Glucocorticoids modulate tumor radiation response through a decrease in tumor oxygen consumption.

PURPOSE: We hypothesized that glucocorticoids may enhance tumor radiosensitivity by increasing tumor oxygenation (pO(2)) through inhibition of mitochondrial respiration. EXPERIMENTAL DESIGN: The effect of three glucocorticoids (hydrocortisone, dexamethasone, and prednisolone) on pO(2) was studied in murine TLT liver tumors and FSaII fibrosarcomas. At the time of maximum pO(2) (t(max), 30 min after administration), perfusion, oxygen consumption, and radiation sensitivity were studied. Local pO(2) measurements were done using electron paramagnetic resonance. The oxygen consumption rate of tumor cells after in vivo glucocorticoid administration was measured using high-frequency electron paramagnetic resonance. Tumor perfusion and permeability measurements were assessed by dynamic contrast-enhanced magnetic resonance imaging. RESULTS: All glucocorticoids tested caused a rapid increase in pO(2). At t(max), tumor perfusion decreased, indicating that the increase in pO(2) was not caused by an increase in oxygen supply. Also at t(max), global oxygen consumption decreased. When irradiation (25 Gy) was applied at t(max), the tumor radiosensitivity was enhanced (regrowth delay increased by a factor of 1.7). CONCLUSION: These results show the potential usefulness of the administration of glucocorticoids before irradiation.

Botulinum toxin potentiates cancer radiotherapy and chemotherapy.

PURPOSE: Structural and functional abnormalities in the tumor vascular network are considered factors of resistance of solid tumors to cytotoxic treatments. To increase the efficacy of anticancer treatments, efforts must be made to find new strategies for transiently opening the tumor vascular bed to alleviate tumor hypoxia (source of resistance to radiotherapy) and improve the delivery of chemotherapeutic agents. We hypothesized that Botulinum neurotoxin type A (BoNT-A) could interfere with neurotransmitter release at the perivascular sympathetic varicosities, leading to inhibition of the neurogenic contractions of tumor vessels and therefore improving tumor perfusion and oxygenation. EXPERIMENTAL DESIGN: To test this hypothesis, BoNT-A was injected locally into mouse tumors (fibrosarcoma FSaII, hepatocarcinoma transplantable liver tumor), and electron paramagnetic resonance oximetry was used to monitor pO(2) in vivo repeatedly for 4 days. Additionally, contrast-enhanced magnetic resonance imaging was used to measure tumor perfusion in vivo. Finally, isolated arteries were mounted in wire myograph to monitor specifically the neurogenic tone developed by arterioles that were co-opted by the surrounding growing tumor cells. RESULTS: Using these tumor models, we showed that local administration of BoNT-A (two sites; dose, 29 units/kg) substantially increases tumor oxygenation and perfusion, leading to a substantial improvement in the tumor response to radiotherapy (20 Gy of 250-kV radiation) and chemotherapy (cyclophosphamide, 50 mg/kg). This observed therapeutic gain results from an opening of the tumor vascular bed by BoNT-A because we showed that BoNT-A could inhibit neurogenic tone in the tumor vasculature. CONCLUSIONS: The opening of the vascular bed induced by BoNT-A offers a way to significantly increase the response of tumors to radiotherapy and chemotherapy.

Tumor radiosensitization by antiinflammatory drugs: evidence for a new mechanism involving the oxygen effect.

We hypothesized that nonsteroidal antiinflammatory drugs (NSAIDs) might enhance tumor radiosensitivity by increasing tumor oxygenation (pO2), via either a decrease in the recruitment of macrophages or from inhibition of mitochondrial respiration. The effect of four NSAIDs (diclofenac, indomethacin, piroxicam, and NS-398) on pO2 was studied in murine TLT liver tumors and FSaII fibrosarcomas. At the time of maximum pO2 (t(max), 30 minutes after administration), perfusion, oxygen consumption, and radiation sensitivity were studied. Local pO2 measurements were done using electron paramagnetic resonance. Tumor perfusion and permeability measurements were assessed by dynamic contrast-enhanced magnetic resonance imaging. The oxygen consumption rate of tumor cells after in vivo NSAID administration was measured using high-frequency electron paramagnetic resonance. Tumor-infiltrating macrophage localization was done with immunohistochemistry using CD11b antibody. All the NSAIDs tested caused a rapid increase in pO2. At t(max), tumor perfusion decreased, indicating that the increase in pO2 was not caused by an increase in oxygen supply. Also at t(max), global oxygen consumption decreased but the amount of tumor-infiltrating macrophages remained unchanged. Our study strongly indicates that the oxygen effect caused by NSAIDs is primarily mediated by an effect on mitochondrial respiration. When irradiation (18 Gy) was applied at t(max), the tumor radiosensitivity was enhanced (regrowth delay increased by a factor of 1.7). These results show the potential utility of an acute administration of NSAIDs for radiosensitizing tumors, and shed new light on the mechanisms of NSAID radiosensitization. These results also provide a new rationale for the treatment schedule when combining NSAIDs and radiotherapy.

Preclinical safety and antitumor efficacy of insulin combined with irradiation.

BACKGROUND AND PURPOSE: We have previously reported that insulin significantly enhances tumor oxygenation (pO(2)) and increases radiation-induced tumor regrowth delay in experimental models. Considering the large radiosensitizing effect, clinical trials might be envisioned. The aim of the present pre-clinical study was to obtain a more complete set of safety and efficacy data which would further justify the commencement of such clinical trials. MATERIAL AND METHODS: Toxicity on normal (early and late-responding) tissues was measured by the intestinal crypt regeneration assay and the late leg contracture assay. Efficacy in terms of enhancement of pO(2) (measured by in vivo EPR oximetry) and increase in radiation-induced tumor regrowth delay was evaluated with a dose-response study on mice bearing FSaII fibrosarcoma. RESULTS: The effect on regrowth delay was directly correlated with the effect on the tumor pO(2), with a maximal effect using 400 mU kg(-1) insulin. Importantly, there was no increase in the radiation toxicity for normal tissues. Finally, we found that the hypoglycaemia induced by insulin can be corrected by simultaneous glucose infusion without modification of efficacy. CONCLUSION: Insulin here demonstrated a therapeutic gain and a lack of toxicity to normal tissues. The results of this study fully justify further larger preclinical assays such as the use of fractionated irradiation and a tumor control dose assay, before determining the utility of insulin as a radiosensitizer for human patients in the clinic.

Thalidomide radiosensitizes tumors through early changes in the tumor microenvironment.

PURPOSE: The aim of this work was to study changes in the tumor microenvironment early after an antiangiogenic treatment using thalidomide (a promising angiogenesis inhibitor in a variety of cancers), with special focus on a possible "normalization" of the tumor vasculature that could be exploited to improve radiotherapy. EXPERIMENTAL DESIGN: Tumor oxygenation, perfusion, permeability, interstitial fluid pressure (IFP), and radiation sensitivity were studied in an FSAII tumor model. Mice were treated by daily i.p. injection of thalidomide at a dose of 200 mg/kg. Measurements of the partial pressure of oxygen (pO(2)) were carried out using electron paramagnetic resonance oximetry. Three complementary techniques were used to assess the blood flow inside the tumor: dynamic contrast-enhanced magnetic resonance imaging, Patent Blue staining, and laser Doppler imaging. IFP was measured by a "wick-in-needle" technique. RESULTS: Our results show that thalidomide induces tumor reoxygenation within 2 days. This reoxygenation is correlated with a reduction in IFP and an increase in perfusion. These changes can be attributed to extensive vascular remodeling that we observed using CD31 labeling. CONCLUSIONS: In summary, the microenvironmental changes induced by thalidomide were sufficient to radiosensitize tumors. The fact that thalidomide radiosensitization was not observed in vitro, and that in vivo radiosensitization occurred in a narrow time window, lead us to believe that initial vascular normalization by thalidomide accounts for tumor radiosensitization.

Changes in the tumor microenvironment during low-dose-rate permanent seed implantation iodine-125 brachytherapy.

PURPOSE: There is a lack of data regarding how the tumor microenvironment (e.g., perfusion and oxygen partial pressure [pO2]) changes in response to low-dose-rate (LDR) brachytherapy. This may be why some clinical issues remain unresolved, such as the appropriate use of adjuvant external beam radiation therapy (EBRT). The purpose of this work was to obtain some basic preclinical data on how the tumor microenvironment evolves in response to LDR brachytherapy. METHODS AND MATERIALS: In an experimental mouse tumor, pO2 (measured by electron paramagnetic resonance) and perfusion (measured by dynamic contrast-enhanced magnetic resonance imaging) were monitored as a function of time (0-6 days) and distance (0-2 mm and 2-4 mm) from an implanted 0.5 mCi iodine-125 brachytherapy seed. RESULTS: For most of the experiments, including controls, tumors remained hypoxic at all times. At distances of 2-4 mm from radioactive seeds ( approximately 1.5 Gy/day), however, there was an early, significant increase in pO2 within 24 h. The pO2 in that region remained elevated through Day 3. Additionally, the perfusion in that region was significantly higher than for controls starting at Day 3. CONCLUSION: It may be advantageous to give adjuvant EBRT shortly (approximately 1 to 2 days) after commencement of clinical LDR brachytherapy, when the pO2 in the spatial regions between seeds should be elevated. If chemotherapy is given adjuvantly, it may best be administered just a little later (approximately 3 or 4 days) after the start of LDR brachytherapy, when perfusion should be elevated.

Early reoxygenation in tumors after irradiation: determining factors and consequences for radiotherapy regimens using daily multiple fractions.

PURPOSE: To characterize changes in the tumor microenvironment early after irradiation and determine the factors responsible for early reoxygenation. METHODS AND MATERIALS: Fibrosarcoma type II (FSaII) and hepatocarcinoma transplantable liver tumor tumor oxygenation were determined using electron paramagnetic resonance oximetry and a fiberoptic device. Perfusion was assessed by laser Doppler, dynamic contrast-enhanced MRI, and dye penetration. Oxygen consumption was determined by electron paramagnetic resonance. The interstitial fluid pressure was evaluated by the wick-in-needle technique. RESULTS: An increase in oxygen partial pressure was observed 3-4 h after irradiation. This increase resulted from a decrease in global oxygen consumption and an increase in oxygen delivery. The increase in oxygen delivery was due to radiation-induced acute inflammation (that was partially inhibited by the antiinflammatory agent diclofenac) and to a decrease in interstitial fluid pressure. The endothelial nitric oxide synthase pathway, identified as a contributing factor at 24 h after irradiation, did not play a role in the early stage after irradiation. We also observed that splitting a treatment of 18 Gy into two fractions separated by 4 h (time of maximal reoxygenation) had a greater effect on tumor regrowth delay than when applied as a single dose. CONCLUSION: Although the cell cycle redistribution effect is important for treatment protocols using multiple daily radiation fractions, the results of this work emphasize that the oxygen effect must be also considered to optimize the treatment strategy.

Enhancement of quinone redox cycling by ascorbate induces a caspase-3 independent cell death in human leukaemia cells. An in vitro comparative study.

Since the higher redox potential of quinone molecules has been correlated with enhanced cellular deleterious effects, we studied the ability of the association of ascorbate with several quinones derivatives (having different redox potentials) to cause cell death in K562 human leukaemia cell line. The rationale is that the reduction of quinone by ascorbate should be dependent of the quinone half-redox potential thus determining if reactive oxygen species (ROS) are formed or not, leading ultimately to cell death or cell survival. Among different ROS that may be formed during redox cycling between ascorbate and the quinone, the use of different antioxidant compounds (mannitol, desferal, N-acetylcysteine, catalase and superoxide dismutase) led to support H2O2 as the main oxidizing agent. We observed that standard redox potentials, oxygen uptake, free ascorbyl radical formation and cell survival were linked. The oxidative stress induced by the mixture of ascorbate and the different quinones decreases cellular contents of ATP and GSH while caspase-3-like activity remains unchanged. Again, we observed that quinones having higher values of half-redox potential provoke a severe depletion of ATP and GSH when they were associated with ascorbate. Such a drop in ATP content may explain the lack of activation of caspase-3. In conclusion, our results indicate that the cytotoxicity of the association quinone/ascorbate on K562 cancer cells may be predicted on the basis of half-redox potentials of quinones.

Nitric oxide-mediated increase in tumor blood flow and oxygenation of tumors implanted in muscles stimulated by electric pulses.

PURPOSE: Oxygen deficiency in tumors reduces the efficacy of nonsurgical treatment modalities. We tested the hypothesis that electrical stimulation of the sciatic nerve could modify the oxygenation status and the blood flow of tumors implanted in the thigh of mice. MATERIALS AND METHODS: The sciatic nerve was electrically stimulated at 5 Hz. Local transplantable liver tumor (TLT) and fibrosarcoma (FSaII) tumor oxygen pressure (pO(2)) and perfusion measurements were carried out using electron paramagnetic resonance (EPR) oximetry and the OxyLite/OxyFlo technique. The radiosensitizing effect of the protocol was assessed by irradiating FSaII tumors with X-rays. RESULTS: Tumor pO(2) increased from approximately 3 mm Hg to approximately 8 mm Hg, and relative tumor blood flow was increased by 241% and 162% for TLT and FSaII tumor models, respectively. The effect on the tumor oxygenation was inhibited by a nitric oxide synthase (NOS) inhibitor, and an increase in the tumor nitric oxide (NO) content was observed using EPR spin-trapping. The tumor oxygen consumption rate was decreased after the stimulation protocol. In addition, the electrical stimulation of the host tissue increased regrowth delays by a factor of 1.65. CONCLUSIONS: This increase in tumor oxygenation is due to the temporary increase in tumor blood flow, but particularly to a decrease in the tumor oxygen consumption rate (inhibition of respiration) that is mediated by a local production of NO during the protocol. Those tumor hemodynamic changes resulted in a radiosensitizing effect.

Noninvasive in vivo EPR monitoring of the methyl methacrylate polymerization during the bone cement formation.

The curing of poly(methyl methacrylate) (PMMA) bone cement is done by a free radical polymerization. As the amount of free radicals present is a marker of the amount of unpolymerized chains present in the polymer, it is assumed that this could be related to the mechanical properties such as strength or density. In this study, the direct observation of the free radicals produced during the PMMA bone cement formation was obtained for the first time in vivo using low-frequency EPR spectrometers (1.2 GHz). Low frequency permits measurements in live animals due to the increased microwave penetration. The amount of polymerization radicals was carried out noninvasively over days on the same animals. The decay rates obtained in vitro and in vivo were compared: the decay rates were significantly lower when the curing process occurred in vivo compared to the situation in vitro. As the kinetics are rather different in vitro and in vivo, this emphasizes the value of the present method that permits the noninvasive monitoring of the curing process directly in vivo.

Assessment of liver phagocytic activity using EPR spectrometry and imaging.

The aim of the present study was to evaluate the usefulness of electron paramagnetic resonance (EPR) spectroscopy and imaging in assessing the phagocytic activity of the liver after administration of India ink. We conducted experiments on livers from control rodents and from rodents in which the Kupffer cell population had been depleted by pretreatment with gadolinium chloride. The EPR signal intensity recorded in liver homogenates was about two times lower in GdCl(3) treated rats than in control rats. EPR imaging carried out on precision-cut liver slices indicated a good correlation between the depletion of Kupffer cells and the EPR signal intensity.

In vivo detection of inflammation using pegylated iron oxide particles targeted at E-selectin: a multimodal approach using MR imaging and EPR spectroscopy.

OBJECTIVES: Ultrasmall particles of iron oxide (USPIO) possess superparamagnetic properties and are used as negative contrast agent in magnetic resonance imaging (MRI) because of their strong T(2) and T(2)* effects. Besides this method, electron paramagnetic resonance (EPR) offers the unique capability to quantify these particles. The objective of this study was to evaluate a molecular marker for non invasive diagnosis and monitoring of inflammation. During inflammation cell adhesion molecules such as E-selectin are expressed on the endothelial cell surface. An E-selectin ligand was coupled to pegylated USPIO particles. MATERIALS AND METHODS: Inflammation was induced by intramuscular injection of Freund's Complete Adjuvant in male NMRI mice. After intravenous injection of grafted or ungrafted USPIO particles, iron concentration in inflamed muscles was quantified ex vivo by X-band EPR. Particle accumulation was also assessed in vivo by L-Band EPR, as well as by T(2)-weighted MRI. RESULTS: We determined the mean iron oxide concentration in inflamed muscles after injection of grafted or ungrafted UPSIO particles, which was 0.8% and 0.4% of the initially injected dose, respectively. By L-band EPR, we observed that the concentration of the grafted USPIO particles in inflamed muscles was twice higher than for the ungrafted particles. Using MRI experiments, a higher signal loss was clearly observed in the inflamed muscle when grafted particles were injected in comparison with the ungrafted particles. CONCLUSION: Even taking into account a non specific accumulation of iron oxides, the targeting of USPIO particles with E-selectin ligands significantly improved the sensitivity of detection of inflamed tissues.

19F NMR in vivo spectroscopy reflects the effectiveness of perfusion-enhancing vascular modifiers for improving gemcitabine chemotherapy.

Nuclear magnetic resonance spectroscopy of fluorine-19 ((19)F NMR) has proven useful for evaluating kinetics of fluorinated chemotherapy drugs in tumors in vivo. This work investigated how three perfusion-enhancing vascular modifiers (BQ123, thalidomide, and Botulinum neurotoxin type A [BoNT-A]) would affect the chemotherapeutic efficacy of gemcitabine, a fluorinated drug widely used in human cancer treatment. Murine tumor growth experiments demonstrated that only BoNT-A showed a strong trend to enhance tumor growth inhibition by gemcitabine (1.7 days growth delay, P = 0.052, Student t-test). In accord with these results, (19)F NMR experiments showed that only BoNT-A increased significantly the uptake of gemcitabine in tumors (50% increase, P = 0.0008, Student t-test). Further experiments on gemcitabine kinetics (NMR vs time) and distribution ((19)F MRI) confirmed the uptake-enhancing properties of BoNT-A. The results of this study demonstrate that (19)F NMR can monitor modulation of the pharmacokinetics of fluorinated chemotherapy drugs in tumors. The results also show that (19)F NMR data can give a strong indication of the effectiveness of perfusion-enhancing vascular modifiers for improving gemcitabine chemotherapy in murine tumors. (19)F NMR is a promising tool for preclinical evaluation of such vascular modifiers and may ultimately be used in the clinic to monitor how these modifiers affect chemotherapy.

Development and evaluation of biocompatible inks for the local measurement of oxygen using in vivo EPR.

In vivo EPR oximetry is a powerful minimally invasive method that allows the measurement of oxygen in tissues through the use of a paramagnetic probe. In the present study, we investigated new strategies for preparing biocompatible inks containing carbon black particles (Printex U), which could be used as oxygen sensors. The carbon black particles were dispersed in solutions of biocompatible polymers of carboxy methyl cellulose (CMC), hydroxypropyl methyl cellulose (HPMC) or polyvinyl pyrrolidone (PVP). A total of 12 polymers with different molecular weights were tested. A physico-chemical characterization of the inks was carried out to assess the sedimentation of the particles, the rheological behavior of these inks, and the relative diffusion of the inks. The preparations with CMC and PVP had the highest viscosity and stability. The presence of the polymers did not modify the calibration curves (EPR linewidth as a function of the pO2) of the carbon black. In vivo, the oxygen sensors were stable for at least one month in muscles as the EPR linewidth remained fully sensitive to induced ischemia or carbogen challenge. The calibration curve was not modified after this period of implantation. A first study of biocompatibility was carried out in vitro (hemolysis and cytotoxicity assay) and in vivo (histological examination). No sign of toxicity was observed using these inks. These preparations are good candidates for future in vivo studies including clinical trials.

Carbon blacks as EPR sensors for localized measurements of tissue oxygenation.

New electron paramagnetic resonance (EPR) oximetry probes were identified in the class of carbon black materials. These compounds exhibit very high oxygen sensitivity and favorable EPR characteristics for biological applications. At low pO(2), the linewidth is particularly sensitive to changes in oxygen tension (sensitivity of 750 mG/mmHg). The application of the probes for oximetry was demonstrated in vivo: the pO(2) was measured in muscle in which the blood flow was temporarily restricted as well as in tumor-bearing mice during a carbogen breathing challenge. The responsiveness to pO(2) was stable in muscle for at least 3 months. No toxicity was observed using these materials in cellular experiments and in histological studies performed 2, 7, and 28 days after implantation. In view of their EPR characteristics (high sensitivity) as well as the well-characterized production procedure that make them available on a large scale, these probes can be considered as very promising tools for future developments in EPR oximetry.

Potentiation of radiation-induced regrowth delay by isosorbide dinitrate in FSaII murine tumors.

Oxygen deficiency in tumors reduces the efficacy of nonsurgical treatment modalities such as conventional radiotherapy and chemotherapy. Since tumor perfusion is directly affected by the vascular resistance to flow of vessels feeding the tumor, vasodilator drugs might be a way to increase tumor blood flow and oxygenation. The effects of nitric oxide (NO) donor administration on tumor oxygenation, perfusion and radiation sensitivity were studied in the FSaII tumor model. Local tumor oxygenation was measured using electron paramagnetic resonance oximetry and a fiberoptic probe, OxyLite. We concomitantly measured the modulation of tumor blood flow by laser Doppler flowmetry. We determined FSaII tumor regrowth delay after isosorbide dinitrate administration and irradiation compared to carbogen breathing before irradiation and with X-rays alone. Administration of the NO donor improved the FSaII tumor pO(2) concomitant with an increase in tumor blood flow. We also demonstrated an increase in FSaII tumor radiation sensitivity after isosorbide dinitrate administration, which was similar to the effect of carbogen breathing in the same tumor model. Administration of isosorbide dinitrate could be considered in terms of improvement in tumor blood flow and a possible concomitant increase in accessibility of chemosensitizing agents to the tumor, particularly in terms of modification of the tumor response to irradiation.

Contribution of oxygenation to BOLD contrast in exercising muscle.

The potential physiological and therapeutic applications of functional MRI (fMRI) in skeletal muscle will depend on our ability to identify factors that may contribute to fluctuations in the BOLD signal. Until now, interpretations of signal changes in fMRI studies of muscle have mostly relied on the increase in muscle T2 associated with osmotically driven fluid shifts. However, recent studies have documented increases in BOLD signal intensity (SI) after single contractions, coinciding with increases in muscle hemoglobin saturation. In this study, the factors that contribute to variations in the intensity of the BOLD signal in exercising muscle are further addressed. For this purpose, BOLD imaging was performed during and after a moderate electrical stimulation was applied to the sciatic nerve in mice. In addition, oxygen pressure (pO2), blood flow, and skeletal muscle T2 (fast and slow components: T2 and T'2, respectively) were monitored. A comparison between mice lacking eNOS (eNOS-/- mice) and their wild-type (WT) littermates was performed. In WT mice, the BOLD SI, as well as muscle oxygenation and T'2, were significantly increased for a prolonged time in response to this moderate exercise protocol. Blood flow immediately dropped after the electrical stimulation was stopped. In eNOS-/- mice, the high BOLD SI did not persist after the exercise protocol ended. This finding correlates well with the evolution of muscle oxygenation, which progressively decreases after stimulation in eNOS-/- mice. However, T'2 remained high for a prolonged time after stimulation. We therefore concluded that the maintenance of BOLD SI in moderately exercising skeletal muscle depends mainly on changes in pO2, rather than on blood flow or T2 effects.

Nitric oxide as a radiosensitizer: evidence for an intrinsic role in addition to its effect on oxygen delivery and consumption.

Different nitric oxide (NO)-mediated treatments (e.g., isosorbide dinitrate, insulin and electrical stimulation of the host tissue) have been investigated for their effects on tumor oxygenation and radiation sensitivity. We further address the issue of the role played by modulation of the NO-pathway in tumor radiosensitivity. For this purpose, the local concentration of NO was monitored after treatment in FSaII tumors and a comparison between the sensitivity of LLC tumors implanted both on eNOS(-/-) and wild-type (WT) mice was carried out. First, we demonstrate the central role played by eNOS in the radiosensitizing effect after application of insulin treatment and electrical stimulation: a significant increase in tumor NO content is induced by these treatments and the increase in tumor oxygenation, as well as the radiosensitizing effect are abolished in eNOS knock-out mice, in contrast to WT mice. Second, by comparing the level of oxygen and NO achieved in tumors after NO-mediated treatments and carbogen, we provide evidence that these NO-mediated treatments are not simply acting by a single oxygen effect. These treatments induced significant regrowth delays compared to carbogen, despite a smaller increase in tumor oxygenation. For the NO-mediated treatments, there was a direct correlation between the NO content and the radiosensitizing effect. These data strongly suggest that NO is a complementary factor additive to oxygen in determining the sensitivity to irradiation and we therefore propose that NO acts as an intrinsic radiosensitizer in vivo.