Stem cell-derived cardiomyocyte assays and their new role in CV safety testing

Cardiac toxicity is a leading contributor to late-stage attrition in drug discovery programs. It has been estimated that up to 60% of compounds developed as potential drugs have tested positive for cardiac risk by current standards and subsequently terminated from development despite their potential therapeutic benefit versus risk.

In recognition of the need to develop earlier and more predictive cardiac safety testing, an expert working group, sponsored by the Cardiac Safety Research Consortium (CSRC), Health and Environmental Sciences Institute (HESI) and the US Food and Drug Administration (FDA), has proposed a new strategy entitled, “Comprehensive in vitro Proarrhythmia Assay” (CiPA). This proposal shifts emphasis away from QT-interval prolongation (a weak surrogate marker for drug-induced cardiac arrhythmia) and focuses on predicting cardiac risk using a combination of in vitro tests.

CiPA, expected to be implemented by 2016, consists of two components: an evaluation of the effects of test compounds in conventional cardiac ion channel patch clamp assays paired with in silico modeling to interpret the data, and novel, integrative assays that measure drug effects in cardiac muscle cells, otherwise known as cardiomyocytes.

The most relevant cells for integrative assays are live, adult human ventricular cardiomyocytes, but sources are extremely limited—myocytes are infrequently available when hearts go unused in transplanation operations. So the field is eyeing stem cell-derived cardiomyocytes (SC-hCMs) as a potentially new model for safety testing.

This next-generation approach would not have been possible a decade or even five years ago. But we’ve learned a lot since induced pluripotent human stem cells or iPSCs, which have the ability to replicate indefinitely and differentiate into different cell types including cardiomyocytes, were first reported in 2006. Stem-cell derived human cardiomyocytes are now commercially available in sufficient enough quantities to be practical for routine safety pharmacology studies, and there are a number of SC-hCM assays available that evaluate the heart’s viability, contractility and electrical activity.

Here’s a brief synopsis of two SC-hCM assays that we offer.

Multiple Electrode Array Assay

SC-hCMs grown in culture dishes form a spontaneously beating cellular layer that displays electrical activity similar to an intact human heart, depicted in this video.


Video 1. iCell® Cardiomyocytes from Cellular Dynamics International


When grown in contact with a multiple electrode array (MEA) at the bottom of the dish, the cells generate a signal that closely resembles a whole-body electrocardiogram (ECG). Miniaturization of the recording system in the form of multi-well (48 or 96 wells) MEA assay plates (Figure 1) enables simultaneous, parallel measurements. This “heart-in-a-dish” technology, commercialized by several equipment companies in recent years, provides rapid safety testing of compounds at multiple concentrations and time points.

Figure 1
Figure 1. MEA 48-well assay plate (Maestro MEA system, Axion Biosystems)

Figure 2
Figure 2. Examples of electrical field potential recordings at ascending concentrations of test compound. An arrhythmic event was observed at the highest concentration (white trace).

The ECG-like signal, “electrical field potential” provides a read-out of the integrated functioning of intact myocytes (Figure 2, ChanTest validation data). This allows the identification of several electrical irregularities including arrhythmia, premature beats, beat rate slowing, conduction velocity decrease, and beat prolongation.

Benefits of the MEA assay include:

  • Much higher throughput than classical single-cell techniques
  • A decreased need for animal models, while providing a more translatable, physiologically-relevant human cell culture model
  • The ability to administer test compounds over long periods of time, and at higher concentrations than could be tolerated in animals or humans.

Cell Impedance/Contractility Assay

SC-hCMs beat spontaneously in a regular, rhythmic pattern that can be detected as transient changes in impedance or impedance-twitches, i.e., electrical irregularities generated by myocyte contraction, that are recorded with sensitive amplifiers coupled to a multi-electrode array. Impedance-based measurement of SC-hCM contractile activity in multi-well instruments, available from several manufacturers, adds a whole new dimension to cardiac risk assessment. Assay validation data show that key aspects of cardiac excitation-contraction coupling are present in SC-hCMs, the dependence of the twitch contraction on Ca2+ entry being one example(Figure 3, ChanTest validation data).

By accurately recapitulating the heart’s ability to contract in vitro, the impedance recordings allow assessment of drug-induced effects that could translate into compromised cardiac function in vivo. Moreover, the multi-well format offers a significant throughput advantage over current ex vivo or in vivo animal-based testing.

Figure 3
Figure 3. Twitch Impedance regulated by calcium entry. Impedance-twitch traces from three wells (60 s sweeps/well) are superimposed and averaged (Mean ± SEM). The L-type calcium channel agonist FPL64176 (1 µM) increased twitch duration and amplitude, altered kinetics and slowed beat rate. Opposite effects were observed after administration of 100 nM verapamil, a calcium channel blocker. The graph shows the concentration-dependence of the verapamil effect (ΔRel80 = change in duration at 80% relaxation decay time point).

The ability to identify cytotoxic compounds is another important attribute of the impedance assay. Average impedance (‘Cell Index’, CI) is a sensitive indicator of cell attachment to the substrate. Cell death causes detachment and subsequent decrease in CI. Figure 4 (ChanTest validation data) shows a typical example of the effect of long exposure to a cardiotoxic compound (doxorubicin).

Figure 4
Figure 4. Long-term stability of the impedance signal enables detection of cardiotoxicity. Cell Index (CI) was normalized to zero, averaged (Mean ± SEM; n = 6 per condition) and plotted over 54 hours. The reduction in normalized CI in the presence of doxorubicin indicates progressive cytotoxicity. Vehicle control (0.1% DMSO) and pentamidine (known to affect electrical activity of cardiomyocytes without cytotoxicity) did not change the average impedance.

Like any in vitro model, the challenge to stem-cell cardiomyocyte assays will be to translate the results into predictions of clinical outcomes. Two factors are particularly relevant. First, the electrical properties of SC-hCMs more closely resemble fetal than adult characteristics. Second, the contractile properties of two-dimensional stem cell-derived monolayers are likely to be different from those of the three-dimensional heart. Nonetheless, SC-hCMs, thus far, have shown expected pharmacological responses to reference compounds.

As the rollout of CiPA gets closer, it is anticipated that with additional SC-hCM validation, the assays will make a significant contribution to improving the efficiency and predictivity of cardiac safety assessment in drug discovery.

This hopefully will go a long way toward ensuring the safety of marketed drugs.