Theoretically, I would think an existing Anti-Ballistic Missile would be the best way to deal with it. Stick a GBI in a Starship (I checked, it fits and is within the weight capacity for GTO mission profile) and go full send. That way it would have good enough terminal-phase guidance to actually hit it. Bear in mind that DART had a far larger target with a lot less uncertainty.
Unfortunately I couldn't find any data on extra-atmospheric launch of the entire GBI missile - probably because nobody is dumb enough to ask that in the first place. I would hope it would have the fuel to go the ~4m miles from a GTO orbit, considering its range from surface launch. Nukes are probably a bad idea - more likely to just break it up and ensure that earth will be peppered by assorted meteoroids than anything else.
I don't know if any other ABM would have the performance at those velocities without significant modification. Using publicly available data, GBI appears to be able to handle closure rates of something like 18 Km/s, where SM-3, THAAD, Arrow 3, and 51T6 seem to cap out around 10 Km/s. Considering that 2024 YR4 is already travelling at 13 Km/s it would appear to preclude them. However, considering the angle it would approach from given a 2028 interception, it is possible that closure rate performance may not be so important.
The EKV used in GBI missiles would have to be going 11.5 m/s directly opposite the flight path of the meteor on impact. LEAP, used in SM-3, would need to be going 37.7 m/s. I suspect most of the energy would be burned in getting to interception point, then stopping and turning. Accelerating to 40 m/s in space is trivial.