In this study, we use polarized light scattering to study immunotargeted plasmonic nanoparticles which bind to live SK-BR-3 human breast carcinoma cells. Gold nanoparticles can be conjugated to various biomolecules in order to target specific molecular signatures of disease. This specific targeting provides enhanced contrast in scattering-based optical imaging techniques. While there are papers which report the number of antibodies that bind per nanoparticle, there are almost no reports of the key factor which influences diagnostic or therapeutic efficacy using nanoparticles: the number of targeted nanoparticles that bind per cell. To achieve this goal, we have developed a 'negative' method of determining the binding concentration of those antibody/nanoparticle bioconjugates which are targeted specifically to breast cancer cells. Unlike previously reported methods, we collected unbound nanoparticle bioconjugates and measured the light scattering from dilute solutions of these particles so that quantitative binding information can be obtained. By following this process, the interaction effects of adjacent bound nanoparticles on the cell membrane can be avoided simply by measuring the light scattering from the unbound nanoparticles. Specifically, using nanoshells of two different sizes, we compared the binding concentrations of anti-HER2/nanoshell and anti-IgG/nanoshell bioconjugates targeted to HER2-positive SK-BR-3 breast cancer cells. The results indicate that, for anti-HER2/nanoshell bioconjugates, there are approximately 800-1600 nanoshells bound per cell; for anti-IgG/nanoshell bioconjugates, the binding concentration is significantly lower at nearly 100 nanoshells bound per cell. These results are also supported by dark-field microscopy images of the cells labeled with anti-HER2/nanoshell and anti-IgG/nanoshell bioconjugates.