FIELD OF THE INVENTION The present invention describes a new method whereby magnetic resonance imaging (MRI) of the heart in patients with ischemic heart disease or heart failure can be improved with intravenous (iv) infusion of contrast media that release paramagnetic manganese (Mn) ions and applying a physical or pharmacological stress.

BACKGROUND Ischemic heart disease (IHD) as manifested by angina pectoris, myocardial infarction and ensuing heart failure accounts for the majority of deaths in the western world. In ischemic heart disease atherosclerosis, inflammation and thrombosis narrow coronary arteries gradually or suddenly and reduce blood supply to the heart muscle. If ischemia becomes severe, myocardial infarction develops with impaired cardiac function and loss of cellular viability. Cell death can be prevented if blood flow is restored by drug treatment, percutaneous coronary interventions or surgery before the onset of myocardial infarction.

The seriousness and the epidemic nature of ischemic heart disease call for effective diagnostic methods to identify at an early stage the individual patients at risk and to improve the follow-up of their treatment. Furthermore, there is a need to diagnose heart failure caused by ischemic heart disease or by any other form of heart disease, and to monitor the efficacy of drug treatment for heart failure. It is within this context that imaging modalities and methods are most crucial for overall patient handling and for planning and follow-up of therapy.

Magnetic resonance (MR) imaging (MRI) is a young cardiac imaging technique in rapid development. By intravenous infusion of contrast media that release divalent and paramagnetic manganese (Mn) ions, these ions will be easily taken up into cardiac cells and act as intracellular contrast agents.

Diagnostic imaging and stress testing. In cardiac imaging with use of ultrasound, nuclear medical techniques, and now also magnetic resonance (MRI) (see Nagel E. Left ventricular function in ischemic heart disease. In Higgins CB, de Roos A: Cardiovascular MRI. Lippincott Williams & Wilkins. Philadelphia 2003. pp 190-208), stress tests are particularly useful. Stress tests are designed to make defects in function and perfusion in ischemic regions of the myocardium more visible since these are not easily detected under resting conditions. The stress may be elicited by any kind of physical exercise, like in bicycle ergometry and ECG testing, or by the use of pharmaceuticals. Intravenous infusion of drugs that increase cardiac work or decrease myocardial blood flow to the ischemic region are commonly applied irrespective of imaging modality. Imaging or intravenous injection of a myocardial marker are undertaken at the height of stress.

In the present invention cardiac MRI and intracellular Mn ions are combined optimally with physical or pharmaceutical stress. The Mn ions are strongly paramagnetic and induce changes in basic MR parameters like the relaxation times, longitudinal T1 and transversal T2, or the relaxation rates (R1=I / T1; R2=I / T2) of protons in water and lipids. This again leads to changes in signal intensity and in contrast in images.

Cell Mn ion uptake and properties in cardiac MRI. Mn is a trace metal and Mn ions are important for normal intracellular biochemistry. The use of free Mn ions in the body is however limited by their potential toxicity in the brain and, partly also, in the cardiovascular system. Complexing of these cations with a variety of ligands to delay the release of Mn ions serves to greatly reduce their toxicity whilst retaining their paramagnetic properties.

WO 99/01162 describes some basic principles for how (Mn) containing and Mn ion- releasing compounds can be used as contrast media for the heart. This Mn ion-release-and- uptake based MRI, MnMRI, was shown (Jynge P. et al., Myocardial manganese elevation and proton relaxivity enhancement with manganese dipyridoxyl diphosphate. Ex vivo assessments in normally and ischemic guinea pig hearts. NMR Biomed 1999; 12: 364- 372) to be particularly promising for the diagnosis of myocardial function and viability in ischemic heart disease. This was based on the knowledge that the cell uptake of Mn ions occurs via physiological channels in the cell membrane for calcium (Ca) ions which are main conductors of cell physiology and metabolism. Further, that the intracellular Mn ion retention lasts for hours and that Mn ions induce paramagnetic effects inside the cardiac cell. Accordingly, enhancement of relaxation rates, particularly OfR1, and thus of signal intensity and contrast in T1 weighted images depend greatly on cell membrane Ca channels that allow Mn ions to enter cardiac cells. Since the activity of these Ca channels and Mn ion entry is sensitive to changes in normal physiology and metabolism of the heart, MnMRI reveals fundamental information about myocardial function and viability, and indirectly also of myocardial blood supply as perfusion. An additional factor favouring cell uptake and retention is the function of mitochondria which are responsible for cell energy production and also act as an intracellular sink for cellular Mn.

Concerning imaging of cardiac function in intact animals a further evidence for these basic MnMRI principles was described in WO 99/01162.

After cell uptake into the cytosol, the main cell water pool, Mn ions may distribute to various intracellular organelles and compartments (see Keen CL et al., Frieden (ed) Biochemistry of the Essential Ultra-trace Elements. New York. Plenum Press 1984; pp 89- 132). From the cytosol a large portion of Mn ions enter and are trapped transiently within normally functioning mitochondria. This Mn ion influx occurs via Ca channels in the mitochondrial membrane which are activated by a rise in cytosolic Ca, by an increase in cardiac work and or by an adrenergic stimulus. Thus the fraction of cell Mn content in mitochondria was greatly increased following infusion of the β-adrenergic agonist isoprenaline in isolated perfused rat hearts (see Hunter DR et al., Cellular manganese uptake by the isolated perfused rat heart: a probe for the sarcolemma calcium channel. J MoI Cell Cardiol 1981; 13: 823-832).

Intracellular efficacy of Mn ions. After uptake Mn ions bind to different molecules inside the cell. Binding to small molecules may reduce proton R1 compared to nonliganded Mn aqua ions. However, binding to large molecules like proteins or to other biomolecules enhances proton R1 by reducing the rotational correlation time (TR). Additionally, if the Mn ions are bound at a molecular site that allows fast water exchange and inner plus outer sphere relaxation, further enhancement OfR1 and signal intensity may be expected (see Aime S et al., Relaxometric evaluation of novel manganese(II) complexes for application as contrast agents in magnetic resonance imaging. J Biol Inorg Chem. 2002; 7: 58-67). The overall degree with which the intracellular bulk of Mn ions influence proton relaxation in the myocardium has recently been described (see Jynge P. et al., Manganese ions as intracellular contrast agents: Proton relaxation and calcium interactions in rat myocardium. NMR in Biomedicine, 2003; 16 (2): 82-95) and (see Jynge P. et al., Intracellular manganese ions provide strong T1 relaxation in rat myocardium Experience with manganese dipyridoxyl diphosphate (MnDPDP). Magn Reson Med.). T1 was here measured by MR relaxography (20 MHz, 370C) of myocardial tissue removed from isolated rat hearts perfused with Mn ion-releasing compounds. Then the longitudinal molar relaxivity (Jr1), an index of paramagnetic efficacy, of Mn ions in intracellular water was calculated. Remarkably high T1 values of 60 (mM s)"1 and 57 (mM s)'1, with intracellular Mn ions in Mn enriched hearts were observed . These Jr1 values are about one order of magnitude higher than the in vitro values of Mn ions in pure water and are as high as the most efficient protein-liganded T1 contrast agents yet described by the pharmaceutical industry. These results showed that Mn ions inside cardiac cells bind to specific molecules that enhance T1 relaxation strongly. Mitochondrial proteins and other large biomolecules are the most likely candidates for these effects.

The above shows that the overall cell handling of Mn ions and thus T1 relaxation enhancement depends on both cell membrane Ca channel function and probably also on the overall function of mitochondria. Furthermore, it is amply demonstrated that proton relaxation enhancement also depends on optimal intracellular ligands for Mn ions.

A most important consequence is that after an initial cell uptake into cytosol, the subsequent redistribution of Mn ions from cytosol to mitochondria may enhance the overall Mn-protein- water relaxation properties of cardiac cells. Thus a rise in mitochondrial Mn content may increase T1 weighted signal intensity and contrast. In ischemic or failing myocardium such contrast enhancement may then be reduced or even absent.

SUMMARY OF THE INVENTION

The present invention thus describes a two-step method of dynamic stress of the heart in a human or non-human body, said method comprising separation of administration of a formulation of manganese from the stress procedure by first under resting conditions performing a slow intravenous infusion to said body of a physiologically acceptable manganese ion (Mn) containing and releasing compound in which manganese is slowly released from an organic chelator or ligand, thereafter applying stress and finally subjecting said body to an imaging procedure.

Further, the present invention describes a one-step method of dynamic stress testing of the heart in a human or non-human body, said method comprising administration of a formulation of manganese during the stress procedure by applying stress and simultaneously performing a slow intravenous infusion to said body of a physiologically acceptable manganese ion (Mn) containing and releasing compound in which manganese is slowly released from an organic chelator and finally subjecting said body to an imaging procedure.

Another aspect of the present invention is a method of detecting a disease or disorder in a human or non-human body, said method comprising administering to said body a physiologically acceptable manganese ion (Mn) containing and manganese ion (Mn) releasing compound or salt thereof, applying stress and subjecting said body to an imaging procedure.

Further, the present invention describes a method for assessement of function, viability or perfusion of the cardiac muscle.

Another aspect of the invention is a method wherein said stress is applied by physical exercise or infusion of a pharmacological stressor, whereby subjecting the heart muscle to an increased work load.

Further, the present invention describes a method wherein said imaging procedure involves applying radiofrequency pulse sequences for proton longitudinal (Tl) and transversal (T2) relaxation, for relaxation in the rotating frame reference (Tl D) and for magnetization transfer (MT) that produces changes in signal intensity and thus contrast in images of the myocardium in proton MRI.

Another aspect of the invention is a method wherein said Mn-ion containing and releasing compounds or salts thereof are administrated intravenously at a preferably low dose, preferably 0.1 to 20.0, more preferably 0.1 to 15.0, and most preferably 0.1 to 10.0 μmol/kg body weight.

Further, it is described use of a physiologically acceptable manganese compound or salt thereof for the manufacture of a contrast medium for use in a method as claimed in any one of the claims 1-21.

Other aspects of the present invention are as stated in the present set of claims.

DETAILED DESCRIPTION OF THE INVENTION In particular, the method according to the present invention describes how magnetic resonance imaging (MRI) with paramagnetic Mn ions can be optimized by dynamic stress testing. The heart muscle is thus subjected to an increased work load, so that existing deficiencies in myocardial function and perfusion can be revealed more effectively in MR images.

The tests can be performed in a one- or two-step procedure. The one-step procedure is performed by simultaneous Mn infusion and stress application, hi the preferable two-step procedure the first step comprises intravenous infusion of the Mn ion-releasing contrast medium, and in the second step physical exercise or intravenous infusion of a pharmacological stressor is applied.

Imaging can be performed at all stages, preferably during and most preferably after stress application. The imaging may further be performed immediately after or within 6 hours, preferably within 4 hours, and most preferably within 2 hours. Furthermore the imaging procedure is by MRI, radionuclide scanning, and preferably MRI. hi proton MRI the imaging technique involves applying radiofrequency (rf) pulse sequences for proton longitudinal (T1) and transversal (T2) relaxation, for relaxation in the rotating frame reference (Tip) and for magnetization transfer (MT). These rf techniques produce changes in signal intensity and thus contrast in images of the myocardium. The method is thus based on our new knowledge regarding the remarkably high intracellular relaxivity and extensive macromolecular binding of Mn ions. The method may have a wide clinical application for the assessment of myocardial function and viability, and indirectly also of myocardial perfusion, in cardiac disease.

Particular advantages of the method according to the present invention is that the first step or both steps may be undertaken out-of-magnet. In this way patient safety can be greatly improved and MR examination time and costs can be reduced. A major advantage is slow infusion of a manganese containing contrast agent over 5 to 30 min, preferably 10-20 min, thus avoiding or reducing potential side-effects. Preferably, said Mn-containing and Mn- releasing compound has a dynamic formation constant in the range of 1012 to 1018 which ensures a slow release of manganese ions from the complex. Most preferably, the compound is a slow Mn ion releaser like Mn dipyridoxyl diphosphate (MnDPDP) with a formation constant of 1015'1 (WO 99/011162). Said Mn-containing compounds are to be administrated intravenously at these dosages in preferably 0.1 to 20.0, more preferably 0.1 to 15.0, and most preferably 0.1 to 10.0 μmol/kg body weight.

The disease or disorder that may be detected according to the present invention is ischemic heart disease as angina pectoris, myocardial infarction, myocardial stunning and hibernation or postichemic heart failure, or any other form of heart failure and heart disease.

The method according to present invention may thus be used for assessment of function, viability or indirectly of perfusion of the cardiac muscle.

Stress testing by use of dynamic MnMRI. The dynamic stress tests according to the present invention by use of for example adrenergic agonists of both β and αtype, preferably β-adrenergic agonists like isoprenaline and dobutamine, together with infusion of a Mn ion-releasing contrast agent can exploit these physiological properties greatly, whether the test is undertaken in a combined one- step procedure or, as also proposed below, in a two-step procedure.

The above indicates that mitochondrial function may be essential for the outcome of myocardial stress tests in more than one way. The reasons for a probable improvement of the efficacy of a stress test are based, as stated above, on five potentially additive factors working together in normal myocardium (1-5). These include the activity of cell membrane Ca channels which is the major factor in cell uptake of Mn ions from the extracellular space to the cytosol (l).Further is mitochondrial Mn entry of importance. Enhanced mitochondrial Mn influx via Ca channels may be promoted by a rise in cardiac work during exercise or during an adrenergic stimulus (2). Another important factor is that normally functioning mitochondria act as a cellular sink for Mn ions allowing more Mn ions to be drawn from the extracellular space into the cytosol of cardiac cells and Mn ions to be maintained intracellularly for hours (3). Further of importance are the remarkably strong R1 enhancing properties of Mn-protein adducts compared to free Mn ions in water. This may be compounded by the high protein and biomolecule content in mitochondria (4). And finally, mitochondria occupy as much as 1/3 of the cardiac cell volume and 1/4 of the myocardial tissue volume and have a higher density, closer packing and tighter ordering than in the cytosol (5).

In the one-step procedure the intravenous infusions of a Mn ion-releasing compound and a pharmaceutical stressor might be expected to act simultaneously by promoting cell Mn uptake to cytosol and the simultanous distribution of Mn ions from cytosol to mitochondria. The latter requires the use of a sufficiently high level of cardiac work or dosage of the stressor agent. The simultaneous action of Mn infusion and stress application may draw more Mn ions into cardiac cells by allowing direct transfer of Mn ions into the mitochondrial sink.

Imaging in a one-step procedure can be undertaken simultaneously with Mn infusion and stressor infusion or preferably at a defined interval thereafter due to cell retention of Mn ions. The first option requires that the overall procedure takes place within the magnet. With the other option of delayed imaging of the myocardium with a memory of the conditions at the height of stress, Mn infusion and stress test may be undertaken out-of- magnet which is a great advantage for patient safety and for MR economy reasons.

It is described that during infusion OfMnCl2 in anesthetized mice kept within a MRI instrument, simultaneous infusion of dobutamine enhanced myocardial signal intensity considerably (see Hu TCC et al., Manganese-enhanced MRI of mouse heart during changes in inotropy. MRM 2001; 46: 884-890). However, no attempts were made to identify and separate the effects of a primary Mn uptake into the cytosol of cardiac cells from those of a potential secondary Mn redistribution to mitochondria, and delayed imaging was not applied. Thus by use of principles described in the present invention the one-step procedure is greatly improved and patient safety increased.

WO 2004/054623 Al describes a method in which Mn administration is combined with physical or pharmacological stress in which Mn ion releasers are injected during the final phase of the stress period (peak stress injection). In order to promote a rapid Mn ion uptake into cardiac cells, the core invention involves the use of rapid-to-intermediate Mn ion releasing agents with dynamic formation constants in the range 10 - 10 ' . This combined procedure is different from the present invention which employs a slow infusion of a slow Mn ion releaser and either apply a simultaneous stress (one-step procedure) or subsequent stress application after a resting period (two step procedure). Thus the present invention is more safe for the patient by providing a more gradual release of Mn ions to the blood stream in the one-step procedure and even safer when the Mn infusion precedes the stress induction of myocardial ischemia in the two-step procedure.

The two-step stress test according to the present invention is thus performed by myocardial (Mn) ion loading by slow intravenous infusion of a slow Mn-ion-releasing compound under resting conditions and thereafter to undertake infusion with a stressor or to undertake an exercise-based stress test.

In this way manganese (Mn) infusion can be separated from the later stress and side-effects resulting from both interventions can be greatly reduced. In clinical terms this means that a patient can receive an intravenous infusion of a Mn ion releasing contrast medium out-of- magnet, like in the ward or in the preparation room. Thereafter, while still being out-of- magnet, the patient can receive a pharmacological stressor or undergo a physical stress test by exercise under similar conditions with excellent possibilities for patient monitoring and surveillance.

The method according to the present invention represents many advantages. The imaging may be performed at different stages during the entire procedure. Preferably, imaging can be performed immediately prior to and during the second step to follow the kinetics of signal intensity changes. More preferably, imaging can be undertaken at a chosen time within hours after the second step. Since the time required for intravenous infusion of contrast media matters less, slow Mn ion releasers can be applied as contrast media in a preferably low dose. This again may also improve patient safety and prevent the possibility to avoid potential contrast media related side-effects. Of particular importance is that Mn ions or their releasing agents may promote nitric oxide mediated vasodilation, and hypotension may present a safety hazard to cardiac patients. Whereas hypotension can be balanced by autonomic reflexes in the body, it is necessary to keep the dosage of Mn ion releasers and more so the plasma free concentration of released Mn ions at a preferably low level. In addition to this, it is well known that parenteral nutrition with essential Mn ions over time may cause serious neurotoxic symptoms (e.g. Fell et al. Manganese toxicity in children receiving long-term parenteral nutrition. Lancet 1996;84:295). Although the corresponding tolerable dose of a relatively stable Mn2+ compound, such as MnDPDP, of course is higher, repeated doses may over time represent an increased risk of unwanted side effects.

Furthermore, different pharmacological stressor agents and dosages of stressors can be applied so that these can produce a predictable and graded response. Alternatively, an exercise and ergometry based stress test may also be tuned to produce a graded response. Since the stress manoeuvres accelerate the heart rate to more than twice the normal or higher and also cause changes in blood pressure with the intention to provoke ischemia in a region of interest in the myocardium, it follows that stress itself leads to a considerable safety risk to the patient. It is therefore an advantage of the present invention that the patient can be closely followed up and vital parameters recorded during the stress period and, furthermore, that Mn infusion is undertaken in a slow manner or even completely separated from the stress itself. A further advantage is that the final MR imaging may be planned in advance, take less time and reduce examination costs.

The procedure of undertaking Mn infusion prior to and separated from the following stress, as in the present invention, appears as most safe to the patient. Further, according to the present invention it was surprisingly found that such an approach will be effective in raising signal intensity in MR images, ie. the subsequent stress increase the efficacy in MR terms. As shown in the examples below the final stress itself caused a substantial, 20-45 %, rise in the the Rl values above what was obtained by slow Mn- infusion (5-15 min) only. Thus the method according to the invention confers both an added safety and an added efficacy. For both the one-step and the two-step tests imaging of the heart, especially of myocardium of right and left ventricle and of septum, will reveal reduced signal intensity enhancement in the ischemic regions where myocardial function and perfusion are jeopardized. Similarly in failing hearts cell function may be reduced regionally or globally and the accompanying changes can be assessed by imaging. For both ischemic heart disease and heart failure the myocardial stress tests described here are particularly suitable for the early diagnosis and the long term follow-up of disease and of treatment.

The method according to the present invention is particularly amenable for T1 weighted imaging. Transversal (T2) proton relaxation is a much faster process, but shows a parallel response to Mn ion enrichment in rat myocardium. The present invention therefore includes the potential use of T2 weighted imaging in cardiac MnMRI. Other included imaging techniques involve relaxation in the rotating frame reference (T1 p) and magnetization transfer (MT). The invention is not limited to any particular instrument or hardware techniques for MRI, but requires the use of Mn containing contrast media that release Mn ions slowly after intravenous infusion.

The present invention of stress tests as above presented for MRI is also applicable for radionuclide scanning with use of different Mn isotopes and slow infusion of the marker as proposed for MRI. In nuclear imaging of the myocardium the Mn-52m isotope has proved effective in animals (see Atkins HL et al., Myocardial positron tomography with manganese-52m. Radiology 1979; 133: 769-774).

The following examples are for illustrative purposes only and are not intended, nor should they be constructed as limiting the invention in any manner. Those skilled in the art will appreciate that variations and modifications can be made without violating the spirit or scope of the invention.

EXAMPLES. Experiments were undertaken with use of isolated perfused rat hearts. Example 1 The heart isolated from a Wistar rat was perfused continually through the coronary arteries with normal perfusion medium (Krebs buffer) at normal flow rate (10 ml/min) at 37° C. The perfused heart was kept within the center of a 20 MHz relaxometer and was subjected to the following perfusion sequence intervals: • 20 min control perfusion. • 15 min perfusion with Krebs buffer plus MnCl2 7.5 μM (Mn wash-in). • 15 min perfusion with Krebs buffer only (Mn wash-out). • 15 min perfusion with Krebs buffer plus isoprenaline 2.0 μM. Mn ions were taken up into the heart cells during the wash-in period and all residual extracellular Mn ions were removed during the following wash-out.

T1 was measured by an inversion recovery method directly in the myocardial tissue. Analysis by a monoexponential function revealed these T1 values during perfusion: • Control perfusion period 1100 ms. • After step 1, Mn loading with wash-in and wash-out, 935 ms. • After step 2, isoprenaline infusion, 760 ms.

This experiment shows a rise in intracellular Mn content after step 1, leading to T1 reduction by 15 % and R1 elevation by 18 %. This reflects the overall efficacy of cardiac cell uptake of Mn ions under resting conditions and the efficacy of Mn ions, most probably located in the cytosol.

In step 2, a strong /3-adrenergic stimulus was induced and the work load on the heart was increased by isoprenaline. In comparison to values during the control period T1 was now reduced by 31 % and R1 was increased by 45 %. Thus the stress in step 2 caused a further enhancement in relaxation compared to step 1, and now T1 decreased by 16 % and R1 increased by 30 %. These changes took place without any further rise in the total cell content of Mn ions. This amply demonstrates the efficacy of Mn ion redistribution to mitochondria which have a particularly high content of proteins and other biomolecules that may form exceedingly relaxation-effective complexes with Mn ions. Example 2. Hearts isolated from Wistar rats were perfused continually through the coronary arteries with normal perfusion medium (Krebs buffer) at normal flow rate (10 ml/min) at 370C. The following perfusion sequences were applied: • 20 min control perfusion. • 5 min perfusion with Krebs buffer plus MnCl2 25 μM (Mn wash-in). • 15 min perfusion with Krebs buffer only (Mn wash-out). • 5 min perfusion with Krebs buffer plus isoprenaline 2.0 μM.

At the end of experiments ventricular myocardium was rapidly removed for measurement Of T1 by an inversion recovery method at 20 MHz and 37° C. Thereafter Tl data were analyzed by use of a biexponential function revealing the apparent intracellular (ic) and extracellular (ec) relaxation times (Tl-zc, Tl-ec) and relaxation rates (Rl -zc, Rl -ec), as described previously (Jynge P. et al., Manganese ions as intracellular contrast agents: Proton relaxation and calcium interactions in rat myocardium. NMR in Biomedicine, 2003; 16 (2): 82-95).

Three groups (n=4) were included: • Control, with tissue removed after control perfusion. • Manganese, with tissue removed after Mn wash-out. • Manganese + Stress, with tissue removed after perfusion with isoprenaline.

Table 1. Intracellular and extracellular relaxation times and rates following manganese uptake and following manganese uptake plus pharmacological stress.

* significant from control group, ** significant from manganese group Mean values ± SD are shown, hitergroup differences assessed by Students t-test. Tl-zc was reduced significantly in two steps compared to control: by 52 % in step 1 with a rise in the contents of Mn ions in cardiac cells; and following stress in step 2 there was a further reduction by 17 %. Changes in Tl-ec were less prominent. A 24 % reduction in Tl- ec followed stepl is consistent with slow water exchange across the cardiac cell membrane.

The experiments show that with specific techniques to separate resonances from respectively intracellular and extracellular water compartments, it is possible to demonstrate the effects of an applied stressor to enhance intracellular relaxation (Rl -ic) rates, most probably caused by distribution of Mn ions from cytosol to mitochondria.