'Oxygen Level in a Tissue' - What Do Available Measurements Really Report?

The aim of the paper is to discuss what currently is feasible clinically to measure the level of oxygen and how that measurement can be clinically useful. Because oxygen in tissues is quite heterogeneous and all methods of measurement can only provide an average across heterogeneities at some spatial and temporal resolution, the values that are obtained may have limitations on their clinical utility. However, even if such limitations are significant, if one utilizes repeated measurements and focuses on changes in the measured levels, rather than 'absolute levels', it may be possible to obtain very useful clinical information. While these considerations are especially pertinent in cancer, they also pertain to most other types of pathology.

Clinical and Statistical Considerations when Assessing Oxygen Levels in Tumors: Illustrative Results from Clinical EPR Oximetry Studies.

The success of treatment for malignancies, especially those undergoing radiation therapy or chemotherapy, has long been recognized to depend on the degree of hypoxia in the tumor. In addition to the prognostic value of knowing the tumor's initial level of hypoxia, assessing the tumor oxygenation during standard therapy or oxygen-related treatments (such as breathing oxygen-enriched gas mixtures or taking drugs that can increase oxygen supply to tissues) can provide valuable data to improve the efficacy of treatments. A series of early clinical studies of tumors in humans are ongoing at Dartmouth and Emory using electron paramagnetic resonance (EPR) oximetry to assess tumor oxygenation, initially and over time during either natural disease progression or treatment. This approach has the potential for reaching the long-sought goal of enhancing the effectiveness of cancer therapy. In order to effectively reach this goal, we consider the validity of the practical and statistical assumptions when interpreting the measurements made in vivo for patients undergoing treatment for cancer.

Dynamic contrast-enhanced MRI in mice at high field: estimation of the arterial input function can be achieved by phase imaging.

PURPOSE: Quantitative dynamic contrast-enhanced MRI requires an accurate arterial input function (AIF). At high field, increased susceptibility effects and decreased longitudinal relaxivity of contrast agents lead to predominant T2* effects in blood vessels, producing a dip in signal during passage of the contrast agent bolus. This study determined phase-derived AIFs in mice at 11.7 T. METHODS: AIFs were measured in aorta/vena cava for five FBV/N mice and in iliac arteries/veins for five NMRI mice with a fast low angle shot sequence, simultaneously with tumor imaging (temporal resolution: 1.19 s). Gadoterate was injected into the tail vein as a bolus (0.286 mmol Gd/kg). An in vitro study was also performed to calculate the relationship between ΔΦ and gadolinium concentration. RESULTS: The phantom system confirmed the linear relationship between measured ΔΦ and gadolinium concentration. In vivo, a dip in arterial magnitude signal made it impossible to quantify the AIF. With phase imaging, a clear quantifiable bolus peak was obtained; peak measured concentration in plasma was 4.9 ± 0.9 mM for FBV/N mice and 8.0 ± 0.6 mM for NMRI mice, close to the expected concentration of 6.8 mM. CONCLUSION: Phase imaging seems to be an appropriate means to measure the AIF of mice at high field.

Dynamic contrast-enhanced MRI in mouse tumors at 11.7 T: comparison of three contrast agents with different molecular weights to assess the early effects of combretastatin A4.

Dynamic contrast-enhanced (DCE)-MRI is useful to assess the early effects of drugs acting on tumor vasculature, namely anti-angiogenic and vascular disrupting agents. Ultra-high-field MRI allows higher-resolution scanning for DCE-MRI while maintaining an adequate signal-to-noise ratio. However, increases in susceptibility effects, combined with decreases in longitudinal relaxivity of gadolinium-based contrast agents (GdCAs), make DCE-MRI more challenging at high field. The aim of this work was to explore the feasibility of using DCE-MRI at 11.7 T to assess the tumor hemodynamics of mice. Three GdCAs possessing different molecular weights (gadoterate: 560 Da, 0.29 mmol Gd/kg; p846: 3.5 kDa, 0.10 mmol Gd/kg; and p792: 6.47 kDa, 0.15 mmol Gd/kg) were compared to see the influence of the molecular weight in the highlight of the biologic effects induced by combretastatin A4 (CA4). Mice bearing transplantable liver tumor (TLT) hepatocarcinoma were divided into two groups (n = 5-6 per group and per GdCA): a treated group receiving 100 mg/kg CA4, and a control group receiving vehicle. The mice were imaged at 11.7 T with a T1 -weighted FLASH sequence 2 h after the treatment. Individual arterial input functions (AIFs) were computed using phase imaging. These AIFs were used in the Extended Tofts Model to determine K(trans) and vp values. A separate immunohistochemistry study was performed to assess the vascular perfusion and the vascular density. Phase imaging was used successfully to measure the AIF for the three GdCAs. In control groups, an inverse relationship between the molecular weight of the GdCA and K(trans) and vp values was observed. K(trans) was significantly decreased in the treated group compared with the control group for each GdCA. DCE-MRI at 11.7 T is feasible to assess tumor hemodynamics in mice. With K(trans) , the three GdCAs were able to track the early vascular effects induced by CA4 treatment.

Tumour hypoxia determines the potential of combining mTOR and autophagy inhibitors to treat mammary tumours.

BACKGROUND: Hypoxia can activate autophagy, a self-digest adaptive process that maintains cell turnover. Mammalian target of rapamycin (mTOR) inhibitors are used to treat cancer but also stimulate autophagy. METHODS: Human mammary cancer cells and derived xenografts were used to examine whether hypoxia could exacerbate autophagy-mediated resistance to the mTOR inhibitor rapamycin. RESULTS: Rapamycin exerted potent antitumour effects in MCF-7 and MDA-MB-231 mammary tumours through a marked inhibition of angiogenesis, but the autophagy inhibitor chloroquine (CQ) failed to further sensitise tumours to mTOR inhibition. Rapamycin treatment actually led to tumour reoxygenation, thereby preventing the development of autophagy. Chloroquine alone, however, blocked the growth of MCF-7 tumours and in vitro blunted the hypoxia-induced component of autophagy in these cells. Finally, when initiating CQ treatment in large, hypoxic tumours, a robust antitumour effect could be observed, which also further increased the antiproliferative effects of rapamycin. CONCLUSION: The mTOR inhibitor rapamycin significantly contributes to tumour growth inhibition and normalisation of the tumour vasculature through potent antiangiogenic effects. The resulting reduction in hypoxia accounts for a lack of sensitisation by the autophagy inhibitor CQ, except if the tumours are already at an advanced stage, and thus largely hypoxic at the initiation of the combination of rapamycin and CQ treatment.

Antiproliferative effects of phenylaminonaphthoquinones are increased by ascorbate and associated with the appearance of a senescent phenotype in human bladder cancer cells.

Quinone-containing molecules have been developed against cancer mainly for their redox cycling ability leading to reactive oxygen species (ROS) formation. We have previously shown that donor-acceptor phenylaminonaphthoquinones are biologically active against a panel of cancer cells. In this report, we explored the mechanisms involved in cancer cell growth inhibition caused by two phenylaminonaphthoquinones, namely Q7 and Q9, with or without ascorbate (ASC). The results show that Q7 and Q9 are both redox cyclers able to form ROS, which strongly inhibit the proliferation of T24 cells. Q9 was a better redox cycler than Q7 because of marked stabilization of the semiquinone radical species arising from its reduction by ascorbate. Indeed, ASC dramatically enhances the inhibitory effect of Q9 on cell proliferation. Q9 plus ASC impairs the cell cycle, causing a decrease in the number of cells in the G2/M phase without involving other cell cycle regulating key proteins. Moreover, Q9 plus ASC influences the MAPK signaling pathways, provoking the appearance of a senescent cancer cell phenotype and ultimately leading to necrotic-like cell death. Because cellular senescence limits the replicative capacity of cells, our results suggest that induction of senescence may be exploited as a basis for new approaches to cancer therapy.

Endotoxin-induced basal respiration alterations of renal HK-2 cells: a sign of pathologic metabolism down-regulation.

To study the mechanism of oxygen regulation in inflammation-induced acute kidney injury, we investigate the effects of a bacterial endotoxin (lipopolysaccharide, LPS) on the basal respiration of proximal tubular epithelial cells (HK-2) both by high-resolution respirometry and electron spin resonance spectroscopy. These two complementary methods have shown that HK-2 cells exhibit a decreased oxygen consumption rate when treated with LPS. Surprisingly, this cellular respiration alteration persists even after the stress factor was removed. We suggested that this irreversible decrease in renal oxygen consumption after LPS challenge is related to a pathologic metabolic down-regulation such as a lack of oxygen utilization by cells.

Multimodal assessment of early tumor response to chemotherapy: comparison between diffusion-weighted MRI, 1H-MR spectroscopy of choline and USPIO particles targeted at cell death.

The aim of this study was to determine the value of different magnetic resonance (MR) protocols to assess early tumor response to chemotherapy. We used a murine tumor model (TLT) presenting different degrees of response to three different cytotoxic agents. As shown in survival curves, cyclophosphamide (CP) was the most efficient drug followed by 5-fluorouracil (5-FU), whereas the etoposide treatment had little impact on TLT tumors. Three different MR protocols were used at 9.4 Tesla 24 h post-treatment: diffusion-weighted (DW)-MRI, choline measurement by (1) H MRS, and contrast-enhanced MRI using ultrasmall iron oxide nanoparticles (USPIO) targeted at phosphatidylserine. Accumulation of contrast agent in apoptotic tumors was monitored by T(2) -weighted images and quantified by EPR spectroscopy. Necrosis and apoptosis were assessed by histology. Large variations were observed in the measurement of choline peak areas and could not be directly correlated to tumor response. Although the targeted USPIO particles were able to significantly differentiate between the efficiency of each cytotoxic agent and best correlated with survival endpoint, they present the main disadvantage of non-specific tumor accumulation, which could be problematic when transferring the method to the clinic. DW-MRI presents a better compromise by combining longitudinal studies with a high dynamic range; however, DW-MRI was unable to show any significant effect for 5-FU. This study illustrates the need for multimodal imaging in assessing tumor response to treatment to compensate for individual limitations.

Noninvasive mapping of spontaneous fluctuations in tumor oxygenation using 19F MRI.

PURPOSE: Acute hypoxia (transient cycles of hypoxia-reoxygenation) is known to occur in solid tumors and may be a poorly appreciated therapeutic problem as it can be associated with resistance to radiation therapy, impaired delivery of chemotherapeutic agents, or metastasis development. The objective of the present study was to use MR 19F relaxometry maps to analyze the spontaneous fluctuations of partial pressure of oxygen (pO2) over time in experimental tumors. METHODS: The pO2 maps were generated after direct intratumoral administration of a fluorine compound (hexafluorobenzene) whose relaxation rate (1/T1) is proportional to the % O2. The authors used a SNAP inversion-recovery sequence at 4.7 T to acquire parametric images of the T1 relaxation time with a high spatial and temporal resolution. Homemade routines were developed to perform regions of interest analysis, as well as pixel by pixel analysis of pO2 over time. RESULTS: The authors were able to quantify and probe the heterogeneity of spontaneous fluctuations in tumor pO2: (i) Spontaneous fluctuations in pO2 occurred regardless of the basal oxygenation state (i.e., both in oxygenated and in hypoxic regions) and (ii) spontaneous fluctuations occurred at a rate of 1 cycle/12-47 min. For validation, the analysis was performed in dead mice for which acute changes did not occur. The authors thereby demonstrated that 19F MRI technique is sensitive to acute change in pO2 in tumors. CONCLUSIONS: This is the first approach that allows quantitative minimally invasive measurement of the spontaneous fluctuations of tumor oxygenation using a look-locker approach (e.g., SNAP IR). This approach could be an important tool to characterize the phenomenon of tumor acute hypoxia, to understand its physiopathology, and to improve therapies.

Experimental determination of the radial dose distribution in high gradient regions around 192Ir wires: comparison of electron paramagnetic resonance imaging, films, and Monte Carlo simulations.

PURPOSE: The experimental determination of doses at proximal distances from radioactive sources is difficult because of the steepness of the dose gradient. The goal of this study was to determine the relative radial dose distribution for a low dose rate 192Ir wire source using electron paramagnetic resonance imaging (EPRI) and to compare the results to those obtained using Gafchromic EBT film dosimetry and Monte Carlo (MC) simulations. METHODS: Lithium formate and ammonium formate were chosen as the EPR dosimetric materials and were used to form cylindrical phantoms. The dose distribution of the stable radiation-induced free radicals in the lithium formate and ammonium formate phantoms was assessed by EPRI. EBT films were also inserted inside in ammonium formate phantoms for comparison. MC simulation was performed using the MCNP4C2 software code. RESULTS: The radical signal in irradiated ammonium formate is contained in a single narrow EPR line, with an EPR peak-to-peak linewidth narrower than that of lithium formate (approximately 0.64 and 1.4 mT, respectively). The spatial resolution of EPR images was enhanced by a factor of 2.3 using ammonium formate compared to lithium formate because its linewidth is about 0.75 mT narrower than that of lithium formate. The EPRI results were consistent to within 1% with those of Gafchromic EBT films and MC simulations at distances from 1.0 to 2.9 mm. The radial dose values obtained by EPRI were about 4% lower at distances from 2.9 to 4.0 mm than those determined by MC simulation and EBT film dosimetry. CONCLUSIONS: Ammonium formate is a suitable material under certain conditions for use in brachytherapy dosimetry using EPRI. In this study, the authors demonstrated that the EPRI technique allows the estimation of the relative radial dose distribution at short distances for a 192Ir wire source.

Radioimmunotherapy with radioactive nanoparticles: biological doses and treatment efficiency for vascularized tumors with or without a central hypoxic area.

PURPOSE: Radioactive atoms attached to monoclonal antibodies are used in radioimmunotherapy to treat cancer while limiting radiation to healthy tissues. One limitation of this method is that only one radioactive atom is linked to each antibody and the deposited dose is often insufficient to eradicate solid and radioresistant tumors. In a previous study, simulations with the Monte Carlo N-Particle eXtended code showed that physical doses up to 50 Gy can be delivered inside tumors by replacing the single radionuclide by a radioactive nanoparticle of 5 nm diameter containing hundreds of radioactive atoms. However, tumoral and normal tissues are not equally sensitive to radiation, and previous works did not take account the biological effects such as cellular repair processes or the presence of less radiosensitive cells such as hypoxic cells. METHODS: The idea is to adapt the linear-quadratic expression to the tumor model and to determine biological effective doses (BEDs) delivered through and around a tumor. This BED is then incorporated into a Poisson formula to determine the shell control probability (SCP) which predicts the cell cluster-killing efficiency at different distances "r" from the center of the tumor. BED and SCP models are used to analyze the advantages of injecting radioactive nanoparticles instead of a single radionuclide per vector in radioimmunotherapy. RESULTS: Calculations of BED and SCP for different distances r from the center of a solid tumor, using the non-small-cell lung cancer as an example, were investigated for 90Y2O3 nanoparticles. With a total activity of about 3.5 and 20 MBq for tumor radii of 0.5 and 1.0 cm, respectively, results show that a very high BED is deposited in the well oxygenated part of the spherical carcinoma. CONCLUSIONS: For either small or large solid tumors, BED and SCP calculations highlight the important benefit in replacing the single beta-emitter 90Y attached to each antibody by a 90Y2O3 nanoparticle.

Electron Paramagnetic Resonance Imaging of Melanin in Honey Bee.

Honey bees play a crucial role in the nature by pollinating wild flowers. Over the past years, there has been an increasing concern regarding the honey bee colony decline. Pesticides or environmental effects targeting the biochemistry of insect chitin and cuticle coating may be in part responsible for honey bee pathologies. We here propose the use of electron paramagnetic resonance imaging (EPRI) as a tool to image the melanin-chitin complexes as part of the exoskeleton of the honey bee. EPRI at 9.65 GHz was applied on intact freeze-dried bees. The imaging data were collected on the melanin peak. High-resolution images revealed that this compound is extensively distributed in the periphery of the animal, data consistent with the localization in the cuticle of the bee. While EPR of melanin has been so far explored in the context of melanoma characterization, it may offer new opportunities in research on honey bees and other insects.

Acidosis-induced metabolic reprogramming in tumor cells enhances the anti-proliferative activity of the PDK inhibitor dichloroacetate.

Altered metabolic pathways in cancer such as exacerbated glycolytic flux and increased glutamine metabolism are promising targets for anti-cancer therapies. While commonly observed in glycolytic tumors, extracellular acidosis has never been considered as a potential modulator of anti-metabolic drug activity such as dichloroacetate (DCA). Using cancer cells from various origins selected for their ability to proliferate under acidic conditions, we found that DCA exerts greater inhibitory effects on the growth of these acid-adapted cells than in parental cells. Moreover, daily DCA administration to mice led to a significant decrease in tumor growth from acid-adapted cells but not from parental cells. 13C-tracer studies revealed that DCA induced a double metabolic shift, diminishing glycolysis and increasing intracellular glutamine in acid-adapted cells. As a consequence, DCA reduced the pentose phosphate pathway activity more extensively and increased apoptosis in acid-adapted cells. Finally, the combination of DCA with a glutaminase inhibitor significantly enhanced the cytotoxic effects of DCA. Overall, the interplay between acidosis and DCA exposure leads to metabolic reprogramming that considerably alters cellular fitness.

Retrievable micro-inserts containing oxygen sensors for monitoring tissue oxygenation using EPR oximetry.

Tissue oxygenation is a crucial parameter in various physiopathological situations and can influence the therapeutic response of tumours. EPR oximetry is a reliable method for assessing and monitoring oxygen levels in vivo over long periods of time. Among the different paramagnetic oxygen sensors available for EPR oximetry, lithium phthalocyanine (LiPc) is a serious candidate for in vivo applications because of its narrow linewidth and its high signal-to-noise ratio. To enhance the biocompatibility of the sensors, fluoropolymer Teflon AF2400 was used to make cylindrical micro-inserts containing LiPc crystals. This new micro-pellet design has several advantages for in vivo studies, including the possibility of being able to choose the implant size, a high sensor content, the facility of in vivo insertion and complete protection with preservation of the oxygen sensor's characteristics. The response to oxygen and the kinetics of this response were tested using in vivo EPR: no differences were observed between micro-inserts and uncoated LiPc crystals. Pellets implanted in vivo in muscles conserved their responsiveness over a long period of time (approximately two months), which is much longer than the few days of stability observed using LiPc crystals without protection by the implant. Finally, evaluation of the biocompatibility of the implants revealed no inflammatory reaction around the implantation area.

Characterization of self-assembling copolymers in aqueous solutions using Electron Paramagnetic Resonance and Fluorescence spectroscopy.

Electron Paramagnetic Resonance and fluorescence spectroscopy have been used to determine the micropolarity and microviscosity of self-assembling systems based on mmePEG-p(CL-co-TMC) having different PEG chain lengths and different CL/TMC ratios and PEG/MOG/SA (45/5/50) polymers with different PEG chain lengths. Four reporter probes have been used: two spin probes, 16-doxyl stearic acid and 5-doxylstearic acid, and two fluorescent probes, pyrene and 1,3-bis(1-pyrenyl) propane (P3P). We found that the micelles based on mmePEG-p(CL-co-TMC) polymers are of a biphasic nature. The micelles are made of a hydrophilic corona with low viscosity while the core of the micelle is more hydrophobic and more viscous. The outer shell is made up of PEG chains, the hydrophobic part of the chains making the core. The partial hydration of the shell seems to lead to a looser chain network than that associated with deeper domains in the micelles. By contrast, in micelles composed of PEG/MOG/SA, there is no clear domain separation. This is consistent with a spatial configuration of random polymeric chains forming a loose network. In these micelles, the microviscosity is low and the hydrophobicity is high.

Radioimmunotherapy with radioactive nanoparticles: first results of dosimetry for vascularized and necrosed solid tumors.

Radioimmunotherapy uses monoclonal antibodies that are still labeled with only one radioactive atom. The aim of this paper is to assess, by means of MCNPX simulations, the doses delivered around and throughout a solid tumor when the radioactive atom linked to each antibody is replaced by a 5 nm diameter nanoparticle composed of numerous radionuclides. A new model for a spherical vascularized tumor has been developed in which the antibody distributions inside the tumor can be uniform or heterogeneous. It is also possible to simulate a central necrotic core inside the tumor where the concentration of radiolabeled antibodies is assumed to be zero. Dosimetry calculations have been performed for the beta-emitting radionuclide (90)Y2O3. Preliminary results show that the irregularity of vasculature and the presence of a necrotic core have a noticeable influence on the deposited dose profiles. Moreover, with a total activity of 5 and 34 MBq for tumor radii of 0.5 and 1.0 cm, respectively, viable tumor cells can receive doses of up to 50 Gy, even if high nonuniformity of the total activity is observed in the tumor. These simulations still require accurate information about antibody characteristics and necrosis sizes but clearly confirm that the use of monoclonal antibodies conjugated to nanoparticles could lead to a considerable enhancement of treatment efficacy against cancer.

[Development of new methods of magnetic resonance for the characterization of oxygenation and perfusion of tumors. Diagnostic and therapeutic applications]. / Développement de nouvelles méthodes de résonance magnétique en vue de la caractérisation de l'oxygénation et de la perfusion de tumeurs. Applications diagnostiques et thérapeutiques.

Tumor hypoxia is a key factor of the tumor growth and response to therapy. In this short note, we describe our recent contribution to the development of new non-invasive methods for measuring tumor oxygenation and perfusion by magnetic resonance techniques (NMR and EPR). After validation, these techniques have been applied to study the influence of physiological and pharmacological factors on tumor hemodynamics. The non-invasive study of the changes in the tumor microenvironment allowed the elaboration of rational schedules to potentiate the anti-cancer therapy when combining treatments.

Injectable nanomedicine hydrogel for local chemotherapy of glioblastoma after surgical resection.

Glioblastoma (GBM) treatment includes, when possible, surgical resection of the tumor followed by radiotherapy and oral chemotherapy with temozolomide, however recurrences quickly develop around the resection cavity borders leading to patient death. We hypothesize that the local delivery of Lauroyl-gemcitabine lipid nanocapsule based hydrogel (GemC12-LNC) in the tumor resection cavity of GBM is a promising strategy as it would allow to bypass the blood brain barrier, thus reaching high local concentrations of the drug. The cytotoxicity and internalization pathways of GemC12-LNC were studied on different GBM cell lines (U251, T98-G, 9L-LacZ, U-87 MG). The GemC12-LNC hydrogel was well tolerated when injected in mouse brain. In an orthotopic xenograft model, after intratumoral administration, GemC12-LNC significantly increased mice survival compared to the controls. Moreover, its ability to delay tumor recurrences was demonstrated after perisurgical administration in the GBM resection cavity. In conclusion, we demonstrate that GemC12-LNC hydrogel could be considered as a promising tool for the post-resection management of GBM, prior to the standard of care chemo-radiation.

Development and evaluation of biocompatible films of polytetrafluoroethylene polymers holding lithium phthalocyanine crystals for their use in EPR oximetry.

Electron paramagnetic resonance (EPR) oximetry is a powerful technology that allows the monitoring of oxygenation in tissues. The measurement of tissue oxygenation can be achieved using lithium phthalocyanine (LiPc) crystals as oxygen reporters. In order to have biocompatibility for the sensing system and to assure long-term stability in the responsiveness of the system, we developed films of Teflon AF 2400 with embedded LiPc crystals. These systems can be used as retrievable inserts or parts of an implantable resonator or catheter. Atomic force microscopy studies revealed that the surface of the films was regular and planar. The response to oxygen of the sensor (EPR linewidth as a function of pO(2)) remained unchanged after implantation in mice, and was not affected by sterilization or irradiation. The use of resonators, holding LiPc embedded in Teflon AF 2400, implanted in the gastrocnemius muscle of rabbits allowed the monitoring of oxygen during several weeks. Several assays also demonstrated the biocompatibility of the system: (1) no hemolytic effect was noted; (2) no toxicity was found using the systemic injection test of extracts; (3) histological analysis in rabbit muscle in which the films were implanted for 1 week or 3 months was similar to standard polyethylene biocompatible devices. These advanced oxygen sensors are promising tools for future pre-clinical and clinical developments of EPR oximetry. These developments can be applied for other applications of biosensors where there is a need for oxygen permeable membranes.

The relationship between partial pressure of oxygen and perfusion in two murine tumors after X-ray irradiation: a combined gadopentetate dimeglumine dynamic magnetic resonance imaging and in vivo electron paramagnetic resonance oximetry study.

Changes of partial pressure of oxygen (pO2) and blood perfusion were studied in MTG-B and RIF-1 tumors (n = 5 each) before and after a single 20-Gy dose of X-ray irradiation. Using electron paramagnetic resonance oximetry, we have observed an initial fast decrease of pO2 after irradiation, followed by a slow increase. The time course of these changes was faster in the MTG-B tumors than in the RIF-1 tumors. Gadopentetate dimeglumine (Gd-DTPA) dynamic magnetic resonance imaging studies showed a reduction in uptake of Gd-DTPA at the time of minimum pO2 and a recovery at the time of maximum pO2 in each tumor. Previous work indicates that there is microscopic heterogeneity in tumors, with well-vascularized "capillary regions" being closer to capillaries than poorly vascularized "noncapillary regions." We propose a two-component (slow and fast) model of Gd-DTPA uptake that is designed to quantify the kinetics of these two compartments by analyzing the total tumor uptake kinetics without having to identify specific regions of interest. Total perfusion in the tumors was greatly reduced at the time of minimum oxygenation, and the volume of the slow component increased after irradiation. We conclude that a decrease in blood perfusion is one of the main causes of the decline in pO2 observed after irradiation.