Joint first author: Qiong-lan Tang contributed equally to this work.
Reprint request to: Dr Qiong-lan Tang, Department of Pathology, Sun Yat-sen Memorial Hospital of Sun Yat-sen University, Yanjiang Xilu 107, Guangzhou 510120, China. Email: firstname.lastname@example.org
It has been well established that tumor-associated macrophages (TAMs) play a tumor promoting role in endometrial endometrioid adenocarcinoma (EEC). But the association with TAMs and sex hormone receptor expression, and progression of precancerous endometrial lesions in EEC has been little reported.
Material and Methods
We used immunohistochemistry to examine the expression of CD68, CD34, vascular endothelial growth factor (VEGF), estrogen receptor (ER) and progesterone receptor (PR) in 95 cases of EEC, as well as 35 cases of endometrial hyperplasia (including 15 atypical hyperplasia, 10 complex hyperplasia and 10 simple hyperplasia). We also correlated TAMs count with various clinicopathological factors, sex hormone receptor, and prognostic value in patients with EEC.
We identified that TAMs count increased linearly with disease progression (mean count per case at ×200 magnification: simple hyperplasia, 6.30; complex hyperplasia, 11.20; atypical hyperplasia, 29.40; EEC 55.81, respectively; P < 0.001), that microvascular density (MVD) also increased accordingly (27.50, 30.20, 50.13 and 59.94, respectively; P < 0.001). The expression of progesterone receptor, not of estrogen receptor, significantly decreased with disease progression (P < 0.05). Moreover, histopathologic grades, International Federation of Gynecology and Obstetrics (FIGO) stage (2009), depth of myometrial invasion, pelvic lymph node metastasis, lymphovascular space invasion, and expression of PR and VEGF were associated with TAMs count (P = 0.0001, P = 0.004, P = 0.0001, P = 0.04, P = 0.0001, P = 0.0001, P = 0.0001, respectively). Progesterone receptor expression was also associated with histopathologic grades, lymphovascular space invasion, VEGF and high TAMs (P = 0.035, P = 0.022, P = 0.014, P = 0.001, respectively). The estimated 5-year survival rate of patients with low TAMs was significantly higher than those with high TAMs (96.4% vs 69.8%, P = 0.002).
TAMs are potentially related to PR loss and progression of precancerous endometrial lesions in EEC.
Recent studies indicate that the cancer microenvironment is important for tumor progression. Among the microenvironment components, tumor-associated macrophages (TAMs) are the major inflammatory component of the stroma in a tumor. They are located in tumor epithelium, stroma, necrotic areas, and the tumor invasive margin. It was also reported that TAMs play a crucial role in tumor carcinogenesis in cervical intraepithelial neoplasia and that CD68+ macrophages are associated with cervical carcinogenesis from intraepithelial lesions to invasive stages. Macrophages increase linearly with the progression of cervical intraepithelial neoplasia (CIN), migrating from the stroma into the epithelium, and are influenced not only by inflammation itself, but also by the dysplastic cells.
In endometrial carcinoma, several reports have demonstrated that TAMs play an important role in the promotion of angiogenesis[4-7] and are correlated with worse prognosis.[8, 9] However, few reports have described the possible role of TAMs in sex hormone receptor expression and malignant transformation in endometrial endometrioid adenocarcinoma (EEC). Thus, we investigated TAMs, sex hormone receptor, tumor angiogenesis in endometrial cancer and endometrium hyperalia, and their associations with the clinicopathologic features of EEC. The procedures were approved by the Ethical Committee of Human Experimentation of the First Affiliated Hospital of Jinan University in China.
Material and Methods
A total of 130 cases of paraffin-embedded endometrial tissue samples diagnosed as 95 cases of endometrioid adenocarcinomas, 35 cases of endometrial hyperplasia including 10 simple hyperplasia, 10 complex hyperplasia and 15 atypical hyperplasia were selected from surgical pathology files of the Department of Pathology of the First Affiliated Hospital of Jinan University and Sun Yat-sen Memorial Hospital of Sun Yat-sen University, Guangzhou, China, between January 1999 and January 2008. Hematoxylin–eosin (HE) sections were reviewed to confirm the pathological diagnosis. All patients received no hormone therapy. All patients were followed up from the date of surgery to the date of cancer-related death or last contact for the patients who were still alive on 30 June 2011. Cases of EEC were selected in this study if a follow-up was obtained and clinical data were available. The follow-up period for the 95 patients with EEC ranged from 5 to 150 months (mean 66.7 months) after surgery. All patients with endometrioid adenocarcinomas underwent modified radical hysterectomy, bilateral salpingo-oophorectomy and pelvic lymphadenectomy and para-aortic lymph nodes sampling. The age of subjects ranged from 27 to 70 years with an average of 55.6 years. The following histopathologic prognostic factors were correlated with staining intensity of TAM expression by immunohistochemistry: the International Federation of Gynecology and Obstetrics (FIGO) (2009) stage, histopathologic grade, depth of myometrial invasion, lymphovascular space invasion (LVSI), and pelvic lymph node metastasis (LNM). All specimens were collected after obtaining informed consent from the patients in accordance with the institutional guidelines.
For immunohistochemistry, representative tumor areas with the deepest myometrial invasion were selected. In cases without myometrial invasion, one slide showing a representative tumor was selected.
Histological material fixed in 10% formalin and embedded in paraffin was cut into 4-μm sections. Streptavidin peroxidase immunohistochemical staining was performed using the following panel of antibodies: monoclonal mouse anti-human CD68 antibody (PG-M1, ZM-0464, dilution 1:100, Beijing Zhongshan Biotechnology, CN), CD34 antibody (H-140, SC-9095, dilution 1:200, Santa Cruz, USA); mouse monoclonal estrogen receptor (ER) (1D5, ZM-0104, dilution 1:200, Beijing Zhongshan Biotechnology, CN) and mouse monoclonal progesterone receptor (PR) (1A6, ZM-0215, dilution 1:200, Beijing Zhongshan Biotechnology, China), and a general ultrasensitive streptavidin peroxidase immunohistochemistry system (Beijing Zhongshan Biotechnology, China).
Sections were deparaffinized in xylene and hydrated using distilled water according to standard methods. Antigens were retrieved by immersing the sections in citric acid buffer at pH 6.0 and by boiling them in a microwave for 15 min. The sections were stained with anti-CD68, anti-CD34, anti-vascular endothelial growth factor (VEGF), anti-ER or anti-PR antibodies at different working dilution and then incubated at 4°C overnight. General biotin-linked secondary antibody was added and incubated at 37°C for 30 min. Streptavidin peroxidase solution was added and incubated at 37°C for 30 min. 3.3'-diaminobenzidine (DAB) (Dako, Carpinteria, CA, USA) stain was applied for 5 min for chromogenesis. Hematoxylin was used to counterstain the nuclei. The sections were then mounted. For a negative control, pre-immune serum was used in place of primary antibody. For a positive control, a known positively expressing tissue (mammary carcinoma) was used.
Evaluation of immunohistochemistry
According to the Soeda method, we determined the number of CD68+ cells that had infiltrated into cancer nests or stroma (intratumor TAMs) and the number that had become distributed along the tumor-myometrial junction (margin TAMs). Each section was scanned at low (×40 and ×100) magnifications, and three representative areas were identified. Necrotic areas were excluded when calculating the number of intratumor TAMs. An average of the values of the three representative areas at ×200 magnification for intratumor TAMs and margin TAMs, respectively, was used for statistical analysis.
The Weidner method was used for capillary quantification. First, by using a low-power lens, the entire CD34-stained section was scanned to search for regions of high capillary density (hot spots). A × 200 lens was then used to quantify the number of vascular endothelial cell clusters stained with brown dye. The mean of five capillary quantification observed with the ×200 lens, was used as the final microvascular density (MVD).
Immunostaining staining was evaluated by two independent observers who were blinded to clinical data. Yellow-brown granules indicated a positive result. Five high-power fields were randomly selected for observation. Under a high-power objective lens, a section was scored according to staining intensity and staining area, as described previously. The scoring for positive cells was set arbitrarily as follows: <10% = 0 points, 10–25% = 1 point, 26–50% = 2 points, and >50% = 3 points. Staining strength was scored as follows: weak = 1 point, moderate = 2 points, strong = 3 points. The sum of the two values determined the result: a total of 1 point was negative, and 2–5 points was positive for the expression of VEGF, ER and PR.
All of the experiments were repeated three times. Scoring differences were discussed to reach consensus.
Statistical analyses were performed using windows SPSS version 13.0 software. Statistical analysis included the Mann–Whitney U-test of variance, chi-square test and the Spearman rank correlation test. The Spearman rank correlation test was used to investigate the relationships between the number of TAMs and clinicopathological variables. The analysis of the correlation between TAMs count and VEGF expression was performed by chi-square test. Survival analysis was carried out using the log–rank test in association with Kaplan-Meier analysis and Cox proportional hazards model analysis. The forward stepwise method was used to examine the effect of each variable. A P-value less than 0.05 was considered significant.
Association between TAMs and malignant cell transformation of precancerous endometrial lesions (Table 1)
Table 1. Association between tumor-associated macrophages (TAMs) and cell transformation in different types of endometrium
Type of endometrium
One-way ANOVA, Student Newman–Keuls test; ‡Pearson chi-square test, χ2 = 0.926, P = 0.819; §Pearson chi-square test, χ2 = 8.394, P = 0.039. SD, standard deviation. MVD, microvascular density.
A total of 130 cases of different types of endometrium were evaluated for TAMs, MVD, ER, PR and VEGF expression. CD68+ cells were observed in all of the specimens examined and mainly localized to the cytoplasm. Most of them were distributed along the invasive margin of the tumor and in cancer cell nests and tumor stroma (Figs 1-3–3). A direct relationship was found between the increasing grade of the endometrial lesion and the number of macrophages in the epithelium, in the stroma. The mean number of TAMs per case at ×200 magnification in simple hyperplasia, complex hyperplasia, atypical hyperplasia and EEC was 6.30, 11.20 29.40, and 55.81 respectively, TAMs mean count per case at ×200 magnification increased linearly with disease progression. There was a significant difference between different types of endometrium (P<0.001), moreover, mean MVD also increased accordingly: 27.50, 30.20, 50.13 and 59.94, respectively (P<0.001); There was also a significant difference of PR expression between different types of endometrium (P<0.05), but not for ER (P>0.05) (Table 1).
Correlation of number of TAMs with clinicopathological factors of endometrioid adenocarcinoma (Table 2)
Table 2. Association between tumor-associated macrophages (TAMs) and clinicopathological characteristics of the patients with endometrial endometrioid adenocarcinoma
To obtain a better understanding of the clinical significance of TAMs in endometrioid adenocarcinoma, we correlated its expression with a series of clinicopathological factors and the expression of ER, PR and VEGF. As shown in Table 2, the number of TAMs was significantly correlated with the FIGO stage (P = 0.004), histopathological grade (P = 0.0001) (Figs 2, 3), and the depth of myometrial invasion (MI) (P = 0.0001), lymph node metastases (P = 0.04), and lymphovascular space invasion (P = 0.0001). PR-negative cases showed higher number of TAMs than in positives ones (P = 0.0001), but VEGF positive cases showed higher TAMs than in negative ones (P = 0.0001). The number of TAMs was not associated with age (P = 0.100) and estrogen receptor expression (P = 0.446) in this cohort.
Survival and TAMs and other variables
Because the value of TAMs ranged from 13.0 to 35.0 and from 50.0 to 110.5, the patients were intuitively divided into two groups: those with high TAMs (n = 63) and those with low TAMs (n = 32) with a threshold of 50. The median TAM density was 73.8 (range 50.0 to 110.5) for the high TAMs group and 26.5 (range 13.0 to 35.0) for the low TAMs group.
Patients with high TAMs had significantly worse overall survival (OS) than those with low TAMs (log–rank test, P = 0.0002) (Fig. 4). The estimated 5-year OS of patients with high TAMs was 69.8%, while that of patients with low TAMs was 96.4%. Furthermore, PR (P=0.001), high TAMs (P = 0.005), FIGO stage (P=0.039), pelvic lymph node metastasis (P=0.031) and lymphovascular space invasion (P=0.040) were significantly related to OS on univariate analysis. But on multivariate analysis, using cox proportional hazards model (Table 3), high TAMs (P=0.010) and negative PR (P=0.023) were significantly associated with OS.
Table 3. Multivariate analysis on the prognostic factors for endometrioid adenocarcinoma
ER, estrogen receptor; FIGO, International Federation of Gynecology and Obstetrics; MVD, microvascular density; PR, progesterone receptor; TAMs, tumor-associated macrophages; VEGF, vascular endothelial growth factor.
Depth of myometrial invasion
Pelvic lymph node metastasis
Lymphovascular space invasion
To evaluate the combined effect of TAMs and PR on the survival rate, the 95 patients were divided into four groups based on high or low TAMs and the positive or negative of PR showed an estimated 5-year survival rate of 100.0% for low TAMs PR(+) (n = 17, group a), 93.8% for low TAMs PR(−) (n = 16, group b), 89.7% for high TAMs PR(+) (n = 29, group c), and 54.5% for high TAMs PR(−) (n = 33, group d) (Fig. 5). There was a significant difference among the four groups (P=0.0001).
Endometrial carcinoma is the most common malignant tumor of the female genital tract in Western countries. EEC accounts for three-quarters of endometrial cancers and is thought to develop following a continuum of premalignant lesions ranging from endometrial hyperplasia without atypia, to hyperplasia with atypia and finally to well-differentiated carcinoma, which may be involved in distinct molecular mechanisms along with the cell malignant transformation.[13-15]
The connection between inflammation and cancer is now generally accepted. Experimental and epidemiological studies indicate a strong link between chronic inflammation and tumor progression. Cancer-associated inflammation promote cells malignant transformation in CIN and affect many aspects of malignancy, including the proliferation and survival of malignant cells, angiogenesis, invasion and tumor metastasis[7-10] in endometrial carcinoma, which has been also verified in our data.
In the present study, we detected that TAMs count as well as MVD increased linearly with disease progression, along with the decrease in the level of progesterone receptor expression. Our study clearly demonstrates a strong association between the cells malignant transformation of the endometrial lesion and an increase in the number of tumor-associated macrophages in the stroma as well as in the glandular epithelium.
In recent years, tumor-associated macrophages have become one of the most important biological markers. TAMs are the major component in tumor microenvironment. When acting as normal macrophages, they excrete immune-regulatory factors or enzymes, which inhibit growth of tumor. However, when influenced negatively by tumor cells, they may excrete inflammation factors and extracellular matrix (ECM) regulatory factors, inhibit antitumor immune reaction, accelerate tumor angiogenesis and destroy the barrier of the local basement membrane.[7-9]
There was a positive correlation between CD68+ macrophages and microvessel density in primary tumors and their corresponding regional lymph node metastases. These findings link increased microvessel proliferation to stromal macrophage infiltration, and suggest that enhanced tumor angiogenesis triggered by stromal macrophages, regulates the progression of endometrioid carcinomas.
In general, chronic local inflammation may predispose to tumor development by generating free radicals and up-regulating COX-2 and PGE2, which in turn can damage DNA and induce cell proliferation, thus initiating and promoting neoplastic transformation. Chronic inflammation can also dysregulate the NF-kB pathway, thereby inhibiting apoptosis, blocking cell cycle arrest, and further stimulating production of proinflammatory cytokines.[15, 16]
We further correlated TAMs and clinicopathological variables. The result showed PR loss was significantly associated with high TAMs count. To the best of our knowledge, this is the first report describing the relationship with the number of TAMs and sex hormone receptor. As discussed, a proinflammatory milieu can also increase estrogen production. In particular, IL-6 can stimulate estrogen synthesis and can act synergistically with TNF-a to enhance the activities of aromatase. Thus, it can contribute to an estrogen–progestogen imbalance, possibly predisposing the endometrium to the neoplastic process. In particular, the absence of progesterone can lead to an inflammatory milieu, which can both increase estrogen production and further induce proinflammatory cytokines, thereby potentially creating an environment susceptible to tumor initiation and promotion. Recently, it has been suggested that progesterone inhibits metastatic spread of endometrial cancer by inhibiting epithelial-to-mesenchymal cell transition (EMT) and by stimulating T-cell infiltration. Loss of PR expression correlates with loss of immunosuppression and increased EMT in progressive endometrial cancer. Besides stimulating tumor-infiltrating lymphocytes (TILs), progesterone can also inhibit Wnt/β-catenin signaling and loss of progesterone signaling may be involved in tumor onset and progression towards a more invasive disease.[15, 20] MPA can inhibit EMT in the Ishikawa endometrial cancer cell line. Recent studies have suggested that TAMs may be involved in the EMT program.[21, 22] Thus, progesterone may inhibit TAMs infiltration through EMT, which will need to be verified in future studies.
Progesterone receptor expression was also associated with histopathologic grades, lymphovascular space invasion and VEGF expression. Progesterone receptor status has been examined in endometrial cancer specimens and demonstrated significant correlation between PR-positive tumors and grade, histology, adnexal spread, and recurrence.[23, 24]
The relationship between VEGF and PR has reported no consistent results: Kim et al. reported that VEGF was expressed in the absence of treatment with E2 or MPA, and expression was unaltered by continuous treatment with E2, but Mueller et al. considered that progestin's had a direct effect on VEGF gene transcription. In the present study, there was significantly decreased expression of PR in positive VEGF expression group, which may be involved in angiogenesis and be consistent with TAMs in EEC.
We also performed multivariate analysis of prognostic factors and found that TAMs was an independent prognostic factor for patients with endometrioid adenocarcinoma as well as PR loss. Creasman analyzed the correlation between hormone receptors and survival using biochemical methods and demonstrated that PR status significantly predicted disease-free survival of endometrial carcinomas.
Therefore, to evaluate the combined effect of TAMs and PR on survival rate, patients were divided into four groups based on high or low TAMs and the presence or absence of PR. Overall survival (OS) was the worst in high TAMs PR(−) cases and achieved the highest OS in low TAMs PR(+) cases. These results show that high TAMs are associated with PR loss and poor prognosis. Patients with high TAMs may not be sensitive to progestin therapy. In addition, patients with high TAMs PR(−) should received comprehensive treatment or adjuvant therapy as soon as possible.
In summary, high expression of TAMs was related to PR loss and malignant transformation in EEC. Further studies will be needed to determine the biological role of progesterone on TAMs infiltration and further examine the possibility as a therapeutic target in endometrial cancer.
The authors have no conflicts of interest to declare.