In clinical and real world trials, China’s Sinovac underperforms

THE LATEST results for China’s CoronaVac vaccine, developed by Sinovac Biotech, a Beijing-based pharmaceutical company, were disappointing for the aspiring scientific and technological powerhouse. Phase-three trials, which were conducted on health-care workers in Brazil, yielded an efficacy rate of just 50.7% (with a 95% confidence interval of 35.7% to 62.2%), just barely above the 50% threshold set by the World Health Organisation for covid-19 vaccines (see chart). The results of a real-world trial released a week earlier were even worse: the vaccine was estimated to be just 49.6% effective (11.3% to 71.4%) against symptomatic covid-19 cases; when asymptomatic infections were included, this figure dropped to a dismal 35.1%.

The Chinese authorities’ reaction did little to boost confidence. After news broke of the discouraging results, Gao Fu, head of the Chinese Centre for Disease Control and Prevention, admitted at a conference on April 10th that current vaccines “don't have very high rates of protection”, and suggested that vaccines could be mixed to improve efficacy. Mr Gao later backtracked from the comments, claiming that it was “a complete misunderstanding”.

Sinovac can take some comfort from the fact that predicting the effectiveness of vaccines is devilishly hard. In an ideal world, vaccines would be tested and compared in a head-to-head trial, in which several vaccines are administered in the same setting using the same trial procedures. In the real world, however, vaccines are rigorously tested as quickly as possible, which means that trial results are not strictly comparable. Trials of the same vaccine conducted in different places, under different conditions, often have different results. In the case of covid-19 vaccines, results can differ depending on the strains of the virus that are most prevalent, the populations tested, and sometimes even the way the results are measured (eg, whether all infections are counted, or only symptomatic ones).

That the Sinovac trials were conducted in Brazil, which harbours a variant of the virus thought to be particularly nasty, suggests some caution in assessing the outcome is in order. Indeed, preliminary results from the firm’s phase-three trials in Turkey suggest an efficacy rate of 83.5%. Similar phase-three trials are under way in Chile, Indonesia, and the Philippines.

Effective or not, Sinovac has already exported approximately 120m doses to 19 different countries, according to Airfinity, a London-based science analytics company (see chart). So far, the largest shipments have gone to Brazil, Indonesia and Turkey. In China, nearly 180m people have already been vaccinated.

Despite the apparently poor results in stopping infections, the Sinovac vaccine, like its competitors, appears to offer remarkably good protection against severe disease and death. In the Brazil phase-three trials, nearly 4% of the unvaccinated participants who were infected got severely ill; among the vaccinated, none did.