Concern around the safety of rivaroxaban (Xarelto) for atrial fibrillation is a “storm in a teacup”, according to cardiologists. The highest selling NOAC in the country came under fire after a BMJ article, “Rivaroxaban: can we trust the evidence?”, drew attention to a faulty device that might have undermined the results of the key […]
Concern around the safety of rivaroxaban (Xarelto) for atrial fibrillation is a “storm in a teacup”, according to cardiologists.
The highest selling NOAC in the country came under fire after a BMJ article, “Rivaroxaban: can we trust the evidence?”, drew attention to a faulty device that might have undermined the results of the key trial behind the drug’s approval.
The point of care device used to measure the INR in patients in the warfarin arm of the study potentially delivered “clinically significantly lower” results than other tests, and was recalled four years later in 2014.
This faulty reading may have led to patients in the warfarin arm having their dose unnecessarily increased, putting them at a greater risk of bleeding.
Former FDA cardiovascular and renal drug reviewer, Dr Thomas Marciniak, said in the BMJ that he believed that this had compromised the trial.
And Deborah Cohen, author and associate editor of the BMJ, said the study may have caused rivaroxaban to appear safer than it was in terms of bleeding risk.
The European Medicines Agency and the FDA are currently investigating whether the faulty test led to erroneous results and therefore mischaracterised the safety of rivaroxaban.
However, Dr Harry Gibbs, cardiologist and general physician at Melbourne’s Alfred Hospital, labeled the concerns “a storm in a teacup”.
Dr Gibbs pointed to a reanalysis of the data published in the NEJM on the same day as the BMJ article. This sensitivity reanalysis of the data was consistent with the initial findings.
Over 14,000 participants were included in the ROCKET AF trial, which found that rivaroxaban was non-inferior to warfarin in preventing ischaemic stroke and systemic emboli, and accounted for fewer intracranial and fatal bleeding episodes.
Other studies since, including community studies, have shown that rivaroxaban is equal or better to warfarin, Dr Gibbs said.
“Don’t be concerned and don’t stop your patients who already on rivaroxaban,” he said. “The danger is that patients might get the wrong message, that the drug is unsafe, and discontinue. And that will put them at risk of stroke.”
Bayer, the drug’s manufacturer, said they were not informed of the recall of the device almost a year after, in September 2015.
It would be ideal to get more data and do more analyses, said Professor David Brieger, a cardiologist at Concord Hospital.
He echoed Ms Cohen’s call for trial datasets to be made open, in order to allow external and independent researchers to do these analyses.
But in the meantime, Professor Brieger said he was reassured by the analysis done by the reputable Duke University researchers.
“I think we can relax,” Professor Brieger said. “I certainly wouldn’t be changing my practice in light of this.”
Professor Andrew Sindone, a Sydney cardiologist, also said he was reassured by the secondary analysis and is comfortable with other research since showing the incidence of bleeding on rivaroxaban is very low.
“[With NOACs] you can eat whatever food you want, drink alcohol, and the incidence of interactions are much lower,” he said.
Ms Cohen criticized the system in which clinical trials are not required to specify the devices used.
The TGA confirmed they are evaluating whether any action is needed around the drug, which cost the PBS $38.5 million last year.
Dr Harry Gibbs, Professor David Brieger and Professor Andrew Sindone have all received honoraria and speaking fees from the three NOAC companies.
BMJ 2016; online 3 Feb; NEJM 2016; online 3 Feb
UPDATE: The European Medicines Agency concluded that the INR device’s defect did not change the overall safety or benefit-risk balance of the ROCKET AF study.
