Next Article in Journal
CENPE Inhibition Leads to Mitotic Catastrophe and DNA Damage in Medulloblastoma Cells
Next Article in Special Issue
Variability in Breast Cancer Biomarker Assessment and the Effect on Oncological Treatment Decisions: A Nationwide 5-Year Population-Based Study
Previous Article in Journal
Quantitative Assessment of CD200 and CD200R Expression in Lung Cancer
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Cancers
Volume 13
Issue 5
10.3390/cancers13051022
cancers-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Granulocyte Colony Stimulating Factor Expression in Breast Cancer and Its Association with Carbonic Anhydrase IX and Immune Checkpoints

by
Shawn C. Chafe
1,†,
Nazia Riaz
2,3,†,
Samantha Burugu
2,
Dongxia Gao
2,
Samuel C. Y. Leung
2,
Anna F. Lee
2,
Cheng-Han Lee
2,
Shoukat Dedhar
1,4 and
Torsten O. Nielsen
2,*
1
Department of Integrative Oncology, BC Cancer Research Centre, Vancouver, BC V5Z 1L3, Canada
2
Department of Pathology and Laboratory Medicine, University of British Columbia, Vancouver, BC V6H 3Z6, Canada
3
Centre for Regenerative Medicine and Stem Cell Research, Aga Khan University, Sindh 74800, Pakistan
4
Department of Biochemistry and Molecular Biology, University of British Columbia, Vancouver, BC V6T 1Z3, Canada
*
Author to whom correspondence should be addressed.
†
These authors are Co-first authors.
Cancers 2021, 13(5), 1022; https://doi.org/10.3390/cancers13051022
Submission received: 8 January 2021 / Revised: 10 February 2021 / Accepted: 23 February 2021 / Published: 1 March 2021
(This article belongs to the Special Issue Novel Biomarkers: Validity, Evaluation and Biological Roles for Breast Cancer)
Browse Figures
Review Reports Versions Notes

Abstract

:

Simple Summary

Preclinical studies suggest that interactions between granulocyte colony-stimulating factor (G-CSF) and hypoxia-induced carbonic anhydrase IX regulate the trafficking and function of immune cells in the tumour microenvironment. We investigated the clinical significance of this crosstalk by analyzing the protein expression of G-CSF and macrophage markers by immunohistochemistry on a well-characterized tissue microarray series of invasive breast cancers. We report that high expression of G-CSF on breast carcinoma cells is linked with significantly improved survival in an important group of breast cancers that do not respond to hormonal therapy. These tumours were infiltrated by immune cells expressing biomarkers that can be targeted with immune checkpoint inhibitor drugs. In contrast, carbonic anhydrase IX expression was associated with unfavourable outcomes.

Abstract

Purpose: Granulocyte colony-stimulating factor (G-CSF) and hypoxia modulate the tumour immune microenvironment. In model systems, hypoxia-induced carbonic anhydrase IX (CAIX) has been associated with G-CSF and immune responses, including M2 polarization of macrophages. We investigated whether these associations exist in human breast cancer specimens, their relation to breast cancer subtypes, and clinical outcome. Methods: Using validated protocols and prespecified scoring methodology, G-CSF expression on carcinoma cells and CD163 expression on tumour-associated macrophages were assayed by immunohistochemistry and applied to a tissue microarray series of 2960 primary excision specimens linked to clinicopathologic, biomarker, and outcome data. Results: G-CSFhigh expression showed a significant positive association with ER negativity, HER2 positivity, presence of CD163+ M2 macrophages, and CAIX expression. In univariate analysis, G-CSFhigh phenotype was associated with improved survival in non-luminal cases, although the CAIX+ subset had a significantly adverse prognosis. A significant positive association was observed between immune checkpoint biomarkers on tumour-infiltrating lymphocytes and both G-CSF- and CAIX-expressing carcinoma cells. Immune checkpoint biomarkers correlated significantly with favourable prognosis in G-CSFhigh/non-luminal cases independent of standard clinicopathological features. Conclusions: The prognostic associations linking G-CSF to immune biomarkers and CAIX strongly support their immunomodulatory roles in the tumour microenvironment.

1. Introduction

Granulocyte colony-stimulating factor (G-CSF) is a member of the colony-stimulating factor superfamily and exerts its biological effects by binding to the G-CSF receptor (G-CSFR). The archetypal G-CSF/G-CSFR signalling pathway is critical for proliferation and survival of myeloid precursors and their subsequent differentiation into neutrophils with augmentation of their effector functions [1]. Recognition of these biological functions led to the establishment of G-CSF as a prophylactic and therapeutic agent for chemotherapy-induced febrile neutropenia [2]. Considering the robust relationship between inflammation and cancer, the role of G-CSF- and G-CSF-mobilized immune cells has been investigated in preclinical models of several non-hematopoietic malignancies, including breast cancer [3,4]. These studies have implicated aberrant G-CSF/G-CSFR signalling with altered hematopoiesis, leading to the recruitment of immunosuppressive cells in the tumour microenvironment that potentiate migration, invasion, angiogenesis and metastasis [5,6,7,8,9,10,11,12,13]. In line with preclinical data, clinical studies evaluating G-CSF expression on tumour cells have also shown an association with aggressive clinicopathological features and poor prognosis in solid organ malignancies such as clear cell renal cell carcinomas and cervical cancers [14,15].
Tumour hypoxia is a hallmark of invasive tumours and is associated with genomic instability, the emergence of therapy-resistant clones, inhibition of anti-tumour immune responses, and cancer progression [16,17]. Carbonic anhydrase IX (CAIX) is a cell surface membrane-bound enzyme expressed by hypoxic tumours. CAIX catalyzes the reversible hydration of CO2 into bicarbonate ions and protons. The bicarbonate is transported into the cells via bicarbonate transporters, thereby aiding in the buffering of intracellular pH, whereas the protons remain in the extracellular space, contributing to extracellular acidification [18]. We and others have previously shown that immunohistochemical expression of CAIX is frequently associated with features of aggressive disease and adverse survival in breast cancer [19,20,21]. In preclinical models of breast cancer, G-CSF production by the hypoxic tumour cells, in a CAIX-dependent manner, facilitated the formation of a lung premetastatic niche by recruiting myeloid-derived suppressor cells [22]. Furthermore, in the same model, CAIX inhibition enhanced the efficacy of immune checkpoint blockade [23]. Thus, there is strong evidence to support that the hypoxic, acidic microenvironment of the tumours is immunosuppressive and enriched in a range of cells, including myeloid-derived suppressor cells, T lymphocytes, and tumour-associated macrophages (TAMs), all of which contribute to immune tolerance by dampening anti-tumour effector cell functions [16,24,25,26].
Tumour-associated macrophages comprise heterogeneous populations of cells that exhibit remarkable plasticity and play an important role in modulating adaptive immune responses [27]. However, their biological role and prognostic significance are dependent upon phenotype and localization, which can be assessed using biomarkers. The classically activated M1 macrophages are characterized by the production of proinflammatory cytokines, enhanced antigen presentation, tumouricidal effects and improved cancer outcomes [28,29]. In contrast, alternatively activated M2 macrophages are associated with anti-inflammatory cytokines, immunosuppressive responses and protumoural properties. M2 macrophages are reliably identified by the expression of CD163, which is a cell surface glycoprotein and a member of the scavenger receptor superfamily class B [27]. Tumour infiltration with CD163+ M2 macrophages is associated with features of aggressive disease and poor outcomes in breast cancer patients, notably those diagnosed with HER2 and basal subtypes [30,31]. Preclinical studies have shown that G-CSF secretion by triple-negative breast cancer cell lines potentiates macrophage differentiation into an immunosuppressive phenotype associated with enhanced migratory capacity [32]. In addition, it has been shown that hypoxic tumour cells also promote M2-like polarization of tumour-infiltrating macrophages [33].
Thus, existing data from preclinical studies highlight hypoxia and inflammation as critical modulators of the immune microenvironment of solid tumours where a significant interplay between G-CSF and CAIX has been found to play a role. Presently, it is not known whether there is an association amongst G-CSF, CAIX, and immune biomarkers in breast cancer clinical samples. Furthermore, to date, only limited studies (with insufficient statistical power) have evaluated the prognostic significance of G-CSF expression in breast cancer [32]. Hence, we hypothesized that high expression of G-CSF would correlate with features associated with breast cancer aggressiveness, including CAIX expression, and provide prognostic information across breast cancer subtypes. The objectives of this study were to investigate the immunohistochemical expression of G-CSF in a large series of breast cancers powered for correlation with clinicopathological features, CAIX expression, and immune checkpoint biomarkers; and to examine the prognostic significance of G-CSF in relation to CAIX and CD163+ M2 macrophages using breast cancer-specific survival as the primary endpoint, and overall survival and relapse-free survival as secondary endpoints.

2. Materials and Methods

2.1. Study Cohorts

We first established the immunohistochemical (IHC) staining procedure, scoring methodology, and interpretation of G-CSF protein expression on a series of female breast cancer patients diagnosed with stage I–III disease during 1998–2002 (cohort I: n = 330). The clinicopathological characteristics, adjuvant treatments, and inclusion/exclusion criteria have been described previously [34]. The median follow-up of this cohort was 13 years. We conducted validation and exploratory analyses on an independent, much larger series (cohort II: n = 2960) of female patients diagnosed with stage I–III breast cancer in the province of British Columbia at the British Columbia Cancer Agency between 1986–1992, for which formalin-fixed paraffin-embedded tumour blocks were accessible. The adjuvant systemic and endocrine therapies recommended in the specified years were in accordance with the provincial standards. No patients received immune checkpoint inhibitor therapy. The clinicopathological characteristics of this cohort have been formerly described [35,36]. All cases diagnosed with ductal carcinoma in-situ only, stage IV disease at presentation, and those receiving neoadjuvant chemotherapy regimens were excluded from this study.
The Clinical Research Ethics Board of the University of British Columbia and the Breast Cancer Outcomes Unit of BC Cancer approved access to the clinical samples and to corresponding deidentified clinical data. The study was conducted in accordance with the Reporting Recommendations for Tumor Marker Prognostic Studies (REMARK) guidelines [37].

2.2. Tissue Microarrays and Immunohistochemistry

The construction of the tissue microarrays used in this study has been described previously [34,38,39]. In all, there were 20 tissue microarray blocks (0.6 mm cores) built from paraffin-embedded primary surgical specimens. Of these, 3 blocks represented cohort I (2 cores per case), and 17 were built from cohort II (1 core per case). Other biomarkers included in studies of cohort II (ER, PR, HER2, Ki-67, CK5/6, EGFR, CD8, FOXP3, LAG3, TIM3, PD-L1, PD-1, CAIX) and IHC-based intrinsic breast cancer subtyping have been described in previous publications [19,34,36,40,41,42,43]. The assessment criteria for stromal tumour-infiltrating lymphocytes (sTILs) on hematoxylin- and eosin-stained sections was in accordance with the recommendations of the International TIL Working Group [44] and the cases were categorized into two groups using <10% vs. ≥10% score as described previously [45].
The IHC protocols are detailed in Supplementary Table S1. The stained slides were digitally scanned using a BLISS system (Bacus Laboratories/Olympus America, Lombard, IL, USA) and visually scored by the pathologists (DG, AFL, CHL) who were blinded to the clinical characteristics and outcome data.
The scoring of G-CSF expression was adapted from Hollmen et al. [32]. The intensity of cytoplasmic expression in breast carcinoma cells was classified into four categories: score 0, no reactivity; score 1, weak cytoplasmic reactivity; score 2, moderate cytoplasmic reactivity; score 3, strong cytoplasmic reactivity. For cohort I, the mean intensity score of the duplicate cores was estimated. Cases scored as ≤ 1 were considered low expressors of G-CSF, >1 high. For CD163 and CAIX, we used previously published criteria as follows. Membranous or cytoplasmic expression of CD163 was scored on tumour-associated macrophages as previously described [46] and was classified into three categories: score 1 (sparse infiltrates, ≤5 stained macrophages); score 2 (moderate infiltrates, >5 but ≤25 positively stained macrophages); and score 3 (dense infiltrates, >25 positively stained macrophages). For CAIX, membranous expression on carcinoma cells was scored as 1 (any reactivity) and 0 (no reactivity), as described previously [19]. Cores with insufficient diagnostic material or with equivocal staining results were omitted from further analysis. Images from these slides are available for public access via the website of the Genetic Pathology Evaluation Center (http://www.gpec.ubc.ca/gcsf accessed on 1 January 2021), and representative images are shown in Figure 1. Representative images from serial section of a single core are shown in Supplementary Figure S1.

2.3. Statistical Analysis

IBM® SPSS software (version 25) and R (v 3.3.2) were used for statistical analyses. Relevant descriptive statistics were computed for continuous and categorical variables. The correlation of G-CSF with clinicopathological factors and key biomarkers was assessed by chi-square. The primary endpoint for the clinical outcome was breast cancer-specific survival (BCSS), specified as the time interval between the date of diagnosis and the date of breast cancer-associated mortality. Patients who were alive at the end of follow-up or died due to non-breast cancer-related causes were censored. Overall survival (OS) and relapse-free survival (RFS) were used as secondary endpoints. OS was defined as the time period from the date of diagnosis till the date of last follow-up or death due to any cause, whereas RFS was defined as the time period from the date of diagnosis until any breast cancer-related relapse, whether local, regional, distant, or contralateral. Univariate analysis for survival probabilities was computed by the Kaplan–Meier method, and differences in the survival rates between analyzed groups were estimated by log-rank test. Cox proportional hazard modelling was used for multivariable analysis, and adjusted hazard ratios with 95% confidence intervals were reported for each variable included in the model. A p-value of <0.05 was considered statistically significant. To address the concerns of multiple comparisons while assessing associations between G-CSF expression and clinicopathological features, a Bonferroni correction was applied, making the criterion for statistical significance p < 0.003.

3. Results

3.1. Correlation of G-CSF Expression with Clinicopathological Features and Survival

Cohort I was used to evaluate the performance of G-CSF immunostaining, finalize the scoring methodology, and explore cut points such that these could be locked down prior to assessment of the main study cohort II. The correlative analyses with clinicopathological features and association with primary and secondary endpoints for cohort I are presented in supplementary data (Supplementary Tables S2–S3 and Supplementary Figure S2). On this smaller series, no significant clinicopathologic associations were observed. On multivariable analysis, after adjusting for standard clinicopathological features, high expression of G-CSF was significantly associated with better prognosis (Supplementary Table S3).
We describe hereafter the results of our detailed analysis on cohort II (n = 2960), data from which has been previously published for expression of immune biomarkers and CAIX [19,40,42,43].
For cohort II, the mean age of the patients at the time of diagnosis was 58.9 years (range: 23–95 years), and the median duration of follow-up was 12.5 years (range: 0.1–18.5 years). A total of 1956 deaths were recorded in the entire cohort, of which 58.4% were attributed to breast cancer. The clinicopathological characteristics of the study cohort are summarized in Table 1.
Of the 2960 cores, high expression of G-CSF was observed in 46.7% of tumours. After correcting for multiple comparisons, these cases demonstrated significantly higher rates of estrogen receptor (ER) negativity, HER2 positivity, CD163+ M2 macrophages, CAIX expression, and (IHC-defined) HER2 and basal intrinsic breast cancer subtypes, relative to the expression of these biomarkers among the G-CSFlow cases that comprised 53.3% of the study population (Table 1).
In univariate analysis, G-CSF expression was not prognostically informative on the whole cohort (HR 0.95, CI 0.83–1.08; p = 0.43) nor when the analysis was restricted to the luminal subtype, which comprised 74% of the G-CSF interpretable cases (HR 0.99, CI 0.84–1.16; p = 0.90). In contrast, amongst the non-luminal subtype, high expression of G-CSF correlated with a significantly improved BCSS (HR 0.74, CI 0.58–0.95; p = 0.02) (Figure 2). Similarly, high G-CSF was also associated with significantly improved OS and a trend toward a better RFS in non-luminal cases (Supplementary Figure S3).
Consistent with the univariate analysis, G-CSF expression was not an independent prognostic indicator in the whole cohort (n = 2960) using Cox proportional hazards model for multivariate analysis (HR 0.92, CI 0.80–1.06; p = 0.26). However, within the non-luminal cases (n = 665), G-CSFhigh tumours correlated with a 28% reduced risk of breast cancer-related deaths compared to G-CSFlow tumours, independent of the standard clinicopathological factors including age, tumour size, grade, lymphovascular invasion, and axillary lymph node metastasis (HR 0.72, CI 0.55–0.93; p = 0.01) (Figure 3). Within non-luminal subgroups, G-CSFhigh expression was significantly associated with better BCSS in HER2 subtype (n = 196); triple-negative subgroup (n = 469) and basal (n = 274) subgroups of non-luminal breast cancers did not generate hazard ratios that remained significant in multivariable analysis (data not shown).

3.2. Correlation and Prognostic Significance of G-CSF and CAIX with CD163+ M2 Macrophages and Immune Biomarkers

Non-luminal breast cancers contain regions of hypoxia and acidosis [47,48]. Preclinical studies have shown that G-CSF derived from the hypoxic tumour cells is crucial for mobilization of myeloid-derived suppressor cells to visceral organs, thus increasing the metastatic potential of breast cancer cells [22]. In line with this preclinical evidence, we observed a significant positive correlation between expression of G-CSF and CAIX (Table 1). Amongst non-luminal cases with a G-CSFhigh phenotype, significantly adverse BCSS (HR 1.74, CI 1.18–2.56; p = 0.004) was observed in CAIX-expressing tumours (Figure 4A). Similar results were observed for OS and RFS (Figure 4B,C), indicating that the expression of hypoxia-induced CAIX adversely impacted the prognosis of patients with non-luminal breast cancer with high tumoural expression of G-CSF. G-CSFhigh phenotype was associated with significantly better BCSS in CAIX-negative non-luminal cases expressing moderate to high CD163+M2 macrophages; however, no significant difference in survival was observed when analysis was restricted to CAIX-positive cases (Supplementary Figure S4).
Tumour-associated macrophage infiltration in invasive breast cancer is associated with adverse prognostic parameters. In agreement with previous reports [49], we also found that the presence of moderate to dense infiltrates of CD163+ M2 macrophages was associated with poor prognosis (Supplementary Figure S5).
Hypoxic and acidic tumour microenvironments are host to immunosuppressive cells including CD163+ M2 macrophages. Moreover, it has been shown that lactic acidosis induces a phenotypic switch from M1 to M2 macrophages, which supports tumour cell proliferation [33]. Considering the association of CAIX with hypoxia and acidosis, we performed an exploratory analysis to assess the prognostic significance of combinatorial expression of CAIX, CD163+ M2 macrophages, and G-CSF in non-luminal cases. We found that compared to the expression of all three biomarkers, the CAIXnegative/G-CSFhigh phenotype was associated with significantly improved relapse-free survival (HR 0.59, CI 0.4–0.86; p = 0.007), with a similar favourable trend observed for breast cancer-specific and overall survival (Figure 5), suggesting that hypoxia and acidosis are influential to the prognostic association of M2 macrophages in non-luminal breast cancer.
Exhausted T cells either overexpress inhibitory receptors (including programmed death receptor-1 (PD-1), lymphocyte activation gene-3 (LAG-3), and T-cell immunoglobulin domain and mucin domain-3 (TIM3)) or downregulate normal T-cell responses by increasing FOXP3 regulatory T cells [50]. To address this relationship in our cohort, we evaluated the correlation of CAIX and G-CSF with the above immune biomarkers, which have been previously assessed in this tissue microarray series [40,41,42,43]. We found that tumours with G-CSFhigh and CAIX+ phenotypes independently displayed highly significant positive correlations with the presence of intratumoural lymphocytes expressing CD8, PD-1, FOXP3, TIM3, and LAG3, with carcinoma cells expressing PD-L1, and with CD163+ M2 macrophages (Table 2).
Considering the immune-modulating role of G-CSF, we performed a multivariable analysis to investigate the individual prognostic significance of immune checkpoint biomarkers in G-CSF-expressing non-luminal cases. We found that intratumoural lymphocytes expressing CD8, PD-1, FOXP3, TIM3, and LAG3, as well as PD-L1-expressing carcinoma cells, were each associated with better survival, independent from standard clinicopathologic factors (Table 3). Similarly, most immune biomarkers maintained prognostic significance when the analysis was restricted to CAIX-expressing non-luminal cases (Supplementary Table S4).
Taken together, our results demonstrate that the prognostic value of G-CSF in non-luminal breast cancers is influenced by tumour microenvironmental features associated with CAIX positivity.

4. Discussion

We herein report the prognostic significance of G-CSF expression on breast carcinoma cells in a large population-based cohort of stage I–III invasive breast cancers. The most salient findings of our study were (a) a survival advantage associated with high expression of G-CSF in non-luminal subtypes of breast cancer; (b) a novel association between G-CSF, CAIX, and markers of immune exhaustion on tumour-infiltrating lymphocytes, which is consistent with the presence of an immune-suppressive hypoxic and acidic tumour microenvironment; (c) the identification of a patient subset among non-luminal breast cancers with high G-CSF and CAIX expression that is associated with worse survival compared to those that did not express CAIX; and (d) a positive correlation between tumour-infiltrating lymphocytes expressing immune checkpoint biomarkers and carcinoma cells expressing G-CSF, such that their concurrent presence is associated with a survival advantage in non-luminal immune-enriched subtypes.
The G-CSF/G-CSFR signalling axis has been implicated in stimulating neoangiogenesis, tumour cell proliferation, enhanced migratory and metastatic potential, and expansion of the cancer stem cell pool in preclinical models of several solid organ malignancies [51]. Additionally, there is an abundance of evidence to support that bidirectional signalling cues between G-CSF-mobilized immune cells and tumour-infiltrating lymphocytes can either inhibit or stimulate tumour progression by modulating both the innate and adaptive immune responses. For instance, using transgenic mouse models, it has been shown that mammary tumour-derived cytokines upregulate G-CSF production and facilitate tumour dissemination through expansion and reprogramming of neutrophils, which in turn restrain the effector functions of CD8+ cytotoxic T cells through production of inducible nitric oxide synthase [3]. Likewise, in co-culture experiments, G-CSF secretion by MDA-MB-231 breast cancer cells induces a phenotypic switch in peripherally derived monocytes toward immunosuppressive TAMs with enhanced migratory and metastatic potential, which could be abrogated by anti-G-CSF antibodies [32]. In another independent study, using 4T1 murine breast cancer cells, tumour-derived G-CSF was instrumental in promoting hematopoietic stem cell differentiation toward myeloid lineages, with expansion and activation of CD11b+ Ly6G+ neutrophils, which induce T-cell suppression through the production of reactive oxygen species [52]. Our data on clinical specimens confirms that G-CSF is associated with immune infiltration and supports an immunomodulatory role of G-CSF in the breast tumour microenvironment.
It is well documented that tumour hypoxia and ensuing acidification of the microenvironment promote immunological escape and resistance to immunotherapy by several mechanisms [53,54]. Firstly, hypoxic stress creates nutrient competition between the tumour cells and T lymphocytes such that the resulting nutrient deficit profoundly suppresses the T-cell effector functions leading to a state of hyporesponsiveness even in highly antigenic tumours [55]. Secondly, hypoxia-induced chemokine (CCL28) production [56] and FOXP3 transcriptional upregulation enriches for FOXP3+ T regulatory cells, which play a crucial role in self-tolerance [57]. Thirdly, hypoxic stress, via induction of hypoxia-inducible factor 1-α, increases the expression of exhaustion markers such as PD-L1 on the tumour cells, which bind to the PD-1 receptor on the surface of T cells, causing effector cell dysfunction and apoptosis [58]. Moreover, hypoxia-induced CAIX expression has been found in tumours resistant to anti-PD-1 therapy [59]. Consistent with this, we have previously demonstrated, in preclinical models of basal-like breast cancer and malignant melanoma, that CAIX expression is associated with an altered anti-tumour immune response and that its inhibition enhances the efficacy of immune checkpoint blockade [23]. Our data here reinforces our previous findings by identifying that CAIX expression at the protein level is associated with the expression of immune checkpoint molecules by the carcinoma cells and by infiltrating immune cells. Furthermore, the hypoxia and acidosis commonly associated with CAIX expression, and the immune modulation associated with these features, may be critical factors driving the prognostic dichotomization of G-CSFhigh and CD163+ M2 macrophage-infiltrated non-luminal breast cancers.
The identification of a soluble marker, G-CSF, whose expression is associated with the expression of CAIX in a clinical cohort, is an important finding. Our preclinical findings linked CAIX and G-CSF through the NF-κB signalling pathway [22]. Non-luminal breast cancers are known to rely on an active NF-κB signalling pathway; it is therefore plausible that this relationship exists in clinical samples to a certain degree [60,61]. Furthermore, several studies have described a role of G-CSF in tumour progression using the same hypoxic, CAIX-expressing 4T1 model [62,63,64] or additional models containing significant levels of hypoxia such as Lewis lung carcinoma [65] and MMTV-PyMT [66], suggesting a critical role of hypoxia in regulating G-CSF biology in these models. Importantly, cytokine networks in the hypoxic tumour microenvironment, including those orchestrated by TGF-β, may cooperate with G-CSF to potentially influence neutrophil polarization [67]. However, further experimental work is needed to fully elucidate these mechanisms [68].
Differences between the prognostic significance of G-CSF in our study and those previously reported in other tumour types may be attributed to fundamental differences in the underlying tumour biology. For example, inactivation of tumour suppressor pVHL is observed in up to 80% of clear cell renal cell carcinomas, yet this is rare in breast carcinomas [69,70]. The association of G-CSF with poor prognosis in these tumours [14] may be related to a predominant acidotic microenvironment due to constitutive activation of hypoxia-inducible factor 1-α and upregulation of CA9, which encodes CAIX [69]. Similarly, tumour hypoxia and CAIX expression are critical drivers [71,72] of the aggressive biology in patients diagnosed with cervical cancer and are likely to influence the prognostic capacity of G-CSF [15]. Hence, it is plausible that in the context of low hypoxic stress and consequently low CAIX, high tumoural G-CSF confers a survival advantage in breast cancer, as observed in our non-luminal cohort (Figure 5).
To date, limited studies have investigated the significance of G-CSF protein expression in breast cancer. Considering the strong association with neutrophilic mobilization and recruitment in carcinomas, serum G-CSF has been evaluated as a surrogate biomarker for prognostication. In this context, a few studies have shown a higher plasma level of G-CSF in breast cancer patients compared to healthy controls and post-surgical wound-healing fluids [73,74,75]. Since these studies were limited to smaller cohorts, meaningful associations with breast cancer subtypes or prognosis could not be addressed. To the best of our knowledge, only a single study has examined the immunohistochemical expression of G-CSF in a reasonably large breast cancer cohort (548 cases), reporting a significant positive correlation between G-CSF expression and CD163+ tumour-associated macrophages [32]. In a further subgroup analysis, it was shown that triple-negative breast cancer cases (n = 127) with high G-CSF expression were associated with poor overall survival [32]. Though the prognostic significance of G-CSF in this study is opposite of our observation, these discrepancies may be attributed to the small sample size of the earlier study, differences in the endpoints for survival analysis, the IHC protocols, and perhaps the level of CAIX expression in their cohort.
In a recent study, using publicly available data sets, CSF3R was identified as one of the differentially expressed genes in a subset of immune-rich ER+ breast cancers [76]. Although expression analysis of G-CSFR was not included in our study, we performed an exploratory analysis to address whether G-CSF expression contributed to prognostication in a subset of ER+ breast cancers with ≥10% TIL count (n = 284), and we observed no significant difference in survival between G-CSFhigh and G-CSFlow tumours (data not shown).
Our study has major strengths, including the use of analytically validated antibodies, predefined scoring criteria (CD163 and CAIX), and the use of an independent cohort (cohort I) to confirm the performance of G-CSF as a biomarker on breast cancer tissue microarrays. However, there are some important limitations worthy of mention. First, we had hoped to see significant findings in cohort I (which, being a training set, is at risk for overfitting) that could guide prespecified formal hypotheses testing using cohort II as an independent validation set. However, the results on the training set were not significant for our primary endpoint in univariable analysis. Instead, we were able to use the findings from cohort I to lock down the staining methodology, interpretation, and cutpoints, but our observations using the more powered cohort II with longer-term follow-up will need independent validation. Secondly, we did not evaluate the presence of tumour-associated neutrophils or myeloid-derived suppressor cells, which are the two major immune cell types mobilized and recruited to the tumour site by G-CSF. Instead, we took into consideration the expression of immune checkpoints on tumour-infiltrating lymphocytes as a related biomarker, hypothesizing that G-CSF mobilizes immune infiltrates and promotes an immune exhaustion phenotype on tumour-infiltrating lymphocytes. Thirdly, we limited our analysis to G-CSF expression by the carcinoma cells, and these results do not take into account the expression of G-CSF on immune cells, such as macrophages, endothelial cells, or other stromal cells that are part of the tumour immune microenvironment. The use of tissue microarrays may be considered a potential limitation of our study. While there are more than a few studies showing considerable agreement between cores derived from source tumour blocks and the full-face sections [77,78], this may not be the case for biomarkers that show high intratumoural heterogeneity and/or predominant expression in the tumour’s microenvironment. Furthermore, this study preceded the time when HER2-targeted therapies and taxanes were routinely used in adjuvant settings. This may limit the extrapolation of prognostic associations to patients receiving the current standard of care therapeutics and merits validation on a more contemporary or prospective series.

5. Conclusions

In conclusion, our results demonstrate the prognostic significance of G-CSF in invasive breast cancer, whereby high expression serves as an indicator of better survival in aggressive non-luminal subtypes of breast cancer in the absence of CAIX. Our identified associations between CAIX, G-CSF, and immune biomarkers provide a rationale for additional prospective investigations to understand the underlying mechanisms and their role as potential biomarkers for predicting responses to immune checkpoint inhibitor therapy.

Supplementary Materials

The following are available online at https://www.mdpi.com/2072-6694/13/5/1022/s1, Figure S1: Representative photomicrographs for immunohistochemical staining of G-CSF, CD163 and CAIX in serial sections in core # 1715 from breast carcinoma tissue microarray (cohort II). Black arrows indicate (A) cytoplasmic expression of G-CSF (>1) on breast carcinoma cells; (B) membranous or cytoplasmic expression of CD163 on tumor associated macrophages (>5 but ≤25); and (C) membranous expression of CAIX on breast carcinoma cells (Images acquired at 200X), Figure S2: Cohort I: Kaplan Meier curves: Association of G-CSF expression with breast cancer specific survival (A) and overall survival (B), Figure S3: Cohort II: Kaplan Meier curves: Association of G-CSF expression with overall survival (A) and relapse free survival (B) in non-luminal cases, Figure S4: Kaplan Meier curves for association of G-CSF in CD163+ non-luminal tumors with positive (A) and negative (B) expression of CAIX, Figure S5: Kaplan Meier curves: Presence of moderate and dense infiltrates of CD163+ M2 tumor associated macrophages is associated with poor breast cancer specific survival, Table S1: Details of the antibodies and staining protocols, Table S2: Cohort I: Correlation of G-CSF expression with clinicopathological features & other biomarkers, Table S3: Cohort I: Multivariable analysis for breast cancer specific survival and overall survival, Table S4: Cohort II: Multivariable analysis for prognostic significance of immune biomarkers within non-luminal cases with CAIX positive expression.

Author Contributions

Conceptualization, S.C.C., S.B., S.D., and T.O.N.; methodology, S.C.C., N.R., S.B., A.F.L., C.-H.L., S.D., and T.O.N.; formal analysis, S.C.C., N.R., S.C.Y.L., D.G., A.F.L., C.-H.L., and T.O.N.; investigation, S.C.C., N.R., D.G., A.F.L., C.-H.L., and T.O.N.; resources, T.O.N.; data curation, S.C.Y.L. and T.O.N.; writing—original draft preparation, S.C.C., N.R., S.B., S.D., and T.O.N.; writing—review and editing S.C.C., N.R., S.B., S.C.Y.L., D.G., A.F.L., C.-H.L., S.D., and T.O.N.; visualization, S.C.C., N.R., S.C.Y.L., and T.O.N.; supervision, S.D. and T.O.N.; project administration, S.D. and T.O.N.; funding acquisition, S.D. and T.O.N. All authors have read and agreed to the published version of the manuscript.

Funding

This research was supported by funds from the Canadian Institutes of Health Research (CIHR) Foundation Scheme grant (FDN-143348) awarded to S.D. and by an operating grant from the Canadian Cancer Society (#705463) to T.O.N.

Institutional Review Board Statement

The Clinical Research Ethics Board of the University of British Columbia and British Columbia Cancer Agency approved this research project and the access to the deidentified data.

Informed Consent Statement

In accordance with the Canadian Tri-Council Policy Statement for ethical research involving human subjects, informed consent was waived as anonymized archival specimens were used in this study.

Data Availability Statement

The data presented in this study will be available on request from the corresponding author. Images from the tissue microarray slides are available for public access via the website of the Genetic Pathology Evaluation Center (http://www.gpec.ubc.ca/gcsf accessed on 1 January 2021).

Conflicts of Interest

Nielsen reports grants from the Canadian Cancer Society during the conduct of the study; personal fees from Bioclassifier LLC, personal fees from NanoString Technologies, outside the submitted work; In addition, Nielsen has a patent gene expression profile to predict breast cancer outcomes with royalties paid by licensee Veracyte. Other authors have nothing to disclose. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

References

  1. Hareng, L.; Hartung, T. Induction and regulation of endogenous granulocyte colony-stimulating factor formation. Biol. Chem. 2002, 383, 1501–1517. [Google Scholar] [CrossRef] [PubMed]
  2. Edelsberg, J.; Weycker, D.; Bensink, M.; Bowers, C.; Lyman, G.H. Prophylaxis of febrile neutropenia with colony-stimulating factors: The first 25 years. Curr. Med. Res. Opin. 2020, 36, 483–495. [Google Scholar] [CrossRef] [Green Version]
  3. Coffelt, S.B.; Kersten, K.; Doornebal, C.W.; Weiden, J.; Vrijland, K.; Hau, C.S.; Verstegen, N.J.M.; Ciampricotti, M.; Hawinkels, L.; Jonkers, J.; et al. IL-17-producing gammadelta T cells and neutrophils conspire to promote breast cancer metastasis. Nature 2015, 522, 345–348. [Google Scholar] [CrossRef] [PubMed]
  4. Wellenstein, M.D.; Coffelt, S.B.; Duits, D.E.M.; van Miltenburg, M.H.; Slagter, M.; de Rink, I.; Henneman, L.; Kas, S.M.; Prekovic, S.; Hau, C.S.; et al. Loss of p53 triggers WNT-dependent systemic inflammation to drive breast cancer metastasis. Nature 2019, 572, 538–542. [Google Scholar] [CrossRef]
  5. Kumar, J.; Fraser, F.W.; Riley, C.; Ahmed, N.; McCulloch, D.R.; Ward, A.C. Granulocyte colony-stimulating factor receptor signalling via Janus kinase 2/signal transducer and activator of transcription 3 in ovarian cancer. Br. J. Cancer 2014, 110, 133–145. [Google Scholar] [CrossRef]
  6. Chakraborty, A.; Guha, S. Granulocyte colony-stimulating factor/granulocyte colony-stimulating factor receptor biological axis promotes survival and growth of bladder cancer cells. Urology 2007, 69, 1210–1215. [Google Scholar] [CrossRef]
  7. Morris, K.T.; Khan, H.; Ahmad, A.; Weston, L.L.; Nofchissey, R.A.; Pinchuk, I.V.; Beswick, E.J. G-CSF and G-CSFR are highly expressed in human gastric and colon cancers and promote carcinoma cell proliferation and migration. Br. J. Cancer 2014, 110, 1211–1220. [Google Scholar] [CrossRef] [Green Version]
  8. Dobrenis, K.; Gauthier, L.R.; Barroca, V.; Magnon, C. Granulocyte colony-stimulating factor off-target effect on nerve outgrowth promotes prostate cancer development. Int. J. Cancer 2015, 136, 982–988. [Google Scholar] [CrossRef]
  9. Phan, V.T.; Wu, X.; Cheng, J.H.; Sheng, R.X.; Chung, A.S.; Zhuang, G.; Tran, C.; Song, Q.; Kowanetz, M.; Sambrone, A.; et al. Oncogenic RAS pathway activation promotes resistance to anti-VEGF therapy through G-CSF-induced neutrophil recruitment. Proc. Natl. Acad. Sci. USA 2013, 110, 6079–6084. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  10. Gutschalk, C.M.; Herold-Mende, C.C.; Fusenig, N.E.; Mueller, M.M. Granulocyte colony-stimulating factor and granulocyte-macrophage colony-stimulating factor promote malignant growth of cells from head and neck squamous cell carcinomas in vivo. Cancer Res. 2006, 66, 8026–8036. [Google Scholar] [CrossRef] [Green Version]
  11. Liu, L.; Wu, Y.; Zhang, C.; Zhou, C.; Li, Y.; Zeng, Y.; Zhang, C.; Li, R.; Luo, D.; Wang, L.; et al. Cancer-associated adipocytes-derived G-CSF promotes breast cancer malignancy via Stat3 signaling. J. Mol. Cell Biol. 2020. [Google Scholar] [CrossRef] [Green Version]
  12. Pickup, M.W.; Owens, P.; Gorska, A.E.; Chytil, A.; Ye, F.; Shi, C.; Weaver, V.M.; Kalluri, R.; Moses, H.L.; Novitskiy, S.V. Development of Aggressive Pancreatic Ductal Adenocarcinomas Depends on Granulocyte Colony Stimulating Factor Secretion in Carcinoma Cells. Cancer Immunol. Res. 2017, 5, 718–729. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  13. Gabrilovich, D.I.; Ostrand-Rosenberg, S.; Bronte, V. Coordinated regulation of myeloid cells by tumours. Nat. Rev. Immunol. 2012, 12, 253–268. [Google Scholar] [CrossRef] [Green Version]
  14. Liu, Z.; Zhu, Y.; Wang, Y.; Fu, Q.; Fu, H.; Wang, Z.; Zhang, J.; Li, G.; Xu, J.; Dai, B. Prognostic value of granulocyte colony-stimulating factor in patients with non-metastatic clear cell renal cell carcinoma. Oncotarget 2017, 8, 69961–69971. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  15. Kawano, M.; Mabuchi, S.; Matsumoto, Y.; Sasano, T.; Takahashi, R.; Kuroda, H.; Kozasa, K.; Hashimoto, K.; Isobe, A.; Sawada, K.; et al. The significance of G-CSF expression and myeloid-derived suppressor cells in the chemoresistance of uterine cervical cancer. Sci. Rep. 2015, 5, 18217. [Google Scholar] [CrossRef]
  16. Vito, A.; El-Sayes, N.; Mossman, K. Hypoxia-Driven Immune Escape in the Tumor Microenvironment. Cells 2020, 9, 992. [Google Scholar] [CrossRef]
  17. Hanahan, D.; Weinberg, R.A. Hallmarks of cancer: The next generation. Cell 2011, 144, 646–674. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  18. Alterio, V.; Hilvo, M.; Di Fiore, A.; Supuran, C.T.; Pan, P.; Parkkila, S.; Scaloni, A.; Pastorek, J.; Pastorekova, S.; Pedone, C.; et al. Crystal structure of the catalytic domain of the tumor-associated human carbonic anhydrase IX. Proc. Natl. Acad. Sci. USA 2009, 106, 16233–16238. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  19. Lou, Y.; McDonald, P.C.; Oloumi, A.; Chia, S.; Ostlund, C.; Ahmadi, A.; Kyle, A.; Auf dem Keller, U.; Leung, S.; Huntsman, D.; et al. Targeting tumor hypoxia: Suppression of breast tumor growth and metastasis by novel carbonic anhydrase IX inhibitors. Cancer Res. 2011, 71, 3364–3376. [Google Scholar] [CrossRef] [Green Version]
  20. Hussain, S.A.; Ganesan, R.; Reynolds, G.; Gross, L.; Stevens, A.; Pastorek, J.; Murray, P.G.; Perunovic, B.; Anwar, M.S.; Billingham, L.; et al. Hypoxia-regulated carbonic anhydrase IX expression is associated with poor survival in patients with invasive breast cancer. Br. J. Cancer 2007, 96, 104–109. [Google Scholar] [CrossRef]
  21. Chia, S.K.; Wykoff, C.C.; Watson, P.H.; Han, C.; Leek, R.D.; Pastorek, J.; Gatter, K.C.; Ratcliffe, P.; Harris, A.L. Prognostic significance of a novel hypoxia-regulated marker, carbonic anhydrase IX, in invasive breast carcinoma. J. Clin. Oncol. 2001, 19, 3660–3668. [Google Scholar] [CrossRef] [PubMed]
  22. Chafe, S.C.; Lou, Y.; Sceneay, J.; Vallejo, M.; Hamilton, M.J.; McDonald, P.C.; Bennewith, K.L.; Moller, A.; Dedhar, S. Carbonic anhydrase IX promotes myeloid-derived suppressor cell mobilization and establishment of a metastatic niche by stimulating G-CSF production. Cancer Res. 2015, 75, 996–1008. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  23. Chafe, S.C.; McDonald, P.C.; Saberi, S.; Nemirovsky, O.; Venkateswaran, G.; Burugu, S.; Gao, D.; Delaidelli, A.; Kyle, A.H.; Baker, J.H.E.; et al. Targeting Hypoxia-Induced Carbonic Anhydrase IX Enhances Immune-Checkpoint Blockade Locally and Systemically. Cancer Immunol. Res. 2019, 7, 1064–1078. [Google Scholar] [CrossRef] [Green Version]
  24. Noman, M.Z.; Hasmim, M.; Messai, Y.; Terry, S.; Kieda, C.; Janji, B.; Chouaib, S. Hypoxia: A key player in antitumor immune response. A Review in the Theme: Cellular Responses to Hypoxia. Am. J. Physiol. Cell Physiol. 2015, 309, C569–C579. [Google Scholar] [CrossRef] [Green Version]
  25. Jeong, H.; Kim, S.; Hong, B.J.; Lee, C.J.; Kim, Y.E.; Bok, S.; Oh, J.M.; Gwak, S.H.; Yoo, M.Y.; Lee, M.S.; et al. Tumor-Associated Macrophages Enhance Tumor Hypoxia and Aerobic Glycolysis. Cancer Res. 2019, 79, 795–806. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  26. Chaturvedi, P.; Gilkes, D.M.; Takano, N.; Semenza, G.L. Hypoxia-inducible factor-dependent signaling between triple-negative breast cancer cells and mesenchymal stem cells promotes macrophage recruitment. Proc. Natl. Acad. Sci. USA 2014, 111, E2120–E2129. [Google Scholar] [CrossRef] [Green Version]
  27. Qiu, S.Q.; Waaijer, S.J.H.; Zwager, M.C.; de Vries, E.G.E.; van der Vegt, B.; Schroder, C.P. Tumor-associated macrophages in breast cancer: Innocent bystander or important player? Cancer Treat. Rev. 2018, 70, 178–189. [Google Scholar] [CrossRef] [Green Version]
  28. Ong, S.M.; Tan, Y.C.; Beretta, O.; Jiang, D.; Yeap, W.H.; Tai, J.J.; Wong, W.C.; Yang, H.; Schwarz, H.; Lim, K.H.; et al. Macrophages in human colorectal cancer are pro-inflammatory and prime T cells towards an anti-tumour type-1 inflammatory response. Eur. J. Immunol. 2012, 42, 89–100. [Google Scholar] [CrossRef] [PubMed]
  29. Honkanen, T.J.; Tikkanen, A.; Karihtala, P.; Makinen, M.; Vayrynen, J.P.; Koivunen, J.P. Prognostic and predictive role of tumour-associated macrophages in HER2 positive breast cancer. Sci. Rep. 2019, 9, 10961. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  30. Sousa, S.; Brion, R.; Lintunen, M.; Kronqvist, P.; Sandholm, J.; Monkkonen, J.; Kellokumpu-Lehtinen, P.L.; Lauttia, S.; Tynninen, O.; Joensuu, H.; et al. Human breast cancer cells educate macrophages toward the M2 activation status. Breast Cancer Res. 2015, 17, 101. [Google Scholar] [CrossRef] [Green Version]
  31. Klingen, T.A.; Chen, Y.; Aas, H.; Wik, E.; Akslen, L.A. Tumor-associated macrophages are strongly related to vascular invasion, non-luminal subtypes, and interval breast cancer. Hum. Pathol. 2017, 69, 72–80. [Google Scholar] [CrossRef]
  32. Hollmen, M.; Karaman, S.; Schwager, S.; Lisibach, A.; Christiansen, A.J.; Maksimow, M.; Varga, Z.; Jalkanen, S.; Detmar, M. G-CSF regulates macrophage phenotype and associates with poor overall survival in human triple-negative breast cancer. Oncoimmunology 2016, 5, e1115177. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  33. Colegio, O.R.; Chu, N.Q.; Szabo, A.L.; Chu, T.; Rhebergen, A.M.; Jairam, V.; Cyrus, N.; Brokowski, C.E.; Eisenbarth, S.C.; Phillips, G.M.; et al. Functional polarization of tumour-associated macrophages by tumour-derived lactic acid. Nature 2014, 513, 559–563. [Google Scholar] [CrossRef]
  34. Bortnik, S.; Choutka, C.; Horlings, H.M.; Leung, S.; Baker, J.H.; Lebovitz, C.; Dragowska, W.H.; Go, N.E.; Bally, M.B.; Minchinton, A.I.; et al. Identification of breast cancer cell subtypes sensitive to ATG4B inhibition. Oncotarget 2016, 7, 66970–66988. [Google Scholar] [CrossRef] [PubMed]
  35. Chia, S.K.; Speers, C.H.; Bryce, C.J.; Hayes, M.M.; Olivotto, I.A. Ten-year outcomes in a population-based cohort of node-negative, lymphatic, and vascular invasion-negative early breast cancers without adjuvant systemic therapies. J. Clin. Oncol. 2004, 22, 1630–1637. [Google Scholar] [CrossRef] [PubMed]
  36. Cheang, M.C.; Voduc, D.; Bajdik, C.; Leung, S.; McKinney, S.; Chia, S.K.; Perou, C.M.; Nielsen, T.O. Basal-like breast cancer defined by five biomarkers has superior prognostic value than triple-negative phenotype. Clin. Cancer Res. 2008, 14, 1368–1376. [Google Scholar] [CrossRef] [Green Version]
  37. Sauerbrei, W.; Taube, S.E.; McShane, L.M.; Cavenagh, M.M.; Altman, D.G. Reporting Recommendations for Tumor Marker Prognostic Studies (REMARK): An Abridged Explanation and Elaboration. J. Natl. Cancer Inst. 2018, 110, 803–811. [Google Scholar] [CrossRef]
  38. Cheang, M.C.; Treaba, D.O.; Speers, C.H.; Olivotto, I.A.; Bajdik, C.D.; Chia, S.K.; Goldstein, L.C.; Gelmon, K.A.; Huntsman, D.; Gilks, C.B.; et al. Immunohistochemical detection using the new rabbit monoclonal antibody SP1 of estrogen receptor in breast cancer is superior to mouse monoclonal antibody 1D5 in predicting survival. J. Clin. Oncol. 2006, 24, 5637–5644. [Google Scholar] [CrossRef] [Green Version]
  39. Voduc, D.; Kenney, C.; Nielsen, T.O. Tissue microarrays in clinical oncology. Semin. Radiat. Oncol. 2008, 18, 89–97. [Google Scholar] [CrossRef] [Green Version]
  40. Liu, S.; Lachapelle, J.; Leung, S.; Gao, D.; Foulkes, W.D.; Nielsen, T.O. CD8+ lymphocyte infiltration is an independent favorable prognostic indicator in basal-like breast cancer. Breast Cancer Res. 2012, 14, R48. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  41. Burugu, S.; Gao, D.; Leung, S.; Chia, S.K.; Nielsen, T.O. TIM-3 expression in breast cancer. Oncoimmunology 2018, 7, e1502128. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  42. Liu, S.; Foulkes, W.D.; Leung, S.; Gao, D.; Lau, S.; Kos, Z.; Nielsen, T.O. Prognostic significance of FOXP3+ tumor-infiltrating lymphocytes in breast cancer depends on estrogen receptor and human epidermal growth factor receptor-2 expression status and concurrent cytotoxic T-cell infiltration. Breast Cancer Res. 2014, 16, 432. [Google Scholar] [CrossRef] [Green Version]
  43. Burugu, S.; Gao, D.; Leung, S.; Chia, S.K.; Nielsen, T.O. LAG-3+ tumor infiltrating lymphocytes in breast cancer: Clinical correlates and association with PD-1/PD-L1+ tumors. Ann. Oncol. 2017, 28, 2977–2984. [Google Scholar] [CrossRef] [PubMed]
  44. Salgado, R.; Denkert, C.; Demaria, S.; Sirtaine, N.; Klauschen, F.; Pruneri, G.; Wienert, S.; Van den Eynden, G.; Baehner, F.L.; Penault-Llorca, F.; et al. The evaluation of tumor-infiltrating lymphocytes (TILs) in breast cancer: Recommendations by an International TILs Working Group 2014. Ann. Oncol. 2015, 26, 259–271. [Google Scholar] [CrossRef]
  45. Zhu, M.M.T.; Burugu, S.; Gao, D.; Yu, J.; Kos, Z.; Leung, S.; Horst, B.A.; Nielsen, T.O. Evaluation of glucocorticoid-induced TNF receptor (GITR) expression in breast cancer and across multiple tumor types. Mod. Pathol. 2020. [Google Scholar] [CrossRef]
  46. Lee, C.H.; Espinosa, I.; Vrijaldenhoven, S.; Subramanian, S.; Montgomery, K.D.; Zhu, S.; Marinelli, R.J.; Peterse, J.L.; Poulin, N.; Nielsen, T.O.; et al. Prognostic significance of macrophage infiltration in leiomyosarcomas. Clin. Cancer Res. 2008, 14, 1423–1430. [Google Scholar] [CrossRef] [Green Version]
  47. Tan, E.Y.; Yan, M.; Campo, L.; Han, C.; Takano, E.; Turley, H.; Candiloro, I.; Pezzella, F.; Gatter, K.C.; Millar, E.K.; et al. The key hypoxia regulated gene CAIX is upregulated in basal-like breast tumours and is associated with resistance to chemotherapy. Br. J. Cancer 2009, 100, 405–411. [Google Scholar] [CrossRef]
  48. Gatza, M.L.; Kung, H.N.; Blackwell, K.L.; Dewhirst, M.W.; Marks, J.R.; Chi, J.T. Analysis of tumor environmental response and oncogenic pathway activation identifies distinct basal and luminal features in HER2-related breast tumor subtypes. Breast Cancer Res. 2011, 13, R62. [Google Scholar] [CrossRef] [Green Version]
  49. Zhang, Q.W.; Liu, L.; Gong, C.Y.; Shi, H.S.; Zeng, Y.H.; Wang, X.Z.; Zhao, Y.W.; Wei, Y.Q. Prognostic significance of tumor-associated macrophages in solid tumor: A meta-analysis of the literature. PLoS ONE 2012, 7, e50946. [Google Scholar] [CrossRef] [Green Version]
  50. Pardoll, D.M. The blockade of immune checkpoints in cancer immunotherapy. Nat. Rev. Cancer 2012, 12, 252–264. [Google Scholar] [CrossRef] [Green Version]
  51. Yeo, B.; Redfern, A.D.; Mouchemore, K.A.; Hamilton, J.A.; Anderson, R.L. The dark side of granulocyte-colony stimulating factor: A supportive therapy with potential to promote tumour progression. Clin. Exp. Metastasis 2018, 35, 255–267. [Google Scholar] [CrossRef]
  52. Casbon, A.J.; Reynaud, D.; Park, C.; Khuc, E.; Gan, D.D.; Schepers, K.; Passegue, E.; Werb, Z. Invasive breast cancer reprograms early myeloid differentiation in the bone marrow to generate immunosuppressive neutrophils. Proc. Natl. Acad. Sci. USA 2015, 112, E566–E575. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  53. Brand, A.; Singer, K.; Koehl, G.E.; Kolitzus, M.; Schoenhammer, G.; Thiel, A.; Matos, C.; Bruss, C.; Klobuch, S.; Peter, K.; et al. LDHA-Associated Lactic Acid Production Blunts Tumor Immunosurveillance by T and NK Cells. Cell Metab. 2016, 24, 657–671. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  54. Pilon-Thomas, S.; Kodumudi, K.N.; El-Kenawi, A.E.; Russell, S.; Weber, A.M.; Luddy, K.; Damaghi, M.; Wojtkowiak, J.W.; Mule, J.J.; Ibrahim-Hashim, A.; et al. Neutralization of Tumor Acidity Improves Antitumor Responses to Immunotherapy. Cancer Res. 2016, 76, 1381–1390. [Google Scholar] [CrossRef] [Green Version]
  55. Chang, C.H.; Qiu, J.; O’Sullivan, D.; Buck, M.D.; Noguchi, T.; Curtis, J.D.; Chen, Q.; Gindin, M.; Gubin, M.M.; van der Windt, G.J.; et al. Metabolic Competition in the Tumor Microenvironment Is a Driver of Cancer Progression. Cell 2015, 162, 1229–1241. [Google Scholar] [CrossRef] [Green Version]
  56. Facciabene, A.; Peng, X.; Hagemann, I.S.; Balint, K.; Barchetti, A.; Wang, L.P.; Gimotty, P.A.; Gilks, C.B.; Lal, P.; Zhang, L.; et al. Tumour hypoxia promotes tolerance and angiogenesis via CCL28 and T(reg) cells. Nature 2011, 475, 226–230. [Google Scholar] [CrossRef]
  57. Westendorf, A.M.; Skibbe, K.; Adamczyk, A.; Buer, J.; Geffers, R.; Hansen, W.; Pastille, E.; Jendrossek, V. Hypoxia Enhances Immunosuppression by Inhibiting CD4+ Effector T Cell Function and Promoting Treg Activity. Cell Physiol. Biochem. 2017, 41, 1271–1284. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  58. Barsoum, I.B.; Smallwood, C.A.; Siemens, D.R.; Graham, C.H. A mechanism of hypoxia-mediated escape from adaptive immunity in cancer cells. Cancer Res 2014, 74, 665–674. [Google Scholar] [CrossRef] [Green Version]
  59. Gide, T.N.; Quek, C.; Menzies, A.M.; Tasker, A.T.; Shang, P.; Holst, J.; Madore, J.; Lim, S.Y.; Velickovic, R.; Wongchenko, M.; et al. Distinct Immune Cell Populations Define Response to Anti-PD-1 Monotherapy and Anti-PD-1/Anti-CTLA-4 Combined Therapy. Cancer Cell 2019, 35, 238–255. [Google Scholar] [CrossRef] [Green Version]
  60. Biswas, D.K.; Shi, Q.; Baily, S.; Strickland, I.; Ghosh, S.; Pardee, A.B.; Iglehart, J.D. NF-kappa B activation in human breast cancer specimens and its role in cell proliferation and apoptosis. Proc. Natl. Acad. Sci. USA 2004, 101, 10137–10142. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  61. Bailey, S.T.; Miron, P.L.; Choi, Y.J.; Kochupurakkal, B.; Maulik, G.; Rodig, S.J.; Tian, R.; Foley, K.M.; Bowman, T.; Miron, A.; et al. NF-kappaB activation-induced anti-apoptosis renders HER2-positive cells drug resistant and accelerates tumor growth. Mol. Cancer Res. 2014, 12, 408–420. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  62. Swierczak, A.; Cook, A.D.; Lenzo, J.C.; Restall, C.M.; Doherty, J.P.; Anderson, R.L.; Hamilton, J.A. The promotion of breast cancer metastasis caused by inhibition of CSF-1R/CSF-1 signaling is blocked by targeting the G-CSF receptor. Cancer Immunol. Res. 2014, 2, 765–776. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  63. Waight, J.D.; Hu, Q.; Miller, A.; Liu, S.; Abrams, S.I. Tumor-derived G-CSF facilitates neoplastic growth through a granulocytic myeloid-derived suppressor cell-dependent mechanism. PLoS ONE 2011, 6, e27690. [Google Scholar] [CrossRef]
  64. Kowanetz, M.; Wu, X.; Lee, J.; Tan, M.; Hagenbeek, T.; Qu, X.; Yu, L.; Ross, J.; Korsisaari, N.; Cao, T.; et al. Granulocyte-colony stimulating factor promotes lung metastasis through mobilization of Ly6G+Ly6C+ granulocytes. Proc. Natl. Acad. Sci. USA 2010, 107, 21248–21255. [Google Scholar] [CrossRef] [Green Version]
  65. Clement-Colmou, K.; Potiron, V.; Pietri, M.; Guillonneau, M.; Jouglar, E.; Chiavassa, S.; Delpon, G.; Paris, F.; Supiot, S. Influence of Radiotherapy Fractionation Schedule on the Tumor Vascular Microenvironment in Prostate and Lung Cancer Models. Cancers 2020, 12, 121. [Google Scholar] [CrossRef] [Green Version]
  66. Adochite, R.C.; Moshnikova, A.; Carlin, S.D.; Guerrieri, R.A.; Andreev, O.A.; Lewis, J.S.; Reshetnyak, Y.K. Targeting breast tumors with pH (low) insertion peptides. Mol. Pharm. 2014, 11, 2896–2905. [Google Scholar] [CrossRef]
  67. Fridlender, Z.G.; Sun, J.; Kim, S.; Kapoor, V.; Cheng, G.; Ling, L.; Worthen, G.S.; Albelda, S.M. Polarization of tumor-associated neutrophil phenotype by TGF-beta: “N1” versus “N2” TAN. Cancer Cell 2009, 16, 183–194. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  68. Hsu, B.E.; Tabaries, S.; Johnson, R.M.; Andrzejewski, S.; Senecal, J.; Lehuede, C.; Annis, M.G.; Ma, E.H.; Vols, S.; Ramsay, L.; et al. Immature Low-Density Neutrophils Exhibit Metabolic Flexibility that Facilitates Breast Cancer Liver Metastasis. Cell Rep. 2019, 27, 3902–3915.e6. [Google Scholar] [CrossRef] [Green Version]
  69. Schodel, J.; Grampp, S.; Maher, E.R.; Moch, H.; Ratcliffe, P.J.; Russo, P.; Mole, D.R. Hypoxia, Hypoxia-inducible Transcription Factors, and Renal Cancer. Eur. Urol. 2016, 69, 646–657. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  70. Sourvinos, G.; Miyakis, S.; Liloglou, T.L.; Field, J.K.; Spandidos, D.A. Von Hippel-Lindau tumour suppressor gene is not involved in sporadic human breast cancer. Tumour Biol. 2001, 22, 131–136. [Google Scholar] [CrossRef] [PubMed]
  71. Höckel, M.; Schlenger, K.; Aral, B.; Mitze, M.; Schäffer, U.; Vaupel, P. Association between Tumor Hypoxia and Malignant Progression in Advanced Cancer of the Uterine Cervix. Cancer Res. 1996, 56, 4509–4515. [Google Scholar]
  72. Loncaster, J.A.; Harris, A.L.; Davidson, S.E.; Logue, J.P.; Hunter, R.D.; Wycoff, C.C.; Pastorek, J.; Ratcliffe, P.J.; Stratford, I.J.; West, C.M. Carbonic anhydrase (CA IX) expression, a potential new intrinsic marker of hypoxia: Correlations with tumor oxygen measurements and prognosis in locally advanced carcinoma of the cervix. Cancer Res. 2001, 61, 6394–6399. [Google Scholar] [PubMed]
  73. Park, S.; Kim, E.S.; Noh, D.Y.; Hwang, K.T.; Moon, A. H-Ras-specific upregulation of granulocyte colony-stimulating factor promotes human breast cell invasion via matrix metalloproteinase-2. Cytokine 2011, 55, 126–133. [Google Scholar] [CrossRef] [PubMed]
  74. Lawicki, S.; Bedkowska, G.E.; Wojtukiewicz, M.; Szmitkowski, M. Hematopoietic cytokines as tumor markers in breast malignancies. A multivariate analysis with ROC curve in breast cancer patients. Adv. Med. Sci. 2013, 58, 207–215. [Google Scholar] [CrossRef] [Green Version]
  75. Agresti, R.; Triulzi, T.; Sasso, M.; Ghirelli, C.; Aiello, P.; Rybinska, I.; Campiglio, M.; Sfondrini, L.; Tagliabue, E.; Bianchi, F. Wound Healing Fluid Reflects the Inflammatory Nature and Aggressiveness of Breast Tumors. Cells 2019, 8, 181. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  76. O’Meara, T.; Marczyk, M.; Qing, T.; Yaghoobi, V.; Blenman, K.; Cole, K.; Pelekanou, V.; Rimm, D.L.; Pusztai, L. Immunological Differences Between Immune-Rich Estrogen Receptor–Positive and Immune-Rich Triple-Negative Breast Cancers. JCO Precis. Oncol. 2020, 767–779. [Google Scholar] [CrossRef] [PubMed]
  77. Torhorst, J.; Bucher, C.; Kononen, J.; Haas, P.; Zuber, M.; Kochli, O.R.; Mross, F.; Dieterich, H.; Moch, H.; Mihatsch, M.; et al. Tissue microarrays for rapid linking of molecular changes to clinical endpoints. Am. J. Pathol. 2001, 159, 2249–2256. [Google Scholar] [CrossRef] [Green Version]
  78. Kyndi, M.; Sorensen, F.B.; Knudsen, H.; Overgaard, M.; Nielsen, H.M.; Andersen, J.; Overgaard, J. Tissue microarrays compared with whole sections and biochemical analyses. A subgroup analysis of DBCG 82 b&c. Acta Oncol. 2008, 47, 591–599. [Google Scholar] [CrossRef]
Cancers 13 01022 g001 550
Figure 1. Representative photomicrographs for immunohistochemical staining of G-CSF, CD163, and carbonic anhydrase IX (CAIX) in breast carcinoma tissue microarray (cohort II) (Images acquired at 200X). (A) Cytoplasmic expression of G-CSF on breast carcinoma cells; (A-i) low (≤1); (A-ii) high (>1); (B) Membranous or cytoplasmic expression of CD163 on tumour-associated macrophages; (B-i) sparse infiltrates, ≤5 stained macrophages; (B-ii) moderate infiltrates, >5 but ≤25 positively stained macrophages; (B-iii) dense infiltrates, >25 positively stained macrophages; (C) Membranous expression of CAIX on breast carcinoma cells; (C-i) negative; (C-ii) positive.
Figure 1. Representative photomicrographs for immunohistochemical staining of G-CSF, CD163, and carbonic anhydrase IX (CAIX) in breast carcinoma tissue microarray (cohort II) (Images acquired at 200X). (A) Cytoplasmic expression of G-CSF on breast carcinoma cells; (A-i) low (≤1); (A-ii) high (>1); (B) Membranous or cytoplasmic expression of CD163 on tumour-associated macrophages; (B-i) sparse infiltrates, ≤5 stained macrophages; (B-ii) moderate infiltrates, >5 but ≤25 positively stained macrophages; (B-iii) dense infiltrates, >25 positively stained macrophages; (C) Membranous expression of CAIX on breast carcinoma cells; (C-i) negative; (C-ii) positive.
Cancers 13 01022 g001
Cancers 13 01022 g002 550
Figure 2. Cohort II: Kaplan–Meier curves: Association of G-CSF expression with breast cancer-specific survival (BCSS). No significant difference in BCSS was observed in the whole cohort (A) and cases with luminal subtype (B). An improved BCSS was observed in cases with non-luminal subtype (C).
Figure 2. Cohort II: Kaplan–Meier curves: Association of G-CSF expression with breast cancer-specific survival (BCSS). No significant difference in BCSS was observed in the whole cohort (A) and cases with luminal subtype (B). An improved BCSS was observed in cases with non-luminal subtype (C).
Cancers 13 01022 g002
Cancers 13 01022 g003 550
Figure 3. Cohort II: Forest plots based on the results of multivariable analysis for risk factors associated with breast cancer-specific survival in whole series (A) and in non-luminal cases (B).
Figure 3. Cohort II: Forest plots based on the results of multivariable analysis for risk factors associated with breast cancer-specific survival in whole series (A) and in non-luminal cases (B).
Cancers 13 01022 g003
Cancers 13 01022 g004 550
Figure 4. Kaplan–Meier curves for association of CAIX expression in non-luminal cases with G-CSFhigh phenotype. Expression of CAIX is associated with adverse breast cancer-specific survival (A), overall survival (B), and relapse-free survival (C).
Figure 4. Kaplan–Meier curves for association of CAIX expression in non-luminal cases with G-CSFhigh phenotype. Expression of CAIX is associated with adverse breast cancer-specific survival (A), overall survival (B), and relapse-free survival (C).
Cancers 13 01022 g004
Cancers 13 01022 g005 550
Figure 5. Kaplan–Meier curves for association of CAIX and G-CSF in CD163+ non-luminal cases. Breast cancer-specific survival (A), overall survival (B), relapse-free survival (C). CAIXnegative/G-CSFhigh phenotype is associated with significantly better relapse-free survival compared to CAIXpositive/G-CSFlow phenotype (C).
Figure 5. Kaplan–Meier curves for association of CAIX and G-CSF in CD163+ non-luminal cases. Breast cancer-specific survival (A), overall survival (B), relapse-free survival (C). CAIXnegative/G-CSFhigh phenotype is associated with significantly better relapse-free survival compared to CAIXpositive/G-CSFlow phenotype (C).
Cancers 13 01022 g005
Table
Table 1. Correlation of G-CSF expression with clinicopathological features and other biomarkers.
Table 1. Correlation of G-CSF expression with clinicopathological features and other biomarkers.
Clinicopathological VariablesG-CSF Expressionp-Value
Low (≤1)High (>1)
Age at diagnosis 0.02
<50452 (28.7)451 (32.6)
≥501125 (71.3)932 (67.4)
Menstrual status 0.02
Premenopausal434 (28.1)434 (32)
Postmenopausal1111 (71.9)922 (68)
Tumour size (cm) 0.06
≤2784 (50)736 (53.4)
>2785 (50)641 (46.6)
Tumour grade 0.92
1 & 2683 (45.5)606 (45.3)
3818 (54.5)731 (54.7)
Axillary lymph node status 0.91
Negative888 (56.5)783 (56.7)
Positive684 (43.5)598 (43.3)
Lymphovascular invasion 0.31
Negative837 (55.5)714 (53.6)
Positive671 (44.5)618 (46.4)
ER expression <0.001 *
Negative359 (22.8)449 (32.5)
Positive1213 (77.2)932 (67.5)
PR expression 0.57
<1%699 (47.1)642 (48.2)
≥1%785 (52.9)691 (51.8)
HER2 overexpression/amplification <0.001 *
Negative1392 (90.5)1125 (82.6)
Positive146 (9.5)237 (17.4)
CK5/6 expression <0.001 *
Negative1301 (93.2)1112 (89.2)
Positive95 (6.8)135 (10.8)
EGFR expression <0.001 *
Negative1281 (90.3)1059 (83.2)
Positive137 (9.7)214 (16.8)
Ki-67 proliferation index 0.03
<14%800 (56.2)672 (52)
≥14%624 (43.8)621 (48)
CAIX expression <0.001 *
Negative1271 (86.5)1066 (81)
Positive199 (13.5)250 (19)
CD163+ TAMs <0.001 *
Sparse (≤5)567 (41.4)431 (34)
Moderate (>5 ≤ 25)443 (32.4)417 (32.9)
Dense (>25)359 (26.2)419 (33.1)
Treatment 0.001 *
No systemic therapy659 (41.8)594 (43)
Tamoxifen only; no chemotherapy536 (34)407 (29.4)
Chemotherapy only; no hormonal therapy260 (16.5)292 (21.1)
Chemotherapy + Tamoxifen110 (7)88 (6.4)
Others12 (0.8)2 (0.1)
Breast cancer subtypes (IHC-based) <0.001 *
Luminal-NOS96 (6.1)38 (2.7)
Luminal A680 (43.1)523 (37.8)
Luminal B380 (24.1)302 (21.8)
Luminal/HER2+82 (5.2)97 (7)
HER261 (3.9)135 (9.8)
Basal106 (6.7)168 (12.1)
Unassignable61 (3.9)36 (2.6)
Additional Basal if by TNP111 (7)84 (6.1)
* Denotes differences between low and high G-CSF groups that are significant at the Bonferroni-corrected p-value of <0.0031 (=0.05/16). G-CSF, granulocyte colony-stimulating factor; ER, estrogen receptor; PR, progesterone receptor; EGFR, epidermal growth factor receptor; HER2; human epidermal growth factor receptor 2; CK, cytokeratin; CAIX, carbonic anhydrase IX; TAMs, tumour-associated macrophages; TNP, triple-negative phenotype (ER-, PR- and HER2-); NOS, not otherwise specified; IHC, immunohistochemistry.
Table
Table 2. Correlation of immune biomarkers with G-CSF and CAIX expression (whole cohort).
Table 2. Correlation of immune biomarkers with G-CSF and CAIX expression (whole cohort).
VariablesG-CSF Expressionp-ValueCAIX Expressionp-Value
Low (≤1)High (>1)NegativePositive
H & E sTIL count (%) <0.001 <0.001
<101244 (85.7)1029 (79)2539 (83)170 (71.4)
≥10207 (14.3)274 (21)520 (17)68 (28.6)
CD8 iTIL count <0.001 <0.001
<11058 (70)830 (62)2001 (67.9)128 (56.1)
≥1454 (30)508 (38)945 (32.1)100 (43.9)
PD-1 iTIL count <0.001 <0.001
<11337 (94.1)1161 (88.2)2346 (92.4)146 (75.6)
≥184 (5.9)155 (11.8)192 (7.6)47 (24.4)
PDL1+ tumour cells (%) <0.001 <0.001
01332 (94.2)1166 (88.9)2367 (92.4)150 (79.8)
≥182 (5.8)146 (11.1)194 (7.6)38 (20.2)
FOXP3 iTIL count <0.001 <0.001
<21087 (71.9)831 (61.6)1951 (68.4)133 (59.9)
≥2425 (28.1)518 (38.4)900 (31.6)89 (40.1)
TIM3 iTIL count <0.001 <0.001
<11360 (90)1182 (87.4)2453 (89.8)165 (79.3)
≥1151 (10)171 (12.6)279 (10.2)43 (20.7)
LAG3 iTIL count <0.001 <0.001
<11297 (91.1)1136 (85.9)2283 (89.4)148 (75.5)
≥1126 (8.9)186 (14.1)271 (10.6)48 (24.5)
CD163+ TAMs <0.001 <0.001
Sparse (≤5)567 (41.4)431 (34)1048 (37.8)44 (21.1)
Moderate (>5 ≤ 25)443 (32.4)417 (32.9)907 (32.7)64 (30.6)
Dense (>25)359 (26.2)419 (33.1)820 (29.5)101 (48.3)
H & E, hematoxylin and eosin-stained; sTIL, stromal tumour-infiltrating lymphocytes; iTIL, intratumoural tumour-infiltrating lymphocytes; CD, cluster of differentiation; PD-1, programmed cell death protein-1; PD-L1, programmed death ligand-1; FOXP3, forkhead box protein P3; TIM3, T-cell immunoglobulin and mucin domain-containing molecule 3; LAG-3, lymphocyte activation gene 3; TAMs, tumour-associated macrophages.
Table
Table 3. Multivariable analysis for prognostic significance of immune biomarkers in non-luminal cases with G-CSFhigh phenotype.
Table 3. Multivariable analysis for prognostic significance of immune biomarkers in non-luminal cases with G-CSFhigh phenotype.
CovariatesBCSS
Non-Luminal Cases/G-CSFhigh
HR (95% CI)p-Value
Age at diagnosis 0.16
<501
≥500.76 (0.52–1.12)
Tumour size (cm) 0.004
≤21
>21.82 (1.22–2.71)
Tumour grade 0.04
1 & 21
31.80 (1.02–3.17)
Axillary lymph node status 0.04
Negative1
Positive1.59 (1.03–2.46)
LVI 0.58
Negative1
Positive1.14 (0.73–1.77)
H & E stromal TILs (%) 0.001
<101
≥100.48 (0.31–0.72)
Age at diagnosis 0.06
<501
≥500.69 (0.48–1.01)
Tumour size (cm) 0.01
≤21
>21.72 (1.17–2.55)
Tumour grade 0.02
1 & 21
32.06 (1.15–3.70)
Axillary lymph node status 0.10
Negative1
Positive1.43 (0.93–2.19)
LVI 0.07
Negative1
Positive1.49 (0.97–2.31)
CD8 iTIL count 0.01
<11
≥10.59 (0.40–0.90)
Age at diagnosis 0.17
<501
≥500.77 (0.52–1.12)
Tumour size (cm) 0.05
≤21
>21.50 (1.01–2.23)
Tumour grade 0.02
1 & 21
32.03 (1.13–3.64)
Axillary lymph node status 0.05
Negative1
Positive1.55 (0.99–2.43)
LVI 0.18
Negative1
Positive1.37 (0.87–2.16)
PD1 iTIL count <0.001
<11
≥10.36 (0.20–0.63)
Age at diagnosis 0.08
<501
≥500.71 (0.49–1.04)
Tumour size (cm) 0.02
≤21
>21.59 (1.08–2.33)
Tumour grade 0.01
1 & 21
32.13 (1.19–3.83)
Axillary lymph node status 0.02
Negative1
Positive1.64 (1.08–2.50)
LVI 0.18
Negative1
Positive1.35 (0.88–2.07)
FOXP3 iTIL count 0.002
<21
≥20.55 (0.37–0.80)
Age at diagnosis 0.12
<501
≥500.75 (0.51–1.08)
Tumour size (cm) 0.01
≤21
>21.67 (1.14–2.46)
Tumour grade 0.03
1 & 21
31.85 (1.05–3.24)
Axillary lymph node status 0.04
Negative1
Positive1.57 (1.03–2.40)
LVI 0.2
Negative1
Positive1.33 (0.86–2.05)
TIM3 iTIL count 0.01
<11
≥10.48 (0.28–0.84)
Age at diagnosis 0.16
<501
≥500.76 (0.52–1.12)
Tumour size (cm) 0.01
≤21
>21.69 (1.14–2.51)
Tumour grade 0.01
1 & 21
32.09 (1.16–3.75)
Axillary lymph node status 0.04
Negative1
Positive1.58 (1.02–2.44)
LVI 0.23
Negative1
Positive1.31 (0.84–2.04)
LAG3 iTIL count 0.001
<11
≥10.45 (0.28–0.73)
Age at diagnosis 0.05
<501
≥500.67 (0.46–0.99)
Tumour size (cm) 0.01
≤21
>21.68 (1.13–2.50)
Tumour grade 0.03
1 & 21
31.92 (1.07–3.45)
Axillary lymph node status 0.03
Negative1
Positive1.62 (1.04–2.51)
LVI 0.27
Negative1
Positive1.29 (0.82–2.01)
PD-L1+ tumour cells (%) 0.01
01
≥10.46 (0.26–0.83)
Age at diagnosis0.79 (0.54–1.15)0.21
<50
≥50
Tumour size (cm) 0.01
≤21
>21.17 (1.16–2.53)
Tumour grade 0.04
1 & 21
31.85 (1.03–3.30)
Axillary lymph node status 0.11
Negative1
Positive1.42 (0.92–2.19)
LVI 0.15
Negative1
Positive1.39 (0.89–2.16)
CD163+ M2 macrophages
Sparse1
Moderate1.12 (0.63–2.0)0.70
Dense0.78 (0.44–1.35)0.37
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Chafe, S.C.; Riaz, N.; Burugu, S.; Gao, D.; Leung, S.C.Y.; Lee, A.F.; Lee, C.-H.; Dedhar, S.; Nielsen, T.O. Granulocyte Colony Stimulating Factor Expression in Breast Cancer and Its Association with Carbonic Anhydrase IX and Immune Checkpoints. Cancers 2021, 13, 1022. https://doi.org/10.3390/cancers13051022

AMA Style

Chafe SC, Riaz N, Burugu S, Gao D, Leung SCY, Lee AF, Lee C-H, Dedhar S, Nielsen TO. Granulocyte Colony Stimulating Factor Expression in Breast Cancer and Its Association with Carbonic Anhydrase IX and Immune Checkpoints. Cancers. 2021; 13(5):1022. https://doi.org/10.3390/cancers13051022

Chicago/Turabian Style

Chafe, Shawn C., Nazia Riaz, Samantha Burugu, Dongxia Gao, Samuel C. Y. Leung, Anna F. Lee, Cheng-Han Lee, Shoukat Dedhar, and Torsten O. Nielsen. 2021. "Granulocyte Colony Stimulating Factor Expression in Breast Cancer and Its Association with Carbonic Anhydrase IX and Immune Checkpoints" Cancers 13, no. 5: 1022. https://doi.org/10.3390/cancers13051022

APA Style

Chafe, S. C., Riaz, N., Burugu, S., Gao, D., Leung, S. C. Y., Lee, A. F., Lee, C. -H., Dedhar, S., & Nielsen, T. O. (2021). Granulocyte Colony Stimulating Factor Expression in Breast Cancer and Its Association with Carbonic Anhydrase IX and Immune Checkpoints. Cancers, 13(5), 1022. https://doi.org/10.3390/cancers13051022

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop