https://www.biorxiv.org/content/10.1101/2021.04.13.439482v1.full.pdf
Today we have a preprint showing data from the Moderna variant booster testing in mice. They are currently testing it in humans. There's nothing too surprising here. When they give the mice the regular Moderna vaccine, then seven months later give them a B.1.351 (SA variant) booster, they get increases in antibodies that neutralize both the older and SA variant. At least for mice, there does appear to be an effect where the immune system remembers the immunity from the first vaccine and boosts it when the variant vaccine is given.
It looks so far that humans given vaccine after infection get more of a boost than what we are seeing here with the third shot in the mice, but that could just be a species difference.
They also gave mice who had never had a vaccine before the original Moderna, the SA variant Moderna, or a mixture of the two. Unsurprisingly, the mixture vaccine gave antibodies that could neutralize either one. But it was not very good at neutralizing the CAL20.C (California) variant, which increased dramatically in Los Angeles between November and January and is now the predominant variant there. So if you're looking to get a head start on worrying about a new variant, the CA variant might be the next big variant of concern. But remember, these experiments were done with mice and human immune systems might respond differently.
The mixed vaccine was better against the P.1 (Brazil) variant than either alone.
The SA vaccine alone was good against the SA strain, but not great against any of the others they tested (older strain, Brazil, or California). So it would probably not be a good idea to switch to giving people just the new variant vaccine, at least not without testing it in humans first.
At least in mice, the immunity to the SA variant was wearing off faster than immunity to the older strain.