Conflicting COVID-19 results for Roche arthritis drug show more trials needed

By Kate Kelland

LONDON (Reuters) - Larger trials are needed to assess whether Roche's arthritis drug tocilizumab can cut death rates among the sickest COVID-19 patients, scientists said on Wednesday, after a small study found it was no better than standard care in severe cases.

The Brazilian study's results, published in the BMJ British Medical Journal, contradict findings of some earlier trials, including an 800-patient analysis this month that found a significant benefit of tocilizumab, also known as Actemra, for the critically ill.

Among 129 patients in the Brazilian study, which compared tocilizumab plus standard care with standard care alone in people hospitalised with severe or critical COVID-19, clinical outcomes after 15 days were no better in those who got the drug.

The trial was stopped due to safety concerns after preliminary data showed deaths in the tocilizumab group were higher, the researchers said. But they added that in both groups, deaths were attributed to COVID-19-related acute respiratory failure or multiple organ dysfunction.

Experts not involved in this research, but who are examining potential effects of Actemra in other COVID-19 trials, said that because the Brazilian study involved such a small number of patients, the confidence range of the result was wide and it should not be seen as statistically significant.

Peter Horby, a professor of emerging infectious diseases at Britain's Oxford University who co-leads a COVID-19 treatment trial known as the RECOVERY trial, noted what he described as "surprisingly low fatality" among patients in the trial, with only 3% of those in the control group dying.

"Such a low rate is unusual in severely ill COVID patients and suggests that the death rate in the control group may be a chance finding which has affected the result," he said.

"Much larger trials are needed to resolve the continuing uncertainty about the role of tocilizumab in the treatment of (COVID) patients."

Anthony Gordon, a professor at Imperial College London who led the REMAP-CAP trial which found a benefit earlier this month, said there was a "strong possibility" the Brazilian findings were susceptible to the play of chance, "especially considering the low numbers in the overall trial".

Other previous studies of the drug - a type of anti-infammatory medicine known as an IL-6 receptor antagonist - also had mixed results.

In November, Roche said research showed Actemra helped the sickest COVID-19 patients, but it was unclear if it kept people alive or shortened how long they needed intensive care support such as mechanical ventilation, or both.

"This ... illustrates why it is important to conduct large randomised controlled trials, and that evidence across all studies should be considered when making treatment decisions for different groups of patients," Gordon said.

(Reporting by Kate Kelland. Editing by Mark Potter)