My answer is - yes, gas engines are affected in the same ways by hills and wind... but that doesn't mean surface streets are the better option.
When you look at the question as a raw physics question you find the answer that all resistive forces do not discriminate between drive type. Doesn't matter if you all electric, hybrid, gasoline, diesel, car, truck , motorcycle or bicycle - a hill will apply the same amount of resistance in ratio to mass while the conditions are the same.
The big difference comes in "How does each system apply the energy it generates?"
You'll notice is most, if not all, gas vehicles that the "City MPG" is lower than highway. That I due largely to the fact that you're always burning gas, even when you're not moving. All types of EVs (to my knowledge) use a very small amount of power when they are stopped - all of which is required in some way - so there's very little waste comparatively.
Now, with EVs, you're driving an electric motor using electricity from a battery or generator type source. Electric motors always draw more amperage when they come against large resistance, and since EVs don't have a transmission, there's no way to change the forces being fed back to the motor. This mean, hill = resistance = increase in wattage used. However, when you cruise downhill, many vehicles recharge!! So there is a benefit to hills as well.
Aside from these things you have other factors like aerodynamics of the vehicle, friction ratios, tire pressure, power-loss ratios, driving style, ambient temperature and on and on. They all play a part in the complex mechanical system that happens when we drive our vehicles and they all factor into the cost of driving.
In Summary
Yes, all vehicles benefit from the least resistive path which means flattest or steepest downhill, but also no, most gas vehicles won't save gas by using streets on the same commute.
Update to comment
It can happen.
As I noted and was mentioned in a comment, aerodynamics falls into play. You vehicle may just be more efficient at 45 mph.
Consider for example the angle of the hood. Depending on overall geometry, the hood can produce a good amount of downforce at higher speeds. When you add downforce to a hill (rolling or otherwise) you start having force vectors that are stronger.
Downforce is created by peeling energy off of one vector (forward) and redirecting it (down). You can show this on paper quite easily. The wind force (backward) is pushed against an angled surface (spoiler) and is redirected up. Using Newton's third law we know that forcing air up causes force to be applied down. However, also, forcing a redirection of a vector requires and external force, bringing Newton's second law into play as well. This means a force going forward, which is then in conflict with the other forces.
I'm sure I could ramble more on that, but I can't seem to find a good image illustrating my point... I could draw it, but we'll see.
The point is that different vehicles will have different aerodynamic properties. As an example, my wife's Honda Civic seems to get the best mileage in the 40-60 mph range. I did a cross country trip in a Chrysler T&C van and found peak 29 mpg at 85 mph and 21 at 65! So, it varies a lot, and 30% I can easily see.