The study dissolved it in DMSO, then that in polysorbate 80, then that in polyethylene glycol. Then injected that intraperitoneally. That seems doable.
Intraperitoneal injection isn't really done in humans, the closest equivalent would be subcutaneous or maybe intramuscular injection. I don't know about the safety of injecting DMSO into humans but it's probably not too bad since it seems perfectly safe when used for transdermal drug delivery. The bigger problem is that you're injecting a water-insoluble compound mixed in a small amount of solvent into what is essentially a big bag of water. Once that DMSO/PEG/whatever is injected, it will just mix with the water in your body, so the ISRIB will just precipitate out of solution since it is now surrounded by 99% water molecules. So it will never get distributed to all your organs, which is the goal.
DMSO is soluble in both polar and non polar solvents. I assume the insolubility in water is precisely what they used it to solve.
Im any case, their delivery mechanism worked in mice, so it would work in people. The part that might not carry over to people is the effect once in the cell.