Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Filters applied. Clear all
. 2016 Oct 18;68(16):1756-1764.
doi: 10.1016/j.jacc.2016.07.761.

Coupling Data Mining and Laboratory Experiments to Discover Drug Interactions Causing QT Prolongation

Affiliations
Free PMC article

Coupling Data Mining and Laboratory Experiments to Discover Drug Interactions Causing QT Prolongation

Tal Lorberbaum et al. J Am Coll Cardiol. .
Free PMC article

Abstract

Background: QT interval-prolonging drug-drug interactions (QT-DDIs) may increase the risk of life-threatening arrhythmia. Despite guidelines for testing from regulatory agencies, these interactions are usually discovered after drugs are marketed and may go undiscovered for years.

Objectives: Using a combination of adverse event reports, electronic health records (EHR), and laboratory experiments, the goal of this study was to develop a data-driven pipeline for discovering QT-DDIs.

Methods: 1.8 million adverse event reports were mined for signals indicating a QT-DDI. Using 1.6 million electrocardiogram results from 380,000 patients in our institutional EHR, these putative interactions were either refuted or corroborated. In the laboratory, we used patch-clamp electrophysiology to measure the human ether-à-go-go-related gene (hERG) channel block (the primary mechanism by which drugs prolong the QT interval) to evaluate our top candidate.

Results: Both direct and indirect signals in the adverse event reports provided evidence that the combination of ceftriaxone (a cephalosporin antibiotic) and lansoprazole (a proton-pump inhibitor) will prolong the QT interval. In the EHR, we found that patients taking both ceftriaxone and lansoprazole had significantly longer QTc intervals (up to 12 ms in white men) and were 1.4 times more likely to have a QTc interval above 500 ms. In the laboratory, we found that, in combination and at clinically relevant concentrations, these drugs blocked the hERG channel. As a negative control, we evaluated the combination of lansoprazole and cefuroxime (another cephalosporin), which lacked evidence of an interaction in the adverse event reports. We found no significant effect of this pair in either the EHR or in the electrophysiology experiments. Class effect analyses suggested this interaction was specific to lansoprazole combined with ceftriaxone but not with other cephalosporins.

Conclusions: Coupling data mining and laboratory experiments is an efficient method for identifying QT-DDIs. Combination therapy of ceftriaxone and lansoprazole is associated with increased risk of acquired long QT syndrome.

Keywords: data mining; data science; drug-drug interaction; long QT syndrome.

Figures

Figure 1
Figure 1. Data Science and Experimental Pipeline for Identifying and Validating QT-DDIs
(A) Chemical structures for ceftriaxone (cephalosporin) and lansoprazole (proton pump inhibitor, PPI), which we predicted would have a QT-DDI; we predicted cefuroxime (cephalosporin) and lansoprazole not to interact. (B) QT-DDI discovery in FAERS: data-driven side effect profile containing latent evidence of a QT-DDI (black boxes: positive correlation with QT prolongation; white boxes: negative correlation). Each bar represents the reporting frequency of a given side effect in FAERS for ceftriaxone (green), lansoprazole (blue), cefuroxime (orange), ceftriaxone + lansoprazole (red), and cefuroxime + lansoprazole (purple). (C) Retrospective corroboration in electronic health records (EHR). Left: Difference in QTc interval (mean ± 95% CI) between cases (patients prescribed the drug pair) and controls (patients on only 1 drug). We stratified the analysis by sex (men: gray; women: teal) and evaluated all races combined, as well as whites, blacks, and “other, including Hispanic” separately. Star indicates the change in QTc intervals is statistically significant (Mann-Whitney U test with Bonferroni correction). We obtained 95% CIs by bootstrapping case and control QTc distributions, and calculating the change in median QTc for each iteration. Right: Percentage of patients with a QTc interval ≥500 ms (mean ± 95% CI), stratified by sex and race. Star indicates the combination had a significantly greater proportion of patients with a QTc interval ≥500 ms than either drug alone (independent-samples Student t-test, with Bonferroni correction, comparing means of single drug and combination therapy percentage≥500 distributions generated using bootstrapping). (D) Experimental validation using patch-clamp electrophysiology. Left: Change in hERG current from control (mean and SD) for increasing concentrations of cephalosporin alone (dashed line), and increasing concentrations of cephalosporin in the presence of a single concentration of lansoprazole (solid lines). Right: Representative traces from each patch-clamp electrophysiology experiment. From top to bottom: hERG-channel current in the presence of vehicle only (control), and then cephalosporin at 3 concentrations (0.1 μM, 10 μM, and 100 μM); hERG-channel current in the presence of lansoprazole and then in combination with progressively increasing concentrations of cephalosporin. CI = confidence interval; FAERS = FDA Adverse Event Reporting System; hERG = human Ether-à-go-go Related Gene; QT-DDI = QT-prolonging drug-drug interaction.
Figure 2
Figure 2. Results of the Computational Model of Ventricular Epicardial Myocytes
The APD prolongation (measured as APD70) for each case is 9 ms and 50 ms, simulating 1 μM lansoprazole + 100 μM ceftriaxone and 10 μM lansoprazole + 100 μM ceftriaxone, respectively. Briefly, the model was run for a ventricular action potential paced at 1 Hz with baseline conditions (black) and 10% or 55% block of peak hERG current (brown and red, respectively). APD70 = action potential duration at 70% of repolarization; hERG = human Ether-à-go-go Related Gene.
Figure 3
Figure 3. Analysis of Class Effects Between Cephalosporins and PPIs
We analyzed each cephalosporin and PPI pair for evidence of an interaction in FAERS (black box: drug pair matches side effect profile), EHR (red: patients on combination have significantly prolonged QT intervals compared with those on either drug alone; blue: no significant change between cases and controls; white: no patients on the drug pair in the EHR), and that the change seen in the EHR was not due to concomitant medications (red star). We stratified the analysis between men and women. Only ceftriaxone and lansoprazole in men passed each of these criteria. EHR = electronic health records; FAERS = FDA Adverse Event Reporting System; PPI = proton-pump inhibitor.
Central Illustration
Central Illustration. Ceftriaxone and Lansoprazole are Associated With Acquired LQTS
We combined mining of adverse event reports, corroboration in electronic health records, and experimental validation using single cell patch clamp to discover and validate a QT-DDI between ceftriaxone and lansoprazole. We used a data-driven profile of side effects that are predictive of LQTS to prioritize drug pairs pairs in FAERS. We corroborated these findings in the electronic health records by comparing the QTc intervals of patients administered the prioritized drug pair to patients exposed to either drug alone. We then validated our top prediction (ceftriaxone/lansoprazole) by measuring the dose-dependent changes in hERG channel current using patch-clamp electrophysiology. AFib = atrial fibrillation; FAERS = FDA Adverse Event Reporting System; hERG = human Ether-à-go-go-Related Gene; LQTS = long QT syndrome; QT-DDI = QT-prolonging drug-drug interaction; VTach = ventricular tachycardia.

Comment in

Similar articles

See all similar articles

Cited by 13 articles

See all "Cited by" articles

Publication types

MeSH terms

Feedback