I am not familiar with the formulas, but I can tell you that it is not from Coefficient of drag. The fact that the exponent is under 1 means that the mass has a relation that factors it in less than what it would factor in as if you were looking purely at acceleration--where weight matters.
This equation is very simplified. If you were to develop an equation for range that assumed that the car were traveling on a flat surface at a constant speed, weight would not matter at all, but drag area would drive the results.
Conversely, if you were to base range on a stop and go city type environment where it is mostly acceleration and aerodynamic losses are relatively minimal compared to acceleration and
Rolling Resistance, then the weight of the car would matter more, and that exponent would be 1. so it looks like some assumptions about an average driving cycle have been made and the 0.6 exponent has been used to reflect that compromise. It looks as though it uses some stop and go or some hills.
The fact that Cd is not in the equation suggests to me that this assumed driving cycle doesn't have much traveling over 45 mph.
So I am really curious. What is the assumed cycle? Is there an assume value for drag area? (Drag area is just Cd times frontal area, the two variables specific to a car's design.)