancer
.
Clin Cancer Res
2009
;
15
:
284
90
.
58.
Mullen
AC
,
High
FA
,
Hutchins
AS
,
Lee
HW
,
Villarino
AV
,
Livingston
DM
, et al
Role of T-bet in commitment of TH1 cells before IL-12-dependent selection
.
Science
2001
;
292
:
1907
10
.
59.
Galon
J
,
Costes
A
,
Sanchez-Cabo
F
,
Kirilovsky
A
,
Mlecnik
B
,
Lagorce-Pages
C
, et al
Type, density, and location of immune cells within human colorectal tumors predict clinical outcome
.
Science
2006
;
313
:
1960
4
.
60.
Ladoire
S
,
Arnould
L
,
Mignot
G
,
Apetoh
L
,
Rebe
C
,
Martin
F
, et al
T-bet expression in intratumoral lymphoid structures after neoadjuvant trastuzumab plus docetaxel for HER2-overexpressing breast carcinoma predicts survival
.
Br J Cancer
2011
;
105
:
366
71
.
61.
Arnould
L
,
Gelly
M
,
Penault-Llorca
F
,
Benoit
L
,
Bonnetain
F
,
Migeon
C
, et al
Trastuzumab-based treatment of HER2-positive breast cancer: an antibody-dependent cellular cytotoxicity mechanism?
Br J Cancer
2006
;
94
:
259
67
.
62.
Dhodapkar
KM
,
Krasovsky
J
,
Williamson
B
,
Dhodapkar
MV
. 
Antitumor monoclonal antibodies enhance cross-presentation ofcCellular antigens and the generation of myeloma-specific killer T cells by dendritic cells
.
J Exp Med
2002
;
195
:
125
33
.
63.
Lee
M
,
Yea
SS
,
Jeon
YJ
. 
Paclitaxel causes mouse splenic lymphocytes to a state hyporesponsive to lipopolysaccharide stimulation
.
Int J Immunopharmacol
2000
;
22
:
615
21
.
64.
Tsavaris
N
,
Kosmas
C
,
Vadiaka
M
,
Kanelopoulos
P
,
Boulamatsis
D
. 
Immune changes in patients with advanced breast cancer undergoing chemotherapy with taxanes
.
Br J Cancer
2002
;
87
:
21
7
.
65.
Carson
WE
 III
,
Shapiro
CL
,
Crespin
TR
,
Thornton
LM
,
Andersen
BL
. 
Cellular immunity in breast cancer patients completing taxane treatment
.
Clin Cancer Res
2004
;
10
:
3401
9
.
66.
Carson
WE
,
Parihar
R
,
Lindemann
MJ
,
Personeni
N
,
Dierksheide
J
,
Meropol
NJ
, et al
Interleukin-2 enhances the natural killer cell response to Herceptin-coated Her2/neu-positive breast cancer cells
.
Eur J Immunol
2001
;
31
:
3016
25
.
67.
Mullins
DW
,
Koci
MD
,
Burger
CJ
,
Elgert
KD
. 
Interleukin-12 overcomes paclitaxel-mediated suppression of T-cell proliferation
.
Immunopharmacol Immunotoxicol
1998
;
20
:
473
92
.
68.
Chan
OT
,
Yang
LX
. 
The immunological effects of taxanes
.
Cancer Immunol Immunother
2000
;
49
:
181
5
.
69.
Apetoh
L
,
Ghiringhelli
F
,
Tesniere
A
,
Obeid
M
,
Ortiz
C
,
Criollo
A
, et al
Toll-like receptor 4-dependent contribution of the immune system to anticancer chemotherapy and radiotherapy
.
Nat Med
2007
;
13
:
1050
9
.