Several reasons have been proposed as explanations why so many S. aureus vaccines have failed in clinical trials. Clearly, the organism has a battery of virulence factors which may require a multipronged approach and neutralizing just a single mechanism with a specific mAb may not suffice.
There is no (or only incomplete) protection from recurrent infection even against the very same S. aureus strain that cause the after primary infection when all the ‘right’ organism-specific antibodies should be in place.
Some data suggest a role for antibodies as colonized individuals have ‘milder’ infections than non-colonized patients. However, one could argue that this is a rather soft quantitative finding reminiscent of the logic in favor of pneumococcal serum therapy in the days before we had effective antibiotics.
Importantly, patients with agammaglobulinemia provide the strongest evidence that humoral immunity is not central to the defense against S. aureus. These patients rarely suffer from S. aureus infections or bacteremia despite having severe recurrent bacterial infections caused by S. pneumoniae, H. influenzae type B, S. pyogenes and P. aeruginosa.  By contrast, patients with isolated T-cell deficiencies often battle with invasive S. aureus infections.
Taken together, it seems that vaccines that trigger just an IgG response are less likely to succeed than those that also stimulate opsonization, upregulate T-cells and activate cellular immunity. Hence, the case for new adjuvanted vaccines in order to mobilize both arms of the immune system.
A recent article brings another hypothesis to the fore. Spaulding et al. tested various antibody preparations against S. aureus virulence factors in a rabbit pneumonia model and found – quite surprisingly – that antibodies against circulating virulence factors were protective, while antibodies against bacterial cell surface antigens were actually harmful.
Is this the explanation why previous vaccine approaches failed? Is this the end of MSCRAMMs  as vaccine targets? It seems the last word has not been spoken. There are just too many “unknowns” of the Rumsfeld kind as we extrapolate animal data to the human situation.  Or, to quote an FDA source: “Everything cures cancer in the mouse”. By extension, let’s include rabbits.
 J. Winkelstein. Agammaglobulinemia. 2014 http://www.uptodate.com/contents/agammaglobulinemia?source=search_result&search=agammaglobulinemia&selectedTitle=1~53
 Bagnoli. Inferring reasons for the failure of Staphylococcus aureus vaccines in clinical trials. Front. Cell. Infect. Microbiol. 2012; 2: 1
 Spaulding. Vaccination against S. aureus pneumonia. JID 2014; 209: 1955
 MSCRAMM = Microbial Surface Components Recognizing Adhesive Matrix Molecules
 McQuinn. Oncology Drug Development. A reviewers personal observations. http://www.toxicology.org/isot/rc/NorCal/docs/2010Spring/2010_1BOncologyDrugDevelop.pdf