The APM's THD measurements are corroborated in this review :
https://www.0db.co.kr/REVIEW_0DB/1682002
Rtings's distortion measurements use a different methodology but seem to show very little degradation going from 90dB to 100dB :
https://www.rtings.com/headphones/graph#16092/4029
DIYaudio's measurements of the Bose QC35 II suggest that it can use its DSP to indeed lower distortion vs. running fully passive :
https://diyaudioheaven.wordpress.com/headphones/measurements/bose/qc35-ii/
Presumably the inward facing mic system can be used in a sort of feedback way ?
What I'd be interested to know is whether or not this behaviour also works when the headphones are
on, but ANC is
off. That isn't the case with the XM3 which seems to abandon this behaviour whenever the ANC circuit isn't on (DIYaudio's review), but I'm not sure that it's the case with the Bose 700 or APM (don't know about the QC 35 II, I never used them, only their predecessor).
BTW, thanks to DIYaudio for performing these measurements (sealing variation, ANC on / off / passive, THD in various modes, etc.), I don't think that anyone else really does it quite as thoroughly.
