Bradyarrhythmias and direct acting antivirals (DAAs).

The postmarketing reports of bradycardia (1) under sofosbuvir-including regimen raises several issues.

The first one is the incidence of such events and their clinical relevance in the daily practice. In our own experience, we calculated the incidence of symptomatic cardiac events in 1.2% of treated patients (2). This calculation does not include some cases of sudden deaths where it is not possible to exclude the occurrence of undiagnosed severe bradyarrhytmias. If the frequency is undoubtedly higher than in the general population, we do consider that it does not modify the therapeutic indication for HCV infection, especially in “priority” patients who need to achieve a sustained virologic response to decrease the risk of cirrhosis complications, including hepatocellular carcinoma, but also of extra-hepatic complications, especially cryoglobulinemic vasculitis.

The second issue is whether we have to make recommendations for routine monitoring of cardiac rhythm during the initiation of sofosbuvir. Even if it is our “conservative” advice (2), we agree that it may be premature (3) especially because we do not know whether pre-exisiting cardiac abnormalities could significantly increase the risk of cardiac events.

The third issue is the causative relationship between sofosbuvir and the bradycardia. The interrogation of the pacemakers that had to be implemented strongly suggests a causative link: 26% atrial pacing of the exposure time during sofosbuvir exposure (with daclatasvir or simeprevir or ribavirin), which decreased to 4% of the exposure time after treatment discontinuation (2).

The fourth issue is the one of drug-drug interaction. The 9 reported cases (bradycardia was fatal in one patient, needed a pacemaker insertion in three and resolved in the others with drug discontinuation) resulted in a safety announcement by the US Food and Drug Administration (FDA) of a serious slowing of the heart rate when the class III antiarrhythmic drug amiodarone was used with hepatitis C treatment containing sofosbuvir (SOF) in combination with another direct-acting antiviral (DAA) (1). The announcement was based on a total of 9 postmarketing reports., including 2 cases recently published (4). An elegant editorial suggests that the plausible mechanisms of the interaction between sofosbuvir and amiodarone could be related to P-glycoprotein inhibition by amiodarone as sofosbuvir is a P-glycoprotein substrate or an increase of free (active) amiodarone concentrations by a DAAs-related displacement of amiodarone from its binding site (5). Unfortunately, there is no data on serum concentrations of amiodarone in those patients and we do not have data regarding the number of patients treated by DAAs with anti-arythmic drugs, including amiodarone. In our data file, only one of the five patients that we prospectively observed with dysrhythmia in 2014 was given amiodarone, two patients were given betablockers, spironolactone and furosemide, while the two other patients were treated with antiretroviral drugs. That is why we do consider that cardiac events are not systematically associated with amiodarone and that their incidence could be increased by underlying cardiac dysfunction which could be worsened by DAAs exposure.

A potential for cardiac toxicity of oral antivirals has been discussed and initially reported with BILN 2061 (the first DAA, a protease inhibitor) which has been withdrawn because the occurrence of cardiac steatosis in monkeys. Cardiotoxicity has been described with another NS5B inhibitor (BMS 986094) leading to cardiomyopathy without conduction abnormality and withdrawal of the drug deveolpment. While sofosbuvir did not show cardiotoxicity during sudies of carcinogenesis (2 years of sofosbuvir treatment with plasmatic concentrations at 9 fold (rat) and 17 fold (mice) the therapeutic dosage), a cardiotoxicity has been reported for concentration of the sofosbuvir metabolite GS-331007 increased from 16 to 29 fold the therapeutic dosages in rats and from 27 to 41 fold the therapeutic dosages in mice and dogs (inflammation in myocytes).

What are the practical messages at the time we are hoping to be able to treat all the HCV-infected patients ? Amiodarone is formally contra-indicated with sofosbuvir but the durg-drug interaction with other antiarythmics has to be clearly evaluated. An electrocardiogram should be systematically performed at the initiation of DAAs and an expert advice will be mandatory if the electrocardiogram evidences abnormalities. Most of patients, at least in those countries with a « priority to treat policy », will have a risk :benefit ratio favouring the antiviral treatment.

Stanislas Pol
Université Paris Descartes; Hepatology Department, Cochin hospital, APHP
INSERM U-818 and UMS-20, Institut Pasteur, Paris, France

References

1. Anonymous. FDA warns of serious slowing of the heart rate when the antiarrhythmic drug amiodarone us used with hepatitis C treatments containing sofosbuvir (Harvoni) or Sovaldi in combination with another Direct Acting Antiviral drug. Available from: www.fda.gov; accessed May 17, 2015.
2. Fontaine H, Duboc D, Pol S. Bradyarrhythmias Associated with Sofosbuvir Treatment. N Engl J Med 373;19 : 1886-1887
3. Brainard DM, Mc Hutchison JG. Bradyarrhythmias Associated with Sofosbuvir Treatment. Gilead Sciences, the manufacturer of sofosbuvir, replies . N Engl J Med 373;19 : 1888
4. Renet S, Chaumais M-C, Antonini T, et al. Extreme bradycardia after first doses of sofosbuvir and daclatasvir in patients receiving amiodarone: 2 cases including a rechallenge. Gastroenterology 2015;149:1378–1380.
5. Back DJ, Burger DM. Interaction Between Amiodarone and Sofosbuvir-based Treatment for Hepatitis C Virus Infection: Potential Mechanisms and Lessons to be Learned. Gastroenterology 2015;149:1315–1333.