You said the audio came back up when touching the ripple reject cap and you already swapped that high ESR one out, so maybe it's the internal bias. At least we know it can amplify, now to figure out why its not.
What I find interesting is that a cap there just smooths ripple and it should amplify without it, so even a damaged trace or dry cap there shouldn't inhibit amplification. However, the audio coming up then fizzling out makes me think that touching it shifted the internal bias momentarily and suggests to me that the internal bias supply may have been damaged.
Measure the voltage on pin 2 and see if it drifts or rails. It may also be worth touching the leads of a cap from pin 8 gnd to pin 2 which would bypass any unseen trace/pad damage, although I doubt that's the issue here..
There is no detailed schematic of the inside of that 7222 chip, but I would assume the ripple reject pin should be about Vcc/2 (since it would be on the DC bias circuitry). A quick measurement on a different radio would tell us what the voltage should be (because I'm guessing on Vcc/2).
I am guessing about how the ripple reject works, I could be wrong. heck, I'd just swap the chip and see.
