Researchers track the evolution of HIV in a single patient to understand what drives the production of broadly neutralizing antibodies.
So they say they have a strategy to mimic natural infection, where an arms race between generation of antibody binding site diveristy and Env epitopes results in increasingly good neutralising antibodies to Env - and that this could lead to a vaccine.
Um. Yeeeessss...for a single patient, given that this is what they followed?? And now that they will follow other single patients, might they not show that each person's responses evolve completely differently?
So this is good science - in fact, it is GREAT science. However - right now that is all it is; like so much of HIV research, answering hypothesis-generated questions leads to more more hypotheses, and more good science, and more publications....BUT NO VACCINES. In fact, the ONLY vaccines which have made it into Phase III clinical trial are nothing like what people seem to think will work, and only got there because the people who pushed for the trials pretty much ignored the basic scientists.
A potential flaw in the whole approach is looking at what NATURAL infections do. That this may not be relevant is shown by the case of one of the most successful of recent vaccines, which is Human papillomavirus (HPV) VLPs. The virus infects epithelial cells in a topical manner, is cleared by cell-mediated immune responses and elicits only weak antibody responses which are protective, whereas the vaccine is given via injection, and elicits very high neutralising antibody responses which are protective - but are of no use against established infections.
But the trains roll on, and new approaches keep getting unearthed, and maybe we will yet get efficacious HIV vaccines.