TQT Study Assay Sensitivity

Cardiac safety of a drug is normally assessed as early as possible, ideally before the start of a thorough QT/QTc study (TQT study)¹ which serves to confirm the absence of a signal. The International Conference on Harmonization (ICH) guidelines mandates a positive control for these studies and has set a sensitivity target for TQT studies to be able to detect a mean increase in QTc interval of approximately 5 msec.²

Amongst the pharmacological agents, the fluoroquinolone antibiotic moxifloxacin is most commonly used as a positive control to confirm assay sensitivity as mandated by ICH E14 guidelines.^{3, 4, 5} Moxifloxacin is well established and well published in producing average QT prolongation (QTcF) of usually 10 msec or greater following a 400 mg single dose.^{6, 7} This larger than originally anticipated effect is addressed in the ICH E14 implementation group questions and answers document in relation to assessing the adequacy of positive controls in TQT studies.⁸ This highlights the importance of the size of the effect of the positive control when assessing the ability of the study to detect changes around regulatory guidelines of 5 msec; this led to the requirement that at least one of the lower bounds of the confidence intervals must be greater than 5 msec.⁶

Therefore, an alternative method of confirming assay sensitivity which is able to detect small changes around the regulatory threshold has been proposed.⁷ The fluoroquinolone antibiotic, levofloxacin, has been shown to lead to smaller increases in QTc, and thus has the potential to provide a more rigorous evaluation of assay sensitivity by leading to a mean change in QTc of only around 5 msec.⁹

Furthermore, a non-pharmacological approach may be desirable in studies where the use of moxifloxacin is undesirable. The use of postural changes has been proposed¹⁰ but this method has been found to be impractical and ECG measurements may be affected by hysteresis (i.e., the period of QT to RR adaptation after a change in heart rate during which accurate measurements of the QTc interval is impossible).¹¹ More recently a study has demonstrated that food may cause a prolonged and pronounced change in QTc which may warrant its use as a non pharmacological method of confirming assay sensitivity in early phase studies where a simple and robust method of confirming the sensitivity of ECG work is desirable in early phase studies where no signal is seen.

Food effect on QT interval and QTc
A number of studies have demonstrated the effect of food on the QT interval. These include several reports of prolonged QT resulting from low calorie meals and starvation.^{12, 13, 14} Postprandial QT shortening^{15, 16} and postprandial increases in heart rate have also been reported.^{17, 18} The cause of the observed heart rate increases are not clear with one study suggesting sympathetic stimulation¹⁷ and another vagal withdrawal.¹⁸

A recent study has measured the effect of food on QTc in healthy subjects in the resting state.¹⁹ Subjects were randomized to fed or fasting conditions in a cross-over study. A steep increase in mean heart rate (HR) of 9.4 beats per minute (bpm) was observed after the zero time point gradually returning to baseline at the 4 hours time point.¹⁹ A reduction of mean QT duration occurred which was inversely related to the change in HR. The food effect on the uncorrected QT interval reached a maximum of 27 msec at 1.5 hours post-dose or 2 hours from the start of breakfast, which was statistically significant when compared to the baseline data obtained in the fasted state. The heart rate had returned to baseline by 4 hours post dose, whereas the QT interval, whilst prolonged had not fully returned to its baseline value.

Different heart rate correction methods were applied. The QTcF interval showed a maximum shortening at 2 hours post dose (or 2.5 hours from the start of breakfast) of 8.2 msec (95% CI: 6-10).¹⁹ The QTcIP interval showed a maximum shortening at 3 hours post dose (or 3.5 hours from the start of breakfast) of 5.6 msec (95% CI: 4-8).¹⁹ This was the first published study which reported the relationship between ingestion of a standardized meal conducted under the rigorous conditions of a TQT study until 4 hours after a meal. Normally postprandial effects are usually reported up to 2 hours after a meal. It was also the first study to describe a definitive shortening of the QTc interval using a population-based correction method as well as Frediericia’s. Another study has also demonstrated a statistically significant increase in HR of 12 bpm after food intake (500 mL sour milk, 200 mL fruit museli, three cheese sandwiches, and one apple) peaking at 1.3 hours and lasting for a duration of at least 3 hours²⁰ as well as a reduction in the area under the T-wave. Similar findings have been described by Bloomfield et al.¹⁶

These increases in HR and subsequent QTc shortening in response to food, if reproducible, can be used as a basis for a non-pharmacological method for confirming assay sensitivity.

Food content: impact on QT interval change
Meals of high carbohydrate content have been associated with transient endogenous physiological insulinaemia.¹⁷ One study by Scott et al. ¹⁷ has shown a 10 bpm increase in heart rate combined with peak increases, in blood glucose at 0.4 hours and peak insulin levels at 1.2 hours following carbohydrate ingestion. Another study by Gastaldelli et al.²¹ has shown an effect of insulin in a eugylycaemic clamp experiment on heart rate and QTcB. However, the study by Gastaldelli et al. failed to demonstrate an effect on the uncorrected QT and was also inadequate in that only a two-channel ECG recording device had been used to assess QT/QTc effects. Neither study measured the levels of C-Peptide which is excreted in equi molar amounts to insulin and has been associated with a reduction of QTc by Johansson et al. It is possible that the effects of C-Peptide are responsible for the extended shortening of QT beyond the immediate post prandial period.

If postprandial insulinaemia plays a significant role in the observed effects of food on ECG, then meals with high levels of carbohydrates would be expected to show a greater effect. The carbohydrate content of meals were similar for the Taubel et al. study¹⁹ (68%), the Widerlov et al. study20 (57%) and the Nagy et al. study¹⁴ (53%). For all three studies, the food content in these three studies is distinctly different in that they have a much higher carbohydrate and lower fat content than for example an FDA standard breakfast which delivers 950 kcal and contains 58% of fat with a much lower carbohydrate content. This difference may not elicit a similar effect but there is no data to suggest that this is indeed the case. The reduction in calories to approximately 750 kcal and the overall size of the meal is important for example in Japanese female subjects who are unable to consume the standard FDA breakfast.¹⁹ Therefore, standardized meals of high carbohydrate content would be easier to incorporate across different subject populations. There is sufficient evidence in the literature to suggest that the food effect observed by Taubel et al.¹⁹ is well reproducible both in terms of magnitude of effect; time; and course, and robust against short and temporary changes in posture.^{14, 17, 20}

Non-pharmacological methods
An alternative non-pharmacological method for determining assay sensitivity would have advantages compared to the current method of moxifloxacin involving drug induced ion channel blockade. A positive control method which does not involve a drug would:

• Be simple and easy to perform

• Be robust and reproducible

• Be performed in small groups

• Be applicable across different races

• Avoid unnecessary drug exposure

• Have no anticipated adverse events

• Not require ethical approval (no blood samples required)

The changes observed in QT and QTc interval following food appears to be reproducible and fulfill the benefits described above. The food effect observed is much closer to the ICH E14 requirements of 5 msec compared with moxifloxacin.¹⁹ Even though the change is negative (QTcIP/QTcF shortening of 5.6/8.2 msec respectively) the direction of the effect is not important when considering that the assay control is used to confirm that the study is capable of detecting small changes. Therefore, a food arm could be easily included to confirm assay sensitivity in standard TQT studies and also in specialist oncology and pediatric TQT studies, where the use of moxifloxacin has been proven to be problematic. In addition, food can be used to confirm ECG assay sensitivity in early phase studies such as multiple ascending studies (MAD studies) and be used for hypothesis generation and early characterization of cardiac on or off target effects.

Breakfast

Serving

Cals (kcal)

Proteins

Carbohydrates

Fat

Fibre

(g)

kcal

(g)

kcal

(g)

kcal

(g)

kcal

Breakfast Cereal, Kellogg's Cornflakes

30g

111.9

2.1

8.4

25.2

100.8

0.3

2.7

0.9

0.0

Milk

150ml

72.5

5.0

20.0

7.5

30.0

2.5

22.5

0.0

0.0

Sugar

10g

40.0

0.0

0.0

10.0

40.0

0.0

0.0

0.0

0.0

Wholemeal hoagie (1 roll)

79g

178.1

8.6

34.4

27.6

110.4

3.7

33.3

5.5

0.0

Jam

20g

50.0

0.0

0.0

12.5

50.0

0.0

0.0

0.2

0.0

Butter

10g

72.9

0.0

0.0

0.0

0.0

8.1

72.9

0.0

0.0

Apple juice (pure)

200ml

93.1

0.3

1.2

22.3

89.2

0.1

0.9

0.1

0.0

Total

616.7

16.0

64.0

105.1

420.4

14.7

132.3

6.7

0.0

10%

68%

21%

0%

Table 1. The extent of effect on QTc may vary with different types of food with higher or lower
carbohydrate content.

The extent of effect on QTc may vary with different types of food with higher or lower carbohydrate content. Further studies are warranted to investigate the effects of different types of food, particularly to define the magnitude of effect. The food content (high carbohydrate) used by Taubel et al.¹⁹ is presented in Table 1. Subjects were under a “bed-rest” period of 4 hours following breakfast and during the ECG measurements. This can be standardized so that a change of around 5 msec can be consistently reproduced with this breakfast in TQT studies, confirming assay sensitivity.

Outlook and further research
A standardized food arm could be used as an alternative method to demonstrate assay sensitivity in a wide range of studies providing an assurance that these trials would be sufficiently sensitive in detecting an effect on the QTc interval of around 5 to 10 msec. This will certainly help in the planning of TQT studies and may even help avoiding them altogether. The method would be useful in all instances where a pharmacological assay control is not feasible. The method appears to be reproducible and robust even in relatively small study populations given that other groups have found similar effects. It is easy to implement with a simple inclusion of a fasted and fed ECG day to most trials and would not be objectionable to ethics committees.

Additional studies will help further quantify the effects observed with food, which would allow consideration as to whether a well defined food arm may even allow the replacement of moxifloxacin in formal TQT studies. A study (EudraCT number: 20011-002423-17) is currently in progress to investigate the effects of different meal compositions (a carbohydrate rich continental breakfast versus a standard FDA breakfast) on the QT/QTc interval and to assess to what extent the effects of food may cancel out the effects of moxifloxacin on QT/QTc.

References
1. A. J. Camm, “Clinical Trial Design to Evaluate the Effects of Drugs on Cardiac Repolarization: Current State of the Art,” Heart Rhythm, Supplement, 11 (2) S23-S29 (2005).
2. ICH E14, “The Clinical Evaluation of QT/QTc Interval Prolongation and Proarrhythmic Potential for Non-antiarrhythmic Drugs,” International Conference on Harmonization, Step 4, Guideline, EMEA, CHMP/ICH/2/04, (2005).
3. J. L. Demolis, D. Kubitza, L. Tenneze, and C. Funck-Brentano, “Effect of a Single Oral Dose of Moxifloxacin (400mg and 800mg) on Ventricular Repolarization in Healthy Subjects,” Clin Pharmacol Ther, 68 (6) 658-666 (2000).
4. J. Kang, L. Wang, X. L. Chen, D. J. Triggle, and D. Rampe, “Interactions of a Series of Fluoroquinolone Antibacterial Drugs with the Human Cardiac K+ Channel HERG,” Mol Pharmacol, 59 (1) 122-126 (2001).
5. C. Strnadova, “The Assessment of QT/QTc Interval Prolongation in Clinical Trials: A Regulatory Perspective,” Drug Information Journal, 39 407-433 (2005).
6. J. A. Florian, C. W. Tornøe, R. Brundage, A. Parekh, and C. E. Garnett, “Population Pharmacokinetic and Concentration-QTc Models for Moxifloxacin: Pooled Analysis of 20 Thorough QT Studies,” J Clin Pharmacol, 51 (8) 1152-1162 (2011).
7. C. W. Tornøe, C. E. Garnett, Y. Wang, J. Florian, M. Li, and J. V. Gobburu, “Creation of a Knowledge Management System for QT Analyses,” J Clin Pharmacol, 51 (7) 1035-1042 (2011).
8. ICH E14 “The Clinical Evaluation of QT/QTc Interval Prolongation and Proarrhythmic Potential for Non-antiarrhythmic Drugs,” International Conference on Harmonization, E14 Implementation and Working Group, Questions and Answers, (2008).
9. J. Taubel, A. Naseem, T. Harada, D. Wang, R. Arezina, U. Lorch, and A. J. Camm, “Levofloxacin can be Used Effectively as a Positive Control in Thorough QT/QTc Studies in Healthy Volunteers,” British Journal of Clinical Pharmacology, 69 (4) 391-400 (2010).
10. G. C. Williams, K. M. Dunnington, M. Y. Hu, T. R. Zimmerman Jr, Z. Wang, K. B. Hafner, M. Stoltz, E. K. Hill, and J. T. Barbey, “The Impact of Posture on Cardiac Repolarization: More than Heart rate?,” J Cardiovasc Electrophysiol, 17 (4) 352-358 (2006).
11. M. Malik, K. Hnatkova, T. Novotny, and G. Schmidt, “Subject-Specific Profiles of QT/RR Hysteresis,” Am J Physiol Heart Circ Physiol, 295 (6) H2356-2363 (2008).
12. B. C. Thwaites and M. Bose, “Very Low Calorie Diets and Pre-fasting Prolonged QT interval: A Hidden Potential Danger,” West Indian Med J, 41 (4) 169–171 (1992).
13. I. Swenne and P. T. Larsson, “Heart Risk Associated with Weight Loss in Anorexia Nervosa and Eating Disorders: Risk Factors for QTc Interval Prologation and Dispersion, Acta Paediatr, 88 (3) 304–309 (1999).
14. D. Nagy, R. DeMeersman, D. Gallagher, A. Pietrobelli, A. S. Zion, D Daly, and S. B. Heymsfield, “QTc Interval (Cardiac Repolarization): Lengthening After Meals,” Obes Res, 5 (6) 531–537 (1997).
15. R. Hulhoven, D. Rosillon, W. E. Bridson, M. A. Meeus, E. Salas, A. Stockis, “Effect of Levetiracetam on Cardiac Repolarization in Healthy Subjects: A Single-Dose, Randomized, Placebo- and Active-Controlled, Four-Way Crossover Study,” Clin Ther, 30 (2) 260–270 (2008).
16. D. Bloomfield, J. Kost, K. Ghosh, D. Hreniuk, L. Hickey, M. Guitierrez, K. Gottesdiener, and J. Wagner, “The Effect of Moxifloxacin on QTc and Implications for the Design of Thorough QT Studies,” Clin Pharmacol Ther, 84 (4) 475–480 (2008).
17. E. M. Scott, J. P. Greenwood, G. Vacca, J. B. Stoker, S. G. Gilbey, and D. Mary, “Carbohydrate Ingestion, with Transient Endogenous Insulinaemia, Produces both Sympathetic Activation and Vasodilatation in Normal Humans, Clinical Science, 102 (5) 523–529 (2002).
18. C. Lu, X. Zou, W. C. Orr, and J. D. Z. Chen, “Postprandial Changes of Sympathovagal Balance Measured by Heart Rate Variability,” Digestive Diseases and Sciences, 44 (4) 857-861 (1999).
19. J. Taubel, A. H. Wong, A. Naseem, G. Ferber, and A. J. Camm, “Shortening of the QT Interval After Food can be Used to Demonstrate Assay Sensitivity in Thorough QT Studies,” Journal of Clinical Pharmacology, (in press).
20. E. Widerlov, K. Jostell, L. Claesson, E. Odlind, M. Keisu, and U. Freyschuss, “Influence of Food Intake on Electrocardiograms of Healthy Male Volunteers,” Eur J Clin Pharmacol, 55 (9) 619-624 (1999).
21. F. A. Gastaldelli, M. Emdin, F. Conforti, S. Camastra, and E. Ferrannini, “Insulin Prolongs the QTc Interval in Humans,” Am J Physiol Regul Integr Comp Physiol, 279(6) R2022-2025 (2000).

Jorg Taubel* is Chief Executive Officer, e:mail: j.taubel@richmondpharmacology.com, Asif Naseem is Medical Writer, and Georg Ferber is Consultant Statistician all at Richmond Pharmacology Ltd, London, UK. A. John Camm, is BHF Professor of Clinical Cardiology, Department of Cardiological Sciences, St George’s University of London, London, UK.

*To whom all correspondence should be addressed.