Interstitial fluid pressure (IFP) is elevated in many experimental and human tumors, and high IFP is associated with poor prognosis in human cancer. The significance of elevated IFP in the development of metastatic disease was investigated in the present work by using A-07 human melanoma xenografts as models of cancer in humans. IFP was measured with the micropipette technique (tumor periphery) and the wick-in-needle technique (tumor center). Tumor hypoxia was studied by immunohistochemistry using pimonidazole as a hypoxia marker and by using a radiobiological assay. High central tumor IFP was found to be associated with the development of pulmonary (P = 0.000085) and lymph node (P = 0.000036) metastases in small (150–200 mm3) A-07 tumors. Hypoxic cells could not be detected in these tumors. Our study suggests that interstitial hypertension may facilitate tumor cell intravasation and, hence, promote metastasis by mechanisms independent of tumor hypoxia.

Most experimental and human tumors develop elevated IFP3 during growth (1). Measurements of IFP in tumors have yielded values up to 110 mm Hg, whereas most normal tissues show IFP values close to 0 mm Hg (2, 3). The microvascular hydrostatic pressure is the principal driving force for interstitial hypertension in tumors (4, 5). Tumors generally show high resistance to capillary blood flow, low resistance to transcapillary fluid flow, and impaired lymphatic drainage (6). Therefore, the microvascular hydrostatic pressure forces fluid from the microvasculature into the tumor interstitium where the fluid accumulates, distends the elastic extracellular matrix, and causes an increase in the IFP (5, 7). The resistance to fluid flow in the tumor interstitium greatly exceeds the resistance to transcapillary fluid flow (7), leading to a pseudostable state where the central tumor IFP is nearly equal to the microvascular hydrostatic pressure (1, 5).

Tumors may be resistant to various categories of cancer treatment because of interstitial hypertension. Elevated IFP may lead to poor and heterogeneous uptake of macromolecular and nanoparticle therapeutic agents, and, hence, resistance to some forms of immunotherapy and gene therapy (8). Interstitial hypertension may also cause impaired blood flow and oxygen supply (9), and, hence, resistance to radiotherapy (3, 10). Interestingly, a recent prospective study has shown that preradiotherapy IFP can predict disease-free survival in patients with cervical carcinoma independent of clinical prognostic factors and tumor oxygenation (11).

An experimental study demonstrating for the first time an association between IFP and development of metastatic disease is reported in the present study. IFP and hypoxic fraction may be correlated in some tumors (10, 12), and there is experimental and clinical evidence suggesting that hypoxia may promote cancer metastasis (13). Confounding effects of hypoxia were avoided in the present study by using small A-07 human melanoma xenografts as tumor models (14). A-07 tumors are well vascularized and do not develop hypoxic regions at volumes of <200 mm3 as demonstrated here by using immunohistochemical and radiobiological assays in attempts to detect hypoxic cells.

Mice and Tumors.

Adult (8–10 weeks of age) female BALB/c-nu/nu mice were used as host animals for xenografted tumors. Tumors were initiated from exponentially growing A-07 cell cultures. The A-07 cell line, established as described previously (14), was maintained in monolayer culture in RPMI 1640 (25 mm HEPES and l-glutamine) supplemented with 13% bovine calf serum, 250 mg/liter penicillin, and 50 mg/liter streptomycin. Cells (∼2.0 × 105) suspended in 10 μl of Ca2+ and Mg2+-free HBSS were inoculated intradermally in the left mouse flank as described earlier (14). Animal experiments were approved by the Institutional Committee on Research Animal Care and were performed according to the Interdisciplinary Principles and Guidelines for the Use of Animals in Research, Marketing, and Education (New York Academy of Sciences, New York, NY).

IFP Measurements.

The IFP in the periphery of tumors was measured with micropipettes and a servo-nulling device (Vista Electronics, Ramon, CA) by using a procedure similar to that described by Boucher et al.(4). Micropipettes with tip diameters of 2–4 μm, filled with 1 m NaCl solution, were moved progressively into tumors by using a graded micromanipulator under stereomicroscopic guidance. Zero pressure was recorded in a drop of saline at the tumor surface before each micropipette insertion. IFP was measured at intervals of 0.25 mm up to a tumor depth of 2.5 mm. Each IFP value was recorded for at least 10 s. Measurements were accepted as valid when the fluid communication between the micropipette and the tissue could be confirmed electrically and the zero pressure in the saline at the tumor surface was not modified during micropipette insertion and withdrawal.

The IFP in the center of tumors was measured with the wick-in-needle technique by using a procedure similar to that described by Tufto and Rofstad (15). Each IFP value was recorded for ≥5 min. The fluid communication between the pressure transducer and a tumor was tested by compressing and decompressing the tubing between the needle and the transducer after a stable value had been reached. Measurements were accepted as valid when the readings after these tests did not differ by more than 1 mm Hg. Tumor IFP was determined by calculating the mean of these two readings. IFP values in normal tissues, recorded i.m. and subdermally, served as internal controls.

The mice were kept under general anesthesia during IFP measurements. Propanidid (Gedeon Richter, Budapest, Hungary), fentanyl/fluanisone (Janssen Pharmaceutika, Beerse, Belgium), and diazepam (Dumex, Copenhagen, Denmark) were administered i.p. in doses of 400, 0.24/12, and 4 mg/kg body weight, respectively. The body core temperature of the mice, measured with a rectal probe, was kept at 36–38°C by using a heating pad.

Immunohistochemical Detection of Hypoxia.

Pimonidazole was used as a marker of tumor hypoxia. A peroxidase-based immunohistochemical method was used to detect hypoxic tumor regions (16). Pimonidazole hydrochloride, kindly supplied by Professor James A. Raleigh (Department of Radiation Oncology, University of North Carolina School of Medicine, Chapel Hill, NC) was dissolved in 0.9% NaCl and administered i.p. in doses of 30 mg/kg body weight. The tumors were dissected free from the mice 4 h after the pimonidazole administration and fixed in phosphate-buffered 4% paraformaldehyde. Slides with tumor tissue preparations were incubated with polyclonal rabbit antiserum to pimonidazole-protein adducts, a gift from Professor J. A. Raleigh. Visualization of the antibody complex was achieved with the 3,3-diaminobenzidine chromogen. Hematoxylin was used for counterstaining. Large A-07 tumors, ∼1000 mm3 in volume, were used as positive controls. The histology of tumors staining positive for pimonidazole has been displayed elsewhere (16).

Radiobiological Detection of Hypoxia.

A Siemens Stabilipan X-ray unit, operated at 220 kV, 19–20 mA, and with 0.5 mm Cu filtration, was used for irradiation. Cell cultures were irradiated at a dose rate of 3.4 Gy/min and tumors at a dose rate of 5.1 Gy/min. Hypoxic tumors were obtained by occluding the blood supply with a clamp 5 min before irradiation. The detection of hypoxia in tumors with unperturbed blood supply was based on the paired survival curve method (17). Cell survival was measured in vitro by using a plastic surface colony assay. Lethally irradiated (30 Gy) feeder cells were used to ensure a linear relationship between the number of colonies and the number of plated cells. Single-cell suspensions were prepared from tumors by using a combined mechanical and enzymatic procedure. The tumors were minced with scalpels in Ca2+ and Mg2+-free HBSS before being subjected to enzymatic treatment at 37°C for 2 h. The enzyme solution consisted of 0.2% collagenase, 0.05% Pronase, and 0.02% DNase in HBSS. Cells giving rise to colonies with >50 cells were scored as clonogenic. The experimental procedure has been described in detail elsewhere (16).

Metastasis Assay.

Primary tumors were initiated in the left mouse flank as described above. They were removed surgically when the volume was 150–200 mm3, i.e., 7–9 days after initiation. The mice were examined for clinical signs of metastases twice a week. They were killed and autopsied when they were moribund or 3 months after the removal of the primary tumor. The lungs were examined for pulmonary metastases, and mediastinum, abdomen, and the interscapular, submandibular, axillary, and inguinal regions were examined for lymph node metastases. Moribund mice were always positive for metastases. The presence of metastases was confirmed by histological examinations.

Statistical Analysis.

Correlations between two parameters were searched for by linear regression analysis. Statistical comparisons of data were performed by using the Student’s t test. The data sets were verified to comply with the conditions of normality and equal variance. Probability values of P < 0.05, determined from two-sided tests, were considered significant. The statistical analysis was performed by using SigmaStat statistical software (Jandel Scientific GmbH, Erkrath, Germany).

Studies investigating associations between IFP and metastasis should make use of tumors without hypoxic regions, because hypoxia may promote metastasis (13), and IFP and hypoxic fraction may be correlated in some tumors (12). Therefore, radiobiological experiments were performed to search for hypoxic cells in 150–200-mm3 A-07 tumors (Fig. 1). Cell survival curves were determined for aerobic monolayer cultures and for tumors with occluded blood supply, i.e., tumors with a hypoxic fraction of 100%, to establish a framework for the experiments. These reference experiments gave cell survival curves consistent with an oxygen enhancement ratio of ∼3, in agreement with classical radiobiological theory (17). The expected cell survival curve for tumors with a hypoxic fraction of 1%, calculated from these data by using the paired survival curve method (17), is shown as a dashed curve in Fig. 1. Ten tumors with unperturbed blood supply were then irradiated with a dose of 7.5 Gy and assayed for cell survival. The surviving fractions determined for these tumors were similar to those measured for aerobic monolayer cultures exposed to 7.5 Gy and lower than those expected for tumors having a hypoxic fraction of 1% (P = 0.0040). These data demonstrate that the fraction of radiobiologically hypoxic cells in 150–200-mm3 A-07 tumors is <1% and is consistent with the tumors having no radiobiologically hypoxic cells. In comparison, similar experiments have shown that the mean fraction of radiobiologically hypoxic cells in 200–400-mm3 A-07 tumors is 6% (16).

Intratumor heterogeneity in IFP was studied by moving micropipettes progressively into tumors from the outward surface and recording the IFP at intervals of 0.25 mm. The IFP increased with increasing depth until a plateau was reached at a depth of 0.75 mm, measured from the skin surface (Fig. 2,A). Six tumors were included in the study, and all of the tumors showed a steep IFP gradient in the periphery and relatively uniform IFP values at depths beyond 0.75 mm. The IFP profiles did not differ noticeably among the tumors, apart from the fact that the absolute values were highly different, ranging from 5 to 15 mm Hg in the central plateau. Therefore, intertumor heterogeneity in IFP was studied by recording the IFP in the center of tumors with the wick-in-needle technique. The reproducibility of such measurements was investigated by measuring the IFP in 18 tumors twice. The second measurement was performed 3–4 h after the first measurement. The correlation between the first and the second measurement was excellent (P < 0.00001; R2 = 0.96); the difference between the two IFP values never exceeded 2 mm Hg (Fig. 2 B).

An association between IFP and metastasis was searched for by performing an experiment involving 50 mice. The primary tumors were removed when the volume was within the range of 150–200 mm3. IFP was measured in the tumor center with the wick-in-needle technique immediately before tumor removal, and pimonidazole was administered to the host mice 4 h earlier. Twenty-one mice developed metastases, whereas 29 mice did not. Nine of the metastasis-positive mice showed both pulmonary and lymph node metastases, 10 showed pulmonary metastases only, and 2 showed lymph node metastases only. The metastatic primary tumors had ∼1.7-fold higher IFP than the nonmetastatic primary tumors (Fig. 3,A; P = 0.000026). Moreover, IFP was ∼1.7-fold higher in the primary tumors that metastasized to the lungs than in those that did not give rise to pulmonary metastases (Fig. 3,B; P = 0.000085) and ∼1.8-fold higher in the primary tumors that metastasized to lymph nodes than in those that did not form lymph node metastases (Fig. 3 C; P = 0.000036). Hypoxic regions, i.e., positive pimonidazole staining, could not be detected in any of the primary tumors. In contrast, five large A-07 control tumors, ∼1000 mm3 in volume, showed hypoxic fractions of 6–10%, consistent with previous studies (16).

Metastatic spread of malignant cells from the primary tumor to distant sites is the major cause of death of patients with cancer. The metastatic process is composed of a cascade of linked, sequential, and highly selective steps involving multiple host-tumor interactions. These steps include invasion of tumor cells into blood or lymphatic vessels, survival in the circulation, arrest in a secondary organ, extravasation into the secondary organ interstitium and parenchyma, proliferation in the secondary organ, and induction of angiogenesis. Metastatic cell phenotypes are, because of the complexity of the dissemination process, believed to have accumulated several stable and/or unstable genomic changes affecting growth regulation and tissue homeostasis. Associations between metastatic spread and gene expression have been demonstrated for many gene products (18).

The study reported here demonstrates that metastasis also is associated with the IFP of the primary tumor. The development of metastases from small A-07 tumors was studied, and the primary tumors that metastasized to the lungs showed ∼1.7-fold higher IFP than those that did not form pulmonary metastases. Moreover, IFP was ∼1.8-fold higher in the primary tumors that metastasized to lymph nodes than in those that did not give rise to lymph node metastases. However, it should be noticed that a substantial fraction of the nonmetastatic tumors showed IFP values within the same range as those of the metastatic tumors, suggesting that high IFP is not a sufficient condition for metastasis in small A-07 tumors. High IFP is probably not a necessary condition either, because primary tumors with IFP values as low as 5 mm Hg were capable of forming both pulmonary and lymph node metastases.

IFP was measured in a single location in the tumor center by using the wick-in-needle technique. Micropipette studies of the intratumor heterogeneity in IFP justified this procedure. The IFP was relatively uniform throughout the tumors and dropped precipitously to normal tissue values at the tumor surface, in accordance with results from theoretical considerations and experimental studies of murine tumors (4). Moreover, IFP measurements with the wick-in-needle technique gave highly reproducible results, demonstrated by performing repeated measurements in the same tumors, and control measurements in normal tissues gave IFP values ranging from −1 to +1 mm Hg, consistent with the normal tissue values reported by others (19).

Tumors with low resistance to transcapillary fluid flow, such as the A-07 tumors studied here, develop a central IFP that is nearly equal to the microvascular hydrostatic pressure (5, 15). The microvascular hydrostatic pressure is determined by the resistance of the capillary network to blood flow (9, 12), which is elevated in tumors for several reasons, including high blood viscosity, abnormal capillary structure and organization, and capillary compression because of tumor cell proliferation in a confined space (6). Therefore, intertumor heterogeneity in central IFP in tumors with low resistance to transcapillary fluid flow is mainly a consequence of intertumor heterogeneity in the microvascular parameters determining the resistance to capillary blood flow. The A-07 tumors used in the present work were initiated in the same site in identical hosts from the same cell culture. The intertumor heterogeneity in resistance to capillary blood flow and, hence, the intertumor heterogeneity in central IFP was, therefore, most likely a result of stochastic processes in the development of the capillary network rather than genetic differences among the tumors. Consequently, the association between metastasis and IFP in A-07 tumors is unlikely to reflect that the most metastatic cell phenotypes developed the tumors with the highest IFP; the association rather reflects that high IFP, caused by stochastic processes, increased the probability of metastasis in genetically similar tumors.

There is significant evidence from clinical and experimental studies that tumor hypoxia may promote metastasis by activating hypoxia-inducible factor-1 and other transcription factors, and, hence, increase the expression of genes playing an important role in the metastatic process (13). Studies of carcinoma of the uterine cervix have suggested that IFP and hypoxic fraction may be correlated in tumors (3, 10). Confounding effects of hypoxia were avoided in the present work by using small A-07 tumors as models of human cancer. None of the tumors included in the study had developed hypoxic regions at tumor removal, as was revealed by using pimonidazole as a hypoxia marker. Radiobiological experiments showed that the fraction of hypoxic cells in small A-07 tumors is <1%. The radiobiological data were consistent with the assumption that the tumors have no hypoxic cells and were, thus, in agreement with the immunohistochemical observations. Consequently, the association between metastasis and IFP reported here is unlikely to reflect an association between metastasis and hypoxia. If A-07 tumors are representative models of tumors in humans, the present study suggests that interstitial hypertension may promote metastatic disease in human cancer by mechanisms independent of tumor oxygenation.

The mechanisms by which interstitial hypertension promotes metastasis in A-07 tumors cannot be established from the experiments reported here. However, it is tempting to speculate that mechanical forces may facilitate tumor cell intravasation and, hence, metastatic spread in tumors with high IFP. Pulmonary metastases in A-07 tumors most likely originate from tumor cells entering blood vessels within the primary tumor. Studies of experimental tumors have shown that abrupt decreases in the microvascular hydrostatic pressure leads to fluid flow from the interstitium into the microvasculature (7). A-07 tumors show substantial intermittent blood flow at the microvascular level (16), implying that abrupt decreases in the microvascular hydrostatic pressure may be a frequent phenomenon, and it is, thus, possible that the central tumor IFP for short periods of time is substantially higher than the hydrostatic pressure of a microvessel, particularly in tumors with high central IFP, and that the subsequent fluid flow facilitates intravasation of tumor cells adjacent to the microvessel. Lymph node metastases in A-07 tumors most likely originate from tumor cells spread by lymphatics. Functional lymphatics cannot be detected in the interior of A-07 tumors but can be observed close to the tumor periphery in the normal skin tissue surrounding the tumors. Studies of experimental tumors have shown that interstitial fluid is forced from the periphery of tumors into adjacent normal tissue where it is collected and removed by lymphatics (4, 5). This fluid flow is driven by the difference in IFP between the tumor and the adjacent normal tissue. The IFP gradient in the periphery of A-07 tumors is steep, and it is, thus, possible that in tumors with high central IFP, the outward fluid flow is sufficiently strong that the migration of peripheral tumor cells toward lymphatics is promoted by convection.

Only two studies of the IFP of melanomas in humans have been reported thus far, one by Boucher et al.(19) and the other by Curti et al.(2). These studies revealed that the IFP in general is higher in melanoma than in human tumors of other histological types, including cervical carcinoma, breast carcinoma, lymphoma, and colorectal carcinoma (2, 3, 10, 11, 20). Boucher et al.(19) measured IFP values as high as 48 mm Hg in some large melanomas, and even higher values, ≤110 mm Hg, were recorded by Curti et al.(2). In comparison, the highest IFP measured in the A-07 melanomas included in our metastasis study was only 20 mm Hg. Therefore, it is possible that metastasis promotion by interstitial hypertension is an even more extensive problem in human melanomas than the present experimental study suggests. Interestingly, Curti et al.(2) showed that elevated tumor IFP was associated with poor prognosis in melanoma patients treated with interleukin 1α or interleukin 2 immunotherapy.

In summary, pulmonary and lymph node metastasis is associated with the IFP of the primary tumor in A-07 human melanoma xenografts. The association does not reflect that the most aggressive cell phenotypes develop the primary tumors with the highest IFP but suggests rather that interstitial hypertension increases the probability of metastatic dissemination in genetically similar tumors. The mechanism by which elevated IFP promotes metastasis in A-07 tumors is independent of tumor oxygenation.

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

      
1

Supported by grants from The Norwegian Cancer Society.

            
3

The abbreviations used are: IFP, interstitial fluid pressure; pimonidazole, {1-[(2-hydroxy-3-piperidinyl)propyl]-2-nitroimidazole}.

Fig. 1.

•, cell survival levels for A-07 cultures irradiated under aerobic conditions in vitro. Points, means of five experiments; bars, ± SE. ▴, cell survival levels for A-07 tumors irradiated under hypoxic conditions in vivo. Points, means of six tumors; bars, ± SE. ▵, cell survival levels for A-07 tumors irradiated with 7.5 Gy under unperturbed conditions in vivo. Points, single tumors. Dashed curve, calculated cell survival curve for A-07 tumors with a hypoxic fraction of 1% irradiated under unperturbed conditions in vivo.

Fig. 1.

•, cell survival levels for A-07 cultures irradiated under aerobic conditions in vitro. Points, means of five experiments; bars, ± SE. ▴, cell survival levels for A-07 tumors irradiated under hypoxic conditions in vivo. Points, means of six tumors; bars, ± SE. ▵, cell survival levels for A-07 tumors irradiated with 7.5 Gy under unperturbed conditions in vivo. Points, single tumors. Dashed curve, calculated cell survival curve for A-07 tumors with a hypoxic fraction of 1% irradiated under unperturbed conditions in vivo.

Close modal
Fig. 2.

A, intratumor heterogeneity in IFP in A-07 tumors was studied by using the micropipette technique. Normalized IFP versus depth, measured from the skin surface, is shown. IFP was normalized by assigning a value of 100% to the highest IFP value recorded in each tumor. Points, means of 6 tumors; bars, ± SE. B, the reproducibility of the wick-in-needle technique was investigated by measuring the IFP in the center of A-07 tumors twice. The second measurement was performed 3–4 h after the first measurement. IFP (Measurement # 1) versus IFP (Measurement # 2) is shown. Points, individual tumors. Curve, regression line (P < 0.00001; R2 = 0.96).

Fig. 2.

A, intratumor heterogeneity in IFP in A-07 tumors was studied by using the micropipette technique. Normalized IFP versus depth, measured from the skin surface, is shown. IFP was normalized by assigning a value of 100% to the highest IFP value recorded in each tumor. Points, means of 6 tumors; bars, ± SE. B, the reproducibility of the wick-in-needle technique was investigated by measuring the IFP in the center of A-07 tumors twice. The second measurement was performed 3–4 h after the first measurement. IFP (Measurement # 1) versus IFP (Measurement # 2) is shown. Points, individual tumors. Curve, regression line (P < 0.00001; R2 = 0.96).

Close modal
Fig. 3.

IFP in metastatic and nonmetastatic A-07 primary tumors. The lungs were examined for pulmonary metastases, and mediastinum, abdomen, and the interscapular, submandibular, axillary, and inguinal regions were examined for metastases in lymph nodes. Points, single tumors. A, lungs and lymph nodes. B, lungs only. C, lymph nodes only.

Fig. 3.

IFP in metastatic and nonmetastatic A-07 primary tumors. The lungs were examined for pulmonary metastases, and mediastinum, abdomen, and the interscapular, submandibular, axillary, and inguinal regions were examined for metastases in lymph nodes. Points, single tumors. A, lungs and lymph nodes. B, lungs only. C, lymph nodes only.

Close modal

We thank Olav Groven for excellent technical assistance. We also thank Professor James A. Raleigh, Department of Radiation Oncology, University of North Carolina School of Medicine, Chapel Hill, NC, for supplying pimonidazole hydrochloride and antipimonidazole rabbit antiserum.

1
Boucher Y., Leunig M., Jain R. K. Tumor angiogenesis and interstitial hypertension.
Cancer Res.
,
56
:
4264
-4266,  
1996
.
2
Curti B. D., Urba W. J., Alvord W. G., Janik J. E., Smith J. W., Madara K., Longo D. L. Interstitial pressure of subcutaneous nodules in melanoma and lymphoma patients: changes during treatment.
Cancer Res.
,
53
:
2204
-2207,  
1993
.
3
Milosevic M. F., Fyles A. W., Wong R., Pintilie M., Kavanagh M-C., Levin W., Manchul L. A., Keane T. J., Hill R. P. Interstitial fluid pressure in cervical carcinoma: within tumor heterogeneity, and relation to oxygen tension.
Cancer (Phila.)
,
82
:
2418
-2426,  
1998
.
4
Boucher Y., Baxter L. T., Jain R. K. Interstitial pressure gradients in tissue-isolated and subcutaneous tumors: implications for therapy.
Cancer Res.
,
50
:
4478
-4484,  
1990
.
5
Boucher Y., Jain R. K. Microvascular pressure is the principal driving force for interstitial hypertension in solid tumors: implications for vascular collapse.
Cancer Res.
,
52
:
5110
-5114,  
1992
.
6
Vaupel P., Kallinowski F., Okunieff P. Blood flow, oxygen and nutrient supply, and metabolic microenvironment of human tumors: a review.
Cancer Res.
,
49
:
6449
-6465,  
1989
.
7
Netti P. A., Baxter L. T., Boucher Y., Skalak R., Jain R. K. Time-dependent behavior of interstitial fluid pressure in solid tumors: implications for drug delivery.
Cancer Res.
,
55
:
5451
-5458,  
1995
.
8
Jain R. K. The next frontier of molecular medicine: delivery of therapeutics.
Nat. Med.
,
4
:
655
-657,  
1998
.
9
Milosevic M. F., Fyles A. W., Hill R. P. The relationship between elevated interstitial fluid pressure and blood flow in tumors: a bioengineering analysis.
Int. J. Radiat. Oncol. Biol. Phys.
,
43
:
1111
-1123,  
1999
.
10
Roh H. D., Boucher Y., Kalnicki S., Buchsbaum R., Bloomer W. D., Jain R. K. Interstitial hypertension in carcinoma of the uterine cervix in patients: possible correlation with tumor oxygenation and radiation response.
Cancer Res.
,
51
:
6695
-6698,  
1991
.
11
Milosevic M., Fyles A., Hedley D., Pintilie M., Levin W., Manchul M., Hill R. Interstitial fluid pressure predicts survival in patients with cervix cancer independent of clinical prognostic factors and tumor oxygen measurements.
Cancer Res.
,
61
:
6400
-6405,  
2001
.
12
Lee I., Boucher Y., Jain R. K. Nicotinamide can lower tumor interstitial fluid pressure: mechanistic and therapeutic implications.
Cancer Res.
,
52
:
3237
-3240,  
1992
.
13
Rofstad E. K. Microenvironment-induced cancer metastasis.
Int. J. Radiat. Biol.
,
76
:
589
-605,  
2000
.
14
Rofstad E. K. Orthotopic human melanoma xenograft model systems for studies of tumour angiogenesis, pathophysiology, treatment sensitivity, and metastatic pattern.
Br. J. Cancer
,
70
:
804
-812,  
1994
.
15
Tufto I., Rofstad E. K. Interstitial fluid pressure and capillary diameter distribution in human melanoma xenografts.
Microvasc. Res.
,
58
:
205
-214,  
1999
.
16
Rofstad E. K., Måseide K. Radiobiological and immunohistochemical assessment of hypoxia in human melanoma xenografts: acute and chronic hypoxia in individual tumours.
Int. J. Radiat. Biol.
,
75
:
1377
-1393,  
1999
.
17
Hall E. J. .
Radiobiology for the Radiologist
, 5th ed.
91
-111, Lippincott, Williams & Wilkins Philadelphia  
2000
.
18
Yokota J. Tumor progression and metastasis.
Carcinogenesis (Lond.)
,
21
:
497
-503,  
2000
.
19
Boucher Y., Kirkwood J. M., Opacic D., Desantis M., Jain R. K. Interstitial hypertension in superficial metastatic melanomas in humans.
Cancer Res.
,
51
:
6691
-6694,  
1991
.
20
Less J. R., Posner M. C., Boucher Y., Borochovitz D., Wolmark N., Jain R. K. Interstitial hypertension in human breast and colorectal tumors.
Cancer Res.
,
52
:
6371
-6374,  
1992
.