The first thing that comes to mind is that enzymes present a couple of issues that biologicals just don't have.
First is that enzymes ARE biological. They're refined to a high purity, but they aren't synthesized and ultimately come from the bacteria they're meant to replace. This can make them relatively expensive and are an ongoing cost that will be incurred every second the plant is operating. Since wastewater isn't revenue generating, the higher recurring costs of enzymes may not be worth the gain. Fully biological units, properly cared for, should be self-sustaining.
Second is that many enzymes don't do well in continuous process streams and will require a batch operation for various reasons. The usual method of application is to have a large holding tank and introduce a small amount of enzyme with circulation on the fluid. Since the enzyme is just a catalyst, a small amount will act on the entire batch if left long enough under circulation with the time determined by batch size, enzyme amount, and enzyme activity. A huge amount of enzyme could be dumped in to shorten the cycle, but that would be expensive. If you're going to have to batch the stream at that point anyway, why not look at a biological unit?
Third, the enzymes are probably going to simply break the starches down into their component sugars. You're not going to remove the starches from the stream so much as reformulate them. Are the simple sugars that will most likely be produced better or more acceptable than the starches?
An alternative approach may be to investigate a chemical reaction that will precipitate the starches out. If one exists, then you can use a flocculant/lamella/filter/screen/etc. to remove the solid particles from the stream.
Just a few thoughts that popped into my head. Good luck!
