Covid-19: Not all patients develop protective antibodies
medicalxpress.com
medicalxpress.com
First of all, from the paper itself:
> Since we could not perform additional virus-neutralization tests we cannot exclude the possibility that COVID-19 convalescent subjects produce other types of protective antibodies besides those inhibiting RBD-ACE2 binding.
Second: they took antibodies from serum, which would exclude mucosal responses (IgA), observed by other groups.
Third: There's the whole matter of T-cell immunity which is not touched by this paper at all.
The antibody-dependent enhancement test was done in vitro, so there is no guarantee it would happen in vivo. In fact, the authors say so themselves:
> However, studies are needed to investigate if antibody-mediated increases of RBD binding to ACE2 have a clinical relevance.
I'd add that a few vaccines in trial use the full-length spike protein, rather than just the so-called "receptor binding domain" (RBD; the part that binds to the ACE2 receptor), so more data are needed to see what is observed here has implications or not.
Personally, I'm still optimistic.