It's because there are two main types of studies getting published: studies based on the actual virus or real world observations, and studies based on a pseud

It's because there are two main types of studies getting published: studies based on the actual virus or real world observations, and studies based on a pseudo-virus or simplified modeling. This study is the latter- it only tests efficacy against the spike itself, and nothing else. The study that suggested natural immunity plus the vaccine is the best protection was similarly based only on the spike protein. So any immune response based on any part other than the spike gets neglected, and I suspect that's at least somewhat intentional.