The problem is that this phenomenon (funding research where you know there is a good probability it will work before you even start) is not limited to medical r

The problem is that this phenomenon (funding research where you know there is a good probability it will work before you even start) is not limited to medical research, it can be found even in fields where there are no commercial interests involved. Opinions differ as to why that is - I personally suspect the grants system that is used for funding science.