Tumor-Associated Macrophages in Human Breast, Colorectal, Lung, Ovarian and Prostate Cancers.

Prostate cancer (PC) represents the second most frequent malignancy in men with an estimated over 1.5 million new cases diagnosed annually worldwide and ranks as the fifth leading cause of cancer-associated mortality globally ( 241 ). The incidence and mortality rates of PC are trending upwards due to population aging and urbanization, thereby having a significant social and financial burden on global healthcare system ( 242 ). PC belongs to hormonally driven malignancy, whose primary progression relies on functional activity of androgen receptors ( 243 ). Accordingly, three stages in prostate carcinogenesis are distinguished: precancerous intraepithelial neoplasia, androgen-dependent, and followed by aggressive androgen-independent PC ( 244 ). Adenocarcinoma is the most common prostatic tumor, whereas other histological subtypes such as urothelial, small cell, squamous cell, and basal cell carcinomas are diagnosed quite rarely ( 245 ). The major routes for PC progression include extracapsular extension and spread to pelvic lymph nodes, as well as metastasis to lungs and bones ( 246 ). Furthermore, given the abundant innervation of prostate peripheral zone, primary tumors arising in this area tend to escape the organ through perineural invasion ( 247 ). Routine screening of PC involves an evaluation of serum levels of prostate specific antigen (PSA), a serine protease produced by prostate epithelium, while the gold standard for diagnosis confirmation is prostate biopsy analysis ( 248 ). Apart from the TNM staging system, Gleason score is used to characterize the PC metastatic potential on the basis of differentiation patterns. Thus, high-grade PC (Gleason score over 7) has higher risk of metastasis as compared to less aggressive primary tumors with Gleason score below 6 ( 249 ). Given the hormone dependent nature of PC, androgen deprivation therapy (ADT) has been regarded as a standard treatment approach for patients with PC ( 250 ). Despite the initial efficacy and improvement in OS rates, prolonged hormonal treatment is eventually associated with the emergence of aggressive castration-resistant prostate cancer (CRPC) associated with high mortality and poor patient outcomes ( 251 ). Current evidence suggests that inflammatory microenvironment, especially TAMs, is involved in the onset of prostate carcinogenesis and acts as an essential modulator of further malignant progression, metastasis, and overall therapeutic response ( 252 ). In human prostate cancer, the inflammatory component of local TME is considered as an essential modulator of malignant progression and determinant of the overall therapeutic response ( 253 ). To date, a number of investigations have focused on the patterns of macrophage infiltration in prostate cancer specimens in attempts to validate its clinical and pathological significance ( 254 ) ( Table 6 ). The primary analysis of TAMs in 85 prostate carcinomas (Sweden, 2000) demonstrated significant increase of the cell profile area and volume density of CD68+ macrophages in cases with higher Gleason score ( 260 ). A positive correlation was also found between the size of individual macrophage and angiogenesis measured as the number of von Willebrand factor-positive microvessels in the most vascularized area ( 260 ). In the same cohort, increased density and cell profile area of CD68+ TAMs were recognized as predictors of shorter cancer-specific survival (CSS) ( 260 ). Next study of a cohort of 81 prostate cancer patients from USA cohort revealed an increase of macrophage density in tumor versus adjacent benign tissue ( 255 ). Interestingly, a negative association between the amount of CD68+ TAM infiltrate in total tumor tissue and TNM clinical stage was found, while TAM density within cancer cell area positively correlated with Gleason score ( 255 ). Such contradicting results may reflect the heterogeneous distribution of TAMs in the tissue samples and highlights the importance of the compartment-specific macrophages in prostate tumorigenesis. High levels of CD68 in biopsy specimens of 859 patients from the USA cohort with benign prostatic hyperplasia were associated with increased risk for overall clinical progression ( 261 ). Several independent investigations confirmed high expression of CD68 in advanced prostate cancer. Thus, IHC study of 131 Japanese prostate cancer patients detected abundant CD68+ macrophage infiltration in tumor mass in patients with higher serum prostate-specific antigen (PSA) and Gleason score ( 59 ). The relapse-free survival rates in the same cohort were significantly lower in patients with greater TAM counts ( 59 ). Appropriate reporting of methodology, quantitative assessment and statistical analysis in this study could be necessary to ensure the quality of data interpretation in accordance with scientific rigor ( 59 ). Tissue microarray (TMA) containing 332 radical prostatectomy specimens (USA cohort) revealed greater abundance of CD68+ cells in malignant areas in comparison to benign tissues, as well as increase in mean TAM numbers in Gleason grade 4 versus grade 3 ( 262 ). IHC analysis of 100 specimens of prostate adenocarcinoma of the Turkish cohort demonstrated positive correlation between the density of CD68+ TAM infiltration and such clinical–pathological parameters as tumor stage, Gleason score, extracapsular extension, perineural invasion, and positive surgical margins ( 256 ). Furthermore, a study involving 93 prostate cancer patients from the Italian cohort identified that high expression of CD68 in primary tumor identified by IHC was an independent predictor of biochemical recurrence (defined as elevation of PSA level) after radical prostatectomy ( 263 ). Increased CD68+ macrophage count was observed in metastases from the lymph nodes, liver, bladder, rectum, and seminal vesicles in comparison to the corresponding primary tumors collected from 59 prostate cancer patients from the Norway cohort ( 264 ). Recent study of representative TMA collected from over 400 patient cohort from Germany confirmed the increase of CD68+ cell numbers in prostate cancers with Gleason score over 8 ( 265 ). Microarray analysis of 9,393 prostate cancer samples demonstrated that elevated expression signature of TAMs is strongly associated with worse distant metastasis-free survival ( 266 ). Thus, a number of studies indicated that higher CD68+ macrophage abundance in tumor tissue reflects aggressive tumor behavior and unfavorable patient outcomes in prostate cancer. Representative studies demonstrating the association of TAMs with tumor progression parameters in prostate cancer. DFS, disease-free survival; DSS, disease-specific survival; HPF, high-power field; IF, immunofluorescence; IHC, immunohistochemistry; LN, lymph node; TAM, tumor-associated macrophages; OS, overall survival; RFS, recurrence-free survival; TMA, tissue microarray. Not only total macrophage amount but also specific macrophage subtypes were found to be correlated with clinical and pathological characteristics of prostate cancer patients ( Table 6 ). IHC analysis of tissue specimens derived from 93 Italian prostate cancer patients has identified that high amount of CD163+ TAMs was associated with extracapsular extension (Gleason score > 7) and worse biochemical recurrence-free survival rates ( 257 ). Increased infiltration of CD163+ cells correlated with higher Gleason score and incidence of metastasis, as well as lower rates of CSS in a cohort of 234 Swedish prostate cancer patients ( 258 ). These findings were further confirmed in a study involving 592 patients with diagnosed prostate cancer from the Swedish cohort demonstrating greater CD163+ macrophage infiltration in aggressive tumors with Gleason scores ranging from 8 to 10 ( 267 ). The risk of death from prostate cancer in the same cohort was almost twofold higher in patients with high amount of CD163+ TAMs versus those with lower numbers ( 267 ). Positive correlation between the number of CD206+ macrophages and Gleason scores was found in Chinese cohort of 42 prostate adenocarcinoma patients ( 268 ). TMA of 192 prostate cancer samples from the USA cohort revealed greater amount of CD206+ TAMs in primary adenocarcinoma and lymphatic metastases in comparison to benign prostate tissues ( 269 ). IHC analysis of 373 prostate biopsy samples (Japanese cohort) demonstrated significantly lower numbers CD204+ TAMs in cases with prostate cancer in comparison to benign specimens ( 270 ). Negative correlation between the density of CD204+ TAMs and the clinical T stage was confirmed in the retrospective study of 135 PC patients from the Japanese cohort ( 259 ). Inverse association was demonstrated between the expression of MSR-A in primary tumors and the presence of lymph node metastases in the USA cohort of 90 prostate cancer patients ( 271 ). YKL-40 is an emerging TAM biomarker that is produced by both macrophages and cancer cells and enhances inflammation in TME ( 272 ). YKL-40 is also a strong inducer of tumor angiogenesis ( 273 ). In macrophages, YKL-40 is induced by IFN γ and can be considered as M1 biomarker ( 14 , 16 ). Significantly higher concentrations of YKL-40 were detected in the serum of 153 patients (from Denmark) with metastatic prostate cancer compared to healthy donors ( 274 ). Accordingly, elevated plasma YKL-40 levels at the time of diagnosis were predictive of shorter OS rates in the same cohort of patients ( 274 ). To date, androgen deprivation therapy (ADT) is accepted as a standard treatment approach for patients with advanced prostate cancer ( 250 ). Despite initial efficacy and improvement in the OS, prolonged hormonal treatment is eventually associated with aggressive castration-resistant prostate cancer (CRPC) ( 251 ). Multiple lines of evidence indicate crucial role of TAMs in therapeutic response and in post-treatment recurrence of prostate cancer ( 275 ). In comparison with tumor tissues from hormone-naïve prostate cancer patients, CRPC samples displayed higher number of CD68+ macrophages expressing cathepsin S enzyme known to be involved in angiogenesis and remodeling of extracellular matrix ( 276 ). IHC analysis of 75 prostate cancer specimens (Canadian cohort) was performed in two groups of patients—patients pre-treated with Cyproterone (antiandrogen agent) or Leuprolide (gonadotropin-releasing hormone analogue) in combination with Flutamide (nonsteroidal antiandrogen) before radical prostatectomy and patients who underwent surgery only. Increase in the amount of CD68+ TAMs within tumor tissues of pre-treated patients compared to the untreated group was demonstrated ( 277 ). Increased CD68+ and CD163+ macrophage infiltration was found in a cohort of 60 Chinese prostate cancer patients receiving preoperative Bicalutamide-based ADT ( 278 ). TMA analysis was performed for retrospective cohort of 366 prostatectomized patients (Canada) divided into two groups—hormone ablation-treated patients (luteinizing hormone-releasing hormone-agonists and/or antiandrogen prior to surgery) and hormone-naïve patients. This analysis confirmed significantly higher amount of CD163+ TAMs in treated group of patients in comparison with hormone-naïve patients ( 279 ). Mouse model of prostate cancer further confirmed dramatic recruitment of TAMs in response to ADT. Substantial overexpression of VEGF-A, MMP-9, and ARG1 was found in tumors of castrated animals treated by ADT ( 279 ). Also, concentrations of CSF1, major macrophage differentiation, and chemotactic factor, were enhanced in the serum of animals in response to ADT treatment ( 279 ). In parallel, co-culture of Myc-CaP prostate cancer cells and RAW264.7 macrophages treated with antiandrogen Enzalutamide resulted in significant increase in the expression of M2 markers—VEGF-A, MMP- 9, ARG1, IL-10, and CSF1 ( 279 ). Importantly, higher levels of CD163+ macrophages were detected in the prostate cancer sections (Chinese cohort) resected after preoperative ADT in comparison to the corresponding tissues collected before therapy ( 280 ). IHC study of 126 prostate cancer patients (Italian cohort) using pelvic lymph node metastases samples obtained from those patients who received neoadjuvant hormonal treatment flutamide combined with Leuprolide acetate before radical prostatectomy was performed ( 281 ). Double IHC revealed the co-localization of CD68+ TAMs and TARC/CCL17 (thymus- and activation-regulated chemokine), chemokine regulated Treg function, in treated patients in contrast to the untreated group ( 281 ). Clinical trial on 17 patients (USA) with Gleason score 7–10 prostate cancer, treated with anti-PD-1 therapy, revealed significant upregulation of inhibitory molecules PD-L1 and VISTA on CD68+ TAMs in tumor after treatment in comparison with baseline tumor (10-fold and fourfold increase in expression, respectively) ( 282 ). The authors suggested that VISTA expression is a compensatory pathway limiting efficiency of ipilimumab therapy of prostate cancer ( 282 ), and targeting of VISTA on TAMs can be suggested as next therapeutic approach to develop. Monitoring of serum YKL-40 concentrations can also be considered as promising prognostic approach for the management of CRPC. Thus, post-treatment increase of serum YKL-40 was an independent prognostic factor of earlier death in 106 metastatic prostate cancer patients (Denmark cohort) treated with total androgen ablation or parenteral estrogen ( 283 ). Retrospective analysis of 109 patients with CRPC receiving first-line chemotherapy with docetaxel revealed significance of high pre-treatment YKL-40 serum levels as predictive parameter of shorter OS and DSS ( 284 ). These data demonstrate the essential role of TAMs in prostate cancer progression and emphasize on the promise of targeting TAMs to prevent the recurrence of disease and achieve sustained improvements in patient outcomes. Further in-depth investigations must be done to characterize macrophage phenotypes within certain intratumor compartments of prostate cancer and determine their potential diagnostic and therapeutic value.

Tumor-associated macrophages (TAMs) are key innate immune cells in tumor microenvironment (TME) that regulate growth of primary tumors, antitumor adaptive immune response, tumor angiogenesis, extracellular matrix remodeling, intravasation in the vasculature, extravasation in metastatic sites; they establish beneficial conditions for metastatic cells in the secondary organs, and interact with various types of therapies ( 1 , 2 ). Signaling, epigenetic and metabolic mechanisms cooperate to form functional TAM phenotypes ( 3 ). TAMs represent the major component of the innate immune system in TME and can constitute up to 50% of the tumor mass ( 4 ). Two main directions of polarization of TAMs can be defined—classically activated M1 (main markers—HLA-DR, CD80/86) and alternatively activated M2 (main markers—CD206, CD163, CD204, stabilin-1) phenotypes ( 1 , 2 , 5 ) ( Table 1 ). These typical M2 markers are surface receptors that are responsible for the non-inflammatory clearance of microenvironment from apoptotic bodies, ECM components, soluble mediators of activation of cancer cells and angiogenesis ( 6 – 12 ). In addition to scavenging function ( 10 , 11 , 13 ), stabilin-1 acts as an intracellular sorting receptor that targets chitinase-like proteins SI-CLP and YKL-39 to the secretory pathway ( 14 – 19 ). SI-CLP and YKL-39, in turn, regulate monocyte recruitment and angiogenesis ( 15 , 17 , 18 , 20 ). Variety of TAM markers in cancer. All details and references are presented throughout the text. IHC, immunohistochemistry; IF, immunofluorescence; qPCR, quantitative polymerase chain reaction; TAM, tumor-associated macrophage. It is commonly accepted that M1-like macrophages exhibit antitumor activity contributing to the activation of adaptive immune response and inflammation, while M2-like macrophages, in contrast, suppress immune function in tumor microenvironment, induce angiogenesis, and support tumor growth and metastasis ( 21 ). However, this nomenclature is based on the in vitro phenomenon and only schematically reflects vectors of the macrophage polarization in vivo , including their polarization in the complex TME. In each cancer type, TAMs can have cancer-specific phenotypes, and can be represented by the heterogeneous populations. Moreover, TAM subtypes can be distinct in various intratumoral compartments, for example in tumor nest and in tumor stroma. Individual TAM phenotypes can be defined by set of markers that not necessarily give clear classification into the M1 and M2 subtypes. The most common histological markers of macrophages belong to the class of transmembrane receptors (mostly of scavenger type); however, new biomarkers that belong to classes of cytokines, growth factors, enzymes, transcription factors, and chitinase-like proteins allow more precise phenotypic and functional characterization of TAMs ( Table 1 ). TAMs originate from two major sources—tissue-resident macrophages and circulating monocytes recruited in tumor cite by growth factors and chemokines, such as M-CSF, CCL2, and CCL5 ( 21 ). Local resident macrophages can recognize cancer cells, and it is believed that they have intrinsic ability to eliminate sporadically transformed cells. Different origin can define TAM diversity in the TME. Transformed cells can escape local innate immune control and give origin to cancer cell clones that efficiently recruit monocytes from blood circulation and reprogram resident TAMs. The number of experimental model systems demonstrated that growing tumor can program resident and recruit macrophages to support tumor progression ( 22 , 23 ). Both monocyte-derived macrophages and resident macrophages (of adult hematopoietic or embryonic origin) accumulate within an expanding tumor ( 24 , 25 ). Recent study demonstrated that tissue-resident interstitial macrophages in mouse lungs contribute to the pool of TAMs and support tumor growth in vivo , while monocyte-derived TAMs contribute to tumor progression in the form of metastasis ( 26 ). Interestingly, chemotherapeutic treatment resulted in depletion of both resident and monocyte-derived macrophages, but monocyte-derived macrophages were able to recover and provided phagocytosis-mediated tumor clearance ( 26 ). However, not all tumors can do it efficiently, and monocytes and macrophages can also retain their ability to recognize tumor as an unwanted-self structure and inhibit its growth and spread ( 27 , 28 ). In mouse model of ovarian cancer, CD163+ Tim4+ macrophages from omentum, which have embryonic origin and are uniquely independent of bone marrow-derived monocytes, contributed significantly to the metastatic spread ( 29 ). Depletion of CD163+ resident macrophages in tumor-bearing mice with lipid nanoparticles reduced tumor growth and progression ( 29 ). We can hypothesize that TAM heterogeneity is defined both by their high plasticity and by their origin from independent specific lineages. The contribution of each of these factors in the final tumor-specific TAM heterogeneity is a highly relevant topic for the investigation. TAM diversity reflects and defines their specific role in different cancers. A number of studies demonstrated that high infiltration of TAMs in human tumors is associated with poor clinical outcome ( 1 , 2 ). However, the role of TAMs in tumor growth, lymphatic and hematogenous metastasis and treatment outcomes is specific for each type of cancer. By studying patients, the role of TAMs cannot be defined by loss-of-functions and gain-of-function experimentation, and correlation of TAM amounts, their intratumoral localization and functional phenotypes with clinical parameters is a primary source to draw the conclusion. Therefore, precise definition and accurate selection of clinical parameter are essential. Lymphatic and vascular invasions, characterized by cancer cells’ presence within a definite, endothelial-lined space, are parameters that are potential indicators of the ability of cancer cells to metastasize to the lymph nodes and blood vessels, respectively ( 30 , 31 ) Vascular invasion may reflect the risk of recurrent disease and prognosis ( 30 ). There are survival rates that define the probability of the appearance of one or more of tumor progression parameters. For example, progression-free survival (PFS) is calculated as a period of time between the dates of diagnosis and earliest progression (local recurrence or distant metastasis or death) or last follow-up for patients without progression ( 32 ). Similarly, disease-free survival (DFS) is a period of time between the dates of treatment of definite cancer and any signs or symptoms of that cancer; overall survival (OS)—the period where patients still alive for a certain period of time after they were diagnosed with or started treatment for a cancer ( 33 ). In this review we summarize the data about TAM correlation with clinical parameters of widely distributed, dangerous and frequently metastasizing types of cancer: breast, colorectal, lung, ovarian, and prostate. We analyzed the role of TAMs in primary tumor growth and metastasis, and the role of TAMs in the tumor response to therapy with particular focus on tumor relapse and metastatic outbreak. We focus not only on the total amount of TAMs in tumor mass, but we made an accent on the functional TAM biomarkers that can be also distinct in different tumor types.

Conceptualization: IL and JK. Writing—original draft preparation: IL, JK, GT, AP, MS, VP, EC, and NC. Writing—review and editing: IL, GT, and JK. Figure preparation: IL. Supervision: JK. Funding acquisition: JK. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

In our review we compile existing lines of evidence about the clinical role of TAMs in the context of metastasis (including survival rate) and antitumor treatment in different cohorts of patients that come out of a number of courtiers worldwide. We compared the role of TAMs in worldwide leading types of malignant diseases: breast, colorectal, lung, ovarian, and prostate cancers that very frequently give life-threatening distant metastasis. Systematic analysis of TAM biomarkers identified that CD68, and in some cases CD163, are the best markers for the quantification of TAMs in tumor tissue, while several other surface receptors (scavenger receptor stabilin-1, mannose receptor CD206, CD204, MARCO) and chitinase-like proteins (YKL-39, YKL-40) are very informative biomarkers of functional TAM polarization. In patients with breast, ovarian, and prostate cancer, increased amount of TAMs is a clear indicator for rapid tumor growth, aggressive metastatic process, and limited efficiency of therapy ( Tables 2 – 6 ) ( Figure 2 ). In lung and ovarian cancer, the major parameter associated with prognosis was not the total amount of CD68+ macrophages, but M1/M2 index. The prevalence of M1 macrophages was favorable for the patients, indicating that in lung tumor M1 TAMs have the ability to limit tumor progression. Moreover, in lung cancer, high amount of TAMs in tumor nest correlated with the chemotherapy efficiency. The most distinct from other types of cancer was colorectal cancer, where high amounts of TAMs were indicative of the favorable prognosis and restricted ability of primary tumors to grow and to metastasize ( Figure 2 ). In contrast to the total amount of macrophages, M2-like phenotype of TAMs is rather indicative for the negative prognosis for patients with CRC. TAMs in primary tumor growth and metastasis. Role of TAMs in primary tumor growth, hematogenous metastasis, and lymphatic metastasis is illustrated. Green arrow indicates supportive role of TAMs for each process, and orange arrow indicates the suppressive role of TAMs. The role of each specific macrophage marker in the individual type of cancer is indicated within the arrows. TAMs may contribute to resistance to therapy facilitating tumor progression by suppression of T cell immunity, the maintenance of tumor cell survival, and the stimulation of tumor revascularization. Chemotherapy can stimulate antitumor immunity, thereby increasing the pathological complete response (pCR) to the treatment. There is no agreement about the role of TAMs in chemotherapy response. The results are contradictory and depend on the animal model, type of in vitro study, patient cohort, and type of anti-cancer drug ( Table 7 ). Therefore, to achieve the maximum efficiency of chemotherapy, the molecular mechanisms of the interaction of chemotherapeutic agents with TAMs have to be investigated. Understanding of these interactions will also allow developing targeting strategies for TAMs. The investigation of TAM-mediated tumor resistance to therapy is of particular relevance in the era of the development of immunomodulatory approaches aimed to enhance T-cell immunity, to inhibit macrophage recruitment into a tumor, to modify polarization of TAMs, and to enhance phagocytosis of cancer cells by TAMs. The association of TAMs with the effect of chemotherapy in patients. ALN, axillary lymph node; CRC, colorectal cancer; DFS, disease-free survival; DSS, disease-specific survival; FU, fluorouracile; HPF, high-power field; IF, immunofluorescence; IHC, immunohistochemistry; LN, lymph node; NAC, neoadjuvant chemotherapy; NSCLC, non-small cell lung cancer; OS, overall survival; PFS, progression-free survival; pCR, pathological complete response; RFS, recurrence-free survival; TAMs, tumor-associated macrophages; TMA, tissue microarray.

This research was funded by the Russian Science Foundation, grant number #19-15-00151. The publication costs were covered by the Tomsk State University competitiveness improvement program.