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.

'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.

Skeletal Muscle Oxygenation Measured by EPR Oximetry Using a Highly Sensitive Polymer-Encapsulated Paramagnetic Sensor.

We have incorporated LiNc-BuO, an oxygen-sensing paramagnetic material, in polydimethylsiloxane (PDMS), which is an oxygen-permeable, biocompatible, and stable polymer. We fabricated implantable and retrievable oxygen-sensing chips (40 % LiNc-BuO in PDMS) using a 20-G Teflon tubing to mold the chips into variable shapes and sizes for in vivo studies in rats. In vitro EPR measurements were used to test the chip's oxygen response. Oxygen induced linear and reproducible line broadening with increasing partial pressure (pO2). The oxygen response was similar to that of bare (unencapsulated) crystals and did not change significantly on sterilization by autoclaving. The chips were implanted in rat femoris muscle and EPR oximetry was performed repeatedly (weekly) for 12 weeks post-implantation. The measurements showed good reliability and reproducibility over the period of testing. These results demonstrated that the new formulation of OxyChip with 40 % LiNc-BuO will enable the applicability of EPR oximetry for long-term measurement of oxygen concentration in tissues and has the potential for clinical applications.

Calculation of dose conversion factors for doses in the fingernails to organ doses at external gamma irradiation in air.

Absorbed doses to fingernails and organs were calculated for a set of homogenous external gamma-ray irradiation geometries in air. The doses were obtained by stochastic modeling of the ionizing particle transport (Monte Carlo method) for a mathematical human phantom with arms and hands placed loosely along the sides of the body. The resulting dose conversion factors for absorbed doses in fingernails can be used to assess the dose distribution and magnitude in practical dose reconstruction problems. For purposes of estimating dose in a large population exposed to radiation in order to triage people for treatment of acute radiation syndrome, the calculated data for a range of energies having a width of from 0.05 to 3.5 MeV were used to convert absorbed doses in fingernails to corresponding doses in organs and the whole body as well as the effective dose. Doses were assessed based on assumed rates of radioactive fallout at different time periods following a nuclear explosion.

Increased tumor oxygenation and drug uptake during anti-angiogenic weekly low dose cyclophosphamide enhances the anti-tumor effect of weekly tirapazamine.

Metronomic cyclophosphamide treatment is associated with anti-angiogenic activity and is anticipated to generate exploitable hypoxia using hypoxia-activated prodrugs. Weekly administration of tirapazamine (TPZ; 5 mg/kg body weight i.p.) failed to inhibit the growth of 9L gliosarcoma tumors grown s.c. in scid mice. However, the anti-tumor effect of weekly cyclophosphamide (CPA) treatment (140 mg/kg BW i.p.) was substantially enhanced by weekly TPZ administration. An extended tumor free period and increased frequency of tumor eradication without overt toxicity were observed when TPZ was given 3, 4 or 5 days after each weekly CPA treatment. Following the 2(nd) CPA injection, Electron Paramagnetic Resonance (EPR) Oximetry indicated significant increases in tumor pO(2), starting at 48 hr, which further increased after the 3(rd) CPA injection. pO(2) levels were, however, stable in growing untreated tumors. A strong negative correlation (-0.81) between tumor pO(2) and tumor volume during 21 days of weekly CPA chemotherapy was observed, indicating increasing tumor pO(2) with decreasing tumor volume. Furthermore, CPA treatment resulted in increased tumor uptake of activated CPA. CPA induced increases in VEGF RNA, which reached a maximum on day 1, and in PLGF RNA which was sustained throughout the treatment, while anti-angiogenic host thrombospondin-1 increased dramatically through day 7 post-CPA treatment. Weekly cyclophosphamide treatment was anticipated to generate exploitable hypoxia. However, our findings suggest that weekly CPA treatment induces a functional improvement of tumor vasculature, which is characterized by increased tumor oxygenation and drug uptake in tumors, thus counter-intuitively, benefiting intratumoral activation of TPZ and perhaps other bioreductive drugs.

Electron paramagnetic resonance in human fingernails: the sponge model implication.

The most significant problem of electron paramagnetic resonance (EPR) fingernail dosimetry is the presence of two signals of non-radiation origin that overlap the radiation-induced signal (RIS), making it almost impossible to perform dose measurements below 5 Gy. Historically, these two non-radiation components were named mechanically induced signal (MIS) and background signal (BKS). In order to investigate them in detail, three different methods of MIS and BKS mutual isolation have been developed and implemented. After applying these methods, it is shown here that fingernail tissue, after cut, can be modeled as a deformed sponge, where the MIS and BKS are associated with the stress from elastic and plastic deformations, respectively. A sponge has a unique mechanism of mechanical stress absorption, which is necessary for fingernails in order to perform its everyday function of protecting the fingertips from hits and trauma. Like a sponge, fingernails are also known to be an effective water absorber. When a sponge is saturated with water, it tends to restore to its original shape, and when it loses water, it becomes deformed again. The same happens to fingernail tissue. It is proposed that the MIS and BKS signals of mechanical origin be named MIS1 and MIS2 for MISs 1 and 2, respectively. Our suggested interpretation of the mechanical deformation in fingernails gives also a way to distinguish between the MIS and RIS. The results obtained show that the MIS in irradiated fingernails can be almost completely eliminated without a significant change to the RIS by soaking the sample for 10 min in water. The proposed method to measure porosity (the fraction of void space in spongy material) of the fingernails gave values of 0.46-0.48 for three of the studied samples. Existing results of fingernail dosimetry have been obtained on mechanically stressed samples and are not related to the "real" in vivo dosimetric properties of fingernails. A preliminary study of these properties of pre-soaked (unstressed) fingernails has demonstrated their significant difference from fingernails stressed by cut. They show a higher stability signal, a less intensive non-radiation component, and a nonlinear dose dependence. The findings in this study set the stage for understanding fingernail EPR dosimetry and doing in vivo measurements in the future.

Quinoid radio-toxin (QRT) induced metabolic changes in mice: an ex vivo and in vivo EPR investigation.

Radio-toxins are toxic metabolites produced by ionizing irradiation and have toxic effects similar to those caused by direct irradiation. We have investigated the effect of a quinoid radio-toxin (QRT) obtained from gamma-irradiated potato tuber on various organs in mice using ex vivo and in vivo EPR spectroscopy. Results indicate a decrease in the activity of ribonucleotide reductase enzyme in spleen of mice treated with 0.2mg QRT. A dose of 2mg QRT was fatal to mice within 45-60 min of treatment. Nitrosyl hemoglobin complexes alpha-(Fe(2+)-NO)alpha-(Fe(2+))beta-(Fe(2+))(2) were detected from spleen, blood, liver, kidney, heart, and lung tissue samples of mice treated with lethal doses of QRT. A significant decrease of pO(2) in liver and brain was observed after administration of QRT at the lethal dose. The time of the appearance of the nitrosyl hemoglobin complex and its intensity varied with the dose of QRT and the type of tissue. These results indicate that the effect of the QRT is more prominent in spleen and to a lesser extent in liver and blood. The QRT action at the lethal doses resulted in an increased hypoxia over time with disruption of compensatory adaptive response. The results indicate similar outcome of QRT as observed with gamma-irradiation.

Protocol for emergency EPR dosimetry in fingernails.

There is an increased need for after-the-fact dosimetry because of the high risk of radiation exposures due to terrorism or accidents. In case of such an event, a method is needed to make measurements of dose in a large number of individuals rapidly and with sufficient accuracy to facilitate effective medical triage. Dosimetry based on EPR measurements of fingernails potentially could be an effective tool for this purpose. This paper presents the first operational protocols for EPR fingernail dosimetry, including guidelines for collection and storage of samples, parameters for EPR measurements, and the method of dose assessment. In a blinded test of this protocol application was carried out on nails freshly sampled and irradiated to 4 and 20 Gy; this protocol gave dose estimates with an error of less than 30%.

EPR dosimetry in chemically treated fingernails.

By using EPR measurements of radiation-induced radicals it is possible to utilize human fingernails to estimate radiation dose after-the-fact. One of the potentially limiting factors in this approach is the presence of artifacts due to mechanically induced EPR signals (MIS) caused by mechanical stress during the collection and preparation of the samples and the so-called background (non-radiation) signal (BKS). The MIS and BKS have spectral parameters (shape, g-factor and linewidth) that overlap with the radiation-induced signal (RIS) and therefore, if not taken into account properly, could result in a considerable overestimation of the dose. We have investigated the use of different treatments of fingernails with chemical reagents to reduce the MIS and BKS. The most promising chemical treatment (20 min with 0.1 M dithiothreitol aqueous solution) reduced the contribution of MIS and BKS to the total intensity of EPR signal of irradiated fingernails by a factor of 10. This makes it potentially feasible to measure doses as low as 1 Gy almost immediately after irradiation. However, the chemical treatment reduces the intensity of the RIS and modifies dose dependence. This can be compensated by use of an appropriate calibration curve for assessment of dose. On the basis of obtained results it appears feasible to develop a field-deployable protocol that could use EPR measurements of samples of fingernails to assist in the triage of individuals with potential exposure to clinically significant doses of radiation.

Q-band EPR biodosimetry in tooth enamel microsamples: feasibility test and comparison with x-band.

A comparative study of electron paramagnetic resonance dosimetry in Q- and X-bands has shown that Q-band is able to provide accurate measurements of radiation doses even below 0.5 Gy with tooth enamel samples as small as 2 mg. The optimal amount of tooth enamel for dose measurements in Q-band was found to be 4 mg. This is less than 1% of the total amount of tooth enamel in one molar tooth. Such a small amount of tooth enamel can be harmlessly obtained in an emergency requiring after-the-fact radiation dose measurement. The other important advantage of Q-band is full resolution of the radiation-induced EPR signal from the native, background signal. This separation makes dose response measurements much easier in comparison to conventional X-band measurements in which these overlapping signals necessitate special methods for doses below 0.5 Gy. The main disadvantages of Q-band measurements are a higher level of noise and lower spectral reproducibility than in X-band. The effect of these negative factors on the precision of dose measurements in Q-band could probably be reduced by improvement of sample fixation in the resonance cavity and better optimization of signal filtration to reduce high-frequency noise.

Influence of different anesthetics on skin oxygenation studied by electron paramagnetic resonance in vivo.

The effects of two general anesthetics on skin oxygenation in mice are evaluated by electron paramagnetic resonance oximetry. Up to now no data on the effects of different anesthetics on skin oxygenation could be found. In this study animals were anesthetized with ketamine/xylazine or isoflurane, and partial pressure of oxygen (pO(2)) in the skin, heart rate and hemoglobin oxygen saturation were followed as a function of time and inhaled oxygen concentration. The skin pO(2) significantly increased continuously for about 60 min in mice anesthetized with isoflurane and remained constant after that. During ketamine/xylazine anesthesia, the pO(2) in the skin only slightly decreased. The skin pO(2) increased with higher inspired oxygen concentrations for both anesthetics groups. When breathing 21% oxygen, mice anesthetized with isoflurane had two-fold higher pO(2) in the skin compared to mice anesthetized with ketamine/xylazine. The heart rate was significantly lower in animals anesthetized with ketamine/xylazine, while hemoglobin saturation was almost the same in both groups at all inhaled oxygen concentrations. These results show that the type of anesthesia is an important parameter that needs to be considered in experiments where skin pO(2) is followed.

Radiation dose reconstruction from L-band in vivo EPR spectroscopy of intact teeth: Comparison of methods.

In vivo EPR tooth dosimetry is a more challenging problem than in vitro EPR dosimetry because of several potential additional sources of variation associated with measurements that are made in the mouth of a living subject. For in vivo measurement a lower RF frequency is used and, unlike in the in vitro studies, the tooth cannot be processed to optimize the amount and configuration of the enamel that is measured. Additional factors involved with in vivo measurements include the reproducibility of positioning the resonator on the surface of the tooth in the mouth, irregular tooth geometry, and the possible influence of environmental noise. Consequently, in addition to using the theoretical and empirical models developed for analyzing data from measurements of teeth in vitro, other unconventional and more robust methods of dose reconstruction may be needed. The experimental parameter of interest is the peak-to-peak amplitude of the spectrum, which is correlated to the radiation dose through a calibration curve to derive the reconstructed dose. In this study we describe and compare the results from seven types of computations to measure the peak-to-peak amplitude for estimation of the radiation induced signal. The data utilized were from three sets of in vivo measurements of irradiated teeth. Six different teeth with different doses were placed in the mouth of a volunteer in situ and measurements of each tooth were carried out on three different days. The standard error of dose prediction (SEP) is used as a figure of merit for quantifying precision of the reconstruction. We found that many of the methods gave fairly similar results, with the best error of prediction resulting from a computation based on a Lorentzian line model whose center field corresponds to the known parameter of the radiation-induced EPR spectra of teeth, with corrections from a standard sample that was measured as part of the data acquisition scheme. When the results from the three days of measurement were pooled, the SEP decreased dramatically, which suggests that one of the principal sources of variation in the data is the ability to precisely standardize the measurements conditions within the mouth. There are very plausible ways to accomplish improvements in the existing procedures.

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.

EPR spectrometer for clinical applications.

This article describes an EPR spectrometer specifically designed and constructed for EPR spectroscopy in humans. The spectrometer is based on a permanent magnet, suitable for measurements at 1200 MHz. The magnet has a full 50 cm gap between the poles, which facilitates accurate and comfortable placement of the subject for the EPR measurement at any location on the human body. The bridge includes features to facilitate clinical operations, including an indicator for phasing of the reference arm and a 2 level RF amplifier. Resonators with holders for each type and site of measurement have been developed that comfortably position the resonator and the patient and prevent artifacts due to motion. The initial applications for which the spectrometer has been designed are for oximetry using loops on the surface, oximetry using implanted resonators for measuring deep sites, and measurements in the teeth for determination of exposures to clinically significant doses of ionizing radiation.

Spectrum file size optimization for EPR tooth dosimetry.

Spectral acquisition time is one of the limiting factors in electron paramagnetic resonance (EPR) retrospective biodosimetry in teeth. Acquisition times for one sample can be from 2 to 4h. This problem is even more acute for in vivo EPR measurements in L-band. Patients cannot be expected to remain stationary for these lengths of time. In order to overcome this limitation, we investigated the dependence of EPR dose measurements on the number of data points in an EPR spectrum. We have shown that this number could be reduced from 1024 to 256 (factor of 4 reduction in spectral acquisition time) at 5 mT magnetic field sweep without a loss of precision in the dose measurements.

Reduced blood flow and oxygenation in SA-1 tumours after electrochemotherapy with cisplatin.

Electrochemotherapy is an antitumour treatment that utilises locally delivered electric pulses to increase cytotoxicity of chemotherapeutic drugs. Besides increased drug delivery, application of electric pulses affects tumour blood flow. The aim of this study was to determine tumour blood flow modifying effects of electrochemotherapy with cisplatin, its effects on tumour oxygenation and to determine their relation to antitumour effectiveness. Electrochemotherapy of SA-1 subcutaneous tumours was performed by application of electric pulses to the tumours, following administration of cisplatin. Tumour blood flow modifying effects of electrochemotherapy were determined by measurement of tumour perfusion using the Patent blue staining technique, determination of tumour blood volume, and microvascular permeability using contrast enhanced magnetic resonance imaging, and tumour oxygenation using electron paramagnetic resonance oximetry. Antitumour effectiveness was determined by tumour growth delay and the extent of tumour necrosis and apoptosis. Tumour treatment by electrochemotherapy induced 9.4 days tumour growth delay. Tumour blood flow was reduced instantaneously and persisted for several days. This reduction in tumour blood flow was reflected in reduced tumour oxygenation. The maximal reduction in partial oxygen pressure (pO2) levels was observed at 2 h after the treatment, with steady recovery to the pretreatment level within 48 h. The reduced tumour blood flow and oxygenation correlated well with the extent of tumour necrosis and tumour cells apoptosis induced by electrochemotherapy with cisplatin. Therefore, the data indicate that antitumour effectiveness of electrochemotherapy is not only due to increased cytotoxicity of cisplatin due to electroporation of tumour cells, but also due to anti-vascular effect of electrochemotherapy, which resulted in reduced tumour blood flow and oxygenation.

Firefly flashing is controlled by gating oxygen to light-emitting cells.

Although many aspects of firefly bioluminescence are understood, the mechanism by which adult fireflies produce light as discrete rapid flashes is not. Here we examine the most postulated theory, that flashing is controlled by gating oxygen access to the light-emitting cells (photocytes). According to this theory, the dark state represents repression of bioluminescence by limiting oxygen, which is required for bioluminescence; relief from this repression by transiently allowing oxygen access to the photocytes allows the flash. We show that normobaric hyperoxia releases the repression of light emission in the dark state of both spontaneously flashing and non-flashing fireflies, causing continual glowing, and we measure the kinetics of this process. Secondly, we determine the length of the barriers to oxygen diffusion to the photocytes in the aqueous and gas phases. Thirdly, we provide constraints upon the distance between any gas-phase gating structure(s) and the photocytes. We conclude from these data that the flash of the adult firefly is controlled by gating of oxygen to the photocytes, and demonstrate that this control mechanism is likely to act by modulating the levels of fluid in the tracheoles supplying photocytes, providing a variable barrier to oxygen diffusion.

Development of biocompatible oxygen-permeable films holding paramagnetic carbon particles: evaluation of their performance and stability in EPR oximetry.

EPR oximetry using paramagnetic particles relies on the measurement of the EPR linewidth, which is directly related to the pO2. It was previously found that some of the paramagnetic materials with optimal EPR spectroscopic properties in vitro may lose their responsiveness to oxygen in tissues (change of the calibration curve of the EPR linewidth as a function of the pO2). We hypothesized that coating paramagnetic particle materials could improve the stability of response, as well as the biocompatibility. In this study, very thin films holding paramagnetic materials were prepared with different biopolymers (cellulose acetate, cellulose triacetate, cellulose nitrate, silicone, and polyurethane) that already are accepted for clinical applications. Their performance was evaluated in EPR oximetry by measuring the stability of the calibration curves (EPR linewidth as a function of pO2) after a prolonged period in an aqueous environment (1 week in saline) or in vivo (implantation for 3 weeks under the skin of mice). We found that one type of silicone film was able to stabilize the responsiveness of an intrinsically unstable carbon material (a wood char).

High spatial resolution multi-site EPR oximetry. The use of convolution-based fitting method.

We describe a new method to enhance the spatial resolution of multi-site electron paramagnetic resonance (EPR) oximetry. The method is suitable for any shape (density distribution function) of a solid paramagnetic material implanted in tissue. It corrects distortions of lineshapes caused by the gradient and thus overcomes limitations of previous multi-site EPR oximetry methods that restricted the ratio of the particle size to the distance between sites. The new method is based on consecutive applications of magnetic field gradients with the same direction but with a different magnitude and uses a convolution-based fitting algorithm to derive Lorentzian EPR linewidths of each individual peak of the EPR spectrum. The method is applicable for any particulate EPR oxygen sensitive materials whose EPR spectra can be approximated by a Lorentzian function or a superposition of Lorentzian functions. By incorporating this model of the lineshape in the data processing, we are able to decrease significantly the number of parameters needed for the calculations and to recover the oxygen concentration, even from quite noisy spectra. We (i) describe our method and the data-processing algorithm, (ii) demonstrate our approach in model and in vivo experiments, and (iii) discuss the limitations.