This remarkable difference between the best synthetic vector and a typical viral vector leads to a key question: Is there an inherent limitation to polyplexes? Or have we not found the magic polymer yet? In order to address this question we develop a detailed model of the gene delivery pathway which includes information about both the host cell and the vector, from the whole cell level to the single particle level. Firstly, the model shows that there exists a strong effect of cell morphology on the optimum intracellular itinerary of the vector. For flat, adherent, 2D cells in vitro the optimal pathway corresponds to enhancing the rates of all the steps involved. However, this is entirely due to the large number of vectors that enter directly on top the cell nucleus and are thus proximal to it. On the other hand, for ellipsoidal 3D cells in vivo the optimal pathway is much more complicated. There exist strong optima in the rates of endosomal escape and vector unpacking and greater enhancement in these rates actually reduces the delivery efficiency by up to 1000-fold. The optima themselves depend upon other parameters, showing that a systems analysis is essential to arrive at the optimal pathway.
On the absolute scale, the model predicts that the maximum delivery efficiency achievable by polyplexes even after optimization, is still very low compared to adenoviruses. This limit exists because of the low mobility of the vector in cell cytoplasm and its failure to protect DNA until the point of nuclear entry. New strategies for improved efficiency are proposed based on the findings from this study.