I'm going to answer this for a LWR, since I guess that's what you're thinking of.
First of all, the premise of the question is not quite correct. The power distribution of an LWR is generally sinusoidal, with a peak towards the middle of the fuel. This is true with or without control rods inserted. The most common power reactor design in operation--a conventional PWRs operating in full-power baseload operation--does not have control rods deeply inserted into the core for any significant amount of time. This shape is driven by a few factors: first, fast neutrons tend to 'leak' out of the core without thermalizing more towards the extremities. Additionally, because the top of the fuel has a higher moderator temperature (or maybe even some voids!), this also leads to less thermalization there. So, because of the shape of the power distribution, the tops and bottoms of the fuel assembly have far less power (and therefore less relative depletion) than the 'middle' of the fuel.
When control rods are more heavily use for power distribution control (e.g., in a BWR, or some PWR designs), this creates a secondary 'rod shadowing' effect that further shifts the power distribution away from one axial extremity of the core. Also, there are different groups of control rods, and not every fuel assembly location has a control rod (if you google, you should be able to find some patterns online for conventional reactor types.) Some of these groups are used only for shutdown, and some are meant for insertion into the core when it is operating for power distribution control. These control rods used for power distribution control are not the same composition as rods used for shutdown--power distribution control typically uses control rods that are much weaker neutron absorbers than shutdown rods. So while they cause local reductions in the neutron flux when inserted, there is still plenty of fission going on in those areas... just less of it than there would be if the control rods were withdrawn.
(Side note: BWR control rods come from the bottom of the core, and PWR control rods from the top.)
Now, to answer your actual question: there are a few ways the fuel is built to economize around having less fission at the top and the bottom of the core, but all are basically built around reducing the amount of reactivity in those regions. You can do some or all of these things just for fuel in the top and bottom of a fuel assembly:
- Lower the enrichment (right down to using natural U)
- Use annular fuel pellets
- Not using any neutron poison in the fuel (gadolina etc.)