Ok, I just tried to actually read the paper, and in the discussion they address this question:
"The negative estimates in the final period arguably suggest different behaviour and/or exposure patterns in the vaccinated and unvaccinated cohorts causing underestimation of the VE. This was likely the result of Omicron spreading rapidly initially through single (super-spreading) events causing many infections among young, vaccinated individuals."
So I think they're saying that VE calculation may not be so meaningful in that time period because vaxed were behaving differently, exposing themselves to more large social events, so more likely to get infected than unvaxed. This sounds like a reasonable speculation to me, but I'm not an epidemiologist etc, so will wait to hear whether the experts buy it.
In any event, they are saying that immunity basically returns with the booster, even if it's short-lasting against Omicron, which might be all we need, since Omicron seems to be short-lasting as an epidemic, so if we can all boost our immunity for another month or so, we may be over it.