(Reuters) -German biotech firm CureVac saw billions of euros wiped from its market value on Thursday after its COVID-19 vaccine proved only 47% effective in an initial trial read-out, denting investor confidence in its ability to take on rival shots. The disappointing efficacy of its vaccine, CVnCoV, emerged from an interim analysis in a study of about 40,000 volunteers in Europe and Latin America, with CureVac saying on Wednesday that new variants had proved a headwind. CureVac shares listed in Frankfurt and New York fell more than 50% in early trading in those markets and were down 43% at 1451 GMT, their lowest in more than seven months, erasing more than 6 billion euros ($7.2 billion) from the company’s market value. Late-stage trials of BioNTech/Pfizer and Moderna vaccines, which like CureVac’s are based on mRNA technology, had efficacy rates of well above 90% but took place when the original version of the coronavirus was dominant. Data on their products have, however, so far suggested only somewhat weaker protection against new variants. “This could cast doubt on competitiveness of its mRNA platform,” Jefferies said in a research note on CureVac’s study. CureVac, which had registered 134 infections overall for the initial analysis, said it expected to publish final results based on more than 200 cases in two to three weeks’ time. Sufficient additional infections had already occurred among participants but had yet to be vetted for inclusion in the final analysis, it added. Public health representatives across the globe are pushing for a fast deployment of available vaccines to counter highly contagious mutations of COVID-19 such as the Delta variant that first emerged in India. The Jefferies analyst said CureVac’s readout was in stark contrast with results announced by Novavax this week, showing that its experimental COVID-19 vaccine was more than 90% effective, including against a variety of concerning variants of the coronavirus in a U.S.-based clinical trial. Yet, CureVac’s efficacy rate was similar to Sinovac’s vaccine which the World Health Organization said this month was 51% effective in preventing symptomatic disease, taking the Chinese shot one step closer to be deployed in low and middle-income countries under the COVAX platform. LOW DOSE TO BLAME? Peter Kremsner of the University Hospital in Tuebingen, the scientist leading the CureVac study, said a low dosage was likely the reason behind the disappointing efficacy, but tolerability concerns posed limits to increasing the dose. With just 12 micrograms of mRNA per dose, the company has been seeking to position itself as a low-dose alternative to BioNTech and Moderna’s established shots with 30 and 100 micrograms per shot, respectively. CureVac CEO Franz-Werner Haas told Reuters the company still planned to apply for approval from the European Medicines Agency (EMA), provided the shot clears the formal hurdle of 50% efficacy. However the EMA said on Thursday it would not impose a 50% efficacy threshold for vaccines, adding full trial data was necessary for it to make a sound assessment on the benefits and risks of a shot. Under CureVac’s only major supply deal for the product tested in the trial, the European Union secured up to 405 million doses of the vaccine in November, of which 180 million are optional. In a bet on CureVac’s technology, Britain placed a conditional 50 million dose order in February on yet-to-be-developed vaccines that build on the product tested in the trial. After the interim results, the European Commission said it would need to wait for final results and the assessment of the EU drugs regulator before taking any steps, a spokeswoman for the EU executive told a news conference on Thursday. Bayer and Novartis, two of several production partners that CureVac had lined up to mass-manufacture its product, said they were following developments closely and were awaiting further analysis. ($1 = 0.8377 euros) Reporting by Ludwig Burger; additional reporting by Francesco Guarascio in Brussels; Editing by Alexander Smith and Pravin Char Our Standards: The Thomson Reuters Trust Principles.