- Email: [email protected]
Consequence of varying target coverage and dose heterogeneity in accelerated partial breast irradiation brachytherapy modalities on tumor control probability
106
Abstracts / Brachytherapy 7 (2008) 91e194
Purpose: Previous studies of external beam radiotherapy for breast cancer have shown that the risk of chest wall toxicity increases with larger fraction sizes. A recent experience using the MammoSiteÒ (MS) reported a 12.5% incidence of chronic chest wall pain. Reports from several institutions using the multi-catheter (MC) technique have not shown elevated chest wall toxicity. Additionally, recent investigations have suggested that increased toxicity may occur with the MS when the dose to the skin, and perhaps chest wall, exceeds 125% of the prescribed dose. This investigation compares the skin and chest wall doses of a cohort of patients treated with the MC technique to a group treated with the MS. Methods and Materials: The dosimetric data for 43 patients treated with the MC technique and 83 patients treated with the MS at Virginia Commonwealth University (VCU) were reviewed. This cohort represents consecutively treated patients from our most recent experience to minimize the effect of any learning curve on dosimetry. Plans were generated using 3D software (Brachyvision, Varian Medical Systems, Inc., Palo Alto, CA). Multiple dwell positions were used for all MS patients to optimize dose delivery. The minimum distances from the planning target volume to the skin and chest wall were calculated, as well as the maximum doses delivered to the skin and chest wall. Results: The mean skin distances for patients treated with the MC technique and MS were 0.5 and 1.3 cm, respectively (p ! 0.001). Despite the significantly smaller mean skin distance, the mean skin dose for the MC technique was only 2.3 Gy per fraction (67% of prescription). The mean skin dose for the MS was 3.2 Gy per fraction (94% of prescription, p ! 0.001). The mean chest wall distance was 0.9 cm for the MC technique and 1.2 cm for the MS (p 5 0.035). Again, the mean chest wall dose for the MC technique was only 2.3 Gy per fraction (67% of prescription). The mean skin dose for the MS was 3.6 Gy per fraction (105% of prescription, p ! 0.001). The percentage of patients receiving skin doses in excess of 125% for the MC and MS were 0% and 9.6%, respectively. The percentage of patients receiving chest wall doses in excess of 125% for the MC and MS were 0% and 38.6%, respectively. Conclusions: The MC technique results in more conformal dose delivery, with significantly lower mean skin and chest wall doses. Treatment with the MS was associated with significantly more patients receiving doses to the skin or chest wall in excess of 125% of the prescription. Given the limited followup available for the MS, and the significant dose delivered to the skin and chest wall, the use of this device may be associated with a higher incidence of late chest wall toxicity than previously expected. Dr. Cuttino is a consultant for Hologic, Inc. (formerly Cytyc).
OR40 Presentation Time: 10:50 AM Consequence of varying target coverage and dose heterogeneity in accelerated partial breast irradiation brachytherapy modalities on tumor control probability Michael C. Kirk, Ph.D., Alan B. Coon, M.D., Ph.D., Lan Jiang, Ph.D., Damian Bernard, Ph.D., James C.H. Chu, Ph.D. Radiation Oncology, Rush University Medical Center, Chicago, IL. Purpose: The NSABP-B39/RTOG-0413 protocol is randomizing patients to conventional whole breast irradiation versus one of three accelerated partial breast irradiation (APBI) modalities: Mammosite brachytherapy (MB), multi-catheter brachytherapy (MCT) and 3D conformal partial breast treatment. The two brachytherapy based APBI techniques typically result in widely varying dose heterogeneity in the target (with dose heterogeneity typically larger in MB). We studied the effect of this variability in heterogeneity and minimum target dose coverage on the predicted tumor control probability (TCP) based on the linear-quadratic (LQ) model. Methods and Materials: Values for the minimum target coverage-V90 (percent volume receiving 90% of prescription dose), V100 (percent volume receiving 100% of prescription dose)-as well as dose heterogeneity-V150 (percent volume receiving 150% of prescription dose)-were chosen to encompass the range found in typical MB and MCT treatments. For each level of target coverage, V100 5 100%, V100 5
95%, V90 5 95% and V90 5 90%, the value of V150 was allowed to vary between 10 and 40%. The dose volume histograms (DVHs) were generated using a smoothing spline through these points so that the resulting curves exhibited typical features of brachytherapy DVHs. The curve representing the cumulative DVH was converted to a differential DVH using in-house Interactive Data Language (IDL) software. The TCP was calculated for each DVH using the BIOPLAN software package. The LQ parameters a 5 0.3 Gy-1 and a/b 5 10 Gy were used for the calculations. Results: For a target coverage of V100 5 100%, the TCP was 95.5%, 95.8%, 96.1%, 96.6% for V150 values of 10, 20, 30, 40 respectively. For a target coverage of V100 5 95%, the TCP was 88.5%, 88.8%, 89.2%, and 89.6% for V150 values of 10, 20, 30, 40 respectively. For a target coverage of V90 5 95%, the TCP was 84.3%, 84.5%, 84.9%, and 85.3% for V150 values of 10, 20, 30, 40 respectively. For a target coverage of V90 5 90, the TCP was 79.1%, 79.5%, 79.8%, and 80.1% for V150 values of 10, 20, 30, 40 respectively. The mean variation in TCP over all target coverages for any V150 values greater than 16.4%. On the other hand, for a specific minimum target coverage, the mean variation in calculated TCP over all V150 values was less than 1.2%. Conclusions: Large variations in V150 result in less than 1.2% differences in TCP for the same V100 or V90 target coverage. Thus the two APBI brachytherapy modalities can be considered equivalent despite the greater dose heterogeneity found in MB. TCP, however, could vary by more than 16% between V100 5 100% and V90 5 90%. We recommend that these data be considered while analyzing results from the protocol.
PHYSICS ORAL PRESENTATION SESSION 1 (electronic brachytherapy) Monday May 5, 2008 10:00 AMe11:00 AM OR41 Presentation Time: 10:00 AM Depth dose modulation (ddm) for electronic brachytherapy Jessica R. Hiatt, M.S.1,2 Jaroslaw T. Hepel, M.D.1,2 Gene A. Cardarelli, Ph.D.1,2 Mark Carol, M.D., Edward S. Sternick, Ph.D.1,2 David E. Wazer, M.D.1,2 1Radiation Oncology, Rhode Island Hospital, Providence, RI; 2 Radiation Oncology, Tufts-New England Medical Center, Boston, MA. Purpose: To design a system for modulation of depth dose characteristics for the Xoft (Xoft, Inc., Fremont, CA) micro-miniature electronic x-ray source. Utilizing source collimation and indexing, customized dose distributions can be created for a variety of brachytherapy applications. Methods and Materials: A custom collimator was designed for the Xoft (Xoft Inc., Fremont, CA) 50 kV x-ray source. The collimator consists of a cylindrical band for the lower portion of the source and a cap for the source tip and allows for collimation of the source to a circumferential fan beam. The degree of collimation can be adjusted by varying the width of the aperture. Depth-dose experiments were performed evaluating the effect of collimation and indexing. Collimation was varied using no collimation (unshielded source) and collimator aperture sizes of 0.5 mm, 1.0 mm, and 2.0 mm. Each collimator configuration was analyzed using a single source dwell position and indexing using step sizes of 0.5 mm, 1.0 mm, and 2.5 mm. Depth dose profiles were measured using Gafchromic film. Measurements were performed with the film positioned mid-plane directly adjacent to the Xoft catheter. The film and source were surrounded with 3 cm water-equivalent material to provide a 4p measurement geometry. Depth dose data for each of the trials was normalized to generate relative depth dose curves. Results: Depth dose is increased using a multi-step plan and with decreasing step size. At 1 cm, the depth dose is 20% for the single dwell position, whereas it remains at 60% using a 0.5 mm micro-stepped plan. Depth dose is increased by increasing aperture size. At 1 cm, dose falls to 44% with no shielding (no collimation), to 47% with a 0.5 cm aperture, to 56% with a 1 cm aperture, and to 63% with a 2 mm aperture. Conclusions: Using collimation and microindexing, the depth-dose characteristics of the Xoft 50 kV micro-miniature source can be
Abstracts / Brachytherapy 7 (2008) 91e194
Purpose: Previous studies of external beam radiotherapy for breast cancer have shown that the risk of chest wall toxicity increases with larger fraction sizes. A recent experience using the MammoSiteÒ (MS) reported a 12.5% incidence of chronic chest wall pain. Reports from several institutions using the multi-catheter (MC) technique have not shown elevated chest wall toxicity. Additionally, recent investigations have suggested that increased toxicity may occur with the MS when the dose to the skin, and perhaps chest wall, exceeds 125% of the prescribed dose. This investigation compares the skin and chest wall doses of a cohort of patients treated with the MC technique to a group treated with the MS. Methods and Materials: The dosimetric data for 43 patients treated with the MC technique and 83 patients treated with the MS at Virginia Commonwealth University (VCU) were reviewed. This cohort represents consecutively treated patients from our most recent experience to minimize the effect of any learning curve on dosimetry. Plans were generated using 3D software (Brachyvision, Varian Medical Systems, Inc., Palo Alto, CA). Multiple dwell positions were used for all MS patients to optimize dose delivery. The minimum distances from the planning target volume to the skin and chest wall were calculated, as well as the maximum doses delivered to the skin and chest wall. Results: The mean skin distances for patients treated with the MC technique and MS were 0.5 and 1.3 cm, respectively (p ! 0.001). Despite the significantly smaller mean skin distance, the mean skin dose for the MC technique was only 2.3 Gy per fraction (67% of prescription). The mean skin dose for the MS was 3.2 Gy per fraction (94% of prescription, p ! 0.001). The mean chest wall distance was 0.9 cm for the MC technique and 1.2 cm for the MS (p 5 0.035). Again, the mean chest wall dose for the MC technique was only 2.3 Gy per fraction (67% of prescription). The mean skin dose for the MS was 3.6 Gy per fraction (105% of prescription, p ! 0.001). The percentage of patients receiving skin doses in excess of 125% for the MC and MS were 0% and 9.6%, respectively. The percentage of patients receiving chest wall doses in excess of 125% for the MC and MS were 0% and 38.6%, respectively. Conclusions: The MC technique results in more conformal dose delivery, with significantly lower mean skin and chest wall doses. Treatment with the MS was associated with significantly more patients receiving doses to the skin or chest wall in excess of 125% of the prescription. Given the limited followup available for the MS, and the significant dose delivered to the skin and chest wall, the use of this device may be associated with a higher incidence of late chest wall toxicity than previously expected. Dr. Cuttino is a consultant for Hologic, Inc. (formerly Cytyc).
OR40 Presentation Time: 10:50 AM Consequence of varying target coverage and dose heterogeneity in accelerated partial breast irradiation brachytherapy modalities on tumor control probability Michael C. Kirk, Ph.D., Alan B. Coon, M.D., Ph.D., Lan Jiang, Ph.D., Damian Bernard, Ph.D., James C.H. Chu, Ph.D. Radiation Oncology, Rush University Medical Center, Chicago, IL. Purpose: The NSABP-B39/RTOG-0413 protocol is randomizing patients to conventional whole breast irradiation versus one of three accelerated partial breast irradiation (APBI) modalities: Mammosite brachytherapy (MB), multi-catheter brachytherapy (MCT) and 3D conformal partial breast treatment. The two brachytherapy based APBI techniques typically result in widely varying dose heterogeneity in the target (with dose heterogeneity typically larger in MB). We studied the effect of this variability in heterogeneity and minimum target dose coverage on the predicted tumor control probability (TCP) based on the linear-quadratic (LQ) model. Methods and Materials: Values for the minimum target coverage-V90 (percent volume receiving 90% of prescription dose), V100 (percent volume receiving 100% of prescription dose)-as well as dose heterogeneity-V150 (percent volume receiving 150% of prescription dose)-were chosen to encompass the range found in typical MB and MCT treatments. For each level of target coverage, V100 5 100%, V100 5
95%, V90 5 95% and V90 5 90%, the value of V150 was allowed to vary between 10 and 40%. The dose volume histograms (DVHs) were generated using a smoothing spline through these points so that the resulting curves exhibited typical features of brachytherapy DVHs. The curve representing the cumulative DVH was converted to a differential DVH using in-house Interactive Data Language (IDL) software. The TCP was calculated for each DVH using the BIOPLAN software package. The LQ parameters a 5 0.3 Gy-1 and a/b 5 10 Gy were used for the calculations. Results: For a target coverage of V100 5 100%, the TCP was 95.5%, 95.8%, 96.1%, 96.6% for V150 values of 10, 20, 30, 40 respectively. For a target coverage of V100 5 95%, the TCP was 88.5%, 88.8%, 89.2%, and 89.6% for V150 values of 10, 20, 30, 40 respectively. For a target coverage of V90 5 95%, the TCP was 84.3%, 84.5%, 84.9%, and 85.3% for V150 values of 10, 20, 30, 40 respectively. For a target coverage of V90 5 90, the TCP was 79.1%, 79.5%, 79.8%, and 80.1% for V150 values of 10, 20, 30, 40 respectively. The mean variation in TCP over all target coverages for any V150 values greater than 16.4%. On the other hand, for a specific minimum target coverage, the mean variation in calculated TCP over all V150 values was less than 1.2%. Conclusions: Large variations in V150 result in less than 1.2% differences in TCP for the same V100 or V90 target coverage. Thus the two APBI brachytherapy modalities can be considered equivalent despite the greater dose heterogeneity found in MB. TCP, however, could vary by more than 16% between V100 5 100% and V90 5 90%. We recommend that these data be considered while analyzing results from the protocol.
PHYSICS ORAL PRESENTATION SESSION 1 (electronic brachytherapy) Monday May 5, 2008 10:00 AMe11:00 AM OR41 Presentation Time: 10:00 AM Depth dose modulation (ddm) for electronic brachytherapy Jessica R. Hiatt, M.S.1,2 Jaroslaw T. Hepel, M.D.1,2 Gene A. Cardarelli, Ph.D.1,2 Mark Carol, M.D., Edward S. Sternick, Ph.D.1,2 David E. Wazer, M.D.1,2 1Radiation Oncology, Rhode Island Hospital, Providence, RI; 2 Radiation Oncology, Tufts-New England Medical Center, Boston, MA. Purpose: To design a system for modulation of depth dose characteristics for the Xoft (Xoft, Inc., Fremont, CA) micro-miniature electronic x-ray source. Utilizing source collimation and indexing, customized dose distributions can be created for a variety of brachytherapy applications. Methods and Materials: A custom collimator was designed for the Xoft (Xoft Inc., Fremont, CA) 50 kV x-ray source. The collimator consists of a cylindrical band for the lower portion of the source and a cap for the source tip and allows for collimation of the source to a circumferential fan beam. The degree of collimation can be adjusted by varying the width of the aperture. Depth-dose experiments were performed evaluating the effect of collimation and indexing. Collimation was varied using no collimation (unshielded source) and collimator aperture sizes of 0.5 mm, 1.0 mm, and 2.0 mm. Each collimator configuration was analyzed using a single source dwell position and indexing using step sizes of 0.5 mm, 1.0 mm, and 2.5 mm. Depth dose profiles were measured using Gafchromic film. Measurements were performed with the film positioned mid-plane directly adjacent to the Xoft catheter. The film and source were surrounded with 3 cm water-equivalent material to provide a 4p measurement geometry. Depth dose data for each of the trials was normalized to generate relative depth dose curves. Results: Depth dose is increased using a multi-step plan and with decreasing step size. At 1 cm, the depth dose is 20% for the single dwell position, whereas it remains at 60% using a 0.5 mm micro-stepped plan. Depth dose is increased by increasing aperture size. At 1 cm, dose falls to 44% with no shielding (no collimation), to 47% with a 0.5 cm aperture, to 56% with a 1 cm aperture, and to 63% with a 2 mm aperture. Conclusions: Using collimation and microindexing, the depth-dose characteristics of the Xoft 50 kV micro-miniature source can be