Lgp 10.0 Preview: Peformance Of The New Inverse Treatment Planning Algorithm For Pituitary Adenoma Treatments

David Schlesinger1, Chun Po Yen2, Jason P. Sheehan, PhD3

1Charlottesville, VA United States 2University of Virginia, Charlottesville, USA 3Department of Neurosurgery, University of Virginia

Keywords: gamma knife, dose planning, pituitary adenoma, software, technique

Interactive Manuscript

Ask Questions of this Manuscript:



     Treatment planning for the Gamma Knife has traditionally been a forward planning only approach with results that depend significantly on the experience of the dosimetrist. LGP version 10.0, currently in beta testing, introduces a new inverse planning engine that may allow more reproducible results across dosimetrists and individual institutions.
     This purpose of this project is to compare the inverse treatment planning results against actual treatment plans completed using the forward-planning method.
     Fifty randomly selected treatment plans created manually by a dosimetrist for pituitary adenomas treated at the University of Virginia between October 2007 and February 2010 will be retrospectively loaded onto the LGP 10.0 treatment planning system. Treatment plans targeting the entire sellar contents (i.e. – no visible tumor) will be excluded. Measures of plan quality including prescription dose, prescription isodose, target coverage, target selectivity, Paddick Gradient Index, number of isocenters, optic pathway doses, and treatment time will be recorded. New treatment plans will be generated using the inverse planning functionality of LGP 10.0, adjusting inverse planning parameters with the goal of exceeding the plan metrics used in the actual clinical plans. Corresponding metrics for the inverse-planner generated plans will be recorded and compared against the original plan metrics.
     The proposed presentation will present the results of the experiment described above and highlight the similarities and/or differences in plan quality metrics between the actual forward planned treatment plans and the corresponding inverse plans. The results will demonstrate the level of capability of the new inverse planning functionality to create treatment plans of a quality that meets or exceeds those currently used, without subsequent manual intervention.
     This is a retrospective study.
     The conclusions of the talk will discuss the advantages and limitations of using the inverse treatment planning engine currently under beta testing and scheduled to be introduced in LGP version 10.0.


Project Roles:

D. Schlesinger (), C. Yen (), J. Sheehan ()