Room: Track 6
Purpose:
We recently proposed a planning method for single isocenter multiple target (SIMT) radiosurgery entitled Conformal Arc Informed VMAT (CAVMAT). Here we report further development of CAVMAT as a semi-automated technique and evaluate its applicability to large volume cases.
Methods:
The CAVMAT technique comprises three steps: assigning targets to subgroups per arc, arc weighting, and limited inverse optimization to increase target conformity and meet plan specific OAR constraints. Target subgroup assignment was automated via a script within the planning system; field weight normalization was automated using standalone code; limited inverse optimization was performed within the Eclipse optimizer at MR level 4. In addition, CAVMAT was applied to SIMT cases with larger volumes (plans with targets >2cm diameter or total V12Gy > 20cc), whereas prior studies focused on single fraction SRS. Eight large volume intracranial multi-target radiosurgery cases, previously planned with VMAT, were re-planned with CAVMAT. Plans were normalized such that VMAT and CAVMAT had the same target coverage. Plan quality was evaluated by V5Gy, V12Gy, and conformity index.
Results:
Semi-automating CAVMAT decreased planning time from the 4+ hours required for manual CAVMAT to a feasible timeframe for routine planning (~1hr). For large volume cases replanned with CAVMAT, V5Gy and V12Gy decreased by 21.2% (411±171cc for VMAT vs. 339±123 for CAVMAT) and 12.9% (61±18cc vs. 54±16cc), respectively when max target volume was <30cc (n=6). However for plans with >30cc (n=2), V12Gy was smaller for VMAT. Similar trends were observed for conformity index.
Conclusion:
The decreased planning time and versatility of use with volumes <30cc SIMT cases suggest CAVMAT is a practical SRS planning technique for most cases. Future work will entail further automation, improved field weight normalization to account for large volumes, and comparison with other SRS SIMT planning techniques.
Not Applicable / None Entered.