Comparison of RapidArc vs. Conventional Intensity Modulated Radiation Therapy for Stereotactic Body...
1
Results: Significant inter-fraction directional and vector shifts were noted (largest in sup-inf direction). Maximum shifts observed were 10.0 mm (lateral), -31.0 mm (longitudinal), and -12.0 mm (vertical). Shifts $ 5 mm were applied inter-fractionally in 6.8% (lateral), 29.5% (longitudinal), and 6.8% (vertical) of all delivered treatments. For a 10 mm threshold, corresponding percentages were 2.3%, 13.6%, and 2.3%, respectively. Shifts $ 20 mm were applied only in longitudinal direction (11.4% of all delivered treatments). Across all fractions, calculated inter-fractional shift vectors ranged from 0-31.2 mm, with 40.9%, 15.9%, and 11.4% of all fractions having shift vectors $ 5 mm, $ 10 mm, and $ 20 mm. Intra-fraction directional and vector shifts were non-significant. Mean overall reduction in GTV of 21.1% was observed for all tumors over course of SBRT, but no temporal trends were identified. Conclusions: Significant changes in both position and volume occur during SBRT for early stage NSCLC. Shifts (particularly in the sup-inf direction) may exceed applied margins and compromise target coverage. Due to extremely high fraction doses and small number of fractions associated with SBRT, inter-fractional image guidance is essential. Author Disclosure: S.K. Yi, None; J. Perks, None; R. Houston, None; R. Stern, None; J.A. Purdy, None; A.M. Chen, None. 3298 Dosimetric Effect of Extra-Parenchymal Dose Spill in the Stereotactic Radiosurgical Treatment of Metastatic Lesions P. E. Young, R. Jacob, J. Fiveash, R. Meredith, S. Spencer, M. Dobelbower UAB Highlands Gamma Knife, Birmingham, AL Purpose/Objective(s): The Purpose of this project was to investigate the dosimetric effect of neglecting the extra-parenchymal portion of the prescription isodose volume (PIV) applied to the stereotactic radiosurgical treatment of peripheral, metastatic lesions. Materials/Methods: A series of 60, peripheral, metastatic lesions, treated by Gamma Knife Radiosurgery, was investigated. For each lesion, the Leksell Gamma Plan Software [LGP 8.3.1] was used to determine the total PIV and the extra-parenchymal portion, or ‘‘spill’’, of the PIV. This dose spill was subtracted from the total PIV to determine a discounted PIV. Using a dose-volume, iso- effect curve, historically employed in our clinic, the total and discounted PIV were separately employed to determine two prescrip- tion dose values for each lesion. The difference in these two dose values was recorded as a potential prescription dose increase. Similarly, the total and discounted PIV values were separately employed in calculations of dose conformality index (CI = PIV/ GTV where GTV = Gross Tumor Volume). Results: The maximum, potential increase in dose prescription found was 1 Gy, with 10% [6/60] of lesions being assigned a po- tential increase of 0.5 to 1.0 Gy. Of these six, five were surgical resection beds, having a significant portion of their PIV spilling into the resection cavity. The sixth lesion was an unresected met at the extreme edge of treatable frame space. The position of this lesion necessitated the use of larger than usual shots to cover the anterior edge of the lesion. 12% [7/60] of the lesions, had initial CI values . 2. After making adjustments for extra-parenchymal spill, six had their CI reduced to \2. Only 2% [1/60] had a final, adjusted CI . 2. Conclusions: In the treatment of peripheral lesions, especially those with resection cavities, discounting extra-parenchymal dose spill results in effectively improved dose conformality and the potential selection of higher prescription dose values. Author Disclosure: P.E. Young, None; R. Jacob, None; J. Fiveash, None; R. Meredith, None; S. Spencer, None; M. Dobelbower, None. 3299 Comparison of RapidArc vs. Conventional Intensity Modulated Radiation Therapy for Stereotactic Body Radiation Therapy for Pancreatic Cancer K. N. Kielar, T. F. Atwood, C. M. Taniguchi, C. Christman-Skieller, L. Xing, A. C. Koong, D. T. Chang Stanford University, Stanford, CA Purpose/Objective(s): To provide a dosimetric comparison of RapidArcÒ vs. conventional intensity-modulated radiation therapy for stereotactic body radiation therapy (SBRT) for pancreatic adenocarcinoma. Materials/Methods: In this study, 5 patients who were previously treated with SBRT at Stanford University using conventional step-and-shoot IMRT (CIMRT) to a dose of 30-33 Gy in 5 fractions were evaluated. For each patient, a new RapidArcÒ IMRT (RA) plan was generated using 6 MV photons, with two full arcs and alternating collimator angles. The equivalent dose was pre- scribed to the planning target volume (PTV), which consisted of a 3 mm expansion on the gross tumor volume (GTV). All CIMRT and RA plans were normalized so that 100% of the prescription dose was delivered to 95% of the PTV. Each of the patients had implanted fiducial seeds in the GTV for localization, and all RA plans were generated by a single planner (TA). For the main critical structures, duodenum and stomach, the dose to 3 cc (D 3 ), 9 cc (D 9 ), 16 cc (D 16 ), maximum dose (D max ), mean dose (D mean ), volume that receives 5 Gy (V 5 ), 10 Gy (V 10 ), 15 Gy (V 15 ), 20 Gy (V 20 ), and 25 Gy (V 25 ) were compared between the plans. Results: When compared to the original CIMRT plans, the RA technique was able to produce treatment plans with similar target conformality. Both RA and CIMRT plans had a mean conformality index of 1.08. For the duodenum, the RA plans had lower mean D 3 (14.2 Gy vs. 17.3 Gy, p = 0.01), D 9 (11.0 Gy vs. 13.0 Gy, p = 0.045), D mean (10.0 Gy vs. 11.9 Gy, p = 0.03), V 25 (0.1 cc vs. 0.7 cc, p = 0.04), and D max (28.6 Gy vs. 31.4 Gy, p = 0.0002) compared to the CIMRT plans. The mean D 16 (8.5 Gy vs. 9.0 Gy, p = 0.63), V 10 (19.5 cc vs. 13.7 cc, p = 0.06), V 15 (3.1 cc vs. 6.4 cc, p = 0.11), and V 20 (0.6 cc vs. 2.2 cc, p = 0.09) were not different for RA vs. CIMRT. The mean V 5 was higher for the RA vs. the CIMRT plans (35.8 cc vs. 33.0 cc, p = 0.01). For the stomach, the mean D max was again lower for the RA plans compared to the CIMRT plans (27.4 Gy vs. 30.5 Gy, p = 0.006), but no difference was seen for D mean or D 9 . Conclusions: RA was able to significantly reduce the amounts of duodenum and stomach that received high doses of radiation compared to CIMRT for pancreatic SBRT, while lower and intermediate dose regions were larger or similar. Because of the re- duction in radiation dose to the duodenum and stomach, RA treatment should decrease the risk of treatment related complications. Author Disclosure: K.N. Kielar, None; T.F. Atwood, None; C.M. Taniguchi, None; C. Christman-Skieller, None; L. Xing, None; A.C. Koong, None; D.T. Chang, None. Proceedings of the 52nd Annual ASTRO Meeting S785
Transcript of Comparison of RapidArc vs. Conventional Intensity Modulated Radiation Therapy for Stereotactic Body...
Proceedings of the 52nd Annual ASTRO Meeting S785
Results: Significant inter-fraction directional and vector shifts were noted (largest in sup-inf direction). Maximum shifts observedwere 10.0 mm (lateral), -31.0 mm (longitudinal), and -12.0 mm (vertical). Shifts $ 5 mm were applied inter-fractionally in 6.8%(lateral), 29.5% (longitudinal), and 6.8% (vertical) of all delivered treatments. For a 10 mm threshold, corresponding percentageswere 2.3%, 13.6%, and 2.3%, respectively. Shifts $ 20 mm were applied only in longitudinal direction (11.4% of all deliveredtreatments). Across all fractions, calculated inter-fractional shift vectors ranged from 0-31.2 mm, with 40.9%, 15.9%, and11.4% of all fractions having shift vectors $ 5 mm, $ 10 mm, and $ 20 mm. Intra-fraction directional and vector shifts werenon-significant. Mean overall reduction in GTV of 21.1% was observed for all tumors over course of SBRT, but no temporal trendswere identified.
Conclusions: Significant changes in both position and volume occur during SBRT for early stage NSCLC. Shifts (particularly inthe sup-inf direction) may exceed applied margins and compromise target coverage. Due to extremely high fraction doses and smallnumber of fractions associated with SBRT, inter-fractional image guidance is essential.
Author Disclosure: S.K. Yi, None; J. Perks, None; R. Houston, None; R. Stern, None; J.A. Purdy, None; A.M. Chen, None.
3298 Dosimetric Effect of Extra-Parenchymal Dose Spill in the Stereotactic Radiosurgical Treatment of
Metastatic LesionsP. E. Young, R. Jacob, J. Fiveash, R. Meredith, S. Spencer, M. Dobelbower
UAB Highlands Gamma Knife, Birmingham, AL
Purpose/Objective(s): The Purpose of this project was to investigate the dosimetric effect of neglecting the extra-parenchymalportion of the prescription isodose volume (PIV) applied to the stereotactic radiosurgical treatment of peripheral, metastatic lesions.
Materials/Methods: A series of 60, peripheral, metastatic lesions, treated by Gamma Knife Radiosurgery, was investigated. Foreach lesion, the Leksell Gamma Plan Software [LGP 8.3.1] was used to determine the total PIV and the extra-parenchymal portion,or ‘‘spill’’, of the PIV. This dose spill was subtracted from the total PIV to determine a discounted PIV. Using a dose-volume, iso-effect curve, historically employed in our clinic, the total and discounted PIV were separately employed to determine two prescrip-tion dose values for each lesion. The difference in these two dose values was recorded as a potential prescription dose increase.Similarly, the total and discounted PIV values were separately employed in calculations of dose conformality index (CI = PIV/GTV where GTV = Gross Tumor Volume).
Results: The maximum, potential increase in dose prescription found was 1 Gy, with 10% [6/60] of lesions being assigned a po-tential increase of 0.5 to 1.0 Gy. Of these six, five were surgical resection beds, having a significant portion of their PIV spillinginto the resection cavity. The sixth lesion was an unresected met at the extreme edge of treatable frame space. The position of thislesion necessitated the use of larger than usual shots to cover the anterior edge of the lesion. 12% [7/60] of the lesions, had initialCI values . 2. After making adjustments for extra-parenchymal spill, six had their CI reduced to\2. Only 2% [1/60] had a final,adjusted CI . 2.
Conclusions: In the treatment of peripheral lesions, especially those with resection cavities, discounting extra-parenchymal dosespill results in effectively improved dose conformality and the potential selection of higher prescription dose values.
Author Disclosure: P.E. Young, None; R. Jacob, None; J. Fiveash, None; R. Meredith, None; S. Spencer, None; M. Dobelbower,None.
3299 Comparison of RapidArc vs. Conventional Intensity Modulated Radiation Therapy for Stereotactic Body
Radiation Therapy for Pancreatic CancerK. N. Kielar, T. F. Atwood, C. M. Taniguchi, C. Christman-Skieller, L. Xing, A. C. Koong, D. T. Chang
Stanford University, Stanford, CA
Purpose/Objective(s): To provide a dosimetric comparison of RapidArc� vs. conventional intensity-modulated radiation therapyfor stereotactic body radiation therapy (SBRT) for pancreatic adenocarcinoma.
Materials/Methods: In this study, 5 patients who were previously treated with SBRT at Stanford University using conventionalstep-and-shoot IMRT (CIMRT) to a dose of 30-33 Gy in 5 fractions were evaluated. For each patient, a new RapidArc� IMRT(RA) plan was generated using 6 MV photons, with two full arcs and alternating collimator angles. The equivalent dose was pre-scribed to the planning target volume (PTV), which consisted of a 3 mm expansion on the gross tumor volume (GTV). All CIMRTand RA plans were normalized so that 100% of the prescription dose was delivered to 95% of the PTV. Each of the patients hadimplanted fiducial seeds in the GTV for localization, and all RA plans were generated by a single planner (TA). For the main criticalstructures, duodenum and stomach, the dose to 3 cc (D3), 9 cc (D9), 16 cc (D16), maximum dose (Dmax), mean dose (Dmean), volumethat receives 5 Gy (V5), 10 Gy (V10), 15 Gy (V15), 20 Gy (V20), and 25 Gy (V25) were compared between the plans.
Results: When compared to the original CIMRT plans, the RA technique was able to produce treatment plans with similar targetconformality. Both RA and CIMRT plans had a mean conformality index of 1.08. For the duodenum, the RA plans had lower meanD3 (14.2 Gy vs. 17.3 Gy, p = 0.01), D9 (11.0 Gy vs. 13.0 Gy, p = 0.045), Dmean (10.0 Gy vs. 11.9 Gy, p = 0.03), V25 (0.1 cc vs. 0.7cc, p = 0.04), and Dmax (28.6 Gy vs. 31.4 Gy, p = 0.0002) compared to the CIMRT plans. The mean D16 (8.5 Gy vs. 9.0 Gy, p =0.63), V10 (19.5 cc vs. 13.7 cc, p = 0.06), V15 (3.1 cc vs. 6.4 cc, p = 0.11), and V20 (0.6 cc vs. 2.2 cc, p = 0.09) were not different forRA vs. CIMRT. The mean V5 was higher for the RA vs. the CIMRT plans (35.8 cc vs. 33.0 cc, p = 0.01). For the stomach, the meanDmax was again lower for the RA plans compared to the CIMRT plans (27.4 Gy vs. 30.5 Gy, p = 0.006), but no difference was seenfor Dmean or D9.
Conclusions: RA was able to significantly reduce the amounts of duodenum and stomach that received high doses of radiationcompared to CIMRT for pancreatic SBRT, while lower and intermediate dose regions were larger or similar. Because of the re-duction in radiation dose to the duodenum and stomach, RA treatment should decrease the risk of treatment related complications.
Author Disclosure: K.N. Kielar, None; T.F. Atwood, None; C.M. Taniguchi, None; C. Christman-Skieller, None; L. Xing, None;A.C. Koong, None; D.T. Chang, None.
