It would be very unwise for the devs to use a different game engine.
Scawen and Eric knows the ins and outs of their own system and although it's taking time, they are making the system work to their needs and not having to try and make the assets and code conform to another system.
If you use a third party engine, you have to work to their developer's systems and pipelines. There's probably a lot in there that is not required too because they're a bit like a Swiss Army Knife.
If LFS was a new project and the content was still in blockout stages, then using an engine such as Unreal 4 or Unity may make sense if time is an issue and you just want something to work out of the box and refine after initial release.
But for an established IP such as LFS, it would require a crazy amount of work to set up all art assets so they can be exported using the 3rd Party's exporter or to take full advantage of the engine's features.
Also the same assets may need more materials set up, which is fine if you have a large team who can essentially port it all over to use the new engine's rendering potential.
But then there is also the issue of trying to re-write the code that handles the car dynamics. One of the things that makes LFS survive the test of time, even if rendering tech had overtaken it over the last several years until now.
Plus if LFS were to use another developer's engine i.e. Unity or UE4, expect the cost of the licences to increase as Scavier would need to recoup the engine licence fee somehow.