Applying the CiPA approach to evaluate cardiac proarrhythmia risk of some antimalarials used off‐label in the first wave of COVID‐19

Figures

• 

  FIGURE 1. Concentration‐response curves of hERG‐mediated current (I Kr) amplitude for hydroxychloroquine (OH‐CLQ) (a) and azithromycin (AZI) (b) tested alone or in combination. Isobologram for drug combinations of AZI and OH‐CLQ (c) was constructed. Data best fit with a linear equation, indicating no synergistic or antagonistic interaction. Values are mean ± SD

• 

  FIGURE 2. Concentration‐response curves of a selection of AP and calcium transient (CT) electrophysiological biomarkers, simulated in a control (healthy) population of virtual cell models treated with chloroquine (CLQ), hydroxychloroquine (OH‐CLQ), azithromycin (AZI), and erythromycin (ERT), and their combinations. See the main text for biomarker descriptions. In each boxplot, the central mark is the median of the population, box limits are the 25th and 75th percentiles, and whiskers extend to the most extreme data points not considered outliers, which are plotted individually as separate crosses. Only the models not displaying repolarization abnormalities (RAs) or depolarization abnormalities (DAs) are plotted for each concentration (the number of cells is listed in brackets on top of each boxplot). CLQ and, with smaller magnitude, OH‐CLQ affected all biomarkers even at low concentrations, whereas AZI and ERT only showed effects at high concentrations. Combinations of an antimalarial with an antibiotic did not worsen the response. EMw, electromechanical window

• 

  FIGURE 3. (a) Summary of the drug‐induced abnormalities (repolarization abnormalities [RAs], left; depolarization abnormalities [DAs], right) observed in silico for each of the tested drugs and combinations, in the control (top) and high‐risk (bottom) populations. RAs appear at low concentrations of chloroquine (CLQ), the RA prevalence is not increased by combination with an antibiotic but is higher in the high‐risk population. DAs only occur at higher concentrations. (b, c) AP and calcium transient (CT) traces obtained for the control population when simulating CLQ at 5 µM. The in silico models displaying RAs are highlighted in blue, whereas the ones showing normal—even though prolonged—repolarization, are plotted in gray. About half of the cells showed RA. AZI, azithromycin; ERT, erythromycin; OH‐CLQ, hydroxychloroquine

• 

  FIGURE 4. Changes observed in the qNet biomarker following CLQ, OH‐CLQ, AZI, and ERT application. Simulation results are shown for the ORd‐CiPA model (black diamonds) and the corresponding control population (boxplots). (a) Simulation results obtained using the half‐maximal inhibitory concentration (IC50)/Hill coefficients of dataset #1 from Table 1. (b) Simulation results obtained using the IC50/Hill coefficients of dataset #2 from Table 1. In each panel, the thresholds suggested by Li et al. 36 to separate low/medium/high risk drugs are shown as black dotted and dashed lines. CLQ would be classified as high‐risk, OH‐CLQ as medium risk, and AZI and RT as low risk drug. AZI, azithromycin; CLQ, chloroquine; ERT, erythromycin; OH‐CLQ, hydroxychloroquine; ORd‐CiPA, O’Hara‐Rudy Comprehensive In Vitro Proarrhythmia Assay

• 

  FIGURE 5. Composite “QT‐ograms” integrating concentration‐responses to CLQ, OH‐CLQ, AZI, and ERT generated in hERG patch clamp assays at room temperature (dataset #1, triangles) or 37°C (dataset #2, diamonds), in virtual human ventricular myocyte population using ion channel half‐maximal inhibitory concentration (IC50) values from dataset #1 (plain squares for APD90, plain circles for electromechanical window [EMw]) or dataset #2 (open squares for APD90, open circles for EMw). Effects are expressed as absolute percentage change from control conditions (percentage inhibition for hERG, percentage increase in APD90, or percentage decrease in EMw for virtual cardiomyocytes). For CLQ, percentage QTc increase in man (crosses) are extracted from the clinical study GSK protocol TAF106491 (clinicaltrials.gov identifier NCT00871156) and were estimated assuming a baseline corrected QT (QTc) of 400 ms. Vertical bars represent therapeutic free plasma concentration (FPC) of each drug and multiples. AZI, azithromycin; CLQ, chloroquine; ERT, erythromycin; OH‐CLQ, hydroxychloroquine; RT, room temperature