The flu takes a formidable toll each year. Researchers and health workers save lives by routinely rolling out seasonal vaccines and deploying drugs to fight the virus and its secondary infections. But in the U.S. alone the flu still kills tens of thousands of people and hospitalizes hundreds of thousands more.

A big part of the problem has been correctly predicting what strains of the influenza virus health officials should try to combat in a given season. A team of scientists from the U.S. and China now say they have designed a vaccine that could take the guesswork out of seasonal flu protection by boosting the immune system’s capacity to combat many viral strains.

The University of California, Los Angeles–led group reported in this week’s Science that they may have created the “Goldilocks” of flu vaccines—one that manages to trigger a very strong immune response without making infected animals sick. And unlike current flu vaccines, the new version also fuels a strong reaction from disease-fighting white blood cells called T cells. That development is important because a T cell response will likely confer longer-term protection than current inoculations do and defend against a variety of flu strains (because T cells would be on the lookout for several different features of the flu virus whereas antibodies would be primarily focused on the shape of a specific strain). “This is really exciting,” says Kathleen Sullivan, chief of the Division of Allergy and Immunology at The Children’s Hospital of Philadelphia, who was not involved in the work.

So what makes the U.C.L.A. team’s flu approach different than others? Flu vaccines typically include a cocktail of several strains of killed virus. Injecting this mix into the body prompts the development of antibodies that can latch onto any intruder that resembles the flu—helping prevent infection. But that standard method does not lead to a significant T cell reaction because the virus is dead. The new vaccine, by contrast, uses a live virus, so it elicits both an antibody response and T cell immunity—at least in lab ferrets and mice. “It has the magic of both great antibody response and inducing a strong, strong T cell response that will be a safety net—so if a virus breaks through the first line of defense, you will have T cells to make sure you don’t get very sick,” Sullivan says.

The researchers dissected the flu virus in lab dishes and tested how different mutations in each segment responded when exposed to interferon, a protein released by the body when viruses attack that helps keep flu infections in check. The scientists were then able to identify which mutations made the virus most likely to provoke action from protective interferons. Armed with that information, the researchers then designed a mutant flu strain that was powerful enough to replicate well but highly susceptible to our body’s own ability to control the virus—the ideal ingredients for a vaccine.

The resulting inoculation looks promising in both ferrets and mice, the most commonly used models of influenza infection. If this approach is proved to work as well in humans, the authors say their invention could negate the need for annual flu shots. (Although they are not sure how long their vaccine would remain effective in humans, T cell responses tend to confer longer-term immunity.) The scientists believe that because they included eight mutations in their lab-made viral strain, it is unlikely the virus will revert back to its original, more dangerous form (a common concern with any live-virus vaccine). There may also be other applications from this work, they say: Researchers could similarly take other viruses apart in the lab, scour them for important mutations and create vaccines against a plethora of other infections.

Multiple obstacles stand in the way of this becoming a future universal flu vaccine for humans, scientists from The Scripps Research Institute cautioned in an accompanying commentary in Science. Chief among them: although the U.C.L.A. team found there was some cross-protection across a small set of flu strains—H1N1 and H3N2 subtypes—that may not hold true across all forms of the flu. Researchers will also have to examine if triggering a robust immune response to the virus puts people at risk, Sullivan notes, because a frenzied immune system response is what destroys lung tissue and sometimes proves deadly when people are infected with H5N1, a type of avian flu. “There are lots of practical questions about rolling this out for humans,” she says. “But this is hugely innovative and exciting.”