Effect of hydroxychloroquine and azithromycin on cardiac
repolarization in iPSC cardiomyocytes
Data from our screen of hERG potency showed only a subtle
(hydroxychloroquine), or non-significant (chloroquine), additive effect
of azithromycin (Supplementary figure 2). In order to measure the
effects of the combination of drugs directly on cardiac repolarisation,
as well as to assess whether there are any potentially proarrhythmic
effects that manifest with longer term exposure, we measured field
potential duration, a surrogate of the QT interval, in hiPSC derived
cardiomyocytes from three normal patients. Representative field
potentials for hydroxychloroquine alone (Figure 3A) and in combination
with azithromycin (Figure 3B) are shown for (i) baseline; (ii), acutely
after drug addition; (iii) 24 hrs and (iv) 48 hrs after drug addition.
Summary data for all three lines over time is presented in Figure 3C
(individual technical replicates for each cell line are presented in
Supplementary Figure 4). There was no significant difference in the
extent of prolongation of repolarisation between hydroxychloroquine
alone, or in combination with azithromycin (Repeated measures two-way
ANOVA; p = 0.109). However, there was a significant time dependent
effect (Repeated measures two-way ANOVA; p = 0.027), indicating an
accumulation of block with longer term exposure to the drugs.
Additionally, hydroxychloroquine and the combination of
hydroxychloroquine and azithromycin caused cessation of beating or loss
of capture in multiple wells (Table 2). This is consistent with previous
reports showing that profound drug-induced disruption of repolarisation,
corresponding to ~ 75 % or more block of IKr, can
result in cessation of beating or occurrence of arrhythmia (23). Next,
we assessed the effect of kalaemic variation on prolongation of
repolarisation by hydroxychloroquine. 10 µM hydroxychloroquine caused
significantly more prolongation in hypokalaemia (3 mM
K+, 7.7 ± 1.1 % prolongation (SEM; n=16)) than
normokalaemia (5.4 mM K+, 3.3 ± 0.5 % prolongation
(SEM; n=17)) demonstrating that the hypokalaemia-dependent increase in
potency of hydroxychloroquine against hERG, as measured in our
Syncropatch screens (Figure 2), manifests in a more severe prolongation
of cardiac repolarisation.