Intracellular Uptake of Agents That Block the hERG Channel Can Confound Assessment of QT Prolongation and Arrhythmic Risk

Abstract

Background:

Oliceridine is a biased ligand at the μ-opioid receptor recently approved for the treatment of acute pain. In a thorough QT study, QTc prolongation displayed peaks at 2.5 and 60 min following a supratherapeutic dose. Mean plasma concentration peaked at 5 min, declining rapidly thereafter.

Objective:

The present study examines the basis for the delayed effect of oliceridine to prolong QTc.

Methods:

Repolarization parameters and tissue accumulation of oliceridine were evaluated in rabbit left-ventricular wedge preparations over a period of 5-hours. Effects of oliceridine on ion channel currents were evaluated in HEK and CHO cells. Quinidine was used as a control.

Results:

Oliceridine and quinidine produced progressive prolongation of QTc and action potential duration (APD) over a period of 5-hours, paralleling slow progressive tissue uptake of the drugs. Oliceridine caused modest prolongation of these parameters, whereas quinidine produced prominent prolongation of APD and QTc, development of early after-depolarization (after 2-hrs) resulting in a high Torsade-de-Pointes (TdP) score. IC₅₀ values for oliceridine inhibition of I_Kr (I_hERG) and late-I_Na were 2.2 and 3.45μM, when assessed following traditional acute exposure, but much lower following 3 hrs of drug exposure.

Conclusions:

Our findings suggest that a gradual increase of intracellular access of drugs to the hERG channels as a result of their intracellular uptake and accumulation can significantly delay effects on repolarization, thus confounding assessment of QT prolongation and arrhythmic risk when studied acutely. The multi-ion channel effects of oliceridine, late-I_Na inhibition in particular, point to a low risk for development of TdP.

INTRODUCTION

Oliceridine is a biased ligand at the μ opioid receptor recently approved for the treatment of acute pain in adult patients for whom an intravenous opioid is warranted and for whom other treatments have proved inadequate. In a thorough QT (TQT) study designed to assess its QT liability, oliceridine (IV injection for 5 min) caused two peaks in placebo- and baseline-corrected QTc prolongation, the first at 2.5 min and the second at 60 minutes after a supratherapeutic dose of 6 mg (Fig. 1). Plasma concentration of oliceridine reached a peak within minutes (T_max 5 min; C_max 283.9 ng/ml; 6 mg) and declined rapidly thereafter (Fig. 1). The principal aim of the present study was to advance our understanding of the cellular mechanisms underlying the actions of oliceridine. The study was specifically designed to elucidate 1) the mechanism underlying the effect of oliceridine to transiently prolong the QT interval as well as 2) the mechanism(s) underlying the approximately one hour delay in attainment of peak QT interval prolongation observed following a 6 mg (supratherapeutic) bolus of oliceridine in the TQT study (Fig. 1). Our study tests the hypothesis that the delay in attainment of peak QT prolongation is due to progressively increasing intracellular access of the drug to hERG channels secondary to its intracellular uptake and accumulation. The protocols utilized are designed to compare the timecourse of effect of oliceridine to prolong endocardial action potential duration (APD), QT and QRS intervals with the timecourse of intracellular accumulation of oliceridine in rabbit left ventricular wedge preparations. Quinidine was selected as a positive control because it has similarly been reported to cause a delayed effect on ventricular repolarization in both clinical and experimental studies.^1-3

Figure 1.

Results of Thorough QT study of the effect of oliceridine administered as a supratherapeutic dose. The placebo and baseline-corrected QTcF (ΔΔQTcF, including 95% CI) and plasma oliceridine concentrations (+/− SEM) are plotted as a function of time after administration of a single 6 mg dose of oliceridine to healthy subjects. The dashed line represents the regulatory threshold of interest for a change in the ΔΔQTcF interval.

METHODS

Rabbit isolated coronary-perfused left ventricular prpearations

The investigation was performed according to the Guide for Care and Use of Laboratory Animals published by the US National Institute of Health (NIH publication N85-23, revised 1985). The study was approved by the Institutional Animal Use and Care Committee of the Lankenau Institute for Medical Research. Rabbits were initially given xylazine at a dose of 4-8 mg/kg IM. Once sedated, the rabbits were anticoagulated with heparin (human pharmaceutical grade, 1000 U/kg, IV) and euthanized using ketamine/xylazine (80-150 mg/kg / 4.0-8.0 mg/kg IV) administered via an ear vein. The chest was then opened via a left thoracotomy, the heart excised and placed in a cardioplegic solution consisting of cold (4 °C) normal Tyrode’s solution. The left coronary artery was cannulated and perfused with Tyrode’s solution. Unperfused areas of the left ventricle (easily identified from the reddish appearance due to unwashed erythrocytes) were excised. The preparation was then placed in a small tissue bath and coronary perfused with Tyrode's solution containing 3 mM K⁺ bubbled with 95% O₂ and 5% CO₂ (Temperature: 37.0 ± 0.1°C, mean perfusion pressure: 35-45 mmHg). The preparation was stimulated using bipolar silver electrodes insulated except for the tips applied to the endocardial surface at a cycle length of 1 sec. The ventricular wedge preparation was initially perfused with drug-free solution for 60 min, a period sufficient to attain electrophysiological stability.

A transmural electrocardiogram (ECG) was recorded using extracellular silver/silver chloride electrodes placed within the tissue bath at 1.0 to 1.5 cm from the epicardial and endocardial surfaces of the preparation. The QT interval was defined as the time from the onset of the QRS to the point at which the tangent to the final downslope of the T wave crosses the isoelectric line. The QRS duration was measured as the interval from the beginning of the Q wave to the end of the S wave. A transmembrane action potential was recorded from the endocardium of the ventricular wedge preparation using a floating glass microelectrode. The glass microelectrode was made from a glass pipette using a micropipette puller (Sutter Instrument Co - Model P-87, USA). The microelectrode was then filled with 2.7 M KCl and connected to a differential amplifier via a silver wire. Action potentials were recorded for measurement of action potential duration at 90% repolarization (APD₉₀) and for identification of early afterdepolarization (EAD).

Five durations of oliceridine exposure were tested, i.e., 1, 2, 3, 4 and 5 hours. ECG parameters and transmembrane action potentials were recorded every 30 minutes. Following each of these periods, the wedge was perfused with drug-free Tyrode’s solution for a period of 10 minutes before obtaining the final recordings and freezing of the tissue. Time control experiments (negative control) were used to demonstrate the excellent stability of the preparation, showing no significant change in the electrophysiological parameters over a period of 6 hours (Online Fig. S1 and Online Table S1). A positive control involved exposure of the wedge preparations to quinidine HCl. Clinically relevant concentrations of oliceridine (1 μM) and quinidine (3.3 μM) were tested.

Tissue uptake of drugs

To evaluate the possible effects of tissue uptake and accumulation of oliceridine and quinidine (positive control) on QT prolongation bioanalytical methods were developed and utilized to measure oliceridine and quinidine concentrations in homogenized cardiac tissue (n = 4-5 per timepoint) (Frontage Laboratories, Concord, OH). These methods are presented in the Online Supplement.

Ion channel profile of oliceridine

The in vitro effects of oliceridine on ion currents were evaluated using voltage-clamp techniques. The methods for evaluation of the effect of oliceridine on the rapidly activating delayed rectifier current (hERG, I_Kr), calcium channel current (Ca_V1.2), I_Ca), peak and late sodium channel currents (Na_V1.5, I_Na). Acute vs. chronic effects of the drug on I_Kr and late I_Na were evaluated as well. These methods are presented in the Online Supplement.

Drugs

Oliceridine was obtained from Trevena Inc. and quinidine was purchased from Sigma-Aldrich Inc. Both drugs were prepared fresh and dissolved in distilled water before each experiment (10 mM stock sulution).

Statistics

Statistical analysis of the data was performed using one-way repeated measures analysis of variance ANOVA followed by a Bonferroni test or using Student’s t test. Data are reported as mean ± SD or mean ± SE (as indicated).

RESULTS

Ion Channel Current Data

Figure 2 shows the effect of oliceridine to inhibit I_Kr (hERG channels) stably expressed in HEK cells. Analysis of the data yields an IC₅₀ of 2.2 μM. Figure 3 shows the inhibitory effect of oliceridine on late I_Na (IC₅₀ of 3.45 μM). Table 1 summarizes the IC₅₀ values for oliceridine block of hERG, Ca_V1.2, and Na_V1.5 (tonic and phasic) and late I_Na channels. The table includes the effects of oliceridine’s major metabolites. Based on the relatively high IC₅₀ values observed, oliceridine is not expected to have any clinically relevant effect on the cardiac action potential.

Figure 2: Oliceridine inhibition of hERG potassium current (I_Kr).

A: Upper panel shows superimposed, records of hERG potassium currents obtained in a typical cell during application of control, oliceridine (10 μM, TRV130A) and positive control (E-4031). hERG potassium currents were evoked by the voltage protocol shown in the lower panel. B: Concentration response relationship for oliceridine inhibition of I_Kr yielding an IC₅₀ of 2.2 μM.

Figure 3:

A: Oliceridine inhibition of the late sodium channel current (Late I_Na). Effect of 5 nM ATX-II to induce late I_Na and the effect of increasing concentrations of oliceridine to inhibit this current. B: Concentration-response relationship of the effect of oliceridine to inhibit late I_Na yielding an IC₅₀ of 3.45 μM.

Table 1.

IC₅₀ values (μM) for oliceridine and its major inactive metabolites for inhibition of hERG, Ca_V1.2 and Na_V1.5 (tonic and phasic) and late Na_V1.5 ion channel currents

Compound	hERG IKr	NaV1.5 Peak INa	CaV1.2 ICa	NaV1.5 Late INa
Oliceridine	2.2	>10; 19.51* (tonic) 9.023; >10* (phasic)	>10 to 39.63*	3.45
TRVO109662	>300	>300 (tonic) >300 (phasic)	>300	>300
TRV0306954 (M22)	>300	>300 (tonic) >300 (phasic)	>300	>300

^*Results from two independent studies

The two major inactive metabolites of oliceridine, TRV0109662 and TRV0306954 (M22) did not have any clinically relevant effect on any of the ion channels tested (Table 1).

QT Interval and APD₉₀

Oliceridine and quinidine were both observed to produce a progressive prolongation of the QT interval and APD₉₀ over a period of 5 hours, without achieving a steady-state, particularly at pacing CLs of 1000 and 2000 ms (Fig. 4-5; Online Tables S2-11). The degree of APD and QT interval prolongation was much greater during the initial 30-60 minutes of exposure to the drugs (Figs. 4-5; Online Tables S2-11). The effect of oliceridine on ventricular repolarization was modest compared to that of quinidine.

Figure 4.

Modest time-dependent prolongation of repolarization by oliceridine (1 μM) and dramatic time-dependent prolongation of repolarization by quinidine (3.3 μM). Shown are superimposed simulatenously recorded ECGs and endocardial action potentials (AP) before (Cntr), after 30, 60, 120, 180, 240 and 300 min of exposure to oliceridine and quinidine, and after 10 min of wahsout of these agents. Pacing CL = 1000 ms.

Figure 5.

Time- and rate-dependent effects of oliceridine and quinidine to prolong QT interval and action potential duration at 90% repolarization (APD90). * - p<0.05 vs. respective control. † - p<0.001 vs. respective control. ** - p<0.05 vs. 300 min. N= 9 for oliceridine and n = 8 for quinidine for most of the data points, with fewer data points at the slow pacing rates due to spontaneous activity with a CL shorter than 1000 and/or 2000 ms CLs (minimum n = 6 for oliceridine and n = 5 for quinidine; noted in online Tables S6 and S11). Mean±SD

The effect of both drugs was bradycardia-dependent, displaying a much greater prolongation of APD₉₀ and QT interval at slower pacing rates (Figs. 4-5; Online Tables S2-11). Ten minutes of washout of quinidine led to a paradoxical prolongation of the QT interval, at a CL of 1000 ms following 5 hours of quinidine exposure (Fig. 5). Ten minutes of washout of oliceridine caused no statistically significant change in QT or APD₉₀ (Fig. 5).

Early afterdepolarizations (EADs) were observed in 3 preparations exposed to quinidine (online Fig. S2) but in none of the preparations exposed to oliceridine.

QRS Duration

QRS was not significantly affected by oliceridine under any of the experimental conditions studied (Fig. 6; Online Tables S2-6). In contrast, QRS was significantly prolonged by quinidine at CLs of 500 and 1000 ms but to a much lesser extent at a CL of 2000 ms (Fig. 6; Online Tables S7-11).

Figure 6.

Time- and rate-dependent effects of oliceridine and quinidine on QRS duration. * - p<0.05 vs. respective control. † - p<0.001 vs. respective control. N= 9 for oliceridine and n = 8 for quinidine for most of the data points, with fewer data points at the slow pacing rates due to spontaneous activity with a CL shorter than 1000 and/or 2000 ms CLs (noted in Online Supplemental Tables 6 and 11). Mean±SD.

Correlation of Electrical Changes Induced by Oliceridine and Quinidine withTissue Uptake of These Drugs.

Both oliceridine and quinidine dispalyed progressive tissue uptake and accumulation, presumably due to intracellular uptake, over a period of 5 hours without achieving a steady-state (Fig. 7). There was a direct correlation between progressive prolongation of APD and QT intervals and tissue uptake of these pharmacological agents (Figs. 4 and 5).

Figure 7.

Progressive uptake and accumulation of oliceridine and quinidine in rabbit left ventricular tissues. * - p<0.05 vs. respective 1-hour value. ** - p<0.05 vs. respective 2-hour value. † - p<0.05 vs. respective 3-hour value. Mean ± SE. n=8 for all except the quinidine 3-hour value which included an n=10.

Time-dependence of Ion Channel Effects of Oliceridine

Because voltage clamp studies designed to evalaute IC₅₀ values are typcally performed with relatively short exposure of the the cells to the drug (on the order of minutes), it is possible that longer exposure permitting intracellular uptake and accumulation of drug causing progressive time-dependent inhibition of the ion channel current, would yield very different IC₅₀ values.

As a test of this hypothesis we performed additional studies comparing acute (minutes) vs. chronic (hours) effects of olicerdine on both I_Kr and late I_Na. Online Supplment Figure S3 compares the inhibitory effect of oliceridine following several minutes and 3 hours of exposure to the drug showing greater inhibition of I_Kr as a result of intracelular accumulation of the drug. Acute exposure to 1 μM oliceridine resulted in 35% inhibition of I_Kr consistent with an IC₅₀ of approximately 2.2 μM as determiend in a variety of patch clamp approaches, whereas chronic exposure (3 hrs) to 1 μM oliceridine resulted in 55% inhibition, pointing to an IC₅₀ of slightely less than 1 μM.

Online Supplement Figure S4 compares the effects of acute and chronic exposure to oliceridine on inhibition of late I_Na, showing greater inhibition of late I_Na following longer exposure to the drug. Here again, data collected following chronic exposure (3 hrs) to oliceridine point to an IC₅₀ value much lower than the IC₅₀ value of 3.45 μM measured using traditional patch clamp studies involving acute exposure to the drug. These data are consistent with the slow progressive effect of olicerdine to prolong APD and QT interval observed in the isolated coronary-perfused wedge preparations and provide evidence in support of the hypothesis that tissue uptake of oliceridine measured in the ventricular tissues is due to intracellular uptake and accumulation of the drug.

DISCUSSION

Opioid analgesics like morphine, fentanyl and hydromorphone are mainstays of therapy in the management of pain. Their use is hampered by well-known adverse effects, including respiratory depression, nausea, vomiting, prolonged ileus, and sedation. Adverse effects of opioids impact patient recovery in different ways: the threat of respiratory depression may lead to under-dosing of opioids, gastrointestinal adverse effects may delay oral intake and may thereby prolong the patient’s stay in the hospital, and sedation may lead to longer time to ambulation or an increased risk of falls. ⁴

Oliceridine is a μ-opioid receptor agonist, recently approved by FDA for the treatment of acute severe pain. Oliceridine elicits robust G protein signaling, with potency and efficacy similar to morphine, but with far less β-arrestin2 recruitment, resulting in less gastrointestinal (GI) dysfunction and respiratory depression than morphine, at equianalgesic doses. The G protein-biased profile of oliceridine and improved therapeutic window for analgesia relative to respiratory and GI dysfunction demonstrated in nonclinical studies suggest that oliceridine can provide an attractive alternative for the management of moderate to severe acute pain when opioid medications are clinically warranted.

In the thorough QTc study conducted to assess its proclivity to prolong QTc, oliceridine administration was observed to cause two peaks in QTc prolongation, the first at 2.5 minutes and a second peak one hour after initial dosing (Fig. 1); both peaks were modest and present only at the supratherapeutic dose of 6 mg. The initial prolongation of QTc (Fig. 1; about 3% increase of the QTc interval) can be accounted for by the direct effect of the drug to inhibit I_Kr since it is temporally aligned with the rapid rise in plasma concentration of oliceridine; Fig. 1). The QTc abbreviation that follows can likewise be accounted for by the decline in plasma concentrations of oliceridine (Fig. 1). However, the subsequent rise of QTc in the face of continued decline of plasma concentration of oliceridine is unexpected (Fig. 1).

A fundamental principle of clinical pharmacotherapeutics is the existence of a relationship between drug concentration and pharmacological and toxicological effects of drugs. ⁵ Pharmacokinetic-pharmacodynamic (PK-PD) models used in the evaluation of the concentration-QTc response of drugs most commonly assume that plasma concentration is in equilibrium and proportional with the effect site (BioPhase) concentration. In its simplest form the drug effect is directly proportional to drug concentrations at the active site and this relationship is independent of time.⁶ The relationship between drug concentration and pharmacological effect most often follows a sigmoidal E_max model (Hill equation).

The further prolongation of APD₉₀ and QTc following washout of the drug is likely attributable to diminished late I_Na inhibition attending removal of the drug extracellularly.¹ An outward shift in the balance of current is expected, giving rise to a paradoxical prolongation of repolarization during washout of the drug. These findings are consistent with the recognition that oliceridine and quindine like other hERG blockers inhibit the channel via intracellular binding to the hERG channel. ^{7, 8}

A similar delay between peak plasma concertation and maximum QTc prolongation has previously been described for other drugs that prolong QTc interval, including quinidine, sotalol, clazosentan.^{2, 9, 10} In the case of oliceridine , the two major inactive metabolites do not appear to be the cause of the delay since they lack electrophysiological effects (Table 1). In the case of both oliceridine and quinidine, tissue uptake and accumulation of the drugs likely contributes to the plasma concentration-QT effect hysteresis resulting in a modest increase in QTc one or more hours following a supratherapeutic dose (Fig. 1).

We selected quinidine as a positive control because quinidine is known to cause a concentration-QT effect hysteresis and acquired long QT syndrome in clinical studies.² Interestingly, in 1964, quinidine was the first drug recognized to prolong QTc and lead to the development of life-threatening arrhythmias.¹¹ Even in this early study it was recognized that quinidine “syncope” generally occurred with a delay, 1 to 3 hours after the last dose of the drug, and that the arrhythmia responsible could recur even after the dose of quinidine was reduced.

In previous studies, our team probed the basis for this delay in the electrophysiologic effects of quinidine by examining the time course of APD changes in relation to intracellular uptake of the drug in isolated canine Purkinje fibers and ventricular tissues following exposure to 3.3 μM quinidine for a period of several hours.¹ The time course of quinidine-induced changes in APD in canine ventricular epicardium recorded under these conditions closely paralleled the time course of tissue uptake of the drug.

Intracellular uptake of oliceridine can also explain the mismatch between plasma levels of the drug and its ability to prolong QTc. When studied in human HEK expressing hERG, oliceridine was found to inhibit the rapidly activating I_Kr with an IC₅₀ of 2.2 μM (Table 1). In the clinical TQT trial (Study CP130-1008), the mean Cmax at the 6 mg supra-therapeutic dose was 283.9 ng/ml . The unbound concentration at Cmax was 56.1 ng/mL or 0.145 μM, which is more than 15-fold lower than the IC₅₀ for inhibition of I_Kr (2.2 μM). Therefore, it seems clear that the plasma concentration of oliceridine alone cannot explain the effect seen at 1 and 2 hours following a 6 mg dose.

Our results provide support for the hypothesis that the mismatch between Cmax and IC₅₀ for inhibition of I_Kr as well as the delayed peak of QT prolongation observed in the TQT Study are due to gradual uptake and accumulation of oliceridine intracellularly in the myocardial cells resulting in greater interaction of the drug with the hERG channels. Although the mean clinical oliceridine unbound Cmax following a supratherapeutic dose (0.145 μM) is a small fraction of the 2.2 μM IC₅₀ for I_Kr inhibition, with progressive increase in the concentration of the drug intracellularly, myocardial levels of oliceridine could reasonably approach levels capable of reducing repolarization reserve, thus prolonging APD and the QT interval. Although oliceridine also blocks other cardiac currents, inlcuding sodium and calcium (Na_V1.5, peak I_Na, IC₅₀=9.023 μM; and calcium (I_Ca, Ca_V1.2, IC₅₀=39.63 μM) channel currents, at therapeutic concentrations the effect of oliceridine on these currents is likely to be relatively small and is expected to have little effect on repolarization. The effect of oliceridine to block late I_Na (IC₅₀=3.45 μM), however, is expected to counter the effect of the drug to.prolong APD and QT interval. The accumulation of oliceridine intracellularly is expected to significantly inhibit both inward (late I_Na) and outward (I_Kr) currents, which likely explains the plateau in QT and APD prolongation observed. Thus, the multi-ion channel effects of oliceridine limit the extent of QT prolongation and the potential arrhythmogenicity of this I_Kr blocker.

An IC₅₀ of 0.716 uM is reported for quinidine block of I_Kr and 14.62 μM for quinidine block of I_Na.¹² The much lower IC₅₀ for I_Kr inhibition and higher therapeutic plasma concentrations account for the much greater effect of quinidine as compared to oliceridine to prolong the APD and QT intervals and therefore to induce TdP in the clinic.

Oliceridine and quinidine showed tissue uptake of drug, failing to reach a steady state after 5 hours of continuous exposure to the drug (Figure 7). The progressive increase in the intracellular concentration of these drugs likely accounts for the progressive effect of the drug to prolong APD and QT intervals over the 5-hour period of exposure. Oliceridine’s effect on QT and APD was much less accentuated than that of quinidine, reaching a quasi-steady-state after 2 hours, likely due to multi-ion channel effects at higher intracellular concentrations. Clinically, since oliceridine is dosed on an as-needed basis, and its plasma half-life is relatively short (1.5-3 hours), plasma concentrations in the treated patient will show peaks and valleys with intermittent dosing, achieving a steady-state concentration within the myocardium under conditions of actual clinical use.

Protective effect of multi-ion channel actions of oliceridine

Both oliceridine and quinidine showed significant myocardial uptake after several hours of exposure to the drugs. (Fig. 7). Despite a relatively high concentration of oliceridine observed in the rabbit myocardium, its effects on QTc, as well as on APD₉₀ were much more modest than those of quinidine at all pacing cycle lengths. These results may be best explained by the higher IC₅₀ value for oliceridine block of the hERG channels as well as oliceridine’s multi-ion channel actions. Because the IC50 for oliceridine block of late I_Na (3.45 μM, Table 1 and Fig. 3) is similar to the IC₅₀ value for oliceridine block of I_Kr (2.2 μM), this action of the drug is expected to attenuate the effect of I_Kr inhibition to prolong APD and QTc. Inhibition of late I_Na is likely also responsible for the inability of oliceridine to induce EADs and TdP.

Implications of delayed drug-induced prolongtion of repolarization paramaters

Preclinical studies performed in the assessment of arrhythmic risk associated with a reduction of repolarization reserve often examine the effects of drugs over a period of several minutes in the case of voltage clamp studies and over a period of 20-60 minutes in the case of electrophysiological studies performed in tissues such Purkinkje fibers or coronary-perfused wedge preparations. Our results obtained using rabbit ventricular tissues, as those previously reported by us using canine ventricular myocardium and Purkinje fibers, ¹ point to slow progressive intracellular uptake and accumulation of drugs as the basis for the slow progressive rise of APD and QTc following exposure to oliceridine and quinidine as well as delayed appearance of arrhythmias following exposure to quinidine..

Brief exposure to drugs as often employed in the assessment of risk may be insufficient to fully characterize the effects of these agents to delay repolarization and to induce arrhythmogenesis thus significantly underestimating their potential risk. Our findings argue for inclusion of a protocol designed to assess the time-dependent effects of drugs over an extended period of time in preclinical experimental models.

Estimation of the risk of acquired long QT and TdP

Despite progressive tissue uptake and accumulation of the drug over a period of many hours, and conditions designed to accentuate proarrhythmic potential of the drug, i.e., hypokalemic and bradycardia, oliceridine prolongation of repolarization parameters was very modest and EADs and TdP were never observed, suggesting no or very little risk of induction of long-QT mediated pro-arrhythmias. The multi-ion channel effects, inhibition of late I_Na in particular, mitigate the effect of oliceridine to prolong the QT interval and to induce TdP. In marked contrast, quinidine, which is well known to cause acquired long QT syndrome and TdP, induced APD and QTc prolongation that failed to achieve a steady-state over a period of 5 hours. Moreover, EADs were observed at slow pacing rates only after 120 min of exposure to quinidine.

The relative TdP risk of oliceridine and quinidine was estimated according to the previously validated criteria.¹³ The estimated TdP score for oliceridine calculated after 2 hours of exposure to 1 μM oliceridine under conditions known to predispose to TdP was 1. In sharp contrast, the TdP score for 3.3 μM quinidine, calculated after 2 hours of exposure to the drug, was 8, indicating essentially no risk of TdP with oliceridine but a very high predicted risk for quinidine.

Limitations

The electrophysiological actions of oliceridine and quinidine were studied acutely in isolated normal Tyrode’s-perfused ventricular preparations, lacking autonomic and hormonal influences. The presence of autonomic influences and other factors present in vivo may modulate the effect of the drugs, resulting in outcomes different from those observed in the present study. Although we postulate that the effects of the intracellular accumulation of these two drugs is due to their presentation at progressively higher concentrations to the hERG receptors inside the cell, impaired trafficking is another potential explanation. Pharmacologic agents capable of prolonging APD and QT intervals such as probucol and sitagliptin have been reported to produce a delayed response owing in part to their actions to inhibit trafficking of the hERG channels to the plasma membrane.^{14, 15} Despite extensive research over a period of several decades there are no data implicating quinidine in impaired trafficking. ¹⁴ Oliceridine, however, has not been evaluated with respect to its ability to impair trafficking.

It is noteworthy however that drugs that alter trafficking do so with a time-course of 12-36 hours due to the slow turnover of hERG channels.¹⁶ The peak QTc effects that we observe are therefore too rapid to be due to impairment of ion channel trafficking.

Also noteworthy is the fact that in the tQT study reported by Bloomfield et al, as in our tQT study, moxifloxacin also produced a hint of a double peak following administration of a single oral dose.¹⁵ A delayed response or QTc response that is out of phase with plasma drug concentration, unrelated to the presence of a metabolite, is commonly observed with moxifloxacin, as highlighted in the report by Malik et al. ¹⁷

Conclusions

Our experimenal findings indicate that the oliceridine-induced concentration-QT effect hysterisis observed in the clinical thorough QTc study is due to gradual uptake and accumulation of oliceridine in cardiac tissues and cells. Our results point to a very low probability of oliceridine to cause TdP compared to quinidine due to the drug’s multi-ion channel effects, particulary its ability to block late I_Na. Our data also indicate that intracellular uptake and accumulation of agents that block the hERG channel can importantly confound assessment of QT prolongation and arrhythmic risk when studied acutely, calling for incorporation of protocols that measure these parameters over an extended period of time.

Among the clinical lessons to be learned is that when one observes discordance between Tmax and peak plasma concentration of a drug, further investigation is warranted, probing the possibility of as yet uncharacterized metabolites, ion channel trafficking effects, or an influence of intracellular uptake and accumulation on QTc.