Search

Hypoxia is now established as a key factor influencing the pathophysiology of malignant growth. Among other effects, hypoxia modulates the expression of a multitude of genes through the induction of hypoxia-inducible transcription factors. This differential gene expression favors angiogenesis, cell survival, an invasive/metastatic phenotype, and resistance to anticancer therapies. Because benign tumors do not exhibit these traits, one might expect these entities to be neither hypoxic nor to induce the genetic hypoxia response program. To test this hypothesis, an investigation of the oxygenation status of 17 leiomyomas and 1 leiomyosarcoma of the uterus using polarographic needle electrodes (Eppendorf pO(2) sensor) and the expression of hypoxia-related markers in biopsy specimens of the same tumors was carried out. Marker expression in eight additional archival leiomyosarcomas was also assessed. Leiomyoma tissue was generally found to be severely hypoxic, with median oxygen (O(2)) partial pressure values ranging from 1 to 5 mm Hg. In contrast, none of the hypoxia-related markers hypoxia-inducible factor (HIF)-1alpha, HIF-2alpha, glucose transporter-1, or carbonic anhydrase IX were expressed in any leiomyoma. Larger intercapillary distances were correlated with a poorer oxygenation status. Conversely, the expression of hypoxia-related markers was abundant in the leiomyosarcomas and they also exhibited a high-turnover phenotype (significantly increased proliferation and apoptosis). Uterine leiomyoma might therefore represent a state of oxygen-limited proliferation. Malignancy in the same organ system is associated with growth and metabolism beyond tissue-inherent limitations leading to the induction of hypoxia-related markers, thereby contributing to a self-perpetuating aggressive phenotype.

Hypoxia inducible factor-1alpha (HIF-1alpha) has been proposed as a candidate endogenous marker of tumor hypoxia and as a molecular mediator of hypoxia-driven malignant progression and acquired treatment resistance. In this study, HIF-1alpha expression in 68 biopsies of oxygenation measurement tracks from squamous cell carcinomas of the uterine cervix of 38 patients was assessed. Expression of HIF-1alpha was commonly found to increase as a function of distance from microvessels, at the center of tumor cell aggregations, and in the vicinity of necrotic areas. However, there was no correlation of HIF-1alpha expression with median oxygen tension (oxygen partial pressure; pO2) and hypoxic fractions (hypoxic fraction < 2.5 mm Hg, hypoxic fraction < 5 mm Hg). The results indicate that HIF-1alpha should not be used as an endogenous marker of tumor hypoxia in locally advanced squamous cell carcinomas of the uterine cervix. Additionally, no significant prognostic impact of HIF-1alpha expression was found in this group of patients.

Tumor hypoxia has been linked to acquired treatment resistance, tumor progression, and poor prognosis. Because anemia is a major causative factor for the development of hypoxia, the association between blood hemoglobin concentration (cHb) and breast cancer oxygenation was examined in this study. In addition, a novel parameter characterizing the relationship between oxygenation status and rising cHb is introduced: the oxygenation gain factor (OGF). In breast cancer patients, median cHb over the range 8.5-14.7 g/dl correlated positively with the median pO(2) (3-15 mm Hg), yielding an average OGF of 2 mm Hg.dl/g. In contrast, in normal tissues (normal breast, subcutis, and skeletal muscle) the median pO(2) values were substantially higher (52 mm Hg, 51 mm Hg, and 37 mm Hg, respectively) and remained constant irrespective of the hemoglobin level over the range from 10 to 16 g/dl (OGF = 0 in grade I anemia and nonanemic patients). Moderately lower median pO(2) values in subcutis and skeletal muscle were only observed in grade II anemia (8 g/dl < cHb < or =10 g/dl), although this would appear to be of no biological relevance. Conversely, in breast cancers, even mild anemia (grade I anemia) is a major causative factor for the development of hypoxia or anoxia.

The aim of this study was to examine the impact of anemia prevention by recombinant human erythropoietin (rHuEPO) treatment on the cytotoxicity of cyclophosphamide in solid experimental tumors. Anemia was induced using a single dose of carboplatin (50 mg/kg i.v.) resulting in a long-lasting reduction (30%) of the hemoglobin concentration. In a second group, the development of anemia was prevented by rHuEPO (1000 IU/kg) administered s.c. three times/week starting 7 days before carboplatin application. Four days after carboplatin treatment, tumors (DS-sarcoma of the rat) were implanted s.c. onto the hind food dorsum. Neither carboplatin nor rHuEPO treatment influenced tumor growth rate per se. When tumors were treated with a single dose of cyclophosphamide (60 mg/kg i.p.) 5 days after implantation, a growth delay with a subsequent regrowth of the tumors was observed. In the anemia group, the growth delay was significantly shorter compared with nonanemic controls (13.3 days versus 8.6 days). In the group where anemia was prevented by rHuEPO treatment, growth delay was comparable with that of nonanemic controls (13.3 days). These results suggest that chemotherapy-induced anemia reduces cytotoxicity of cyclophosphamide in tumors, whereas correction of anemia by rHuEPO treatment (epoetin alpha) increases the sensitivity, probably as a result of an improved oxygen supply to tumor tissue.

There is evidence from experimental work that hypoxia induces apoptosis in apoptosis-sensitive neoplastic cells and that this apoptotic sensitivity is lost during malignant progression. Oxygenation profiles and apoptotic indices in human squamous cell cancer of the uterine cervix have been determined, and a subgroup of tumors has been identified with low apoptotic index despite pronounced hypoxia representing carcinomas that consist of neoplastic cells with diminished apoptotic potential. These hypoxic low-apoptotic tumors show a high probability for lymphatic spread and for recurrence despite adjuvant treatment with radiation or chemotherapy in addition to radical surgery. The clinical results presented strongly support the hypothesis derived from experimental studies that the selection of apoptosis-insensitive neoplastic cell phenotypes in a hypoxic microenvironment is an important mechanism for malignant progression in solid tumors.

Dynamic studies of Gd-based contrast agents in magnetic resonance imaging (MRI) are increasingly being used for tumor characterization as well as for therapy response monitoring. Because detailed knowledge regarding the pathophysiological properties, which in turn are responsible for differences in contrast enhancement, remains fairly undetermined, it was the aim of this study to: (a) examine the association of standard and pharmacokinetic analysis of time-intensity curves in dynamic MRI with histomorphological markers of tumor angiogenesis [microvessel density (MVD) and vascular endothelial growth factor (VEGF)]; and (b) determine the ultimate value of a histomorphological and a dynamic MRI approach by the correlation of those data with disease outcome in patients with primary cancer of the uterine cervix. Pharmacokinetic parameters (amplitude, A; exchange rate constant, k21) and standard parameters [the maximum signal intensity increase over baseline (SI-I) and the steepest signal intensity-upslope per second (SI-U/s)] were calculated from a contrast-enhanced dynamic MRI series in 37 patients with biopsy-proven primary cervical cancer. On the surgical whole mount specimens, histomorphological markers of tumor angiogenesis (MVD and VEGF) were compared to MRI-derived parameters. For MRI and histomorphological data, Kaplan-Meier survival curves were calculated and compared using log-rank statistics. A significant association was found between MVD and A (P < 0.01) and SI-I (P < 0.05). No significant relationships were observed between VEGF expression and all dynamic MRI parameters. Kaplan-Meier curves based on k21 and SI-U/s showed that tumors with high k21 and SI-U/s values had a significantly (P < 0.05 and 0.001, respectively) worse disease outcome than did tumors with low k21 and SI-U/s values. None of the histomorphological gold standard markers for assessing tumor angiogenesis (MVD and VEGF) had any significant power to predict patient survival. It is concluded that in patients with uterine cervical cancer: (a) the pathophysiological basis for differences in dynamic MRI is MVD but not VEGF expression; (b) a functional, dynamic MRI approach (both standard and pharmacokinetic analysis) may be better suited to assess angiogenic activity in terms of patient survival than are the current histomorphological-based markers of tumor angiogenesis; and (c) compared with standard analysis, a simple pharmacokinetic analysis of time-intensity curves is not superior to assess MVD or patient survival.

The effects of respiratory hyperoxia (RH) and xanthine oxidase (XO) during localized hyperthermia (HT) were investigated by determining markers of oxidative damage to lipids and proteins and tumor growth. Anesthetized rats with s.c. DS-sarcomas underwent one of the following treatments: (a) localized saline-bath HT (60 min, 44 degrees C); (b) HT + RH (100% O2); and (c) HT + RH + XO (15 units/kg i.v.). Sham-treated animals served as controls. Tumors were investigated for: (a) thiobarbituric acid-reactive substance formation and protein-bound 4-hydroxynonenal, as indicators of lipid peroxidation; (b) reactive oxygen-mediated protein modifications; (c) apoptosis; and (d) tumor volume growth. Upon treatment, increases in thiobarbituric acid-reactive substances, protein-bound 4-hydroxynonenal, protein-associated carbonyl functions, and number of cells undergoing apoptosis were found in tumor tissue, together with an inhibition of tumor growth. When treatment groups were compared, effects in the group HT + RH + XO were generally most pronounced. These findings indicate that the antitumor effect of HT is at least partially mediated through the selective induction of lipid peroxidation and oxidative injury in tumor cells, leading to apoptosis. This effect was enhanced by adding RH or RH + XO, presumably due to enhanced tissue damage following an increased formation of reactive oxygen species, with higher levels of lipid peroxidation and protein oxidation.

Dynamic studies of Gd-based contrast agents in magnetic resonance imaging (MRI) are increasingly being used for tumor characterization as well as therapy response monitoring. Because detailed knowledge regarding the pathophysiological properties, which in turn are responsible for differences in contrast enhancement, remains fairly undetermined, it was the aim of this project to: (a) examine the relationship between contrast-enhanced dynamic MRI-derived characteristics and histological microvessel density counts, a recognized surrogate of tumor angiogenesis, from primary or recurrent cancers of the uterine cervix; and (b) correlate these parameters with lymphatic involvement to characterize tumor aggressiveness in terms of lymphatic spread. Pharmacokinetic parameters (amplitude, A; exchange rate constant, k21) were calculated from a contrast-enhanced dynamic MRI series in 55 patients (ages 25-72 years; mean, 50 years) with biopsy-proven primary (n = 42) or recurrent (n = 13) uterine cervical cancer. Both pharmacokinetic parameters were correlated to histologically determined microvessel density counts (factor VIII-related antigen) and other pathological tumor characteristics obtained from the operative specimens after radical surgery. In addition, the magnetic resonance and histological data were correlated to the presence or absence of lymphatic system involvement. Pharmacokinetic MRI-derived parameters (A and k21) increased with increasing histological microvessel density counts with r = 0.41 and 0.50, respectively. Lymphatic involvement was more comprehensibly assessed by the pharmacokinetic parameter k21 compared with histological microvessel density, resulting in a higher sensitivity, overall accuracy, and comparable specificity. Contrast-enhanced MRI parameters might prove to be applicable for estimation of tumor angiogenesis in uterine cervical cancer; thus, MRI may become an additional tool to characterize malignant progression in terms of lymphatic involvement in uterine cervical cancer.

Growth, blood flow, oxygenation, and bioenergetic status of experimental tumors were investigated in normal (control) and anemic animals after administration of recombinant human erythropoietin (rhEPO). DS sarcomas were implanted s.c. onto the hind foot dorsum of Sprague-Dawley rats. Tumor-associated anemia was induced by the development of an i.p. hemorrhagic ascites. rhEPO (1000 IU/kg) was administered s.c. three times per week over 14 days, after which it was found to have significantly increased hematocrit values in both normal and anemic animals. Tumor growth in anemic animals was slower than in normal animals, and rhEPO administration did not influence tumor growth in either group. Tumor blood flow in anemic animals was lower than in control animals and was only increased in larger tumors in animals in which anemia was prevented by prophylactic rhEPO application. Tumor oxygenation, determined using polarographic needle electrodes and oxygen partial pressure histography, was poorer in anemic animals than in normal animals. This reduction could be reversed partially, but not compensated fully by rhEPO treatment in smaller tumors (< or = 1.4 ml). These changes suggest that rhEPO, by improving tumor oxygenation, may increase the efficacy of standard radiotherapy in anemic animals and may be of use in anemic tumor patients in whom the success of radiotherapy or O2-dependent chemotherapy might be limited by tumor hypoxia.

Experimental tumors contain a significant fraction of microregions that are chronically or transiently hypoxic. Experimental evidence showing that hypoxia (and subsequent reoxygenation) may have a profound impact on malignant progression and on responsiveness to therapy is growing. The clinical relevance of tumor oxygenation in human solid malignancies is under investigation. We have developed and validated a clinically applicable method for measurement of tumor oxygenation in locally advanced cancer of the uterine cervix using a computerized polarographic electrode system. Applying this procedure in patients with cervical cancers

The effect of inhibiting DNA synthesis on the cellular O2 consumption rate of tumor cells (DS sarcoma) in vivo was analyzed using a photometric technique. Five days after DS-sarcoma ascites was induced in SD rats, animals were treated either with fludarabine (400 mg/kg i.p., 6 h prior to measurements) or lovastatin (3 x 20 mg/kg i.p., 24, 15, and 3 h prior to measurements), drugs that can inhibit tumor cell proliferation. In addition to cellular O2 consumption, the cell cycle distribution and the fraction of DNA-synthesizing cells in the tumor ascites were measured. Both drugs lowered DNA synthesis significantly, an effect that was more pronounced with fludarabine. The cellular O2 consumption rate following lovastatin application was significantly impaired (approximately 33%), whereas fludarabine had practically no effect on the respiration rate of tumor cells. From these data, it is concluded that a reduction in DNA synthesis does not necessarily result in a decrease in the O2 consumption rate of tumor cells in vivo.

Global blood flow (TBF), tumor vascular resistance, laser Doppler flow in superficial tumor areas, and mean arterial blood pressure were evaluated in rats bearing s.c. DS sarcomas. Measurements were performed before and after i.v. administration of rhTNF-alpha 2 or recombinant human lymphotoxin (rhLT) (1 mg/kg). Upon application of the cytokines a significant drop in TBF was found at t greater than or equal to 90 min with a stronger action following rhLT than rhTNF-alpha. At relatively constant mean arterial blood pressure values following the cytokine injection, the microcirculatory function in the tumor periphery was found to be impaired somewhat earlier than TBF, indicating that the cytokines do not preferentially act on the poorly perfused tumor center in the model chosen. This finding is inconsistent with previous histological studies on murine tumors. Acute flow changes encompassed only substantial reductions (i.e., hypoperfusion) rather than a complete ischemia. TBF was slightly increased during the first hour following rhLT whereas after rhTNF-alpha a continuous drop was observed. This differential response could not be observed during laser Doppler flowmetry. Tumor vascular resistance changes largely reflected alterations in TBF. 31P-Nuclear magnetic resonance spectroscopy on murine Meth-A fibrosarcomas revealed dose- and time-dependent decreases of ATP/Pi and phosphocreatine/Pi ratios following i.v. administration of rhTNF-alpha. From comparisons of dose-response curves rhLT appears to be more detrimental than rhTNF-alpha with respect to the bioenergetic status. The observed changes in tumor energy metabolism are similar to those described for TBF. It may therefore be concluded that most of the cytokine effects on the bioenergetic status are secondary to the inhibition of the microcirculatory function. As a major causative factor for the latter, an arterial hypotension can be excluded in the tumor model chosen.

Direct oxygen partial pressure (pO2) readings in cancers of the cervix and in the normal cervix of nulliparous or parous women were obtained using a computerized pO2 histography system. The oxygenation status of the tumors was evaluated as a function of clinical staging and histological grading. pO2 measurements were performed with a customized electrode system in conscious pre- and postmenopausal, untreated patients with well-defined arterial blood gas status. With this technique, pO2 measurements in the normal cervix of nulliparous women resulted in oxygenation patterns which were characteristic for normal, adequately supplied tissues (median pO2, 48 mm Hg) with approximately 1% of the pO2 values grouped between zero and 2.5 mm Hg, i.e., in a range with less than half-maximum radiosensitivity. As a rule, the mean (and median) pO2 values were distinctly lower in the normal cervix of parous women (most probably due to scar formation following vaginal delivery) and in malignancies. In the normal cervix of parous women the median pO2 value was 13 mm Hg (with approximately 14% of the pO2 readings in the lowest class), 14 mm Hg in International Federation of Gynecologists and Obstetricians I/II tumors (2% of the readings in the lowest pO2 class), and 11 mm Hg in International Federation of Gynecologists and Obstetricians III/IV cancers (1% of the pO2 data in the lowest class). To date, 5 of 18 cervical cancers exhibited pO2 values between zero and 2.5 mm Hg. The oxygenation pattern in cervical cancers and the occurrence of hypoxia and/or anoxia did not correlate with either the clinical stages and histological grades or with a series of clinically relevant parameters (e.g., tumor size). No significant differences were found between pre- and postmenopausal tumors, between squamous cell carcinomas and adenocarcinomas, and between endophytic or exophytic tumors. From these studies there is clear indication that the oxygenation status of individual tumors cannot be predicted on the basis of staging and/or grading, predominantly because of the pronounced tumor-to-tumor variabilities. Evaluation of the tissue oxygenation of individual tumors is thus mandatory to prove that tumor oxygenation can predict the overall prognosis and/or treatment outcome.

Direct oxygen partial pressure (pO2) readings in breast cancers, in fibrocystic disease, and in the normal breast have been obtained using a novel technique which allows for the systematic evaluation of the oxygenation status as a function of pathological staging and histological grading. Measurements were performed in awake pre- and postmenopausal patients with well-defined arterial blood gas status. The measuring procedure encompasses a computerized electrode movement in the tissue which avoids significant compression artifacts and allows routine measurement in human tumors before, during, and after treatment. Using this reliable technique, pO2 measurements in the normal breast and in fibrocystic disease resulted in oxygenation patterns which were characteristic for normal, adequately supplied tissues. The median pO2 values were 65 and 67 mm Hg, respectively, with no pO2 readings below 12.5 mm Hg in the normal breast, and less than or equal to 5 mm Hg in fibrocystic disease, respectively. In contrast, in breast cancers the median pO2 value was 30 mm Hg (pooled data for pathological stages T1-T4). To date, 6 of 15 breast cancers exhibited pO2 values between zero and 2.5 mm Hg, i.e., tissue areas with less than half-maximum radiosensitivity. The oxygenation pattern in breast cancers and the occurrence of hypoxia and/or anoxia did not correlate with either the pathological stages and histological grades or with a series of clinically relevant parameters. No significant differences were found between pre- and postmenopausal tumors and between lobular and ductal carcinomas. Tumor-to-tumor variability in the oxygenation pattern was more pronounced than intra-tumor heterogeneity. pO2 variations within a tumor cannot be predicted, e.g., as a function of the measuring site (tumor center versus periphery).

Experimental tumors growing in irradiated tissue have been used to study the biological differences characteristic of locally recurrent tumors. Since the hypoxic cell fraction of tumors growing in irradiated tissue is increased and growth rate is slowed, these tumors are assumed to be metabolically deprived with hypoperfusion. In this study, we directly measured the effect of tumor bed irradiation on blood flow, growth rate, rate of nucleoside triphosphate (NTP) turnover, and metabolic state using 31P and 2H nuclear magnetic resonance, and an intradermal assay for angiogenesis. (NTP turnover refers to ATP-synthetase mediated NTP turnover that is visible to 31P nuclear magnetic resonance using the technique of saturation transfer.) A decrease in the number of small blood vessels perfusing tumors in a preirradiated bed was found. Most of the decrease was due to a loss of vessels with diameters less than 0.04 mm. When tumors growing in preirradiated tissue reached approximately 100 mm3 in volume, a high frequency of gross and microscopic necrosis and hemorrhage was already observed in most tumors. Consistent with these observations, the phosphocreatine/inorganic phosphate and nucleoside triphosphate/inorganic phosphate ratios were significantly lower in the tumors growing in a preirradiated bed compared with tumors in a nonirradiated bed. The blood flow rate was similar to control for tumors less than 100 mm3 (45.8 versus 40.5 ml/100 g/min, P = not significant), but was significantly lower than control for tumors greater than 100 mm3 (40.4 versus 12.2 ml/100 g/min, P less than 0.01). The NTP turnover rates correlated (P less than 0.005, r = 0.66) with the volume doubling rate (1/tumor volume doubling time), but for tumors approximately 100 mm3 in size neither the volume doubling rate nor the NTP turnover rate of tumors growing in an irradiated bed was statistically lower than control [NTP turnover: 14 +/- 3%/s versus 9 +/- 2%/s; volume doubling rate: 0.47 +/- 0.07/day versus 0.33 +/- 0.04/day (mean +/- SE)]. A large intertumor variability of all metabolic parameters was observed.

The effects of the vasodilator hydralazine (HYD) on microcirculatory function and hyperthermic response were studied in early generation isotransplants of a spontaneous C3Hf/Sed mouse fibrosarcoma (FSall). Red blood cell flux (RBC flux) in superficial tumor regions was assessed using laser Doppler flowmetry. A differential microcirculatory response was seen between tumor and normal skin after 0.25 micrograms/g i.p. HYD, the tumor showing a transient increase in flow and the skin remaining almost stable. At 1.0 micrograms/g i.p., the differential response continued, this time with a transient fall in tumor blood flow but again no change in skin flow. High dose hydralazine (10.0 micrograms/g i.p.) was associated with a dramatic and prolonged decrease in tumor blood flow but a lesser and only transient decline in skin flow. Identical doses of hydralazine were given 30 min prior to heat treatment (43.5 degrees C for 15, 30, or 60 min). Tumor growth was measured daily and compared to controls (HT without hydralazine). Hydralazine at 0.25 micrograms/g i.p. did not affect heat induced growth delay. At 1.0 micrograms/g i.p., it significantly increased growth delay upon heat exposures of 15 min, but not after 30 or 60 min HT. Hydralazine at 10 micrograms/g i.p. increased growth delay for all heat doses (P less than 0.05). Hydralazine alone had no influence on growth delay of sham-heated tumors. The results obtained clearly indicate that tumor and normal tissues have microcirculatory differences in the time-course, degree and/or direction of response after hydralazine, and that hydralazine has potential for increasing the response of tumor to HT.

Glucose uptake, lactate release, ketone body utilization, spatial distribution of glucose, lactate, and ATP concentrations as well as tissue pH distributions were systematically investigated in s.c. and/or "tissue-isolated" human breast cancer xenografts in T-cell-deficient rnu/rnu rats. Large variations in all parameters were detected within and between tumors indicating a very nonuniform substrate turnover. Glucose was taken up by all xenografts. Glucose consumption rates increased with increasing glucose availabilities, implying that the glucose uptake is mainly determined by the efficiency of nutritive tumor blood flow. The average glucose uptake was 0.37 mumol/g/min in medullary and 0.26 mumol/g/min in squamous cell carcinomas of the breast. At wet weights below 5 g, medullary breast cancers consumed more glucose than squamous cell carcinomas (2P less than 0.05). Most tumors (97%) released lactate in an amount linearly related to glucose consumption. The lactate production of medullary (0.33 mumol/g/min) and squamous cell (0.31 mumol/g/min) breast cancers was similar. In general, the xenografts utilized ketone bodies. beta-Hydroxybutyrate was consumed by 82% and acetoacetate by 73% of the tumors, the uptake rates being linearly related to the respective availabilities. The mean uptake of beta-hydroxybutyrate was 3.48 nmol/g/min and that of acetoacetate 2.56 nmol/g/min. No significant differences were seen between medullary and squamous cell breast cancers. The beta-hydroxybutyrate/acetoacetate ratio in the tumor-venous blood rose with decreasing tumor blood flow indicating the development of hypoxia at advanced growth stages. Glucose, lactate, and ATP levels were all very heterogeneously distributed in medullary and squamous cell tumors as compared with normal tissue. No relationship was evident between the spatial distribution of concentrations of these three substrates. The xenografts were acidotic compared with pH values in normal subcutis. The mean tissue pH in medullary breast cancers was 6.81 +/- 0.25 (SD). Compared with these values, the tissue pH distribution in squamous cell breast cancers was shifted to significantly higher values. The mean pH of the latter tumors was 7.04 +/- 0.19 (2P less than 0.001). From the experimental data presented there is clear indication that the metabolism of the xenografts investigated was mainly determined by the efficiency of nutritive blood flow, i.e., by substrate availability, and not by the metabolic demand of the cancer cells.

Responses of tumor microcirculation (RBC flux) to i.p. glucose or mannitol injections were studied in early generation isotransplants of a spontaneous C3Hf/Sed mouse fibrosarcoma (FSaII). RBC flux in superficial tumor microregions was assessed using laser Doppler flowmetry. After administration of glucose or mannitol (a nonmetabolized sugar alcohol), a dose-dependent reduction in laser Doppler flow, and a dose-dependent increase in systemic hematocrit occurred concurrently. Maximum flow reductions induced by i.p. glucose or mannitol were statistically indistinguishable for equal osmotic load. Maximum decreases in RBC flux for glucose or mannitol were 20 and 25% (1.25 mg/g i.p.), 42 and 48% (2.5 mg/g i.p.), 72 and 60% (5 mg/g i.p.), and 80 and 75% (10 mg/g i.p.), respectively. Maximum increases in systemic hematocrit ranged from 18% (1.25 mg/g glucose i.p.) to 33% (10 mg/g glucose i.p.). Examination of RBC count, blood hemoglobin concentration, and fluid accumulation in the abdominal cavity after glucose or mannitol administration were all compatible with a significant shift of intravascular/extracellular water into the abdominal cavity with resultant systemic hypovolemic hemoconcentration. RBC volume and mean hemoglobin content of RBC remained unchanged with glucose loading. The data suggest that reductions in laser Doppler flow are predominantly caused by hypovolemic hemoconcentration following i.p. administration of hyperosmolar sugar solutions. Changes in laser Doppler flow due to specific glucose-mediated or glucose-related phenomena are probably of minor importance in the murine tumor system investigated. Future studies on murine tumors, examining for specific effects of glucose on metabolism and/or therapy, should not use i.p. administration of hyperosmolar solutions.

The objective of this review article is to summarize current knowledge of blood flow and perfusion-related parameters, which usually go hand in hand and in turn define the cellular metabolic microenvironment of human malignancies. A compilation of available data from the literature on blood flow, oxygen and nutrient supply, and tissue oxygen and pH distribution in human tumors is presented. Whenever possible, data obtained for human tumors are compared with the respective parameters in normal tissues, isotransplanted or spontaneous rodent tumors, and xenografted human tumors. Although data on human tumors in situ are scarce and there may be significant errors associated with the techniques used for measurements, experimental evidence is provided for the existence of a compromised and anisotropic blood supply to many tumors. As a result, O2-depleted areas develop in human malignancies which coincide with nutrient and energy deprivation and with a hostile metabolic microenvironment (e.g., existence of severe tissue acidosis). Significant variations in these relevant parameters must be expected between different locations within the same tumor, at the same location at different times, and between individual tumors of the same grading and staging. Furthermore, this synopsis will attempt to identify relevant pathophysiological parameters and other related areas future research of which might be most beneficial for designing individually tailored treatment protocols with the goal of predicting the acute and/or long-term response of tumors to therapy.

Human breast cancer xenografts in T-cell-deficient rnu/rnu rats permit the detailed and systematic study of blood flow, oxygen supply, and characterization of the cellular microenvironment of human tumors in vivo. Using an epigastric pouching technique, it is possible to obtain a tissue-isolated preparation which makes direct studies of blood flow and oxygen supply in human tumors feasible. So far, medullary and squamous cell carcinomas of the breast from patients have been investigated under well-defined systemic conditions. At comparable tumor sizes, the average blood flow rate through human breast cancer xenografts is higher in medullary than in squamous cell carcinomas (0.17 versus 0.10 ml X g-1 X min-1). Blood flow per unit tumor mass significantly decreases with increasing wet weight. No significant differences are obvious when comparing the flow values of pre- and postmenopausal tumors or of cancer tissues with different hormone receptor capacities. On the average, the oxygen consumption rates of human breast cancer xenografts are 10.4 in medullary and 7.7 microliter O2 X g-1 X min-1 in squamous cell carcinomas. With increasing tumor mass, the O2 consumption rate per unit weight significantly decreases. This decrease parallels the respective decline of tumor blood flow, implying that the O2 consumption rate of the cancer cells in vivo is mostly limited by the nutritive blood flow, i.e., by the O2 availability to the tumors. Due to a restricted blood supply, the O2 utilization of human breast cancer xenografts is high. Tissue oxygenation in microareas of human breast cancers xenotransplanted s.c. into nude rats is mostly inadequate. As a consequence, tissue hypoxia and anoxia are common findings even in very early growth stages. Due to marked intra- and intertumor variabilities in blood flow, heterogeneities in the tissue oxygenation are characteristic features of human breast cancer xenografts. From the results obtained it is concluded that human breast cancers growing as xenografts in rnu/rnu rats may be useful tools for cancer research, especially for investigations of blood flow, tissue oxygenation, and substrate turnover.

Severe disturbances in microcirculation during advanced phases of tumor growth lead to restrictions of convective and diffusive transport. In addition, an inhomogeneous distribution of transport conditions develops, resulting in insufficient and heterogeneous substrate supply and an inadequate drainage of wastes. Polarographic measurements of the local tissue oxygen tension (PO2) using gold microelectrodes reveal that very low PO2 values are prevalent in C3H mouse mammary carcinomas. The tissue PO2 frequency distributions are shifted to low PO2 values and limited in variability. The mean PO2 value is 7 mm Hg. The median is 4 mm Hg, the modal class being 0 to 5 mm Hg. Within different microareas of the same tumor, pronounced heterogeneities exist. Due to an elevated rate of lactic acid production and its subsequent inadequate removal, a severe tissue acidosis is evidenced in malignant tumors. For C3H mouse mammary carcinomas, most of the measured pH values are in the range of 6.4 to 7.1, the modal class being 6.7 to 6.8 (mean pH, 6.73; median pH, 6.75). Within different microareas of the same tumor, clear heterogeneities in the pH distribution do occur. Very low pH values (5.8 to 6.3) have been observed in large ulcerated tumors. In extensively necrotic areas, pH values even higher than the arterial pH could be detected.

Better understanding of the micromilieu of human tumors in situ is mandatory for further improvement of diagnostic and therapeutic interventions. Since investigations of untreated tumors of a wide size range are precluded in humans for ethical reasons, size-dependent changes in the pathophysiology of primary and metastatic human tumors were studied using "tissue-isolated" xenografts in nude rats. Tumor types included lung and breast cancers, ovarian and thyroid carcinomas, uterus tumors, and melanomas. A 10-fold variation in weight-adjusted tumor perfusion indicated large variations in angiogenesis which were unrelated to tumor type. Flow values obtained were consistent with data from clinical observations and were comparable to that in isografted rodent tumors. Using actual consumption and supply rates, maximum oxygen and glucose uptake rates were calculated for each tumor type. The capacity to consume oxygen and glucose varied 9-fold and 4-fold, respectively. However, considering actual consumption rates, blood flow was the principal modulator of substrate supply and tumor metabolism in these human tumor xenografts. Consequently, therapeutically relevant parameters of the metabolic micromilieu largely depended on the efficacy of the tumor circulation. Hereby, high metabolic rates concomitant with high flow values coincided with rapid tumor growth. Thus, in order to design the best individualized therapy, flow-related data should supplement histological classification and clinical staging and grading. Further development of relatively noninvasive technologies (magnetic resonance imaging, magnetic resonance spectroscopy, or positron emission tomography) might permit such monitoring.

The effects of i.p. versus i.v. glucose administration on laser Doppler flow (LDF) were studied in peripheral tissue areas of murine FSaII tumors implanted s.c. in the hind foot dorsum and in normal skin of conscious C3Hf/Sed mice. LDF was monitored prior to and continuously for 90 min following the administration of glucose, galactose, or mannitol at doses of 5 or 10 mg/g. Results showed that i.p. administration of hyperosmolar solutions was followed by a substantial, dose-dependent flow reduction which was indistinguishable for the various agents at equal osmotic load, and similar in tumor tissue and normal skin. Reductions in LDF are, therefore, primarily caused by hypovolemic hemoconcentration following i.p. administration of hyperosmolar sugar solutions. In contrast, i.v. administration of these solutions at 5 mg/g caused an initial flow increase (most probably due to a transient hypervolemic hemodilution), with a return to baseline readings within 5-10 min. At 10 mg/g i.v., a biphasic change in LDF occurred with an initial, temporary increase and a significant decline thereafter with no recovery within the observation period. This drop in LDF most probably is due to a decrease in cardiac output and an increase in viscous resistance to flow. Since comparable changes were observed with all agents and in both tissues investigated, it is concluded that the alterations in flow pattern following injection of hyperosmolar solutions are neither glucose nor tissue specific. Glucose- or tumor-specific effects, if present at all, must be of secondary importance in the animal model chosen.

Tumor acidosis and energy deprivation enhance thermal sensitivity. We have used in vivo 31P nuclear magnetic resonance spectroscopy to noninvasively monitor changes in pH and energy metabolism in FSaII mouse tumors after i.p. administration of glucose. A dose of 5 g/kg glucose induced a pH drop of 0.31 units without any statistically significant change in energy status. The pH changes resolved within 2 h. In contrast, administration of 10 g/kg glucose resulted in a severe acidosis (mean nadir pH of 6.19 corresponding to a mean pH drop of 0.96 units) and loss of energy, the latter most probably being due to an acidosis-induced inhibition of glycolysis during ischemic hypoxia. The resulting acidosis and energy loss were more persistent and resolved in 5.5-28 h. In contrast, after an identical dose of mannitol (10 g/kg), a pH drop of approximately only 0.1 units over 72 min was noted. The data suggest that both cleavage of glucose to lactic acid and blood flow inhibition are involved in glucose induced tumor acidosis. In vivo 31P nuclear magnetic resonance spectroscopy may be useful clinically to monitor therapeutic attempts at enhancing thermal sensitivity.