SU‐GG‐T‐159: Has IMRT Delivery Improved in the Last 5 Years?

Purpose: To determine whether institutions that participate in clinical trials are delivering IMRT more accurately than they were five years ago. Method and Materials: Anthropomorphic head phantoms were mailed to institutions wishing to participate in national clinical trials using IMRT. The phantoms consisted of an imagable insert containing a primary PTV, a secondary PTV and an OAR. The insert also housed TLD and radiochromic film. The institutions imaged the phantom and planned and delivered an IMRT treatment as though it were a patient. The phantom was returned and the TLD and film were analyzed. The delivered dose distribution was compared with the institutions' treatment plans and the results judged against 7% dose and 4 mm DTA criteria developed by the RPC and the RTOG. Results: Between 2001 and 2007 the phantom was irradiated 475 times. The overall pass rate was 76%. The pass rate prior to 2005 was 70%. The pass rate for 2005 was 76% and the pass rates for 2006 and 2007 were 76% and 80% respectively. Conclusion: Institutions interested in participating in NCI sponsored IMRT protocols are delivering IMRT more accurately than they were 4 years ago. One of the reasons for this is improved modeling capabilities in treatment planning systems. One of the most common treatment machine/planning system combinations has shown an improvement in pass rates since a new version of software became available. Other reasons could include improvements in the tools for patient‐specific QA and improvements in the commissioning of IMRT systems. Though there has been improvement, there is still room for more. 20% of the irradiations are still failures. Adequate IMRT quality assurance and commissioning is essential now as always. Conflict of Interest: Work supported by PHS grant CA10953 and CA081647 from the NCI, DHHS.