Experimental design, genetics and animal toxicity tests

Can the design of experiments in a major industry be based on unsound principles? A key principle in experimental design is to control all variables except for the ones being studied. But the pharmaceutical industry tests the toxicity of potential new drugs by using genetically heterogeneous rats and mice, even though these can vary hugely in their responses. This reduces the power of their experiments. Michael Festing asks why basic design principles are being ignored; can this be a contributing factor to the high cost of drug development?