Judging by the weight loss differences, this looks more like that could be what is driving the differences in LDL, TMAO and insulin. i.e losing more weight is already known to cause differences in LDL levels and other metrics.
And this isn't really representative of what most people on those diets in the real world would be eating. i.e. most meat and cheese substitutes are considerably more processed than the foods they are replacing. i.e they contain significant amounts of salt and saturated fats like coconut oil.
I am vegetarian but that's only really doable for me over the long term because of those highly processed meat substitutes. I am virtually certain I could sustain a less processed diet if I continued to eat meat. i.e a seasoned turkey burger with little if any added salt is going to be A LOT more palatable than anything vegan or vegetarian that isn't processed significantly. And the health advantages of a healthy eating pattern only matter if that healthy eating pattern is actually followed over the long term.
I would be interested in a more long term study that tracked them over years. Just to see how many of them actually stuck with it and how good the long term adherance actually is.