Educational package Cardiovascular Safety Pharm.

Cardiovascular safety pharmacology in the minipig

Michael Markert Senior Principal Scientist

May 2011

1

2

Cardiovascular safety pharmacology in the minipig This document describes the cardiovascular system of minipigs and the experimental design when performing cardiovascular safety pharmacology in this species. Corresponding author Michael Markert

Senior Principal Scientist Boehringer Ingelheim Pharma GmbH & Co KG

Ellegaard Göttingen Minipigs A/S Soroe Landevej 302 DK-4261 Dalmose

Tel.: +45 5818 5818 Fax: +45 5818 5880

E-mail: ellegaard@minipigs.dk www.minipigs.dk

3

Table of Contents Introduction ................................................................................................................................................ 4 Cardiovascular system of the minipigs ............................................................................................. 5

Surgery: ................................................................................................................................................................... 5 Experimental design ................................................................................................................................ 6

Intrinsic heart rate of the Göttingen Minipig ........................................................................................... 6 Diurnal and feeding effects on haemodynamics .................................................................................... 6 Diurnal effects on body temperature ......................................................................................................... 7 Impact of model on hemodynamic parameters ...................................................................................... 7 Value of myocardial contractility assessment ........................................................................................ 7 ECG evaluation ..................................................................................................................................................... 7

Discussion .................................................................................................................................................... 8 Conclusion ................................................................................................................................................... 8 Reference List ............................................................................................................................................ 9 Figures and tables ................................................................................................................................. 13

Although every effort has been made to ensure that the information contained is accurate no liability for its use is accepted by the author or by the company that published this booklet.

4

Introduction Pigs are being used increasingly in biomedical research, particularly in pharmacological and toxicological testing. Preclinical testing for tolerability and safety is required in both rodents (usually the rat) and in a second, non-rodent species22, 47 The dog has been the most commonly used non-rodent species for toxicity studies.38 Nevertheless, there are cases where the dog is not suitable. This can be due to dog-specific drug malabsorption, metabolism or any other atypical response that can impact drug safety. Furthermore, a pharmacodynamic response to a given agent is needed to qualify a species for toxicological testing. If a given pharmacodynamic effect cannot be demonstrated in the dog, alternative species must be considered. The primate has been widely used when the dog has shown to be inappropriate, but animal availability and ethical considerations limit the use of primates. Thus, further options for non-rodent safety pharmacology and toxicity testing are needed. The domestic pig, due to its large size as an adult (its bodyweight can exceed 250 kg) has not been a commonly used animal for research purposes. In 1949, the first attempts were made to develop smaller pigs (“miniature pigs”) at the University of Minnesota.8 Today, several different minipig breeds are available and suitable for laboratory use. The Göttingen Minipig is the smallest of the minipig breeds available for research and it reaches a weight of only 20–35 kg when it is fully grown. The other minipig breeds grow significantly larger. The Göttingen Minipig has a consistent phenotype and exhibits a friendly behaviour suitable for handling in a laboratory environment.38 In the meantime, the use of pigs for biomedical research has demonstrated that they offer potential advantages when certain comparisons to humans are desirable. For instance, the minipig has a similar heart-to-body weight ratio and a coronary artery distribution similar to humans. Moreover, cardiac anatomy, metabolism and electrophysiology in pigs are comparable to humans.6, 12, 21, 23, 27, 30, 39, 58 Initial analyses show that the major myocardial ion currents responsible for the human myocardial action potential are also present in the Göttingen Minipig.37 At the same time, the use of the dog as an experimental animal has drawn criticism due to its role as a companion animal.6, 27, 30, 55 In contrast, the pig is still viewed primarily as a farm animal with fewer emotional and ethical implications. Consequently, minipigs are being used more extensively for not only cardiovascular biomedical research,8, 32 but as an alternative non-rodent species for toxicological studies.30, 38 In spite of this trend, scientific literature has little cardiovascular and electrocardiographic reference data from minipigs. Most of the published hemodynamic data was collected using invasive measurement techniques.4 ECG data from minipigs are available from animals restrained in a sling with external limb leads.16, 17, 36, 46, 54 To our knowledge there are no published data for left ventricular pressure from unrestrained minipigs. LVP measurements have been done only in anaesthetised animals5, 15, 34, 50, 56 or in awake, sling-restrained animals.56 LVP measurements are particularly useful in that the derivative of LVP, i.e. LVdP/dt, is a well-established parameter for the assessment of cardiac contractility.31, 41, 42 Drug-induced prolongation of the QT interval (a marker for a delay in cardiac repolarisation) of the electrocardiogram (ECG) has drawn increasing attention from regulatory agencies and the pharmaceutical industry, and detecting these effects has had a great impact on drug discovery and development.18 The delayed repolarisation, frequently attributable to blockade of the rapidly activating delayed rectifier potassium channel, IKr, favours the genesis of early after-depolarisation (EAD), which can initiate arrhythmia.57 Additionally, the prolongation of the QT interval by drugs is often associated with increased heterogeneity of cardiac repolarisation,1 a substrate for a re-entrant mechanism responsible for sustained arrhythmia, and preclinical testing for tolerability and safety is required in both rodents (usually the rat) and in a second, non-rodent species.22, 47 Nevertheless, there are some ‘safe’ drugs that inhibit IKr and cause QT prolongation without inducing TdP.18, 48 In any case, it is desirable that models used for testing new drugs for potential effects on the QT interval are shown to have the sensitivity required for detecting subtle changes. In this study, we have tested a known hERG-blocking agent, Moxifloxacin, as well as a ß-adrenoceptor blocker, Propranolol, in the minipig model. Moxifloxacin has been recommended as a positive control for clinical trials assessing QT prolongation potential28 as it produces a consistent, highly reproducible effect on QT interval duration. Propranolol is a non-selective ß-adrenoceptor blocker mainly used in the treatment of hypertension. It is expected to decrease heart rate and reduce myocardial contractility.35

5

Cardiovascular system of the minipigs Due to anatomical and physiological similarities between pigs and humans, the pig is suitable for medical research. Already at 25–30 kg bodyweight, the cardiovascular system of pigs is regarded as sufficiently matured to model the cardiovascular system of humans.24 Pigs conform with humans in key parameters of the cardiovascular system: relative heart weight, cardiac output, systolic and diastolic blood pressure, coronary anatomy, blood volume, haematocrit and haemoglobin content.24 In pigs, the right ventricle forms a sharp angle with the sternum, whereas in dogs it is parallel to the sternum. Body surface ECGs from minipigs are similar to those from Cetacea, pinnipeds and Equidae, but different from primates and dogs. The form of the body surface potentials during QRS indicates that the pathway of ventricular depolarisation in minipigs must be different from dogs.

Surgery Surgery in minipigs is quite comparable to surgery in dogs, with excellent wound healing. However, it is important to remember that dogs and minipigs have differences. Among other things the skin of minipigs is different from dog skin and with skin sutures the thickness of the minipig skin should be taken into consideration. Procedures such as endotracheal intubation of minipigs require practice. When performing thoracotomy, it is recommended to remove one rib (if access to the tip of the left ventricle is needed, the fifth or sixth rib is preferable).

6

Experimental design The study design can have a big influence on data quality, and it is crucial (and a valuable return on one’s investment) to spend sufficient time adapting the animals to the laboratory environment. It could be advantageous to separate the routine housing environment from the telemetry laboratory. This means that on the first day of a study, the animals are transferred to the pens set up for the telemetry measurements and (ideally) housed in pairs. They could be moved in the early morning and left undisturbed for the remainder of the study, except where mentioned below. An initial period of 30–45 minutes allows the minipigs to acclimatise to the measurement pens, although earlier training ensures that they are well familiar with these pens. After this acclimatisation period, the measurements can start. Minipigs like to differentiate between their preferred sleeping area and a separate area for defecation. It can be advantageous to add a space with an infrared light (for heating) above the resting area. However, one should bear in mind that if the telemetry implants are placed subcutaneously only, this could lead to artificial body temperature measurements. The hour preceding the administration procedure can be taken as a control period for baseline measurements. After this hour, the minipigs are taken out of their pens and the testing article can be administered orally with a special dosing catheter and then returned to their pens and left undisturbed for the remainder of the data collection period. Experiments should be performed using a cross-over design so that each minipig receives each treatment and could thereby serve as its own control.

Intrinsic heart rate of the Göttingen Minipig The high level of training, as well as the excellent health status of Göttingen Minipigs, is evident in the low intrinsic average heart rate (HR) of 56 bpm during daylight hours (Fig. 3). Indeed, the low heart rates are surprising if one compares the currently available haemodynamic data in minipigs obtained with invasive measurements, or data from dogs when used for similar studies. Kano29 reported values of 72–76 bpm in 17 kg, freely moving miniature pigs (but not Göttingen Minipigs). The HR in resting miniature pigs was 80 ± 3.5 bpm in investigations from Kuwahara.32 Beglinger and Becker4 report heart rates of 103 ± 14 bpm in sling-restrained Göttingen Minipigs (~20 kg). Thus, we attribute the very low heart rates seen in our laboratory to the low stress levels achieved using well-trained animals in a laboratory environment where measurements have been taken to reduce unnecessary stress factors. Furthermore, a low variability of the measured data is seen throughout the studies which could also be attributed to both training status and an optimised study environment. Low variability in the measurement data is extremely important for detecting and quantifying potential drug-induced effects. Low variability leads to an optimisation of the statistical power of an experimental model and the possibility of reducing the number of animals used. The HRs of our animals are quite comparable to data from freely moving well-trained and group-housed Labrador dogs.31 This indicates that factors other than species are decisive for the heart rates observed and this likely reflects the level of acclimatisation to the laboratory environment, the low-stress experimental procedure and the fact that no restraint is needed.

Diurnal and feeding effects on haemodynamics The diurnal rhythms of minipigs’ heart rate have been investigated by Kuwahara et al.32 They found that heart rate in the daylight phase was higher than in the dark phase when the animals were housed singly, but minipigs housed in pairs had no diurnal variation. Another group29 reported no marked changes in heart rate between periods of daylight and darkness. The present data differ from both of these findings: we see a significant increase in HR at night when the animals were fed shortly before the start of the dark phase. Intuitively, one would expect HR to decrease at night with an increase in parasympathetic activity. Examination of videos taken at night indicate that the minipigs were indeed sleeping most of the time (main sleeping period: 11 p.m. to 5 a.m.), with only short periods of wandering but with no signs of excitement. When the study design was altered to eliminate feeding at 7 p.m., the HR at night was 51 bpm being comparable to the daylight value. These findings indicate a dependency of the dark phase increase in HR on feeding. The pig is known to digest slowly, taking at least 24 hours to empty the digestive tract after a meal (depending on the type of diet).26 A study by Lückmann40 investigated HR in domestic swine and reported an 8–10% increase in heart rate after feeding. The postprandial HR increase can be larger with voluminous diet.39 This phenomenon may be due to autonomic nervous fibres located in the wall of the digestive tract that react to the filling of the gastrointestinal tract. Nevertheless, it is clear that the effect of feeding on HR can extend beyond 3 hours and that this must be taken into account when designing studies involving Göttingen Minipigs.

7

Diurnal effects on body temperature The body temperature of the minipigs also demonstrated differences between daytime and night time. During daytime (before feeding) body temperature is 36.7 °C in our minipigs, whereas during the night it increased to 37.2 °C. The body temperature cited in literature for miniature pigs ranges from 37–38 °C,7 37.7 ±0.2 °C20 to 38.2–39.9 °C.58 It should be mentioned that the temperatures measured using the full implant telemetry system are not actually core temperatures, since the temperature module is in the transmitter of the ITS “T27-F” device located in the abdominal muscle layers of the left flank. However, with the recently introduced T27-G implant series it is possible to have a temperature probe in the abdomen and therefore a true core temperature. The increased heart rate during the night phase may contribute to the increase in body temperature. Temperature also increased in the experiment without feeding in which heart rate is not affected, indicating that factors other than elevated heart rate could be responsible. One possibility is that the prolonged periods of lateral recumbency during sleep may lead to an artificial effect on temperature, since there is an overhead warming lamp and the floor is heated for the well-being of the animals. Systolic and diastolic arterial blood pressure is remarkably stable throughout the protocol without a diurnal effect with or without feeding (Fig. 1). Values for arterial blood pressure from minipigs range from 135 ± 12 to 160 ± 11 for SAP and 88 ± 4 to 96 ± 14 for DAP.4, 10, 21, 39 However, none of the previous studies used freely moving minipigs. Thus, the slightly higher pressures reported previously may be attributable to the restraint used when taking the measurements.

Impact of model on hemodynamic parameters Since blood pressure varies with age and body weight4, 19 and various blood pressure measurement techniques (direct–indirect, awake–anesthetised) are available, comparisons should be done with caution. Values return to pre-dose values very quickly (in about 20 minutes) after dosing indicating that our minipigs were well accustomed to the study conditions. This contrasts sharply with the effects observed upon feeding, which effects lasted 3–4 hours. Left ventricular pressure values in freely moving Göttingen Minipigs are presented in the publication from Stubhan et al.52 for the first time (Fig. 2). Glodek and Oldigs21 reported LVP measurements with pressure-tip catheters in awake, but sling restrained, animals and reported a value of 142 ± 14 mm/Hg. In anesthetised miniature pigs, a LVP of 58 ± 2 mm/Hg was reported in juvenile Yucatan Miniature swine whereas 158.5 ± 36.5 was observed in adult Göttingen Minipigs. This substantial difference is likely explained by the markedly different models used. The mean value measured in our minipigs over 24 hours is 111 ± 15 mm/Hg. This value remained extremely stable except after dosing and feeding.

Value of myocardial contractility assessment The first derivative of LVP over time, LVdP/dt, is a familiar index of myocardial contractility (ref. Braunwald, Ross, etc.). As many drugs influence myocardial contractility, this parameter provides important information for the evaluation of drug candidates and should therefore be included in the testing for cardiovascular side effects. Changes in the inotropic state of the heart can also affect systemic arterial blood pressure and cardiac output. There is a clear dependency of LVdP/dt on HR and we have previously described this phenomenon in Labrador dogs, Rhesus monkeys, Cynomolgus monkeys and minipigs.42 We see a difference in LVdP/dt between the daylight and the darkness periods, which could be related to the increase observed in HR. Values for LVdP/dt similar to that observed in our conscious minipigs were observed in anaesthetised Göttingen Minipigs (2492 ± 350 mmHg/s;5). However, sling-restrained Göttingen Minipigs had higher levels of LVdP/dt, presumably due to higher heart rates and possible stress associated with the restraint (3821 ± 787 mmHg/s;21).

ECG evaluation Reported ECG investigations in the Göttingen Minipig have been done in relatively young (< 198 days) and sling-restrained animals.4, 17, 21, 46 Therefore, all the reported values here are in ranges far below our ECG results. The spontaneous changing of the polarity of the T-wave as we see it regularly was also apparent in other experiments17 and should presumably be seen in the context of vegetative influences. QT is known to be strongly correlated to the RR interval. Various algorithms have been suggested to describe the relationship between QT and the RR interval or heart rate.3, 9, 13, 25, 43, 44, 45, 51 Based on existing data, we conclude that no correction model is applicable to all individuals, suggesting the use of an individual correction of the QT interval for heart rate changes (Table 1).

8

Discussion Good signal quality can be obtained in the conscious minipig instrumented as described for the telemetric collection of hemodynamic and ECG data. Minipigs have stable haemodynamic parameters over long observation periods with a remarkably low intrinsic heart rate. This suggests that the testing environment can be optimised to minimise the stress induced in the animals, and attempts to exclude disturbing external stimuli are helpful. This type of result is very comparable to the data quality from trained Labrador dogs31 and the data from a previously published paper describing the minipig model.52 (Table 2). After oral dosing, the haemodynamic parameters can return quickly (after approximately 20 minutes) to pre-dose baseline values, further indicating that the animals are well adapted to the study conditions and are unstressed. Nevertheless, it is still possible to detect compounds that have beta-blocking properties. Propranolol, as one example, was selected for testing in this model as a prototypical non-selective beta-adrenoceptor blocker and was expected to block sympathetic input to the heart with anticipated decreases in heart rate and myocardial contractility. Indeed, Propranolol at doses of 3, 10 and 20 mg/kg caused a substantial dose-dependent decrease in HR and myocardial contractility and shortened the log-adjusted QT interval. Such effects on heart rate have been well documented in minipig experimental models,29 and it has been reported that Propranolol shortened the QTc interval in healthy volunteers.11, 53 Results from the pilot study noted that when using higher doses (e.g. 30 mg/kg), the minipigs showed adverse effects (vomiting, general discomfort and restlessness) including arrhythmias (2nd degree AV-Block) (Fig. 5). Because it causes a clear QT interval prolongation in clinical studies and has a well-known PK profile, Moxifloxacin has been recommended as a positive control for clinical trials assessing QT prolongation potential.28 It has been reported that QT interval at rest (i.e. RR = 1000 ms) significantly increased from 379 ± 24 ms with a placebo to 394 ± 33 ms with 400 mg of Moxifloxacin (P < 0.05) and to 396 ± 28 ms with 800 mg of Moxifloxacin (P < 0.05). These effects corresponded to increases of 4.0% ± 5.1% and 4.5% ± 3.8%, respectively.14 In clinical use, a peak level of 4.5 μg/ml (5.6 μM unbound) was reported after daily doses of 400 mg for 10 days.2 In 787 patients in phase 3 clinical trials, daily repeated doses of 400 mg of Moxifloxacin increased the QT interval by 6 ± 26 ms (mean ± s.d.) (Prod Info Avelox®, 2003). In over 20 specially designed studies in healthy volunteers using standardised approaches and methods, single oral doses of 400 mg of Moxifloxacin consistently produced a mean increase in the QTc interval ranging from 5 to 10 ms.49 Our study clearly demonstrated that the concentration-dependent effect of Moxifloxacin in the conscious minipig is comparable to clinical outcomes (Fig. 4). With Moxifloxacin, clinically relevant QTc prolongation is seen at a concentration that produces ~10% inhibition of the hERG current, similar to many IKr/hERG blockers (e.g. dofetilide, E-4031, cisapride, terfenadine, and risperidone). In this telemetry minipig study, the serum concentrations @ 7h were 4.1, 10.6, and 19.6 μmol (total drug concentration after dosing of 30 mg/kg, 100 mg/kg and 300 mg/kg, respectively). These exposures were associated with mean maximal QT prolongation of 6, 17, and 22% (vs placebo), respectively.

Conclusion In conclusion, the trained Göttingen Minipig appears to be suitable for cardiovascular safety pharmacology studies.

9

Reference List 1. Antzelevitch, C. (2004)

Arrhythmogenic mechanisms of QT prolonging drugs: Is QT prolongation really the problem? Journal of Electrocardiology, Vol. 37 (SUPPL.), 15-24.

2. Balfour, J.A.B., Wiseman, L.R. (1999) Moxifloxacin Drugs, 1999, Vol. 57, no. 3, 363-374.

3. Batey, A. J. and C. P. A. Doe. (2002) A method for QT correction based on beat-to-beat analysis of the QT/RR interval relationship in conscious telemetred beagle dogs Journal of Pharmacological and Toxicological Methods, 48(1), 11-19.

4. Beglinger, R., et al. (1975) The Gottinger Minipig as a laboratory animal Research in Experimental Medicine, 165(3), 251-63.

5. Benharkate, M., et al. (1993) Hemodynamic parameters of anesthetized pigs: A comparative study of farm piglets and Gottingen and Yucatan miniature swine Laboratory Animal Science, 43(1), 68-72.

6. Bollen, P. and L. Ellegaard. (1997) The Göttingen minipig in pharmacology and toxicology Pharmacology and Toxicology, Vol. 80.Suppl. 2, 3-4.

7. Bollen, P. J. A., Hansen, A. K., and Rasmussen, H. J. (2000) The Laboratory Swine A Volume in The Laboratory Animal Pocket Reference Series. CRC Press.

8. Bustad, L. K. and R. O. McClellan. (1968) Miniature swine: development, management, and utilization Laboratory animal care 18(2).

9. Chiang, Alan Y., David L. Holdsworth, and Derek J. Leishman. (2006) A one-step approach to the analysis of the QT interval in conscious telemetrized dogs Journal of Pharmacological and Toxicological Methods 54(2), 183-88.

10. Cimini, C. M. and E. J. Zambraski. (1985) Non-invasive blood pressure measurement in Yucatan miniature swine using tail cuff sphygmomanometry Laboratory animal science, 35(4), 412-16.

11. Clozel, J.P., Danchin, N., Genton, P., Thomas, J.L., Cherrier, F. (1984) Effects of Propranolol and of verapamil on heart rate and blood pressure in hyperthyroidism Clin Pharmacol Ther, 36(1), 64-9.

12. Crick, S. J., et al. (1998). Anatomy of the pig heart: Comparisons with normal human cardiac structure Journal of Anatomy, 193(1), 105-19.

13. Davey, P. (2002) How to correct the QT interval for the effects of heart rate in clinical studies Journal of Pharmacological and Toxicological Methods, 48(1), 3-9.

14. Demolis, J.L., Kubitza, D., Tenneze, L., Funck-Brentano, C. (2000) Effect of a single oral dose of Moxifloxacin (400mg and 800mg) on ventricular repolarization in healthy subjects Clinical Pharmacology and Therapeutics, 68, 658-666.

15. Diederen, W. and R. Kadatz. (1981) Effects of AR-L 115 BS, a new cardiotonic compound, on cardiac contractility, heart rate and blood pressure in anaesthetized and conscious animals Arzneimittel-Forschung/Drug Research 31(1A), 146-50.

16. Dukes, T. W. and M. Szabuniewicz. (1969) The electrocardiogram of conventional and miniature swine (Sus scrofa) Canadian Journal of Comparative Medicine 33(2), 118-27.

17. Eckenfels, A. and S. Schuler. (1988) On the normal electrocardiogram of the conscious mini-pig Arzneimittel-Forschung/Drug Research 38(2), 253-59.

10

18. Fermini, B., Fossa, A.A. (2004) Pre-Clinical Assessment of Drug-Induced QT Interval Prolongation. Current Issues and Impact on Drug Discovery Annual Reports in Medicinal Chemistry, 39, 323-334.

19. Gauvin, D. V., et al. (2006) Electrocardiogram, hemodynamics, and core body temperatures of the normal freely moving laboratory beagle dog by remote radiotelemetry Journal of Pharmacological and Toxicological Methods 53(2), 128-39.

20. Georgiev, S., A. Schoen, and M. Merkenschlager. (1972) Effect of various environmental temperatures and humidities of some physiologic parameters of the Göttinger minipig in various stages of growth. 3. Adults Berliner und Munchener Tierarztliche Wochenschrift, 85(21), 409-13.

21. Glodek, P. and Oldigs, B. (1981) Das Göttinger Miniaturschwein Beglinger, R. and Becker, M. 6. Herz und Kreislauf. Schriftenreihe Versuchstierkunde 7. Berlin und Hamburg,

22. Guth, B.D., Germeyer, S., Kolb, W., & Markert, M. (2006) Developing a strategy for the nonclinical assessment of proarrhythmic risk of pharmaceuticals due to prolonged ventricular repolarization Journal of Pharmacological and Toxicological Methods, 49 (3), 159-169.

23. Henke, J. Brill, T. and Feußner, H. (2005) Experimentelle Medizin: Vom Tiermodell zur Computeranimation 11, 329-35.

24. Hiebl, B., Müller, C., Hünigen, H., Gemeinhardt, O., Plendl, J., Jung, F., Hamm, B., Niehues, S.M. (2010) Gross anatomical variants of the vasculature of the Göttingen Minipig™ Applied Cardiopulmonary Pathophysiology 14: 236-243, 2010.

25. Holzgrefe, Henry H., et al. (2007) Novel probabilistic method for precisely correcting the QT interval for heart rate in telemetered dogs and cynomolgus monkeys Journal of Pharmacological and Toxicological Methods, 55(2), 159-75.

26. Hossain M., et al., (1990) Gastrointestinal transit of nondisintegrating, nonerodible oral dosage forms in pigs Pharm Res. 7(11): 1163-6

27. Hughes, H. C. (1986) Swine in cardiovascular research Laboratory animal science 36(4), 348-50.

28. ICH (2005) The nonclinical evaluation of the potential for delayed ventricular repolarization (QT interval prolongation) by human pharmaceuticals (S7B)

29. Kano, M., Toyoshi, T., Iwasaki, S., Dato, M., Shimizu, M., & Ota, T. (2005) QT PRODACT: Usability of miniature pigs in safety pharmacology studies: Assessment for drug-induced QT interval prolongation Journal of Pharmacological Sciences, 99(5), 501-11.

30. Khan, M. A. (1984) Minipig: Advantages and disadvantages as a model in toxicity testing Journal of the American College of Toxicology, 3(6), 337-42.

31. Klumpp, A., Trautmann, T., Markert, M., & Guth, B. (2006) Optimizing the experimental environment for dog telemetry studies Journal of Pharmacological and Toxicological Methods, 54(2), 141-49.

32. Kuwahara, M., et al. (1999) Power spectral analysis of heart rate variability for assessment of diurnal variation of autonomic nervous activity in miniature swine Laboratory animal science, 49(2), 202-08.

33. Kuwahara, M., et al. (2004) Effects of pair housing on diurnal rhythms of heart rate and heart rate variability in miniature swine Experimental Animals, 53(4), 303-09.

34. Lal, B., B. K. Bhattacharya, and N. K. Dadkar. (1981) Cardiovascular and antihypertensive activity of the new vasodilator 9,10-dimethoxy-2-mesitylimino-3-methyl-3,4,6,7-tetrahydro-2H-pyrimido (6,1-a) isoquinolin-4-one hydrochloride (HL 725) IRCS Medical Science, 9(4), 325-26.

11

35. Lalonde, R.L., Pieper, J.A., Straka, R.J., Bottorff, M.B., Mirvis, D.M. (1987) Propranolol pharmacokinetics and pharmacodynamics after single dose and at steady state Eur J Clin Pharmacol, 33, 315-318.

36. Larks, S. D., R. B. Wescott, and G. G. Larks. (1971) Electrocardiographic studies of miniature swine: normal values Laboratory animal science, 21(4), 553-57.

37. Laursen, M. et al., (2010) Characterization of cardiac repolarization in the Göttingen minipig Journal of Pharmacological and Toxicological Methods, 63, 186-195.

38. Lehmann, H. (1998) The minipig in general toxicology Scandinavian Journal of Laboratory Animal Science, 25.Sppl. 1, 59-62.

39. Leucht, W., Gregor, G., and Stier, H. (1982) Einführung in die Versuchstierkunde. Band VI: Das Miniaturschwein VEB Gustav Fischer Verlag Jena.

40. Lückmann, J. (1968) Die Herzfrequenz des Schweines Hannover, Veterinary Dissertation.

41. Markert, M., Klumpp, A., Trautmann, T., & Guth, B., (2004) A novel propellant-free inhalation drug delivery system for cardiovascular drug safety evaluation in conscious dogs Journal of Pharmacological and Toxicological Methods, 50(2), 109-19.

42. Markert, M., Klumpp, A. Trautmann, T., Mayer, K., Stubhan, M., & Guth, B. (2007) The value added by measuring myocardial contractility 'in vivo' in safety pharmacological profiling of drug candidates Journal of Pharmacological and Toxicological Methods, 56(2), 203-11.

43. Matsunaga, T., et al. (1997) QT corrected for heart rate and relation between QT and RR intervals in beagle dogs Journal of Pharmacological and Toxicological Methods, 38(4), 201-09.

44. Meyners, M. & Markert, M. (2004) Correcting the QT Interval for Changes in HR in Pre-clinical Drug Development Methods of Information in Medicine, 43, 445-50.

45. Miyazaki, H. and M. Tagawa. (2002) Rate-correction technique for QT interval in long-term telemetry ECG recording in beagle dogs Experimental Animals, 51(5), 465-75.

46. Nahas, K., P. Baneux, and D. Detweiler. (2002) Electrocardiographic monitoring in the Göttingen minipig Comparative Medicine 52(3), 258-64.

47. Pugsley, M.K., & Curtis, M.J. (2006) Safety pharmacology in focus: New methods developed in the light of the ICH S7B guidance document Journal of Pharmacological and Toxicological Methods, 54(2), 94-98.

48. Redfern, W.S., Carlsson, L., Davis, A.S., Lynch, W.G., MacKenzie, I., Palethorpe, S., Siegl, P.K.S., Strang, I., Sullivan, A.T., Wallis, R. (2003)

Relationships between preclinical cardiac electrophysiology, clinical QT interval prolongation and torsade de pointes for a broad range of drugs: evidence for a provisional safety margin in drug development Cardiovascular Research, 58 (1), 32-45.

49. Shah, R. (2005) Drugs, QT Interval Prolongation and ICH E14: The Need to Get it Right Drug Safety, 28(2), 115-125.

50. Smith, A. C., F. G. Spinale, and M. M. Swindle. (1990) Cardiac function and morphology of Hanford miniature swine and Yucatan miniature and Micro swine Laboratory animal science, 40(1), 47-50.

51. Spence, S., et al. (1998) The heart rate-corrected QT interval of conscious beagle dogs: A formula based on analysis of covariance Toxicological Sciences, 45(2), 247-58.

52. Stubhan, M., Markert, M., Mayer, K., Trautmann, T., Klumpp, A., Henke, J., Guth, B. (2008) Evaluation of cardiovascular and ECG parameters in the normal, freely moving Göttingen Minipig Journal of Pharmacological and Toxicological Methods, 57, 2002-211.

12

53. Sundaram, S., Carnethon, M., Polito, K., Kadish, H., Jeffrey, J. (2008) Autonomic effects on QT-RR interval dynamics after exercise Am J Physiol Heart Circ Physiol; 294: H490 - H497.

54. Suzuki, A., et al. (1998) Establishment of a 24-hour electrocardiogram recording system using a Holter recorder for miniature swine Laboratory Animals, 32(2), 165-72.

55. Tumbleson, M. E. Swine in biomedical research. Rollin, B. E. (1986) Moral, social and scientific aspects of the use of swine in research New York, Plenum Press 29-37.

56. Vainio, O. M., B. C. Bloor, and C. Kim. (1992) Cardiovascular effects of a ketamine-medetomidine combination that produces deep sedation in Yucatan mini swine Laboratory animal science, 42(6), 582-88.

57. Zabel, M., Woosley, R.L., Franz, M.R. (1997) Is dispersion of ventricular repolarization rate dependent? PACE - Pacing and Clinical Electrophysiology, 20 (10 I), 2405-2411.

58. Zambraski, E. J. and B. Fuchs. (1980) Resting metabolism of Yucatan miniature swine Laboratory animal science, 30(1), 51-53.

13

Figures and tables

PLACEBO

-2 0 2 4 6 8 10 12 14 16 18 20 22 2450

60

70

80

90

100

110

120

130

140

150

160

170

180

190

Time(h)

AP

syst

olic

and

dia

stol

ic (m

mHg

)

Fig 1: 24h recording of SAP and DAP. @ 7h the minipigs have been fed

14

PLACEBO

-2 0 2 4 6 8 10 12 14 16 18 20 22 24

90

100

110

120

130

140

150

160

170

180

Time(h)

LVP

syst

olic

(mm

Hg)

Fig 2: 24h recording of LVP. @ 7h the minipigs have been fed

15

PLACEBO

-2 0 2 4 6 8 10 12 14 16 18 20 22 2440

50

60

70

80

90

100

110

120

130

Time(h)

hear

t ra

te (

beat

s pe

r m

in)

Fig 3: 24h recording of HR. @ 7h the minipigs have been fed

16

Fig. 4. ECG morphology changes (increase of QT interval) with 30, 100 and 300 mg/kg Moxifloxacin treatment (red lines, arrows) compared to pre-dose (blue/black lines)

Fig. 5. ECG morphology changes (shortening of QT interval and AV-block) with 30 mg/kg Propranolol treatment (red line) compared to pre-dose (blue lines).

17

Model a b c r AIC RMSE Bazett QT =a*RR^(1/2)

341.31 -0.78 2684.00 73.67

Fridericia QT = a*RR^(1/3)

330.96 -0.10 2320.60 41.15

Sarma Exponential QT = a - b*exp(c*RR)

349.36 356.61 -2.82 0.00 2266.03 37.59

Linear QT = a + b*RR

206.02 123.59 0.02 2267.05 37.71

Hyperbolic QT = a + b/RR

391.56 -63.90 0.00 2264.93 37.58

Parabolic QT = a*RR^b

329.43 0.31 0.04 2260.10 37.29

Shifted logarithmic QT = log(a+b*RR)

4.21 1.28 0.73 1814.80 18.27

Exponential QT = a + b*exp(-RR)

429.82 -271.97 -0.02 2270.48 37.92

Table 1. Parameter estimates a, b, c and selection criteria Pearson’s correlation coefficient r, Akaike’s information criterion (AIC) and PRESS RMSE derived from 1,040 observations of an individual QT correction. Parameter Mean

(all time points) ± SD Mean

(HR ≤ 80) ± SD

PR (ms) 125 21 128 22 QRS (ms) 56 9 56 9 QT (ms) 320 38 336 26 RR (ms) 861 272 998 222 HR (ms) 77 26 63 12 Table 2. Summarised values for (manually) measured ECG intervals during 24 hours in the freely moving Göttingen Minipigs.