The spreading of a wave packet is very fast in quantum mechanics: for an electron, a gaussian wave packet spreads from one angström to 600km in one second! In his famous QM book, Sakurai mentions that there are numerical evidence that taking into account gravitation as $$-{\hbar^2\over 2m}\Delta\psi(\vec r) -{\cal G}m^2\int {|\psi(\vec r')|^2\over |\vec r-\vec r'|}d^3\vec r'\psi(\vec r)=i\hbar{\partial\psi\over\partial t}$$ stops the spreading at distances around 500nm. See for example https://arxiv.org/abs/1105.1921
Why should we take into account the gravitational interaction of the wave packet with itself but not the electromagnetic one? In the case of a free electron for example, the Coulomb repulsion of the wave packet by itself would accelerate the spreading.