I can only speak for HFSS: if you want to simulate the same environment (including the antennas etc), then you would need a huge amount of RAM... maybe with the newer IE solver you could get there, but with less accuracy.
If you are okay with some abstraction, maybe something like a waveguide environment, then HFSS could definitively be your weapon of choice here. The main difference is that HFSS is a FEM solver (frequency domain) whereas CST is a FIT (somewhat like FDTD) solver, ie in the time domain. So, if you want to simulate the absorption or reflection over a frequency span, I think HFSS is definitively capable of doing the job.