Why can a high antibody titer fail to predict protection against infection?

Why

Plotkin's review notes that almost all licensed vaccines work through antibodies, but that the function of those antibodies matters as well as their quantity. An ELISA titer to a whole protein includes antibodies to non-protective epitopes. A neutralization or opsonophagocytic assay measures the activity that blocks infection.

Location matters too. Serum IgG may not prevent infection at a mucosal surface where secretory IgA acts. Durability matters, because titers fall, and protection for diseases with long incubation can rely on central memory that responds after exposure.

For some pathogens, cellular immunity carries much of the protection. Plotkin separates true correlates, which mechanistically cause protection, from surrogates, which only track it. A threshold valid for one vaccine, population and endpoint does not transfer to another.

For example, a variant virus can escape neutralization while binding titers to the original antigen stay high. In drug development the same logic applies to anti-drug antibodies, where the FDA's guidance distinguishes binding from neutralizing activity.