There was this interesting study out of the Univ of Chicago 5 years ago showing that partial tumor irradiation for masses over 65 cc was just as effective as total tumor irradiation for masses < 65 cc. The patients received at least one cycle of pembrolizumab. Not all lesions were irradiation, some were monitored for non-target (abscopal) responses.
There weren't abscopal responses per se, but there were a lot of what they called
ADscopal (local bystander responses) in tumors that were only partially irradiated.
The local control rates were statistically the same for partially (88%) and fully (95%) irradiated tumors, even though the partially-treated tumor mean volume was 116.6 mL, vs only 7.2 mL for fully irradiated:
The take home for me was that maybe partial tumor irradiation could work in the immunotherapy era, or at least some form of margin reduction - maybe negative margins are the way to go. Maybe we need to get out of our own way a little bit, I know I'm always messing up my intended results, sometimes from trying too hard to "fix it real good"