6 7

ABSTRACTThis paper describes the development of a non-rodent in vivo cardiovascular safety model: the fentanyl/etomidate-anaesthetised Göttingen (FEAG) minipig model. There is minimal influence from the applied anaesthetic regime on the baseline cardiovascular, pulmonary and haematological parameters within this model. As shown in the results, all measured parameters were comparable to published con-scious Göttingen minipig data. Hemodynamic, cardiac, elec-trophysiological, respiratory and arterial blood parameters were relatively stable for at least three hours, and the QT interval could be corrected for changes in heart rate using a linear formula: QTc = QT – 1.6 (60 – HR). Intravenous infusion of dofetilide induced a significant QTc prolongation of +71 ms (+18%). Further characterisation of the FEAG minipig model is ongoing and will be published soon.

INTRODUCTIONTo date, miniature pigs have been used extensively in car-diovascular research, because both the hemodynamics and electrophysiology of these animals appear to be similar to humans. Consequently, miniature pigs have become the species of choice for evaluating the safety of some new molecular entities (NME’s).[1 and 2] In cardiovascular safety studies, Göttingen minipigs have mainly been used in con-scious animal settings (i.e. static condition in a sling or freely moving with telemetry implants), and less frequently in anaesthetised models. The big advantage of the latter is obviously the freedom to push the doses/exposures way beyond levels where CNS, respiratory or GI side effects would limit profiling in a conscious animal setting. Moreover, it allows measurement of many different electrophysi-ological, hemodynamic and respiratory endpoints under very controlled conditions, in addition to plasma exposure analysis without any interference to the data being gener-ated. Since some effects (e.g. efficacy) of NMEs appear to be species-dependent, it seemed relevant to develop an anaesthetised minipig model as an alternative option to our anaesthetised dog model. For reasons of comparability across species, our objective was to keep the anaesthetic regime as close as possible to the procedure used in dogs, and therefore we chose a regime similar to the recently published FEAB model.[3]

In this paper the tailored anaesthetic regime and surgical procedures are described. Furthermore, the stability of hemodynamic, electrophysiological, respiratory and arterial blood parameters are presented and discussed.

Because conventional correction formulas were not suf-ficient to correct the heart rate dependency of the QT interval in minipigs, and an individual correction formula[4] is difficult to evaluate in anaesthetised animals, we herewith propose a FEAG-specific correction formula. Finally, in this paper, we illustrate the QT-sensitivity of the FEAG minipig to a known and broadly used IKr-blocker (dofetilide).

MATERIAL AND METHODS1. AnimalsAll published experiments have been conducted in accord-ance with “The provision of the European Convention” on the protection of vertebrate animals which are used for experimental and other scientific purposes, and with “the Appendices A and B”, made at Strasbourg on 18 March 1986 (Belgian Act of 18 October 1991) and the Commission recommendation of 18 June 2007 on guidelines for the accommodation and care of animals used for experimental and other scientific purposes (2007/526/EC). In all the experiments, male Ellegaard Göttingen minipigs aged 18 to 24 months, with a body weight ranging from 27.7 to 39.5 kg, were used. The minipigs were housed on bedding (wood shavings) and in pairs in an AAALAC-accredited facility, with a controlled room temperature (18–24 °C) and a day/night cycle of 12 h light/12 h dark. The animals had access to water ad libitum, were fed between 7:00 and 8:00 AM with a standard minipig diet (ssniff® MPig-H) and were allowed to stay in an outdoor playground twice a week to socialise. The animals were easy to handle and underwent routine clinical examination. The animals were found to be healthy and active before use. Food (but not water) was withheld for at least 12 hours prior to anaesthesia and experimentation.

2. Anaesthetic regimeThe minipigs were premedicated with an intramuscular injection of 3 mg/kg of azaperone (Stresnil®, Janssen Animal Health, Beerse, Belgium) behind the ear at the animal house before transportation. In addition, an intramuscular injec-

ACARDIOVASCULARSAFETYMODEL ➤ ➤

➤

Minipigs are walking freely to the outdoor playground to play and socialise.

Intramuscular injection behind the ear.

The Fentanyl/Etomidate-Anaesthetised Göttingen (FEAG) Minipig:

A Cardiovascular Safety Model1Henk van der Linde*, 1Yves Somers, 1Bruno Van Deuren, 2Leen Roefs, 1Ard Teisman, 1David J. Gallacher1Center of Excellence for Cardiovascular Safety Research and Mechanistic Pharmacology and 2Laboratory Animal Medicine, Johnson & Johnson Pharmaceutical Research and Development, Division of Janssen Pharmaceutica NV, Turnhoutseweg 30, B-2340 Beerse, Belgium. *Corresponding author: hvdlinde@its.jnj.com

➤

8 9

tion of 10 mg/kg of ketamine (Ketalar®, Pfizer, Brussels, Belgium) was given behind the ear, ten minutes later, upon arrival in the operating room.

When the animal was sufficiently sedated, an intramuscular injection of a mixture containing scopolamine 0.02 mg/kg (Alcon Laboratories Inc., USA), etomidate 1.5 mg/kg (Janssen Pharmaceutica NV, Beerse, Belgium) and vecuro-nium 0.1 mg/kg (Organon, Oss, the Netherlands) was given. Immediately after endotracheal intubation (7 mm exterior diameter tube), mechanical ventilation was started with a mixture of air/oxygen (70/30) at a ventilation rate of 15 breaths/min.

Complete anaesthesia was induced via an intravenous injec-tion into the ear vein of a mixture containing lofentanil 0.02 mg/kg (Janssen Pharmaceutica NV, Beerse, Belgium), sco-polamine 0.01 mg/kg (Alcon Laboratories Inc., USA) and vecuronium 0.1 mg/kg (Organon, Oss, the Netherlands).

During the experiment the anaesthesia was maintained with a continuous infusion of etomidate (Janssen Pharmaceutica NV, Beerse, Belgium): infusion was started at 1.2 mg/kg/h, but was slightly adapted according to the individual needs of each animal. Slow, hourly bolus injections of 0.025 mg/kg fentanyl (Janssen Pharmaceutica NV, Beerse, Belgium) were given to ensure a pain-free condition. To compensate for fluid loss during the experiment, a continuous infusion of 0.5 ml/kg/h of 5% dextran and 2.5% glucose in saline was given, and the animals received 500 IU/kg heparin (LEO Pharma

NV, Wilrijk, Belgium) as an anticoagulant at the beginning of the experiment.

3. Surgical procedureSurgery was initiated by exposing all blood vessels needed for catheter insertion with an electro-cutting and coagulating equipment (Erbotom ACC450, Erbe, Germany) to mini-mise bleeding. Directly after coagulation, the surface ECG needles were placed and connected to an amplifier (Emka, France), to achieve a lead II ECG signal.

An injection catheter was inserted into the left femoral vein and positioned close to the atrium. A catheter-tip micro-manometer (MTC800, Dräger, Germany) was inserted into the left femoral artery and positioned in the aortic arch for monitoring aortic blood pressure (AoP). A Swan-Ganz catheter (131HF7, Edwards Lifesciences, USA) with a tem-perature sensor-tip was inserted into the right femoral vein and positioned in the pulmonary artery via the right atrium and ventricle. With this catheter core body temperature and cardiac output (thermodilution method) were measured. A Millar microtip catheter was inserted into the Swan-Ganz catheter to accurately measure pulmonary artery pressure (PAP). An open lumen catheter for arterial blood sampling was placed in the right femoral artery. These blood samples were analysed for electrolytes and blood gases using a blood gas analyser (ABL700; Radiometer, Denmark). An open lumen catheter for infusing solutions and test compounds was inserted into the left maxillary vein. A controllable monophasic action potential catheter (Boston Scientific-EP

Technologies, San Jose, USA) was inserted into the left vena linguofacialis and positioned against the inner wall of the right ventricle in such a way that a stable MAP signal was pro-duced. A catheter tip micromanometer with pigtail (Gaeltec, Dunvegan, Scotland) was inserted into the left carotid artery and positioned in the left ventricle for measuring left ven-tricular pressure (LVP). All catheters were positioned under fluoroscopic (Siremobil2000, Siemens, Germany) guidance.

4. Measured and calculated parameters.Hemodynamic parametersHeart rate (HR; retrieved from the LVP), systolic, diastolic and mean aortic blood pressure (SBP, DBP, MBP; retrieved from the AoP) and left ventricular end-diastolic pressure (LVEDP; retrieved from the LVP) were measured, and pressure-rate product (PRP: SBP x HR) was calculated. In the pulmonary circulation, diastolic and systolic pulmonary pressures (DPP, SPP; retrieved from the PAP) were meas-ured. Systemic and pulmonary vascular resistance values were calculated (SVR; mean arterial pressure/CO, PVR; mean pulmonary pressure/CO).

Cardiac parametersCardiac output (CO) and stroke volume (SV) were cal-culated with the thermodilution method. Contractility parameters: left ventricular dp/dtmax (LV dp/dtmax) and left ventricular dp/dtmax/pd (LV dp/dtmax/pd) and relaxation parameters left ventricular dp/dtmin (LV dp/dtmin) and time constant of left ventricular relaxation (tau) were calculated from the LVP signal. (for more information: http://www.notocord.com/Solutions/Pages/Cardiovascular.aspx).

Electrophysiological parametersThe duration of the PQ interval (measured from the onset of the P wave to the onset of the Q wave), QRS complex (measured from the onset of the Q wave to the end of the S wave) and QT interval (measured from the onset of the Q to the end of the T wave) were measured using lead II of the ECG. The QT interval was corrected for heart rate variations using the correction formulae of Bazett (QTcB), Fridericia (QTcF), Van De Water (QTcVDW) and a species-specific correction formula based on data from six anaesthetised minipigs (QTc). The duration of the endocar-dial action potential signal was measured as the interval from the upstroke to 90% repolarisation (APD90 RV en).

Respiratory parametersTidal volume (TV), airway resistance (Raw) and dynamic lung compliance (Cdyn) were calculated from the airway pressure and airway flow signals using the isovolumetric method.(for more information: http://www.notocord.com/Solutions/Pages/Respiratory.aspx).

Blood gases and electrolytesAt regular intervals during the experiment, arterial blood samples were analysed for partial O2 and CO2 tension, base

excess, bicarbonate and oxygen saturation to evaluate met-abolic and respiratory stability. In the same sample, K+, Na+, Ca2+ and Cl- were analysed. The metabolic parameters glu-cose and lactate and total haemoglobin were also measured.

Physiological parametersCore body temperature (cBT) was measured in the right ventricle.

Dispersion of repolarisationAs an indication of the dispersion across the left ventricular wall (transmural dispersion), the duration of the descend-ing T-wave limb was measured by the interval between the T-wave peak to the end of the T wave (Tp-Te). To measure the dispersion between left and right ventricle (LRD, interventricular dispersion), the difference between the QT interval and APD90RV en was calculated. Temporal dispersion of repolarisation, otherwise named short-term instability (QT-STI), was calculated beat-to-beat over 30 consecutive beats, as previously published.[5]

5. Experimental designTo investigate the relationship between the duration of the QT interval and heart rate, we used six minipigs with high basal heart rates. During the stabilisation period (1–2 hours) the animals received bolus injections of saline with 5% dex-tran and 2.5% glucose to lower the heart rate. The QT/RR and QT/HR data of these minipigs were plotted and linear correlation variables were calculated. To evaluate the base-line values and the stability of all parameters, six minipigs were administered with six volumes (0.032, 0.063, 0.125, 0.25, 0.5 and 1 ml/kg i.v.) of saline, infused over 5 minutes at 30-minute intervals, and all parameters were measured or calculated at selected times (pre-dose and at the end of each infusion). In four of these saline-treated minipigs, dofetilide (0.05 mg/kg i.v.) was infused over 10 minutes at the end of the experiment.

RESULTS AND DISCUSSION1. QT correctionTo evaluate the relationship between the duration of the QT interval and changes in heart rate (HR), the heart rate values of 6 minipigs (all male) with a wide range in heart rates (and RR intervals) during the stabilisation period were plotted against the uncorrected QT intervals. This resulted in a non-linear relation between RR and QT, comparable with data from telemetered minipigs.[4] However, a linear relationship was noted after plotting HR against QT data, with comparable slopes and correlation coefficients (R2) for all animals (Figure 1).

These data (Table 1) were used to calculate a mean slope and the following formula was derived: QTc = QT – 1.6 (60 – HR).

To evaluate our proposed formula, three QTc values calculated according to commonly used formulas (Bazett,

ACARDIOVASCULARSAFETYMODEL ➤ ➤

➤ ➤

Set-up for the Fentanyl/Etomidate anaesthetisation of Göttingen (FEAG)

10 11

➤

Fridericia and Van De Water) were plotted against HR (Figure 2). Linear regression lines are displayed in the graphs and correlation coefficients (R2) and slopes are shown in the graphs; the most effective formula will show the low-est R2 and a slope close to zero. As described by others,[4] the Bazett formula shows an over-correction, resulting in a positive slope of the regression line. The other two (Fridericia and Van De Water) showed both an under-correction (slopes were -0.4748 and -0.976, respectively). However, our proposed correction formula showed a hori-zontal regression line (slope = 0.0373), and no correlation between heart rate and QTc (R2 = 0.004). We realise that this formula is constructed on the basis of a small group of six animals and that more experiments will be necessary to prove the utility of this formula. In our vehicle group, a decrease of heart rate (-24 b.p.m.) after 3 hours resulted in an increase of QT (+33 ms) and was totally corrected by this formula, but the value of this formula can only be con-firmed by using it in further studies in our own laboratory and other laboratories.

2. Baseline values and stabilityBaseline values (mean ± SEM) of all measured and calculated parameters and the stability of these parameters (% after i.v. saline administration from 0.032 to 1 ml/kg; duration of 3 hours) are listed in Table 2. Not all parameters were measured in all minipigs due to technical reasons. For exam-ple, cardiac output (CO), stroke volume (SV), systemic vascular resistance (SVR) and pulmonary vascular resist-ance (PVR) were measured in only 2 animals and must be regarded as provisional. Most parameters were stable over a period of 3 hours and changed less than 10%. Heart rate (HR), pressure rate product (P;P) and left ventricular con-traction (LV dp/dtmax and LV dp/dtmax/pd) showed slight to minor decreases (-30%, -24%, -18% and -13%, respectively), and left ventricular end diastolic pressure (LVEDP) showed an increase (+21%) during the saline experiments (≈ 2 ml/kg i.v.) in this study. The HR decrease was accompanied, as expected, by an increase in the duration of the RR, PQ and QT intervals. The QT and QTc intervals were relatively long, compared to values obtained by others;[4 and 6] this was prob-ably caused by the mild hypokalaemia in addition to the low body temperature.[7] The increase in the left-right dispersion

Figure 2: Corrected QT (QTc in ms) versus Heart Rate (HR in beats per minute): a comparison of different QT correction formulae (Bazett, Fridericia, Van De Water and the FEAG minipig-specific formula).

ACARDIOVASCULARSAFETYMODEL ➤

➤

y = -0.4748x + 447.25R² = 0.289

200

250

300

350

400

450

500

550

20 40 60 80 100 120 140 160 180

QT

c F

(m

s)

HR (b.p.m.)

Fridericia

y = 0.2078x + 411.92R² = 0.056

200

250

300

350

400

450

500

550

20 40 60 80 100 120 140 160 180

QT

c B

(m

s)

HR (b.p.m.)

Bazett

y = -0.976x + 471.78R² = 0.7085

200

250

300

350

400

450

500

550

20 40 60 80 100 120 140 160 180

QT

c V

DW

(m

s)

HR (b.p.m.)

Van de Water

y = 0.0373x + 403.52R² = 0.004

200

250

300

350

400

450

500

550

20 40 60 80 100 120 140 160 180

QT

c (

ms)

HR (b.p.m.)

FEAG minipig

EXP NR HR RANGE (B.P.M.) QT RANGE (MS) R2 SLOPE

high low short long

1 140 48 262 400 0.990 -1.56

2 144 55 269 411 0.993 -1.59

3 129 47 305 430 0.963 -1.54

4 148 56 288 447 0.963 -1.67

5 149 85 269 366 0.937 -1.58

6 168 63 253 424 0.978 -1.64

mean 146 59 275 413 0.971 -1.60

TABLE 1

Table 1: Heart rate (HR) and QT interval (QT) ranges, linear correlation coef-ficients (R2) and slopes of six anesthetised male minipigs (data obtained during stabilisation period).

Figure 1: QT (ms) versus RR (ms) - plot (A) and QT (ms) versus HR (b.p.m) - plot (B); data derived from six anaesthetised male Göttingen minipigs (418 data points).

200

250

300

350

400

450

500

550

200 400 600 800 1000 1200 1400 1600

QT

(ms)

RR (ms)

QT versus RR

y = -1.5627x + 499.52R² = 0.8754

200

250

300

350

400

450

500

550

20 40 60 80 100 120 140 160 180

QT

(ms)

HR (b.p.m.)

QT versus HRA B

200

250

300

350

400

450

500

550

200 400 600 800 1000 1200 1400 1600

QT

(ms)

RR (ms)

QT versus RR

y = -1.5627x + 499.52R² = 0.8754

200

250

300

350

400

450

500

550

20 40 60 80 100 120 140 160 180

QT

(ms)

HR (b.p.m.)

QT versus HRA B

➤

12 13

(LRD) was caused by a relatively larger prolongation of the QT interval, com-pared to the increase in the duration of the endocardial action potential of the right ventricle (APD90RV en).

3. Dofetilide infusionAfter the saline administrations, dofe-tilide was infused in four animals over 10 minutes to reach a total dose of 0.05 mg/kg i.v. As expected, dofetilide induced a prolongation of the QT inter-val (+18%; p = 0.017), QTc interval (+18%; p = 0.04), APD90RV en (+19%; p = 0.04) and QT-STI (+119%, p = 0.10) (Table 3). No significant increase in transmural (Tp-Te) and interventricu-lar (LRD) dispersion of the repolarisa-tion was noted. T-wave morphology changes on the ECG or induction of early after-depolarisations on the MAP signal were not observed in this study (Figure 3).

Figure 3: ECG signal (left) and MAP signal (right) before (black) and after 0.05 mg/kg i.v. dofetilide (red).

REFERENCES1 M. Kano, T. Toyoshi, S. Iwasaki, M. Kato, M. Shimizu and T. Ota. QT PRODACT: Usability of miniature pigs in safety pharmacology studies: Assessment for drug-induced QT interval prolongation. Journal of Pharmacological Sciences 99, 501-511 (2005).2 M. Markert, M. Stubhan, K. Mayer, T. Trautmann, A. Klumpp, A. Schuller-Metz, K. Schumacher and B. Guth. Validation of the normal, freely moving Göttingen minipig for pharmacological safety testing. Journal of Pharmacological and Toxicological Methods 60, 79-87 (2009).3 B. Van Deuren, K. Van Ammel, Y. Somers, F. Cools, R. Straetemans, van H.J. van der Linde and D.J. Gallacher. The fentanyl/etomidate-anaesthe-tised beagle (FEAB) dog: A versatile in vivo model in Cardiovascular Safety Research. Journal of Pharmacological and Toxicological Methods 60, 11-23 (2009).4 M. Stubhan, M. Markert, K. Mayer, T. Trautmann and A. Klumpp. Evaluation of cardiovascular and ECG

parameters in the normal, freely moving Göttingen minipig. Journal of Pharmacological and Toxicological Methods 57, 202-211 (2008).5 H.J. van der Linde, A. Van de Water, W. Loots, B. Van Deuren, H.R. Lu, K. Van Ammel and D.J. Gallacher. A new method to calculate the beat-to-beat instability of QT duration in drug-induced long-QT in anaes-thetised dogs. Journal of Pharmacological and Toxicological Methods 52, 168-177 (2005).6 K. Nahas, P. Baneux and D. Detweiter. Electrocardiographic monitoring in the Göttingen minipig. American Association for Laboratory Animal Science 52, 258-264 (2002).7 M. Laursen, O. Kerlund and T. Mow. Relation between the QT interval and temperature in Göttingen min-ipigs: a point to consider. Poster at the 9th annual meeting of the Safety Pharmacology Society in Strasbourg, France. Poster number 32 (2009).

ACARDIOVASCULARSAFETYMODEL

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

0

100

200

300

400

500

600

700

800

mV

ms

control 0.05 ml/kg dofetilide

-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

0

100

200

300

400

500

600

700

800

mV

ms

control 0.05 ml/kg dofetilide

➤

CONCLUSIONThe fentanyl/etomidate-anaesthetised Göttingen (FEAG) minipig is a potential non-rodent, in vivo model that can be used in safety pharmacology. Indeed, a wide range of hemodynamic, cardiac electrophysiological, respiratory and arterial blood parameters can be measured, have baseline

values within a physiological range, and remain stable for a relatively long period. A formula to correct the duration of the QT interval for changes in heart rate is proposed. As expected, dofetilide was found to prolong the QT and QTc intervals.

Table 3: Changes in electrophysiological parameters after an intravenous infusion with dofetilide (0.05 mg/kg). † = p ≤ 0.05, ‡ = p ≤ 0.10 (2-sided, paired t-Test)

Parameter Unit Pre-dose Post-dose %

RR ms 1274 1332 +5

PQ ms 164 163 0

QRS ms 58 62 +7

QT ms 405 477† +18

QTc ms 385 457† +18

APD90RV en ms 354 422† +19

Tp-Te ms 52 67 +29

QT-STI ms 2.3 5.0‡ +119

LRD ms 51 55 +8

TABLE 3

Parameter Unit Mean SEM % n Parameter Unit Mean SEM % n

cardiovascular parameters lung function parameters

HR bpm 80 8 -30 6 TV ml 250 11 +1 6

SBP mmHg 114 6 +8 6 Cdyn ml/mmHg 45 1 -6 6

MBP mmHg 102 5 +5 6 Raw mmHg/l/min 0.078 0.004 +4 6

DBP mmHg 86 6 +3 6 atrial blood parameters

SPP mmHg 31 2 +6 6 pH 7.5 0 0 5

DPP mmHg 23 3 +6 6 paCO2 mmHg 40 3 -4 5

LVEDP mmHg 14 2 +21 6 paO2 mmHg 182 22 -3 5

LV dp/dtmax mmHg/s 2303 117 -18 6 saO2 % 99.8 0.1 0 5

LV dp/dtmax/pd s-1 42 3 -13 6 HCO3- mmol/l 31.2 0.9 0 5

LV dp/dtmin mmHg/s -2802 277 +8 6 ABE mmol/l 7.3 1.0 0 5

Tau ms 30 2 +8 6 K+ mmol/l 3.7 0.1 0 5

CO l/min 3.7 0.7 +10 2 Ca2+ mmol/l 1.4 0 -1 5

SV ml 53 1 +49 2 Na+ mmol/l 137 1 0 5

PRP mmHg.bpm 9011 1000 -24 6 Cl- mmol/l 98 1 +1 5

SVR mmHg/l/min 30 9 -12 2 glucose mmol/l 5.8 0.3 -20 5

PVR mmHg/l/min 7 2 +7 2 lactate mmol/l 1.2 0.2 +13 5

electrophysiological parameters tHb mmol/l 6.4 0.3 +8 5

RR ms 796 85 +42 6 core body temperature

PQ ms 147 7 +10 6 cBT °C 36.4 0.2 0 6

QRS ms 59 1 +3 6 dispersion of repolarisation parameters

QT ms 372 7 +9 6 Tp-Te ms 37 3 +18 6

QTc ms 403 8 -1 6 LRD ms 23 6 +100 6

APD90RV en ms 350 10 +3 6 QT-STI ms 2.2 0.2 1 6

TABLE 2

Table 2: Baseline cardiovascular (hemodynamic & electrophysiology), pulmonary, blood profile, heart rhythm instability and body temperature values in the Fentanyl/Etomidate-Anaesthetised Göttingen minipig. Data presented are mean values, standard error of the mean (SEM), sample size (n) and % changes (%; after saline dose-response experiments of 3 hours).

➤