The work presented here deals with the optimization of a strategy for detection of single nucleotide polymorphisms based on surface plasmon resonance imaging. First, a sandwich-like assay was designed, and oligonucleotide sequences were computationally selected in order to study optimized conditions for the detection of the rs1045642 single nucleotide polymorphism in the gene ABCB1. Then the strategy was optimized on a surface plasmon resonance imaging biosensor using synthetic DNA sequences in order to evaluate the best conditions for the detection of a single mismatching base. Finally, the assay was tested on DNA extracted from human blood which was subsequently amplified using a whole genome amplification kit. The direct detection of the polymorphism was successfully achieved. The biochip was highly regenerable and reusable for up to 20 measurements. Furthermore, coupling these promising results with the multiarray assay, we can foresee applying this biosensor in clinical research extended to concurrent analysis of different polymorphisms.