Induction of tumour hypoxia by a vasoactive agent. A combined NMR and radiobiological study

Measuring oxygenation in vivo with MRS/MRI--from gas exchange to the cell

Oxygen plays a major role as a substrate in metabolic processes in numerous signaling pathways, in redox metabolism, and in free radical metabolism. To study the role of oxygen in normal and pathophysiological states, methods that can be used noninvasively are required. This review examines the potential of nuclear magnetic resonance techniques to study tissue oxygenation. It is written from a systems perspective, looking at detection methods with respect to the path that oxygen takes in the mammalian system-from the lungs, through the vascular system, into the interstitial space, and finally into the cell. Methods discussed range from those that are quantifiable, such as the assessment of spin lattice relaxation time in fluorocarbon solutions, to those that are more correlative, such as assessment of lactate and high energy phosphates. Since the methods vary in their site of application, sensitivity, and specificity to the quantification of oxygen, this review provides examples of how each method has been applied. This may facilitate the reader's understanding of how to optimally apply different methods to study specific biomedical problems.

The Difficulties in Comparing In Vivo Oxygen Measurements

There has been rapid development of effective new tools that provide information on oxygenation in vivo and an increased recognition of how valuable such information can be. Consequently, there also has been considerable interest in comparing and evaluating the accuracy and usefulness of the different types of measurements. The various types of measurements usually do not measure the same thing. They may measure PO2 or [O2] or something less directly related, such as hemoglobin saturation. They may make measurements in different compartments (e.g. intracellular, extracellular, vascular) in the volume that they sample, the time span over which they average, the local perturbation that they may cause, etc. They also differ in their sensitivity, accuracy, ability to measure repetitively. However, these potentially confounding and confusing differences can be made into an outstanding virtue, if their nature is considered carefully. Then a proper model can relate them to each other. The ability to relate the various measurements to each other can be a powerful tool to test the validity of models that attempt to explain fully the distribution of oxygen in real systems and the factors that affect this. We then could have a major advancement in our understanding of oxygen transport in tissues, with an ability to determine accurately the effects of physiological and pathophysiological perturbations on oxygenation at all levels of cells and tissues in vivo.

Effects of nicotinamide and carbogen on tumour oxygenation, blood flow, energetics and blood glucose levels

Both host carbogen (95% oxygen/5% carbon dioxide) breathing and nicotinamide administration enhance tumour radiotherapeutic response and are being re-evaluated in the clinic. Non-invasive magnetic resonance imaging (MRI) and 31P magnetic resonance spectroscopy (MRS) methods have been used to give information on the effects of nicotinamide alone and in combination with host carbogen breathing on transplanted rat GH3 prolactinomas. Gradient recalled echo (GRE) MRI, sensitive to blood oxygenation changes, and spin echo (SE) MRI, sensitive to perfusion/flow, showed large signal intensity increases with carbogen breathing. Nicotinamide, thought to act by suppressing the transient closure of small blood vessels that cause intermittent tumour hypoxia, induced a small increase in blood oxygenation but no detectable change in perfusion/flow. Carbogen combined with nicotinamide was no more effective than carbogen alone. Both carbogen and nicotinamide caused significant increases in the nucleoside triphosphate/inorganic phosphate (betaNTP/Pi) ratio, implying that the tumour cells normally receive sub-optimal substrate supply, and is consistent with either increased glycolysis and/or a switch to more oxidative metabolism. The most striking observation was the marked increase in blood glucose (twofold) induced by both nicotinamide and carbogen. Whether this may play a role in tumour radiosensitivity has yet to be determined.

Hypoxic cell cytotoxin tirapazamine induces acute changes in tumor energy metabolism and pH: a 31P magnetic resonance spectroscopy study

Tirapazamine is a hypoxic cell cytotoxin in phase II/III trials. To further understand its mechanism of action in vivo, we examined the effect of tirapazamine on tumor energy metabolism and pH. RIF-1 and SCCVII tumors were grown subcutaneously in the flanks of C3H mice. Tumor energy metabolism, expressed as the ratio of inorganic phosphate to nucleotide triphosphate (Pi/NTP), and intracellular pH (pHi), were measured by 31P magnetic resonance spectroscopy (MRS). In RIF-1 and SCCVII tumors, tirapazamine increased the Pi/NTP ratio by 2.6-fold and 3-fold, respectively, within the first hour after an intraperitoneal dose of 0.3 mmol/kg. A corresponding decrease in pHi from 7.05+/-0.07 to 6.48+/-0.06, and 7.21+/-0.09 to 6.45+/-0.02 in RIF-1 and SCCVII tumors, respectively, was observed. The decrease in tumor 31P bioenergetics and pH was reversible, as exemplified by RIF-1 tumors, which showed a further increase in Pi/NTP ratio of 3.5-fold by 5-8 hr, returning to normal range at 24 hr. Corresponding pHi of RIF-1 tumors was 6.88+/-0.05 at 5-8 hr and 7.16+/-0.05 at 24 hr. We concluded that tirapazamine induces acute changes in tumor energy metabolism and pHi. These findings are relevant to the rational selection and optimal timing of coadministered therapy.

31P-magnetic resonance spectroscopy and 2H-magnetic resonance imaging studies of a panel of early-generation transplanted murine tumour models

The objective of this study was first to determine whether three slowly growing early-generation murine transplantable tumours, the T40 fibrosarcoma, T115 mammary carcinoma and T237 lung carcinoma, exhibit patterns of energetics and blood flow during growth that are different from those of the faster growing RIF-1 fibrosarcoma. Serial measurements were made with 31P-magnetic resonance spectroscopy (MRS), relating to nutritive blood flow and 2H-magnetic resonance imaging (MRI), which is sensitive to both nutritive and large-vessel (non-nutritive) flow. All four tumour lines showed a decrease in betaNTP/Pi and pH with growth; however, each line showed a different pattern of blood flow that did not correlate with the decrease in energetics. Qualitative histological analysis strongly correlated with the 2H-MRI. Second, their response to 5 mg kg(-1) hydralazine i.v. was monitored by 31P-MRS. A marked decrease in betaNTP/Pi and pH was observed in both the RIF-1 fibrosarcoma and the third-generation T115 mammary carcinoma after hydralazine challenge. In contrast, the fourth generation T40 fibrosarcoma and T237 lung carcinoma showed no change in 31P-MRS parameters. However, a fifth-generation T237 cohort, which grew approximately three times faster than fourth-generation T237 cohorts, exhibited a significant deterioration in betaNTP/Pi and pH in response to hydralazine. These data are consistent with a decoupling between large-vessel and nutritive blood flow and indicate that early-generation transplants that have a slow growth rate and vascular tone are more appropriate models of human tumour vasculature than more rapidly growing, repeatedly transplanted tumours.

Glucose metabolism in RIF-1 tumors after reduction in blood flow: an in vivo 13C and 31P NMR study

Low pH appears to enhance the effectiveness of therapeutic hyperthermia. 13C and 31P NMR spectroscopy have been employed to examine the possibility that elevating glucose in a solid tumor while simultaneously reducing tumor blood flow would induce a more profound acidosis than either treatment alone. When blood flow in RIF-1 tumors was acutely reduced by administration of hydralazine and additional glucose was delivered locally by intratumoral injection, tumor acidosis (as determined by 31P NMR spectroscopy) during the period of reduced blood flow was not enhanced, relative to administration of hydralazine alone. Tumor NTP/P1 ratios decreased significantly within 20 min of hydralazine administration, whether or not glucose was injected, although NTP/P1 ratios were slightly higher in tumors that received extra glucose. Tumor lactate concentrations were not significantly different in glucose-supplemented tumors, despite glucose concentrations that were 4 to 5 times higher. When the added glucose was labeled with 13C, no correlation was detected between the pH in an individual tumor and the intensity of the 3-[13C]-lactate resonance in the same tumor.

Comparison of the effects of hydralazine on tumor and normal tissue blood perfusion by MRI

PURPOSE

The differential effects on blood perfusion of the vasodilator hydralazine (HYD) between tumor and normal muscle have been measured using the dynamic enhanced-magnetic resonance imaging (DE-MRI) technique.

METHODS AND MATERIALS

DE-MRI is a noninvasive method of determining blood perfusion in tumors and normal tissues using the MR contrast agent Gd-DTPA. Hydralazine is currently being used in an attempt to increase tumor response to bioreductive agents and to hyperthermia.

RESULTS

We show that a dose of 1.2 mg/kg HYD causes an increase in tumor perfusion while doses > or = 2.5 mg/kg cause a decrease in tumor perfusion. The latter was accompanied by a dose-dependent increase in normal muscle perfusion consistent with the "steal effect."

CONCLUSION

This study demonstrates the sensitivity of the DE-MRI technique and its capability of providing estimates of blood perfusion in normal and tumor tissue as well as in smaller regions of a solid tumor. Such features would make it clinically useful in the study of tumor response to radiation therapy and chemotherapy in patients.

Enhancement of interstitial photodynamic therapy by mitomycin C and EO9 in a mouse tumour model

The tumoricidal effect of interstitial photodynamic therapy (IPDT) using Photofrin was found to increase when combined with the bioreductive alkylating agent mitomycin C (MMC) and, to a lesser extent, with the indoloquinone EO9. When MMC was given prior to IPDT or RIF1 tumours, the light dose required for a given regrowth time or for 50% cure was reduced by a factor of 2 compared with IPDT alone. MMC given immediately after illumination did not increase the effects of IPDT, although MMC plus illumination without photosensitizer produced a significant increase in regrowth time compared with MMC or light alone. Combination of IPDT with EO9, given directly before illumination, only marginally increased the tumour regrowth times at non-toxic doses. These results demonstrate that combining IPDT with MMC greatly improves the tumour response. Factors such as PDT-induced hypoxia, pH changes, temperature increases and production of toxic reactive oxygen species by both drugs may play a role in the enhanced MMC cytotoxicity.

In vivo 31P MRS of experimental tumours

More than 50% of cancers fail to respond to any individual treatment and tumour follow-up after treatment plays a major role in routine therapy planning and pharmacological research. Today, MRS is the only technological approach providing non-invasive access to tumour biochemistry. Ten years ago, expectations were raised concerning 31P MRS as an exciting and promising technical approach to the study of tumours. However the expectations have not always come to fruition. How close are we now to seeing routine 31P NMR in clinical oncology? This review of the 127 published papers shows spectroscopy results in more than 150 experimental animal tumour models. These tumour/host/treatment systems provide us with a useful basis to evaluate the current state of the art, summarize the basic knowledge presently available, determine the key points underlying the present disappointment of some clinical oncologists and stimulate new basic research. The information collected concerns the discussion of the reliability of experimental models in oncology, the technical improvement of magnetic resonance technology and the monitoring of bioenergetic status, pH regulation and phospholipid metabolism in treated and untreated tumours. Recent advances (two-thirds of the papers have been published in the last 5 years) seem to provide more optimistic perspectives than those generally accepted a few years ago, in the depressing period following early pioneering work.

Assessing the bioreductive effectiveness of the nitroimidazole RSU1069 and its prodrug RB6145: with particular reference to in vivo methods of evaluation

The nitroimidazole, RSU1069, has been shown to have a very high differential toxicity towards hypoxic cells compared to oxic cells both in in vitro and in vivo experimental conditions. However, in the clinic it was found to cause severe emesis and had to be withdrawn. After an extensive drug development programme an analogue of RSU1069, RB6145, which acts as a pro-drug for RSU1069, was found to be the most suitable candidate for further investigation. In in vivo studies with murine tumour models, when RB6145 was used in combination with X-rays it was shown to produce a similar level of toxicity towards hypoxic cells as that observed for RSU1069. Its activity was the same whether it was administered interperitoneally or orally and the same level of anti-tumour effect was observed if the drug was given before or after X-rays. RB6145 is better tolerated systemically in mice than RSU1069 and canine studies have shown that it is less emetic than the parent drug. Bioreductive drugs can also be used in combination with treatments that preferentially increase tumour hypoxia. Photodynamic therapy (PDT) causes extensive vascular damage in tumours. If either RSU1069 or RB6145 are administered during PDT, very large increases in the growth delay induced by PDT alone are seen for the RIF-1 murine tumour. RB6145 has been accepted for clinical toxicity trials with the prospect of using it in combination with X-rays. In the future it may also be of clinical use with treatments such as PDT.

Effects of agents which inhibit the regulation of intracellular pH on murine solid tumours

Cell killing can be achieved in an acidic environment in tissue culture (medium pH less than 7.0) by agents (nigericin, carbonylcyanide-3-chlorophenylhydrazone (CCCP)) which transport protons from the extracellular space into the cytoplasm. Cell killing is enhanced when these agents are used in combination with compounds (amiloride, 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid (DIDS)) which inhibit the membrane-based exchangers responsible for the regulation of intracellular pH (pHi). We describe experiments which assess the ability of these agents to kill tumour cells in spheroids and in vivo. Both nigericin and CCCP were observed to penetrate tissue based on their ability to kill tumour cells in spheroids. The mean extracellular pH (pHe) of the KHT fibrosarcoma and the EMT-6 sarcoma were observed to be 0.21 and 0.32 pH units more acidic than the mean pHe in muscle tissue. Intraperitoneal (i.p.) administration of the vasodilator hydralazine (10 mg kg-1) caused a reduction of the mean pHe of the KHT but not the EMT-6 tumour. Nigericin (2.5 mg kg-1, i.p.) plus amiloride (10 mg kg-1, i.p.) followed 30 min later by hydralazine (10 mg kg-1, i.p.) reduced the surviving fraction of cells in the KHT and EMT-6 tumours, but had minimal effects on growth delay. When KHT tumours were treated with 15 Gy X-rays followed immediately by nigericin plus amiloride and hydralazine a reduced surviving fraction as well as an increase in tumour growth delay was observed compared to radiation alone. The administration of nigericin (2.5 mg kg-1, i.p.) or the combination of nigericin (2.5 mg kg-1, i.p.) followed by hydralazine (10 mg kg-1, intravenous (i.v.)) resulted in reductions of tumour pHi of 0.27 and 0.29 pH units respectively as determined by 31P magnetic resonance spectroscopy (MRS). Our results show that the combination of nigericin and hydralazine (with or without amiloride) can kill cells in rodent solid tumours and that cell killing is associated with a reduction in the mean pHi of tumour cells.

Magnetic resonance spectroscopy studies on experimental murine and human tumors: comparison of changes in phosphorus metabolism with induced changes in vascular volume

The responses of two experimental murine tumors and two human tumor xenografts to the vasodilator hydralazine were compared using two magnetic resonance spectroscopy endpoints. Changes in tumor metabolism were determined using 31P MRS where inorganic phosphate levels relative to total phosphate (Pi/total) were measured, and alteration in tumor blood volume was examined using 19F MRS with perfluorooctylbromide (PFOB) as tracer. The integrated 19F signal from PFOB is dose dependent and stable for at least 2 hr after injection. The murine tumors SCCVII/Ha and KHT both showed changes in tumor metabolism after hydralazine, as an increase in Pi/total. However, hydralazine reduced vascular volume in the KHT tumor, demonstrated by reduced 19F signal from PFOB, but no such reduction was seen in the SCCVII/Ha tumor. In contrast, hydralazine had no effect on phosphorus metabolism in the HT29 and HX118 human tumor xenografts, but reduced vascular volume in both tumors. These results demonstrate that the effects of vasoactive agents such as hydralazine on tumor phosphorus metabolism are only partially consistent with changes in vascular volume, measured by the 19F MRS technique.

The response of spontaneous and transplantable murine tumors to vasoactive agents measured by 31P magnetic resonance spectroscopy

31P magnetic resonance spectroscopy has been used to compare the effects of the vasoactive agents hydralazine and flunarizine on the oxygenation of the transplantable tumors, SCCVII/Ha and 16C, and a range of spontaneous mammary tumors arising in the breeding stock in the Genetics Division at the Radiobiology Unit. The vasodilator hydralazine, previously shown to increase the radiobiological hypoxic fraction of transplantable murine tumors, increased inorganic phosphate to total phosphate (Pi/total) in SCCVII/Ha and 16C tumors. However, only two spontaneous tumors responded to this agent (2/12). The calcium antagonist flunarizine, which sensitizes the SCCVII tumor to X rays, consistent with a reduction in hypoxic fraction, reduced Pi/total in this and the 16C tumor. Further, most spontaneous tumors tested (8/10) responded to this agent, as measured by a reduction in Pi/total. These results point to fundamental differences between transplantable and spontaneously arising tumors in mice in their response to vasoactive agents.

In vivo 31P nuclear magnetic resonance spectroscopy of experimental murine tumours and human tumour xenografts: effects of blood flow modification

The effect of hydralazine on tumours appears to vary depending on tumour type. Blood flow and radiation sensitivity decrease more in murine tumours than human tumour xenografts. In this study a comparison between various tumour types has been made using in vivo 31P nuclear magnetic resonance spectroscopy (NMRS) to follow the metabolic responses occurring after clamping or intravenous administration of hydralazine (5 mg kg-1). Large increases in the Pi/total phosphate ratio were found with the murine sarcomas, KHT and RIF-1 implanted into C3H/He mice. However little or no effect was seen for the two human xenografted tumours, HX118 and HT29 implanted in MFI nu/nu/01a mice. An intermediate response was observed for KHT tumours grown in nu/nu mice. All tumours showed a large response to clamping. The anaesthetic Hypnorm/Hypnovel has a great influence on the response of the tumour metabolism to hydralazine appearing to both prolong and increase the changes induced. There is evidence to support the theory that the changes in 31P spectra are related to the oxygen status of the tumours.

The measurement of radiosensitizer-induced changes in mouse tumor metabolism by 31P magnetic resonance spectroscopy

Flunarizine and nicotinamide have previously been shown to increase blood perfusion to experimental mouse tumors and consequently, to increase their sensitivity to X rays. These agents were examined for their ability to alter metabolism, measured by 31P magnetic resonance spectroscopy, in the SCCVII/Ha carcinoma and the KHT sarcoma. Flunarizine at 5 mg/kg I.P. produced a 45% reduction in the ratio of inorganic phosphate to total phosphate (Pi/total) in the SCCVII/Ha tumor but only a 24% reduction in this ratio in the KHT tumor. These effects were seen 45 min after drug administration, and ratios returned to control levels by 90 min. In the SCCVII/Ha tumor, nicotinamide at 1000 mg/kg I.P. reduced Pi/total by 56% from 30 min to at least 2 hr after injection, and the ratio was reduced by 59% in the KHT tumor at 30 min after injection, returning to control levels by 2 hr. For the SCCVII/Ha tumor, the time course for the effects of flunarizine and nicotinamide on the inorganic phosphate ratio coincided with that previously reported for radiosensitization.

Modification of the 31P magnetic resonance spectra of a rat tumour using vasodilators and its relationship to hypotension

The effects of different doses of hydralazine and prostacyclin on the 31P magnetic resonance spectra of the LBDS1 fibrosarcoma were investigated and related to their effects on mean arterial blood pressure (MABP) and heart rate. The effect of reducing MABP by bleeding the animals, via the tail artery, was also investigated. Tumour spectral changes following high dose drug treatment (an increase in inorganic phosphate, a reduction in nucleotide triphosphates and a reduction in pH) were consistent with nutrient deprivation. These changes were dose dependent. Changes in MABP and heart rate were consistent with vasodilatation in normal tissues. However, for the same fall in MABP, hydralazine produced a greater rise in tumour inorganic phosphate (Pi) and a greater fall in tumour pH than did prostacyclin. Controlled bleeding was effective in reducing MABP. It also reduced tumour pH but had no significant effect on tumour Pi. The clinical application of the two drugs for reducing tumour blood flow and pH for therapy is likely to be limited by the large degree of hypotension necessary to produce an effect. The differential effect of the two drugs for the same fall in MABP may be related to different degrees of direct tumour vasodilatation or to a direct effect of hydralazine on tumour energy metabolism. The observation that controlled bleeding does not change tumour Pi is further evidence indicating that the degree of arterial hypotension is not the sole factor in determining tumour energy status.

Bioreductive drugs and the selective induction of tumour hypoxia

In this work tumour hypoxia is induced by physically occluding the tumour vascular supply by clamping, or by giving mice 5 mg kg-1 hydralazine. These methods have previously been shown to increase the radiobiological hypoxic fraction in tumours close to 100%. Their effectiveness in potentiating the bioreductive toxicity of: misonidazole (800 mg kg-1), RSU1069 (80 mg kg-1), mitomycin C (5 mg kg-1) and SR4233 (50 mg kg-1) is assessed in the RIF-1 and KHT tumours using regrowth delay as an assay. Clamping alone for 120 min gives little or no response, but when RSU1069 is administered 15 min before clamping, large growth delays result. RIF-1 tumours clamped for 90 or 120 min with RSU1069 give cure rates of 12.5% and 37.5% respectively. Less effect with clamping is seen for the other bioreductive agents. The effect of hydralazine with RSU1069 although significant in the RIF-1 tumour, is modest compared to that for clamping. Small enhancements of toxicity are seen with hydralazine in combination with misonidazole in the RIF-1 tumour and mitomycin C in both tumours. The varying effectiveness of these treatments is attributed to several factors which include the level and duration of hypoxia, concentration and contact time of the bioreductive drugs, the microenvironment of the tumour and the nature of the reductive metabolic pathways available in the different tumour cell types.