The Science of Cardiac Safety

There is considerable interest in drug safety among multiple stakeholders in the United States, including the Food and Drug Administration (FDA), the National Institutes of Health, the Institute of Medicine (IOM), biopharmaceutical companies, Congress, individual congressional committees, the Supreme Court, patient advocacy groups, the media, and perhaps most importantly, prescribing physicians and their patients.¹ Given that no drug is immune from the possibility of causing adverse events in certain genetically and/or environmentally susceptible individuals, an operational definition of "safety" is needed. A useful general definition is found in the FDA's Sentinel Initiative:²

Using medical products brings benefits and risks. Although marketed medical products are required by federal law to be safe for their intended use, safety does not mean zero risk. A safe product is one that has acceptable risks, given the magnitude of the benefit expected in a specific population and within the context of alternatives available.

DON FARRALL/GETTY IMAGES

This definition, which encapsulates the concept of benefit-risk assessment, "one of the important facets of the science of safety,"² applies very well in the science of cardiac safety. An investigational drug must be found by an agency to have a favorable benefit-risk balance to receive marketing approval. It is the responsibility of sponsors to provide regulatory agencies with the most accurate and thorough safety (and efficacy) data possible at the time of requesting marketing approval, and it is also in their own best interests, as discussed later.

History of cardiac safety assessments

Between the late 1980s and the early 2000s, a series of high-profile drug marketing withdrawals for cardiac reasons focused regulatory attention on this issue. Of particular relevance is one form of polymorphic ventricular tachycardia, torsades de pointes, that is very rare but can be fatal.³ Several hundred deaths from widely-prescribed drugs for relatively benign conditions (including terfenadine, an antihistamine, for allergies) indicated that the benefit-risk balance for these drugs was unacceptable.⁴

Torsades de pointes is associated with QT interval prolongation as seen on the surface electrocardiogram (ECG). Figure 1 provides a stylistic representation of the QT interval and the phenomenon of QT interval prolongation.

Figure 1: Stylistic representation of the QT interval and QT interval prolongation.

Each segment of the ECG can be assigned a length in the time domain. The QT interval represents the total time of cardiac muscle cell depolarization (contraction) and repolarization (returning to the relaxed state such that contraction can occur again). It is defined as the length in the time domain from the onset of the Q-wave to the off-set of the T-wave, measured in milliseconds (msec). Precise categorization of the "normal QT interval" for any individual is impractical since it changes with every heart beat. However, consideration of the typical ranges observed in groups of individuals following a period of supine rest provides some perspective. Since the QT interval is impacted by heart rate, tending to be shorter as heart rate increases, it is typically "corrected for heart rate" by one or more of several mathematical formulas, resulting in QTc data. In this context, the normal distribution of QTc intervals for healthy adult males suggests a range from around 350-460 msec. For healthy adult females, the distribution is similar but falls around a somewhat higher mean, suggesting a range from around 360-470 msec.⁵

Given the association between torsades de pointes and QT prolongation, regulatory agencies in Europe, Canada, Japan, and the United States now require a dedicated trial to assess the impact (if any) of an investigational drug on the QT interval. This trial is called a Thorough QT/QTc (TQT) Study, and is described in the 2005 ICH Guideline E14.^{6, 7} While assessment of the QT risk of new drugs had certainly been addressed beforehand,^8-10 the adoption of ICH E14 by regulatory agencies formalized clinical cardiac safety assessment.

The TQT is a rigorously controlled clinical trial. This rigor includes using identical monitoring equipment for each subject to collect multiple digital ECGs, transmitting them to a core ECG laboratory (core lab), and having all QT/QTc measurements for a given study performed at a core ECG lab in a blinded manner by a relatively small number of highly experienced technicians or cardiologists. In addition, the Phase I units conducting these trials must ensure that the subjects in each cohort are treated identically to ensure consistency in conduct. These centralized, standardized processes greatly enhance the quality and integrity of the data.¹¹ The logic here is similar to that behind the use of a single central laboratory (central lab) to analyze biological samples collected at multiple investigational sites. Standardization ensures consistency of many factors that can influence readings, including instrumentation, reference ranges, personnel, environment, and reagents.¹² More detailed discussion of the TQT study and torsadogenic liability can be found in many sources.^13-24

Moving beyond QT/QTc interval prolongation

While ICH E14 has made QT/QTc prolongation a high-profile cardiac safety biomarker, it is widely acknowledged to be imperfect, and, by itself, it is not now sufficient for regulatory evaluation of a drug's cardiac safety when requesting marketing approval. Other assessment methodologies include concentration-QT/QTc modeling,^25,26 assessment of additional ECG intervals such as the PR interval and QRS duration, as well as heart rate, and various characteristics of ECG morphology.^27-29 Importantly, as for QT/QTc interval prolongation, all these additional ECG data points benefit considerably from a centralized, standardized analysis strategy.³⁰

Current lack of centralization for Phase III data

In stark contrast to the approach mandated for the TQT study, the majority of ECG data from other trials worldwide is still collected and analyzed in a decentralized manner. This fragmented, non-standardized approach to collecting cardiac safety data is characterized by the use of different ECG machines (and/or different vintages of the same machine) at different sites, and the common use of paper print-outs of the ECGs.¹¹ For historical reasons, many machines that calculate QTc values employ the Bazett correction formula, which is arguably the most inaccurate.^31,32 Heterogeneity of equipment and software from different manufacturers introduces inconsistency. Kligfield et al. confirmed this long-held suspicion and concluded that systematic differences in QT measurements between manufacturers are significant and must be considered when used as markers of arrhythmogenic risk.³³ Individual printers can introduce errors by distorting the paper as ECGs are being printed. Different physicians at different sites then read these print-outs. While undoubtedly highly qualified in their own medical specialty, not all of these physicians will be highly trained and experienced in cardiac electrophysiology, and in the measurement and interpretation of ECG intervals and waveforms (see also Viskin et al.³⁴).

Further confounding late stage developmental studies is the extreme difficulty in overcoming the logistical issues involved. These trials typically engage large numbers of subjects over several years, and are therefore conducted in multiple outpatient settings. ECG acquisition characteristically consists of single tracings unaligned to pharmacokinetic peaks. The high degree of QT interval variability in even normal subjects challenges the significance of solo measurements as opposed to mean QT values. Likewise, strategic sampling at peak drug concentrations would increase reliability.

Centralizing Phase III ECG data

As disclosed at the end of this article, the authors work for commercial vendors of centralized, standardized, cardiac safety services, and these central labs stand to gain financially should wider adoption of the standardized system occur. That acknowledged, in the authors' collective opinion, employment of the centralized system for ECG data from Phase III trials is in the best interest of patients (the primary consideration) and also in the best interests of sponsors and regulators. No one questions that centralization of biological samples (their analysis at central labs) should be used in Phase III trials. Patients benefit every time optimum quality data are used in drug research, and, in the present context, they are best served when a sponsor utilizes a core lab and presents regulators with ECG data that permit the best possible assessment of the degree (if any) of cardiac risk. If a cardiac risk is identified and the drug's likely benefit is deemed to sufficiently outweigh the cardiac risk, the drug can still be approved. In this case, biopharmaceutical sponsors and regulators will employ risk minimization strategies. These approaches can allow the (possibly very large) majority of potential patients to reap the benefits of the newly approved drug while providing maximum protection to patients for whom a different treatment modality is preferable.

A major reason that many sponsors do not currently employ centralized approaches in Phase III trials is the misperception that it is considerably more expensive to do so. Therefore, in the absence of a regulatory mandate, they choose to use the decentralized approach. As Furlong¹¹ observed, "It remains a challenge for core labs to fully demonstrate to sponsors the added value brought about by centralization as well as the cost-savings realized through the complete study management process." Some members of our industry have already written on this topic^{11, 35-38} but all of us who work at core labs need to do more in this regard.

With regard to agency guidance, regulators have not yet addressed centralization of cardiac safety in later phase trials. It appears that cardiac safety core laboratories and the biopharmaceutical sponsors they work with will need to continue to explore the benefits that centralization has to offer in order to drive this process forward. One obstacle to this is the extreme focus on QT as a surrogate for cardiac safety. In Phase III trials, for drugs that have no QT changes in a TQT study, the focus should be on morphology rather than QT.

To these authors, it would seem prudent to use a strategy that is acknowledged to provide higher quality data for a (somewhat) more heterogeneous subject sample as typically employed in a Phase III trial to assess cardiac parameters as carefully as possible. From a workload perspective, reviewing data collected in a centralized manner rather than a decentralized one would not require extra work. To the contrary, higher quality data would potentially decrease the time needed. The downside, therefore, is not readily apparent. In contrast, the upside is clear: Higher quality data would provide greater assurance that potential risks had been looked for as thoroughly as possible in the preapproval data, the only data available at that time (subsequent postmarketing data are always to be considered once available). It is quite conceivable that a risk that would not have been identified with the data acquired from a decentralized approach would indeed be identified in data acquired in the manner advocated in this article, and hence regulators would have the opportunity to protect the public health in the manner they deemed most appropriate in the context of the magnitude of the risk and the magnitude of the drug's likely benefits.

Additional benefits to the centralized approach

The perception of a cardiac and/or cardiovascular risk being associated with a marketed drug—whether or not the risk is real—attracts enormous mass media attention. In the era of sensationalist, sound-bite coverage, clinical science sadly falls very low on the list of points to be covered in the allotted 30 seconds of television coverage (it can do somewhat better in print coverage). When a cardiac or cardiovascular safety concern is "identified" for a marketed drug, there is frequently a "torrent of recriminations."³⁹ Such recriminations are almost always directed at the biopharmaceutical sponsor, and the costs are considerable, ranging from loss of reputation to a decrease of billions of dollars in market capitalization. However, and perhaps increasingly so, they are targeted at regulatory agencies too. In addition to being in the best interests of patients, the paramount concern, doing everything possible to identify any degree of cardiac risk in preapproval trials is in the best interests of sponsors and regulators.

Conclusion

The centralized, standardized data collection process employed for TQT studies greatly enhances the quality and integrity of the data acquired and analyzed. Currently, in contrast to this approach, the majority of ECG data from other trials worldwide is still collected and analyzed in a fragmented, non-standardized manner. Many sponsors do not currently employ centralized approaches in Phase III trials because of the pervasive misperception that it is considerably more expensive to do so. Those of us who work at ECG core labs need to do more to increase sponsors' awareness of the added value brought about by adoption of a centralized approach as well as the cost-savings realized through the complete study management process being placed with a core lab. In addition, and moving beyond the vendor, biopharmaceutical sponsors then need to take joint responsibility with us in furthering the science of cardiac safety.

Authors' Note: The authors work for commercial vendors of centralized cardiac safety services.

J. Rick Turner,* PhD, is Senior Scientific Director, Cardiac Safety Services, at Quintiles, Durham, NC, e-mail: rick.turner@quintiles.com. Lawrence Z. Satin, MD, FACC, is Chief Medical Officer at Cardiocore, Bethesda, MD. Timothy S. Callahan, PhD, is Chief Scientific Officer at Biomedical Systems, St. Louis, MO. Jeffrey S. Litwin, MD, FACC, is Executive Vice President and Chief Medical Officer at ERT, Philadelphia, PA.

*To whom all correspondence should be addressed.

References

1. J.R. Turner, "Drug Safety, Medication Safety, Patient Safety: An Overview of Recent FDA Guidances and Initiatives," Regulatory Rapporteur, 6 (4), 4-8 (2009).

2. Food and Drug Administration, The Sentinel Initiative: A National Strategy for Monitoring Medical Product Safety, (FDA, Rockville, MD, 2008).

3. F. Dessertenne, "La Tachycardie Ventriculaire A Deux Foyers Opposes Variables," Arch des Mal du Cœur, 59:263 (1966).

4. J.R. Turner and T. A. Durham, Integrated Cardiac Safety: Assessment Methodologies for Noncardiac Drugs in Discovery, Development, and Postmarketing Surveillance. (John Wiley & Sons, Hoboken, NJ, 2009).

5. T. J. Bunch and M. J. Ackerman, "Cardiac Channelopathies," in Mayo Clinic Cardiology: Concise Textbook, 3rd Edition, J. G. Murphy, and M. A. Lloyd, eds. (Mayo Clinical Scientific Press/Informa Healthcare USA, Boca Raton, FL), pp. 335-344.

6. ICH Guideline E14, The Clinical Evaluation of QT/QTc Interval Prolongation and Proarrhythmic Potential for Non-antiarrhythmic Drugs, (Brussels, Belgium, 2005).

7. ICH Guideline E14 Implementation Working Group, The Clinical Evaluation of QT/QTc Interval Prolongation and Proarrhythmic Potential for Non-antiarrhythmic Drugs: Questions & Answers, (Brussels, Belgium, 2008).

8. "CPMP: Points to Consider on the Assessment of the Potential for QT Interval Prolongation by Non-Cardiovascular Medicinal Products," (CPMP/SWP/986/96.).

9. J. Morganroth, F. V. Brozovich, J.T. McDonald, and R.A. Jacobs, "Variability of the QT Measurement In Healthy Men: With Implications for Selection of an Abnormal QT Value to Predict Drug Toxicity and Proarrhythmia," American Journal of Cardiology, 67:774-776 (1991).

10. S.E. Steare and J. Morganroth, "A Rational Approach to the Planning and Analysis of Electrocardiogram Safety Data in Clinical Trials," International Journal of Pharmaceutical Medicine, 16:133-149 (2002).

11. A. Furlong, "Cardiac Safety: Cost to Trials," Applied Clinical Trials, January 2010, 58.

12. J.R. Turner, New Drug Development: An Introduction to Clinical Trials, 2nd Ed. (Springer, New York, 2010).

13. W.S. Redfern, L. Carlsson, A.S. Davis, et al., "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, 58:32-45 (2003).

14. G.J. Noel, D. B. Goodman, S. Chien, et al., "Measuring the Effects of Supratherapeutic Doses of Levofloxacin on Healthy Volunteers Using Four Methods of QT Correction and Periodic and Continuous ECG Recordings," Journal of Clinical Pharmacology, 44:464-473 (2004).

15. J.M. Morganroth and I. Gussak, eds, Cardiac Safety of Noncardiac Drugs: Practical Guidelines for Clinical Research and Drug Development, (Humana Press, Totowa, NJ, 2005).

16. C. Strnadova, "The Assessment of QT/QTc Interval Prolongation in Clinical Trials: A Regulatory Perspective," Drug Information Journal, 39:407-433 (2005).

17. P. Kligfield, B. Tyl, M. Marek, and P. Maison-Blanche, "Magnitude, Mechanism, and Reproducibility of QT Interval Differences Between Superimposed Global and Individual Lead ECG Complexes," Annals of Noninvasive Electrocardiology, 12:145-152 (2007).

18. C.M. Beasley, Jr, A. Dmitrienko, and M.I. Mitchell, "Design and Analysis Considerations for Thorough QT Studies Employing Conventional (10s, 12-Lead) ECG Recordings," Expert Review in Clinical Pharmacology, 1:815-839 (2008).

19. J.S. Litwin, R.B. Kleiman, and I. Gussak, "Acquired (drug-induced) Long QT Syndrome," in Electrical Diseases of the Heart: Genetics, Mechanisms, Prevention, I. Gussak I and C. Antzelevitch, eds. (Springer-Verlag, London, 2008) 705-718.

20. V. Salvi, D.R. Karnad, G.K. Panicker, and S. Kothari, "Update on the Evaluation of a New Drug for Effects on Cardiac Repolarization in Humans: Issues in Early Drug Development," British Journal of Pharmacology, 159:34-48 (2010).

21. K.A. Moore, T.S. Callahan, P. Maison-Blanche, et al., "Thorough Cardiac QTc Interval Conductance Assessment of a Novel Oral Tranexamic Acid Treatment for Heavy Menstrual Bleeding," Expert Opinion on Pharmacotherapy, August 2010. [e-publication ahead of print]

22. L.Z. Satin, T.A. Durham, and J.R. Turner, "Assessing a Drug's Proarrhythmic Liability: An Overview of Computer Simulation Modeling, Nonclinical Assays, and the Thorough QT/QTc Study," Drug Information Journal, (in press).

23. J. Morganroth, R. R. Shah, and J.W. Scott, "Evaluation and Management of Cardiac Safety Using the Electrocardiogram in Oncology Clinical Trials: Focus on Cardiac Repolarization (QTc interval)," Clinical Pharmacology and Therapeutics, 87:166-174 (2010).

24. J. Morganroth, H. Leppor, L.A. Hill, W. Volinn, and G. Hoel, "Effects of the Selective Alpha 1a-Adrenoceptor Antagonist Silodosin On ECGs of Healthy Men in a Randomized, Double-Blind, Placebo- and Moxifloxacin-Controlled Study," Clinical Pharmacology and Therapeutics, 87:609-613 (2010).

25. C.E. Garnett, N. Beasley, V.A. Bhattaram, et al., "Concentration-QT Relationships Play a Key Role in the Evaluation of Proarrhythmic Risk During Regulatory Review," Journal of Clinical Pharmacology, 48:13-18 (2008).

26. S. Rohatagi, T.J. Carrothers, J. Kuwabara-Wagg, and T. Khariton, "Is a Thorough QTc Study Necessary? The Role of Modeling and Simulation in Evaluating the QTc Prolongation Potential of Drugs," Journal of Clinical Pharmacology, 49:1284-96 (2009).

27. M.P. Andersen, J. Q. Xue, C. Graff, et al., "New Descriptors of T-Wave Morphology are Independent of Heart Rate" Journal of Electrocardiology, 41:557-561 (2008).

28. F. Extramiana, R. Dubois, M. Vaglio, et al., "The Time Course of New T-Wave ECG Descriptors Following Single- and Double-Dose Administration of Sotalol in Healthy Subjects," Annals of Noninvasive Electrocardiology, 15:26-35 (2010).

29. F. Extramiana, C. Tatar, P. Maison-Blanche, et al., "Beat-to-Beat T-Wave Amplitude Variability in the Long QT Syndrome," Europace, 12 (9), 1302-1307.

30. J.S. Litwin, "Cardiac Safety: Beyond QT," Applied Clinical Trials, Cardiac Safety in Clincial Trials Supplement, November 2010, 16-21.

31. M. Desai, L. Li, Z. Desta, M. Malik, and D. Flockhart, "Variability of Heart Rate Correction Methods for the QT Interval," British Journal of Clinical Pharmacology, 55:511-517 (2003).

32. M. Malik, K. Hnatkova, and V. Batchvarov, "Differences Between Study-Specific and Subject-Specific Heart Rate Corrections of the QT Interval in Investigations of Drug Induced QTc Prolongation," Pacing and Clinical Electrophysiology, 27:791-800 (2004).

33. P. Kligfield, E.W. Hancock, and E.D. Helfenbein, "Relation of QT Interval Measurements to Evolving Automated Algorithms from Different Manufacturers of Electrocardiographs," American Journal of Cardiology, 98:88-92 (2006).

34. S. Viskin, U. Rosovski, A.J. Sands, et al., "Inaccurate Electrocardiographic Interpretation of Long QT: The Majority of Physicians Cannot Recognize a Long QT When They See One," Heart Rhythm, 2:569-574 (2005).

35. J.S. Litwin, "The Merits of ECG Centralization: A Call to Sponsors to Rethink the Role of ECGs in Drug Development and the Use of Central Core Labs," Applied Clinical Trials, June 2008, 111-114.

36. A. Furlong, "Centralizing ECGs for Efficient, Accurate Cardiac Safety Assessments in Clinical Trials," International Pharmaceutical Industry, Winter 2009/2010 issue, 36-38 (2009).

37. A. Furlong, "Optimizing Centralized ECG Data Collection With New System Innovations," Journal for Clinical Studies, January 2010, 44-46.

38. A. Furlong, "Cutting Edge ECG Technology Reinforces the Continued Advancement of Centralised Cardiac Safety," International Pharmaceutical Industry, Summer 2010, 68-71.

39. B. Cobert, Manual of Drug Safety and Pharmacovigilance, (Jones and Bartlett Publishers, Sudbury, MA, 2007).