A competing plan methodology for optimizing the use of IMRT and 3D-CRT

Abstract

Many institutions have the capability to provide both three-dimensional conformal radiation therapy (3D-CRT) and intensity modulated radiation therapy (IMRT) for selected disease sites. For prostate cancer, IMRT offers the potential to improve target dose coverage while further sparing irradiation of bladder and rectum. However, given the added time for planning, treatment delivery and quality assurance, quantifying the actual benefit of IMRT is important. One method commonly employed is comparing dose volume histograms (DVHs) between plans. However, setting dose-volume constraints for target volumes and critical structures in an inverse planning system for IMRT is subjective, especially without standardized methods. Using rigorous definitions for contouring the bladder and rectum, our institution has established benchmarks for bladder and rectum dose volume histograms based on our experience with 3D-CRT. The purpose of this study is to compare bladder and rectum DVHs based on IMRT plans to our established clinical benchmarks to assist physicians and physicists in the plan evaluation for each individual case. Interestingly, as we have gained experience with IMRT planning, we have also noted improvements in our 3D-CRT planning. A total of 97 treatment plans were generated on 66 patients with prostate cancer treated at our institution between April 2000 and December 2003. Patients were divided into 3 groups based on their dates of treatment and type of planning. Group A consisted of 35 patients treated before the routine implementation of IMRT. Once IMRT became available, our policy was to plan each patient with 3D-CRT and IMRT. Group B consisted of 31 patients with IMRT plans and Group C was the same 31 patients with 3D-CRT plans. The target volumes included the prostate with or without seminal vesicles. Patients requiring pelvic fields were not included in this study. Conventional 6-field (2 laterals and 4 obliques) 3D-CRT plans were generated using the Eclipse (Varian Medical Systems) planning system. Five or seven-field coplanar IMRT plans were generated using the same system. A dose of 74 Gy in 37 fractions was prescribed to the 95% isodose surface and the dose was normalized to the isocenter. Bladder and rectum DVH data was summarized to obtain an average DVH for each technique and then compared. For Group C the bladder doses were: mean 28.8 Gy, V60 16.4%, V70 10.9%. Rectal doses were: mean 39.3 Gy, V60 21.8%, V70 13.6%. IMRT plans resulted in similar mean dose values: bladder 26.4 Gy, rectum 34.9 Gy. However, IMRT significantly reduced the values of V70 for the bladder (28%) and rectum (32%). These benchmark DVHs have resulted in critical evaluation of our 3D-CRT techniques over time. The DVHs of Group C patients resulted in better target coverage and lower doses to bladder and rectum than Group A. In fact, in 10% of Group B and C cases, the bladder and rectum DVHs are comparable between 3D-CRT and IMRT. In selected patients, optimization of beam weighting, beam orientation and MLC shaping could result in developing 3D-CRT plans which are comparable to IMRT. Our institution has developed benchmark DVHs based on our clinical experience of 3D-CRT and IMRT. We use these standards as well as differences in individual cases to make decisions on whether patients may benefit from an IMRT plan rather than 3D-CRT. The use of a competing plan methodology to compare IMRT and 3D-CRT has proven to be an effective means of improving the planner’s skill and leading to selecting the best treatment technique. However, despite forward planning optimization of 3D-CRT, IMRT does improve dose conformality of the target volume, while sparing the bladder and rectum in most patients. Whether the reduced doses to the rectum and bladder achieved by IMRT results in lower complication rates remains to be investigated