Results of the 2017 inspection campaign of French phase I/II research sites in Île-de-France following the BIA 10-2474 accident: Medical vs. regulatory relevance.

AIMS: Following the serious adverse events that occurred in January 2016 during the BIA 10-2474 First-in-Human study, the French Ministry of Health asked the Regional Health Agencies to inspect operations at all research sites conducting phase I/II clinical trials of experimental drugs. The aim of this study was to assess the medical relevance of the inspections made in Île-de-France (Paris region) in 2017. METHODS: All 30 sites of Île-de-France region fully authorized to perform phase I/II trials were inspected by a public health physician and a public health pharmacist. Their reported list of observations was submitted to three physicians with longstanding experience of early pharmacology studies performed in academic or private research facilities. These physicians were asked to adjudicate each observation according to their perceived medical importance regarding safety. Adjudications were first performed separately and disagreements were later settled during a final adjudication meeting. RESULTS: At least one disagreement occurred initially among the 3 adjudicators for 84 of the 120 observations (70%) reported by the inspectors. Following reconciliation, the 3 physicians agreed that 20% of the observations were likely to have potentially serious medical consequences. These observations mainly concerned the management of emergencies and of serious adverse events and the continuity of care. CONCLUSIONS: Maintenance of on-site inspections periodically carried out by regulatory authorities granting authorisations to perform phase I/II trials are justified. However, the medical relevance of these inspections can be improved with more emphasis on factors affecting the safety of research participants than on administrative or purely regulatory issues.

Association of Oral Contraceptives With Drug-Induced QT Interval Prolongation in Healthy Nonmenopausal Women.

Importance: Women are at higher risk of drug-induced torsade de pointes (TdP) than men. Androgens are protective. Influence of oral contraception on drug-induced TdP and QT prolongation is controversial. Objective: To determine if the extent of sotalol-induced corrected QT (QTc) prolongation and specific T-wave morphological changes, which are biomarkers for the risk of drug-induced TdP, differ in patients according to the androgenic activity of the type of oral contraceptive (OCs) they take compared with patients who took no pills. Design, Setting, and Participants: A cohort of 498 healthy, nonmenopausal women received 80 mg of oral sotalol, a drug with known risk of drug-induced TdP, during this study in a clinical investigation center. The participants also took either no oral contraception or received OCs with different types of progestin: levonorgestrel (which has high androgenic potency), desogestrel or gestodene (which has intermediate androgenic potency), or drospirenone (which has antiandrogenic properties). Women were enrolled from February 2008 to February 2012, and data analysis took place from September 2014 to May 2018. Main Outcomes and Measures: Electrocardiographic changes 3 hours after sotalol administration. Results: A total of 137 women received levonorgestrel, 41 received desogestrel, 51 received gestodene, and 62 received drospirenone; another 207 received no OCs. Baseline QTc duration, plasma sotalol levels, and potassium levels did not significantly differ among groups. However, 3 hours after sotalol exposure, QTc prolongation was greater in women taking drospirenone (mean [SD] increase, 31.2 [12.6] milliseconds from baseline) than in women taking no OCs (mean [SD] increase, 24.6 [12.5] milliseconds; P = .005) or those taking levonorgestrel (mean [SD] increase, 24.2 [13.7] milliseconds; P = .005). The frequency of sotalol-induced T-wave alteration was higher in women taking drospirenone (n = 13 of 61 [21.0%]) than those taking levonorgestrel (n = 20 of 137 [14.6%]) or women taking no OCs (n = 24 of 207 [11.6%]; P = .01). Disproportionality analysis using the European pharmacovigilance database showed a higher reporting rate of OC-induced prolonged QT and ventricular arrhythmias in women taking drospirenone than levonorgestrel (drug-induced long QT syndrome: reporting odds ratio [ROR], 6.2 [95% CI, 1.3-30.8]; P = .01; ventricular arrhythmia: ROR, 3.3 [95% CI, 1.7-6.3]; P < .001). Conclusions and Relevance: Contraceptive pills are associated with variable drug-induced alterations of ventricular repolarization in healthy nonmenopausal women. Drospirenone, an antiandrogenic pill, was associated with increased sotalol-induced QTc prolongation, although absolute QTc prolongation was modest. This finding was supported by the European pharmacovigilance database, which showed a higher reporting rate of suspected OC-induced ventricular arrhythmias on drospirenone compared with levonorgestrel. More data are required on whether antiandrogenic OCs lead to clinically significant adverse events in patients taking QTc-prolonging drugs.

GENomE wide analysis of sotalol-induced IKr inhibition during ventricular REPOLarization, "GENEREPOL study": Lack of common variants with large effect sizes.

Many drugs used for non-cardiovascular and cardiovascular purposes, such as sotalol, have the side effect of prolonging cardiac repolarization, which can trigger life-threatening cardiac arrhythmias by inhibiting the potassium-channel IKr (KCNH2). On the electrocardiogram (ECG), IKr inhibition induces an increase in QTc and Tpeak-Tend (TpTe) interval and a decrease of T wave maximal amplitude (TAmp). These changes vary markedly between subjects, suggesting the existence of predisposing genetic factors. 990 healthy individuals, prospectively challenged with an oral 80mg sotalol dose, were monitored for changes in ventricular repolarization on ECG between baseline and 3 hours post dosing. QTc and TpTe increased by 5.5±3.5% and 15±19.6%, respectively, and TAmp decreased by 13.2±15.5%. A principal-component analysis derived from the latter ECG changes was performed. A random subsample of 489 individuals were subjected to a genome-wide-association analysis where 8,306,856 imputed single nucleotide polymorphisms (SNPs) were tested for association with QTc, TpTe and TAmp modulations, as well their derived principal-components, to search for common genetic variants associated with sotalol-induced IKr inhibition. None of the studied SNPs reached the statistical threshold for genome-wide significance. This study supports the lack of common variants with larger effect sizes than one would expect based on previous ECG genome-wide-association studies. CLINICAL TRIAL REGISTRATION: ClinicalTrials.gov NCT00773201.