Evaluation of the Quality of Organs-At-Risk Dose Sparing in Anorectal and Prostate IMRT Plans
S Lu*, L Yuan, F Yin, Q Jackie Wu, Duke University Medical Center, Durham, NCSU-E-T-707 Sunday 3:00PM - 6:00PM Room: Exhibit Hall
To evaluate the quality of organs-at-risk (OARs) dose sparing in IMRT plans for anorectal and prostate cancer cases utilizing organ-based knowledge models to predict OAR DVHs.
The quality of OAR dose sparing in IMRT plans are evaluated by organ specific knowledge models. The knowledge models are built upon prior expert plans and accounts for patient specific anatomy. The knowledge models predict the optimal OAR sparing for bladder, small bowel and femoral heads in anorectal cases and bladder, rectum and femoral heads in prostate cases. We tested 20 anorectal plans and 20 prostate plans using the knowledge models for quality assurance. The reference dosimetric parameters for each test plan generated by knowledge models are compared with the corresponding actual plan values. The inter-quartile range of their differences is used as an indication of the quality variation of the dose sparing quality for a specific OAR for each site.
The inter-quartile-ranges of the deviations between reference dosimetric parameters and actual plan values in anorectal plans are (in percent of prescription dose): bladder D50%: 14.6%, D5%: 7.6%, small bowel D50%: 2.4%, D5%: 5.9%, femoral heads D50%: 11.6%, D5%: 9.6%. The inter-quartile ranges in prostate plans are: bladder D50%: 2.5%, D5%: 6.3%, rectum D50%: 9.6%, D5%: 3.9%, femoral heads: D50%: 10.9%, D5%: 13.9%. The significant smaller inter-quartile-range for small bowel indicates highly consistent dose sparing quality for this organ. The larger inter-quartile-range for bladder D50% in anorectal plans compared with that in prostate plans indicates the larger quality variation of bladder dose sparing in anorectal plans and should warrant improvement.
Based on the patient specific OAR dose sparing references generated by knowledge models, we found the dose sparing quality for small bowel is highly consistent while there is larger variation in the quality of bladder dose sparing in anorectal plans.
Funding Support, Disclosures, and Conflict of Interest: Partly supported by NIH/NCI and a Varian master research grant.