COVID has a serious effect on one's lungs. Smoking does as well. I'm sure they would prefer to see how the medicine works on healthy subjects as a baseline. If it doesn't have the same effect on smokers as the baseline, then you can infer from that. If you just get tainted samples from "diseased" lungs of smokers in the original testing, then that could skew the results. That would be my SWAG.