Microenvironment-associated lymph node metastasis of human cervical carcinoma xenografts

Abstract

Background. The prognosis is particularly poor for patients with advanced squamous cell carcinoma of the uterine cervix when the primary tumor has developed severe physiological abnormalities. The impact of the physiological microenvironment of the primary tumor on lymph node metastasis was investigated in this preclinical study. Material and methods. Xenografted tumors of two human cervical carcinoma lines (CK-160 and TS-415) transplanted into BALB/c nu/nu mice were included in the study. The fraction of radiobiologically hypoxic cells (HF_Rad), interstitial fluid pressure (IFP), and extracellular pH (pH_e) were measured in 22 CK-160 tumors and 16 TS-415 tumors and related to the metastatic status of the host mice. Results. In CK-160, HF_Rad was significantly higher in the metastatic than in the nonmetastatic tumors, whereas the metastatic and nonmetastatic tumors did not differ significantly in IFP or pH_e. In TS-415, IFP was significantly higher in the tumors that metastasized than in those that did not metastasize, whereas the tumors of the metastasis-positive and metastasis-negative mice did not differ significantly in HF_Rad or pH_e. Conclusion. Lymph node metastasis is associated with abnormalities in the physiological microenvironment of the primary tumor in cervical carcinoma xenografts, and tumor line-specific mechanisms are probably involved.

Advanced squamous cell carcinoma of the uterine cervix is treated with radiation therapy alone or radiation therapy in combination with surgery and/or chemotherapy [1]. Important clinical prognostic factors include stage of disease, tumor volume, and lymph node status [2]. The outcome of treatment varies significantly among individual patients of the same prognostic group, and it has been suggested that this variability is a consequence of differences in the physiological microenvironment of the primary tumor [3].

Primary tumors may develop an irregular microvascular network during growth, resulting in inadequate blood supply and an abnormal physiological microenvironment [4]. Abnormalities in the physiological microenvironment have been demonstrated in a wide variety of human tumors, including soft tissue sarcomas, prostate adenocarcinomas, colorectal carcinomas, carcinomas of the head and neck, and cervical carcinomas [3 − 5]. The physiological microenvironment of cervical carcinomas is characterized by low oxygen tension (pO₂), regions with hypoxic tissue (pO₂ < 10 mmHg), elevated interstitial fluid pressure (IFP), low extracellular pH (pH_e), and high lactate concentration [6–8]. This microenvironment may cause resistance to treatment and promote invasive growth and metastatic dissemination [9–11]. Thus, extensive hypoxia in the primary tumor has been shown to be associated with locoregional treatment failure and poor disease-free and overall survival rates in advanced cervical carcinoma [12–15], and studies of cervical cancer patients treated with radiation therapy alone have shown that high IFP in the primary tumor is linked to high probability of pelvic recurrence and distant metastases [16,17]. Moreover, the disease-free and overall survival rates have been shown to be particularly poor for cervical cancer patients with high lactate concentration in the primary tumor [18].

A single pathophysiological parameter was studied in most of these investigations. However, Fyles et al. [16] measured both IFP and pO₂ in the primary tumor of more than 100 patients with advanced cervical carcinoma, and this study showed no significant correlation between IFP and hypoxic fraction or between either IFP and hypoxic fraction and clinical prognostic factors, such as stage, tumor size, hemoglobin concentration, and histological grade. Moreover, the independent prognostic effect of IFP for recurrence and survival was strong, whereas the independent prognostic effect of hypoxia was of borderline significance and was limited to patients without nodal metastases [16]. Except for this study, little information is available on the relationship between the various parameters of the physiological microenvironment of cervical carcinomas and their relative impact on radiocurability and metastatic spread.

In the present investigation, we studied the impact of the physiological microenvironment of the primary tumor on lymph node metastasis in cervical carcinoma xenografts. Three pathophysiological parameters (i.e. hypoxic fraction, IFP, and pH_e) were measured in each tumor and related to the metastatic status of the host mouse. Xenografted tumors of two cervical carcinoma lines differing in histological appearance were included in the study.

Material and methods

Tumor models

CK-160 and TS-415 cervical carcinoma xenografts growing in the gastrocnemius muscle of adult female BALB/c nu/nu mice were used as tumor models [19]. Tumors were initiated from established cell lines cultured in RPMI-1640 (25 mmol/l HEPES and l-glutamine) medium supplemented with 13% bovine calf serum, 250 mg/l penicillin, and 50 mg/l streptomycin. The cell lines were established in our laboratory from patients admitted to the Norwegian Radium Hospital for the treatment of FIGO stage IIB squamous cell carcinoma of the uterine cervix. The CK-160 line was derived from a pelvic lymph node metastasis of a 65-year-old woman having developed a well-differentiated keratinizing primary tumor. The tumor responded well to radiation therapy, but was highly metastatic. The TS-415 line was derived from a pelvic lymph node metastasis of a 45-year-old woman having developed a poorly differentiated nonkeratinizing primary tumor. The tumor was moderately metastatic, but resistant to radiation treatment. Comparative studies have shown that the histology of CK-160 and TS-415 xenografts is similar to that of the donor patients’ tumors [19], and this was confirmed in the present study. Thus, the CK-160 xenografts showed well-differentiated keratinizing cells consistent with Grade I squamous cell carcinoma of the uterine cervix (Figure 1a), whereas the TS-415 xenografts were poorly differentiated with a histological appearance consistent with Grade III squamous cell cervical carcinoma (Figure 1b).

Figure 1. Histological preparations of a well-differentiated CK-160 (a) and a poorly differentiated TS-415 (b) tumor stained with hematoxylin and eosin.

Anesthesia and animal care

Tumor irradiation and measurement of IFP and pH_e were carried out with anesthetized mice (0.63 mg/kg of fentanyl citrate, 20 mg/kg of fluanisone, and 10 mg/kg of midazolam) when the tumors had grown to a volume of ∼400 mm³. Animal care and experimental procedures were approved by the Institutional Committee on Research Animal Care and were performed in accordance with the Interdisciplinary Principles and Guidelines for the Use of Animals in Research, Marketing, and Education (New York Academy of Sciences, New York, NY, USA).

IFP

IFP was measured in the center of the tumors by using a Millar SPC 320 catheter equipped with a 2F Mikro-Tip tranceducer with diameter 0.66 mm (Millar Instruments, Houston, TX, USA) [20]. The catheter was connected to a computer via a Millar TC-510 control unit and a model 13-66150-50 preamplifier (Gould Instruments, Cleveland, OH, USA). Data acquisition was carried out by using LabVIEW software (National Instruments, Austin, TX, USA).

pH_e

An MI-407B needle electrode (Microelectrodes, Bedford, NH, USA) connected to a millivolt adaptor (MV-ADPT) was used to measure pH_e. A reference electrode (MI-402; Microelectrodes) was placed subcutaneously in the tumor-bearing leg. The electrodes were calibrated by using calibration buffers with pH values of 5.0, 6.0, and 7.0. A measurement lasted 12–20 min and was accepted only when the pH_e readings during the last 3 min did not change by more than 0.005 pH units/min [19]. LabVIEW software was used for data acquisition.

Irradiation

Tumors were irradiated under air-breathing or hypoxic conditions at a dose rate of 5.1 Gy/min by using an x-ray therapy unit operated at 220 kV, 19 to 20 mA, and with 0.5-mm Cu filtration [21]. Hypoxic tumors were obtained by euthanizing the host mice 5 min before the radiation exposure.

Cell survival assay

The cell survival of irradiated tumors was measured in vitro [22]. Briefly, the tumors were resected immediately after the radiation exposure, minced in cold saline, and treated with an enzyme solution (0.2% collagenase, 0.05% Pronase, and 0.02% DNase) at 37°C for 2 h. Trypan blue-negative cells were plated in 25-cm² tissue culture flasks and incubated at 37°C for 14 days for colony formation. Cell surviving fractions were calculated from the number of colonies counted and the number of cells seeded, corrected for the mean plating efficiency of the cells of six untreated control tumors.

Fraction of radiobiologically hypoxic cells

The fraction of radiobiologically hypoxic cells in a tumor was measured by using a procedure based on the paired survival curve method [21]. The tumor was irradiated with 10 Gy under air-breathing conditions, and the fraction of surviving tumor cells [SF₁₀ (air-breathing)] was determined as described above. The fraction of radiobiologically hypoxic cells (HF_Rad) was calculated as HF_Rad = SF₁₀ (air-breathing)/SF₁₀ (hypoxic), where SF₁₀ (hypoxic) represents the mean cell surviving fraction of 10 tumors irradiated with 10 Gy under hypoxic conditions. A radiation dose of 10 Gy was chosen for assessment of HF_Rad because the fraction of normoxic tumor cells surviving exposure to 10 Gy has been shown to be insignificant in CK-160 and TS-415 xenografts [19,23]. The assessment of HF_Rad in individual tumors from measurements of SF₁₀ has been justified previously [21]. The theoretical basis for the paired survival curve method has been described in detail by Hall in his textbook on basic radiation biology [24].

Metastasis

Metastatic status was assessed as described elsewhere [19]. Briefly, euthanized mice were examined for external lymph node metastases in the inguinal, axillary, interscapular, and submandibular regions and internal lymph node metastases in the abdomen and mediastinum. The presence of metastatic growth in enlarged lymph nodes was confirmed by histological examination.

Statistical analysis

Data are presented as arithmetic mean ± SE unless otherwise stated. The Pearson product moment correlation test was used to search for correlations between parameters. Statistical comparisons of data were carried out by using the Student's t-test when the data complied with the conditions of normality and equal variance. Under other conditions, comparisons were carried out by nonparametric analysis using the Mann-Whitney rank-sum test. Probability values of p < 0.05, determined from two-sided tests, were considered significant. The statistical analysis was performed by using the SigmaStat statistical software (SPSS Science, Chicago, IL, USA).

Results

HF_Rad, IFP, and pH_e were measured in 22 CK-160 and 16 TS-415 tumors and related to the metastatic status of the host mice. After exposure to 10 Gy under air-breathing conditions, the tumors showed cell surviving fractions differing by factors of ∼3.5 (CK-160) and ∼4.8 (TS-415), whereas the cell surviving fractions of the 10 tumors exposed to 10 Gy under hypoxic conditions differed by a factor of ∼1.1 in both lines (Figure 2). The mean values of SF₁₀ (air-breathing) and SF₁₀ (hypoxic) were 0.019 and 0.044, respectively, for the CK-160 tumors and 0.043 and 0.18, respectively, for the TS-415 tumors. The coefficients of variation were 34% [SF₁₀ (air-breathing)] and 2.6% [SF₁₀ (hypoxic)] for the CK-160 tumors and 43% [SF₁₀ (air-breathing)] and 4.5% [SF₁₀ (hypoxic)] for the TS-415 tumors. The differences in SF₁₀ (air-breathing) among the 22 CK-160 tumors and the differences in SF₁₀ (air-breathing) among the 16 TS-415 tumors thus reflected primarily differences in the extent of hypoxia and not experimental uncertainty.

Figure 2. Cell surviving fractions of CK-160 and TS-415 tumors irradiated with 10 Gy under air-breathing (○) or hypoxic (•) conditions. Points and horizontal lines represent individual tumors and mean values, respectively.

Both the CK-160 tumors and the TS-415 tumors differed significantly in HF_Rad, IFP, and pH_e (Figure 3). HF_Rad differed from 19 to 65% with a geometric mean of 41% and an arithmetic mean of 43% (CK-160) and from 8.4 to 40% with a geometric mean of 21% and an arithmetic mean of 24% (TS-415). IFP differed from 2.7 to 40 mmHg with a mean value of 19 mmHg (CK-160) and from 2.4 to 59 mmHg with a mean value of 24 mmHg (TS-415), and pH_e differed from 6.2 to 7.3 with a mean value of 6.8 (CK-160) and from 5.9 to 7.3 with a mean value of 6.7 (TS-415). HF_Rad was significantly higher in the CK-160 tumors than in the TS-415 tumors (p = 0.000084; Figure 3a), whereas the CK-160 and the TS-415 tumors did not differ significantly in IFP (p > 0.05; Figure 3b) or pH_e (p > 0.05; Figure 3c).

Figure 3. HF_Rad (a), IFP (b), and pH_e (c) of CK-160 and TS-415 tumors. Points and horizontal lines represent individual tumors and mean values, respectively.

The three pathophysiological parameters were not correlated with each other in any of the tumor lines. This is illustrated in Figure 4, which shows plots of HF_Rad versus IFP, HF_Rad versus pH_e, and IFP versus pH_e for the CK-160 tumors (p > 0.05; Figure 4a) and the TS-415 tumors (p > 0.05; Figure 4b).

Figure 4. HF_Rad versus IFP, HF_Rad versus pH_e, and IFP versus pHe for CK-160 (a) and TS-415 (b) tumors. Points represent individual tumors.

Twelve of the 22 mice bearing CK-160 tumors and eight of the 16 mice bearing TS-415 tumors developed lymph node metastases. In the CK-160 line (Figure 5a), HF_Rad was ∼1.5-fold higher in the metastatic than in the nonmetastatic tumors (p = 0.0092), whereas the metastatic and nonmetastatic tumors did not differ significantly in IFP (p > 0.05) or pH_e (p > 0.05). In the TS-415 line (Figure 5b), IFP was∼ 2.6-fold higher in the tumors that metastasized than in those that did not metastasize (p = 0.0063), whereas the tumors of the metastasis-positive and metastasis-negative mice did not differ significantly in HF_Rad (p > 0.05) or pH_e (p > 0.05).

Figure 5. HF_Rad, IFP, and pH_e of metastatic and nonmetastatic CK-160 (a) and TS-415 (b) tumors. Points and horizontal lines represent individual tumors and mean values, respectively.

Discussion

The disease-free and overall survival rates of patients with advanced squamous cell carcinoma of the uterine cervix have been shown to be associated with abnormalities in the physiological microenvironment of the primary tumor [3–18]. The possibility that an abnormal physiological microenvironment may facilitate the development of lymph node metastases was investigated in the present preclinical study. Two cervical carcinoma xenograft lines were included in the study, and it has been shown that the tumors of these lines reflect the donor patients’ tumors in several biological properties, including histological appearance, radiation sensitivity and metastatic propensity [19].

Preclinical studies are suitable for studying tumor-specific mechanisms of microenvironment-associated metastasis because large numbers of genetically equal tumors can be produced and included in experiments. A prerequisite is, however, that the individual tumors show highly different microenvironments. This condition was fulfilled for both tumor lines included in our study. The large physiological differences among the individual tumors were most likely a consequence of stochastic processes influencing the initiation of angiogenesis shortly after tumor cell inoculation. The intertumor heterogeneities in HF_Rad, IFP, and pH_e of the CK-160 and the TS-415 lines were similar to those in comparable physiological parameters reported for cervical carcinomas in humans [4,7,15].

Tumor hypoxia is a result of an imbalance between the rate of oxygen supply, determined primarily by the blood perfusion, and the rate of oxygen consumption, determined primarily by the respiratory activity of the tumor cells [25]. Interstitial hypertension in tumors is mainly a consequence of high geometric and viscous resistance to blood flow, low resistance to transcapillary fluid flow, and impaired lymphatic drainage [7,16,17]. Tumor acidity is primarily due to increased anaerobic metabolism, high rate of aerobic glycolysis, and insufficient vascular transport of acid and waste products [26]. Thus, tumor hypoxia, interstitial hypertension, and acidity are partly consequences of microvascular abnormalities and partly consequences of vascular-unrelated tumor abnormalities [4]. It has therefore been suggested that the fraction of hypoxic cells, IFP, and pH_e might be closely related pathophysiological parameters in tumor tissue. However, there was no correlation between HF_Rad and IFP, HF_Rad and pH_e, or IFP and pH_e in any of the tumor lines in this study, implying that these parameters are not governed by a common mechanism in CK-160 or TS-415 tumors. This observation is consistent with the clinical observation that tissue pO₂ is not correlated to IFP in cervical carcinomas [7].

The development of lymph node metastases was found to be associated with the fraction of hypoxic cells in CK-160 tumors and with the level of interstitial hypertension in TS-415 tumors, whereas associations between lymph node metastasis and tumor acidity could not be detected in any of the tumor lines. These observations suggest that abnormalities in the physiological microenvironment of cervical carcinomas may facilitate metastatic dissemination and, furthermore, that different mechanisms may be involved in different tumors, depending on the biological properties of the tumor tissue. Interestingly, the CK-160 tumors were well differentiated (histological Grade I), whereas the TS-415 tumors were poorly differentiated (histological Grade III).

Tumor hypoxia may promote metastatic spread by inducing genomic instability and selecting for hypoxia-resistant, aggressive cell phenotypes [9,11]. However, this process is slow and is therefore not a probable mechanism for rapidly growing experimental tumors like the CK-160 tumors. Furthermore, studies in our laboratory have shown that long-term culturing of CK-160 cells under chronic or cycling (acute) hypoxia in vitro does not induce or select for cell variants with enhanced aggressiveness or increased metastatic potential.

It is far more probable that hypoxia facilitated lymph node metastasis of CK-160 tumors by up-regulating the expression of genes involved in the metastatic process [9–11]. Tumor hypoxia activates several transcription factors including hypoxia-inducible factor-1 (HIF-1), and targets of HIF-1 play critical roles in many steps of the metastatic process, including angiogenesis, cell viability/apoptosis, cell proliferation/growth arrest, tissue remodelling/cell invasion, and glycolysis/metabolism [27,28]. The disease-free survival of cervical cancer patients has been shown to be associated with the expression of HIF-1 [29–31] and several HIF-1 target genes, including the facilitative glucose transporter Glut-1 [32] and the proangiogenic factor vascular endothelial growth factor-A (VEGF-A) [33].

The development of lymph node metastases in TS-415 tumors was associated with the level of interstitial hypertension rather than the extent of hypoxia. Associations between IFP and metastasis have also been observed in nonhypoxic human melanoma xenografts [34], and IFP has been shown to have independent prognostic power for relapse at distant sites in patients with cervical cancer [16]. Mechanisms linking high IFP to increased metastasis have not been clearly identified, but possible mechanisms have been suggested. Thus, the outward fluid flow in the periphery of tumors with high IFP may be sufficiently strong to promote migration of tumor cells toward functional lymphatics surrounding the tumor tissue [34]. Furthermore, the VEGF family of ligands may be up-regulated by interstitial hypertension and hence promote lymphangiogenesis and lymphatic intravasation in the tumor periphery [35].

Tumor acidity may promote metastasis by up-regulating the expression of proteolytic enzymes as well as proangiogenic factors [36,37]. However, the development of lymph node metastases in CK-160 or TS-415 tumors was not associated with pH_e despite the large intertumor heterogeneities in this parameter, possibly because any acidity-induced gene up-regulation was not sufficiently strong to influence the metastatic process significantly.

Our study shows unequivocally that lymph node metastasis in cervical carcinoma xenografts is associated with abnormalities in the physiological microenvironment of the primary tumor. It is thus conceivable that the poor survival rates of cervical cancer patients having developed primary tumors with severe microenvironmental abnormalities may be a consequence of microenvironment-induced metastasis and, hence, that the tumor microenvironment may represent an important target for novel therapeutic strategies in cervical carcinoma. Therapeutic strategies making use of antiangiogenic agents that normalize the tumor microvasculature may be particularly interesting because it has been shown that such agents may decrease tumor IFP as well as improve tumor oxygenation [38]. Preclinical studies investigating this possibility in cervical carcinoma xenografts are highly warranted.

Acknowledgements

Financial support was received from the Norwegian Cancer Society and the South-Eastern Norway Regional Health Authority.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.