I've noticed it mostly when I have the phone next to the amplification device rather than the cables.
I'd have to agree with you there. In one of our projects we had an audio amplifier IC that when powered would emit a terrible GSM buzz whenever a transmitting phone was within 6 feet of the PCB. But we redesigned it to use a more noise immune amp and now do not exprience the problem.
It has a lot to do with the topology of the amplifer as well as the signal routing. A differential signal all the way to the transistors, fets, ICs etc will help immensely. In our case we are using a 1 watt digital amplifer with single-ended connection but were using a BTL (bridge tied load) device.
Stu