If you switch a transistor between cutoff and saturation region, you always do this with another switch, which switches between cutoff and triode region (I call it "full switch"). Example: to switch on/off a constant current source from a current mirror (operates in saturation region), you either fully switch on/off its Vgs, or use a serial (cascoded) full switch.
BTW: Here's a nice English language exercise for pronouncing "switch":
Three swiss witchbitches, which wished to be switched swiss witchbitches, wish to watch three swiss Swatch watch switches. Which swiss witchbitch which wishes to be a switched swiss witchbitch, wishes to watch which swiss Swatch watch switch?
SCNR! ;-)