Within reason. A doc obsessed with homogeneity may spend hours a week tweaking segs on breast plans. I have yet to be convinced that this is correlated with changes in clinical outcomes (or that we should be confident in our homogeneity or dmax for most plans beyond a roughly 5% uncertainty). I do agree that strategy (beam arrangement, technique, prioritization of planning objectives) is key. I am willing to bet the farm that BMI and breast size correlate much more with long term cosmetic outcomes for whole breast treatment than V105 (although these are not independent).
I'm happy to let dosi do their job.
Regarding RA. Such a great tool for so many applications, but I've yet to be convinced that it is the right strategy for breast. I don't even use it for APBI (although, I think this is fine, but benefit here more related to time on table than anything else).
To me, RA provides benefit regarding time on table and in terms of finding dosimetric solutions unique to having nearly infinite gantry angles. It allows for nice homogeneity and robustness when you want it (prostate, most head and neck) and steep fall off for simple shapes when you allow the target to get hot (SBRT lung and brain). It is almost never a useful tool to minimize low dose bath to adjacent structures however, and this is what we are aiming for in breast. (very low heart, low lung and very low contralateral breast). We tolerate some heterogeneity in breast (lots in most comprehensive nodal and chest wall plans), but breast is by definition adjuvant treatment and there is no tumor to dump dose into to help with steep fall off like in SBRT cases.
