Cardiac Safety in Drug Development: CiPA

In the second of a three-part series looking at cardiac safety testing in early drug discovery, Dr Michael Morton of ApconiX considers the CiPA initiative and how its integrated approach sits alongside existing strategies.

What is the CiPA initiative and what does it mean for drug discovery?

In the first blog, we outlined the background to the hERG assay and its adoption as a screening strategy to avoid the development of potentially life-threatening drugs causing Torsades de pointes (TdP).  High throughput hERG screening has since become a cornerstone of safety testing in drug research and has led to a marked reduction in drugs entering clinical trials that prolong the QT interval of the ECG and there have been no recent withdrawals from the market of drugs causing TdP.

However, a focus on inhibition of the hERG channel alone, as a surrogate marker for pro-arrhythmic potential, paints an oversimplified picture of the relationship between hERG inhibition, QT prolongation and TdP.  We know for example, that not all drugs that inhibit hERG prolong the cardiac QT interval, and not all drugs that prolong QT lead to TdP.  In the later stages of a clinical development program, the final assessment of a drugs ability to increase the duration of the QT interval is made in the ‘Through QT study’ (TQT).  A finding in this study of even a small change in QT is often the death-knell for that compound.

Yet we know that the coordinated propagation of an electrical impulse along cardiac muscle tissue relies on a complex array of ionic movements across the plasma membrane, first to depolarise and then repolarise each cell.  This movement of ions across the cell membrane via the opening of specific channels creates the action potential.  Drugs associated with TdP were subsequently shown to delay to the repolarisation phase, by reduction of a potassium current (IKr) flowing through the hERG channel. However, we know that pharmacological modulation at many of these channels can affect the action potential and lead to susceptibility to arrhythmia.

The CiPA (Comprehensive In Vitro ProArrythmia Assay) initiative proposes to introduce an earlier evaluation of proarrhythmic risk based on a mechanistic understanding of the effects of novel compounds on multiple ion channel currents, and to do away with the evaluation of QT interval changes as a surrogate for risk of TdP   This initiative has four primary components:

1. In vitro drug effects on a range of cardiac ion channels
2. In silico reconstruction or modelling of changes to the action potential in a virtual cardiac cell

• Confirmation of predicted effects using human stem-cell derived cardiomyocytes, and

1. An earlier assessment of the clinical ECG

The in vitro approach

CiPA moves us away from the reductionist hERG approach (as described in the first blog) to looking at multiple ion channels, in silico modelling and whole cell effects to derive a more mechanistic basis for the assessment of risk.  The new CIPA paradigm will be driven by a panel of ion channel assays (up to 7 currents) coupled to in silico mathematical reconstructions of the action potential and supported by predicted and observed responses in human-derived stem cell cardiomyocytes. The aspiration is that this CiPA algorithm will become a freely-available tool, however, the data output would likely still require specialist interpretation for each individual context.

The next steps for CiPA

As with any bold new initiative, there are many challenges to be resolved and much work still ahead.  Stem cell-derived cardiomyocytes are an imperfect representation of primary human heart cells.  The relationship between changes in measured parameters and the pro-arrhythmic potential of key compounds in the validation set is unlikely to be a simple one.  Even with the standardisation of protocols and assay conditions, there may well be variability of data from different suppliers.  One benefit of using stem cells is that they allow for longer exposure (up to 48 hours) compared with the ion channel assays typically run over several minutes. To the extent that the ongoing validation program is successful, CiPA will petition for changes to the regulatory requirements for pro-arrhythmia assessment, including eliminating or waiving the need for the ‘Thorough QT study’.

Although the drug industry has been performing hERG, as well as sodium and calcium ion screening for some time, most have been conducting little, if any,  in silico modelling or stem cell work.  Despite the current uncertainty, many companies have already begun to supplement their traditional approaches with much broader datasets on potential nomination candidates, even though specific CiPA proposals have yet to be published.

In the absence of specific guidance, these companies are likely to run their own ‘CiPA-like’ approaches for a number of years to generate parallel datasets, to inform decisions on cardiac safety.

At ApconiX –in conjunction with our partners PhysioStim and QT Informatics – we’re already seeing requests to perform ‘CiPA-like’ ion channel profiling, data modelling and stem cell testing.  We are finding that the combined output from these experimental approaches can be incredibly informative in guiding better decisions in drug discovery.  We recently published an example of this approach The CiPA Profile of Two Adenosine Uptake Inhibitors, dilazep and dipyridamole

CiPA, regulations and the future

It’s been said that: “All models are wrong, but some are useful”*, and this will also apply to CiPA.  There are undoubtedly many challenges and set-backs ahead before a universally agreed set of CiPA proposals are adopted that adequately predict the pro-arrhythmic potential of new drugs prior to first human trials.    Nevertheless, a new paradigm for assessment of clinical potential of TdP, that is not measured exclusively by potency of hERG inhibition and not at all by QT prolongation, is a bold endeavour that deserves our support.

*attributed to George Box, 1975