Since azhdsrchids were mentioned as the potential ancestors and we're talking of large flying predators that don't just strike at much smaller animals, let's take a look at a good example I'll never get tired of using:
This is a Hatzegopteryx. It's slightly shorter than quetzalcoatlus, but also has a bulkier neck and head as well as a similar weight, as this thing was, as far as we know, an apex predator that differently from most of its kind, could afford to hunt and take down relatively large prey, including things larger than humans. I won't whine on how your dragon must look like, but this gives us a good idea of what a more realistic large flying predator would be like in terms of how it behaves, how it hunts and thus how to respond to it.
So firstly, size. While a creature similar to this could hunt and feed on large creatures, including animals bigger than a human, it it's a comfortable thing to do. On a more realistic scenario, your dragon would be rather fragile for a creature it's size,far from easy to beat depending on how big you are, but this thing would probably never win against a healthy animal of similar size that didn't need to adapt in order to still be able to fly. This I worsened by its potential ancestries, since both pterosaurs and birds have pneumatic bones, which is great for staying light, not so good as mammalian bone at avoiding a fracture.
Secondly, flight: this also touches on the previous one, but your dragon is essentially at about the limit of being big and strong while also being able to fly as far as we know from natural examples. In a more realistic scenario, it'd be also well adapted for life on land and spend a large portion of its time on the ground. It'd still be a good flier, but would probably rely on it mostly to move quickly, escape potential threats and patrol for carcasses and potential prey to feed on depending on the environment. As for what you seem to want, the "swoop down and snatch prey from the ground" thing often seen in raptors really doesn't scale well, especially in the case of a pterosaur ancestry.
So from a perspective based on a slightly more "badass" version of one of the best real life equivalent of a giant flying apex predator (and therefore a pterosaur ancestry), a modern, more scientifically correct dragon:
Would be well adapted to live on the ground as well as to fly. It can easily walk for decent distances and live comfortably on the ground without sacrificing its flight ability. It's running speed would likely be of roughly 36 km/h (~22, 3 mph) while it's flight speed would be at around 180 km/h (80 mph), based on quetzalcoatlus estimates.
Would not be the best equipped to do the classic giant eagle/dragon stuff. It can easily pursue and kill prey that's human sized and even go for larger targets, but the swooping down and snatching horse-sized prey from the sky would probably be a no-go. Most examples of large pterosaurs and large raptors (both modern and prehistoric) seem to indicate that it's just not viable at the sizes we want given the many constraints already in place allowing you to be both big and airborne. It'd most likely kill and eat on the ground, maybe taking smaller prey back to its nest by carrying it on its Jaws.
Would probably act like your average predator in terms of what it hunts. Predators aren't stupid to pick fights against the healthiest strongest member of a group, your dragon would look mostly for isolated targets and target mostly infant, sick, elderly or dead animals, since all of those are usually much less capable of fighting back compared to healthy adults. If your Dragons hunt alone, this would be even more important, as a kick to the wing from a horse could easily be a death sentence for your dragon.
So based on this, we can properly tell what adaptations would likely be the most efficient at countering your dragon based on its limitations and habits (and therefore most selected):
1- size and/or speed. Your dragon isn't going to break any sprinting records on land, and it's far too risky for it to target land prey that's similar to it in size. The approach of deer and sauropods both work well here, as deer rapidly making it for more dense forested areas where would mean the dragon wouldn't be able to fly after it, and its superior ground speed would allow the deer to outrun it. Meanwhile, sauropods growing very large very quickly would also mean they'd quickly exit the dragon's prey size range. Land animals that got decently big and strong would also easily become a risky meal for any dragon.
2- herd behavior. When you're in herds, you're not only more protected in case of an attack, especially if you're more to the middle, you also have more people looking around and trying to spot the attacker before it can strike. Chances are that the dragon could ambush a member more separated from the group, especially if it could quickly land and strike, but this would still mean most of the group could survive by sticking together.
3- eyes pointing up. Much like in smaller animals, it wouldn't be crazy if some of your animals developed something like a pineal eye or similar organ for sensing light and dark pointing up. Much like in lizards and insects, being able to form a proper image isn't necessary, but being able to tell when a shadow passes above you can be a good helper when what hunts you is likely to come from above.
And just for fun: some potential traits that could be selected in your Dragons to counter the preys counters:
1- pack mentality. Working in coordinated packs does rely on the species being smart enough, and means you have to share what you hunt, but it also means you're more capable of hunting even larger prey, and can easily mean you're successful more often, something well seen in African wild dogs and their ability to take down much larger prey. A pack can better work together to search for and isolate the ideal prey from its herd, usually by trying to scare it into running away, and at times into another member.
2- venom. This may be harder to pop up, but if it does, it'd be a great advantage. While venom is costly to produce, having it and being able to land a successful bite is a great way to debilitate prey so it's easier to hunt. Depending on how potent the venom is and how the prey heard reacts to sick individuals, your dragon could simply ambush a member of a heard, land a bite and then follow it from afar, as the venom causes it to rapidly become weaker and fall behind, potentially either killing it or leaving it borderline incapable or resisting as the dragon approaches to deal the final blow and enjoy the meal.