Program Information
Dosimetric Impact On Patient Plans From Errors That Can Pass IMRT QA
J Steers1,2*, B Fraass1 , (1) Cedars-Sinai Medical Center, Los Angeles, CA, (2) University of California - Los Angeles, Los Angeles, CA
Presentations
WE-D-BRA-2 (Wednesday, July 15, 2015) 11:00 AM - 12:15 PM Room: Ballroom A
Purpose: To assess the clinical impact of errors that can pass IMRT QA when using various gamma criteria.
Methods: We previously described the error curve method that allows us to quantitatively determine gamma criteria sensitivities to induced errors in IMRT QA plans. Gamma sensitivity is reported as the range of errors that may go undetected for a given criterion. Using 21 IMRT cases and their respective gamma sensitivities, we induced MU errors and MLC errors back onto the patient anatomy to study the effects of these errors on DVH metrics for PTVs and OARs (mean dose, maximum dose to a volume, and PTV D95). Gamma criteria using a dose threshold of 50% (TH50) were included in this study as our previous data showed increased sensitivity for gamma with increasing dose threshold. Since different sites and prescription regimens were used in the clinical plans, effects on DVH metrics were reported as the percent difference between the error-induced plan relative to the clinical plan.
Results: For the studied errors, we found that using 3%/3mm TH10 90% pixels passing (PP) can allow large errors to go undetected, resulting in dosimetric differences of >±5% for both PTVs and OARs. Even when increasing PP to 95%PP, some cases still experienced ±7% and ±5% differences in PTV and OAR dose metrics, respectively. A stricter criterion, 2%/3mm TH50 90%PP, was able to dramatically increase gamma sensitivity to errors, reducing dosimetric differences to <4% for most cases.
Conclusion: Large errors can go undetected with commonly used gamma criteria. Such errors, should they exist and fail to be discovered in routine IMRT QA, can cause clinically significant dose discrepancies. These findings further support the growing body of evidence that stricter/better metrics are required in order to identify errors that may negatively affect clinical outcomes.
Contact Email: