AOA approaches
....who cares about airspeed?)
Expanding Heyjoe's answer:
L=0.5*rho*(V^2)*S*(CL)
CL=f(AoA); more explicitly, CL~2*Pi*AoA + k
Once we're on final, we won't monkey with the flaps, slats, etc. so S (planform area of the lifting surface, read: wing) is fixed, as is the lift curve slope (approx 2*Pi for a thin airfoil in the linear portion of the curve). We're not changing altitude all THAT much, so air density (rho) is constant.
So, in order to keep a nice, constant rate of descent, L=W, meaning that can be assumed constant too. We are assuming it's not a Hornet whose time rate of change in weight is astronomical, owing to fuel consumption.
So what parameters can we monkey with to keep or alter descent rate? CL and V, with velocity being the only one we can directly measure in the cockpit. CL being controlled solely by AoA, for which we have a handy indexer on the panel.
BUT, and I only mention b/c your post sounded like you were thinking this way, AoA is not Pitch. You can be tricked into thinking they are largely the same or similar, but you would be wrong and likely kill yourself (see "Stick and Rudder" by Langewiesche, basically the whole friggen book is about this concept). B/c of this, you might forget that AoA is determined by the vector resolution of the relative wind, caused both by your forward airspeed and your vertical airspeed. So, if you just hold pitch and descent rate, and neglect airspeed, it could wander on you, which will throw AoA off.
Also, pitot static instruments are less laggy and usually better calibrated (larger scale) than your AoA indexer, so you'll be able to see the movements sooner and correct. So bottom line, airspeed is key to a good AoA approach. Leave it in your scan.