- Email: [email protected]
2035
S228
I. J. Radiation Oncology
● Biology ● Physics
Volume 66, Number 3, Supplement, 2006
Results: Coronal tumor bed area was verified to correlate strongly with tumor bed volume, correlation coefficient 0.818 (p⬍0.001). The mean distance between the geometric centers of the seroma and clip defined abnormalities was 6.1 mm. On average, the seroma extends beyond the clips in medial/lateral/superior/inferior directions by 3 mm. Tumor bed areas defined by the two methods correlate strongly (correlation coefficient 0.758, p⬍0.001). The tumor bed area defined by the seroma was significantly larger than the tumor bed area defined by clips, with a mean difference of 9.9 cm2 (p⫽0.003). Conclusions: Tumor bed definition post lumpectomy, as defined by surgical clips and seroma formation, varies significantly. Radiation oncologists should incorporate the seroma and surgical clips in defining the tumor bed. Future studies will be required to define volumetric changes over time. Author Disclosure: Z. Gabos, None; S. Chafe, None; J. Hanson, None.
2034
Pilot Study of a Real-Time Stereovision-Based Image-Guided Radiotherapy for Breast Cancer
S. Li, E. Walker, D. Liu, S. Andrews, K. Aldridge, J. Kim, C. Fraser, J. Dragovic, I. Aref, B. Movsas Henry Ford Health System, Detroit, MI Purpose/Objective(s): To validate a real-time stereovision-based IGRT for breast cancer (BC) patients undergoing IMRT. Materials/Methods: Ten BC patients were accrued over the last six months on an IRB-approved study. The PTV defined by the physician includes the ipsilateral breast tissue with a skin sparing margin of 5 mm that was encompassed by the prescribed isodose in an IMRT plan and the surgical bed was boosted using an electron field. Our in-house IGRT program automatically loaded the approved plan (included CT image and structures as well as beam fluence) and created a reference surface with critical structures and beams attached. After marker-based daily patient setup, a 3D surface image was snapshot captured using a 3D camera mounted on the ceiling of the LINAC vault. The 3D surface image was semi-automatically matched with the reference image and the table-beam corrections along with the PTV coverage and the lung intersection shown on the screen. For this pilot study, only 3D isocenter shifts were implemented. A second surface image was captured to verify the correction. During the tangential field irradiation, two additional images were captured to monitor the intra-fraction impact of thoracic respiration or patient motion. All results were automatically stored in the IGRT computer for individual patients. X-ray images were taken weekly and at any shift ⬎10 mm. Results: Table 1 shows that all patients excepting for the one under treatment have at least 21 days of images. The 3D isocenter setup errors have been improved from 6.3 ⫾ 6.2 mm to 1.7 ⫾ 3.3 mm after the application of IGRT. Isocenter shifts ⬎ 10 mm caused by either large breast (pts 6 and 7) or setup-marking error (pt 10) were detected and corrected. All major changes were confirmed with portal images. Less change for the isocenter shift on the first pt and for the rotational errors on all pts agrees with no correction at the early trial. Importantly, our data shows that respiration only causes 0.8 ⫾ 2.0 mm isocenter shift and 0.3o ⫾ 1.1o orientation changes indicating achievable accuracy with an IGRT of BC. The system works equally well for White and African-American patients and the entire procedure (including the table shifts) takes ⬍ 5 minutes per day. Conclusions: A real-time stereovision-based IGRT system has been successfully implemented at our institution. The clinical results demonstrate significant improvement for breast target positioning. The test is completely performed by the treatment therapists. Thus, this IGRT for BC is clinically feasible, efficient, accurate, precise, and cost-efficient, and involves no additional radiation exposure.
Author Disclosure: S. Li, inventor, E. Ownership Interest; E. Walker, None; D. Liu, co-inventor, E. Ownership Interest; S. Andrews, None; K. Aldridge, None; J. Kim, None; C. Fraser, None; J. Dragovic, None; I. Aref, None; B. Movsas, None.
2035
Association of Women’s Local Treatment Decisions With Local Recurrence Risk and Life Expectancy 1
M. Akbari , S. A. Russo2, J. S. Jacobson3, F. R. Schnabel4, M. B. El-Tamer4, K. A. Joseph4, L. A. Klein4, C. Chen5, V. R. Grann6 1 Columbia University Medical Center, New York, NY, 2Columbia University Medical Center, Department of Radiation Oncology, New York, NY, 3Columbia University Medical Center, Department of Epidemiology, New York, NY, 4Columbia University Medical Center, Department of Surgery, New York, NY, 5Columbia University Medical Center, Department of Biostatistics, New York, NY, 6Columbia University Medical Center, Department of Medicine and Epidemiology, New York, NY Purpose/Objective(s): Women diagnosed with early stage breast cancer (BC) have several local treatment options including mastectomy (M), conservative surgery (CS) ⫹ whole breast radiation therapy (WRT), CS ⫹ accelerated partial breast irradiation (APBI), or randomization in the RTOG 0413 trial. Depending on their age at diagnosis and receptor status, they may be offered oral adjuvant systemic treatments, such as tamoxifen (T), with or without local breast irradiation. The purpose of this study was to evaluate how the effects of age (A), race/ethnicity (RE), education (E), income (I), risks of local recurrence (LR), and life expectancy influence a women’s local treatment preferences.
Proceedings of the 48th Annual ASTRO Meeting
S229
Materials/Methods: Between 6/05 and 9/05, we asked 161 women (106 without a diagnosis of BC and 55 with a diagnosis of BC) to participate in an IRB-approved questionnaire. Women answering the survey were between the ages of 25 and 65. The questionnaire presented women with clinical vignettes that included information about age at diagnosis (50 versus 70), local treatment choices (M, CS ⫹/- {WRT versus APBI} ⫹/- T, participation in RTOG 0413 trial) and risks of LR. It then asked women to choose among these treatment options and to rate them, using a time trade-off method given different local treatments and LR risks. Results: The distribution of all respondents by A, RE, E and I is shown in the table below. In the total population of women surveyed, A and RE were associated with a women’s choice of CS⫹WRT⫹T to decrease her risk of LR (p⬍ 0.009 and p⬍ 0.056). 28.5% of all women surveyed answered they would be willing to participate in the RTOG 0413 trial; less E and lower I were associated with the decision to participate (p⬍0.009 and p⬍0.03). When asked to choose M, CS⫹WRT or CS⫹APBI given related risks of LR, BC patients most often chose CS⫹WRT (64%) with only 9% of women choosing CS⫹APBI; RE and E influenced this choice (p ⬍ 0.04 and p ⬍ 0.05). Stratified analysis showed an association between CS⫹WRT⫹T and E (p⬍0.006) in women with BC. Women with BC were less likely than controls to trade time to receive CS⫹WRT instead of M and were also less likely to give up years of life to receive CS⫹APBI instead of CS⫹WRT. Conclusions: When offered local treatment choices and information on their risks of LR, women chose local therapies that provided the best chance of local control. Local treatment choices were influenced by A, RE, E and I. Women diagnosed with BC were less likely to trade years of life for less aggressive local treatments. Table 1: Age, race/ethnicity, education, income of respondents
Age (A) 25-34 35-45 46-55 56-65
N (%) 43 40 41 37
(27) (25) (25) (23)
Race/Ethnicity (RE) Asian Black Hispanic White Other
N (%) 11 21 47 75 7
(7) (13) (29) (47) (4)
Education (E) High School (HS) or below Above HS, some college Bachelors degree Master’s or higher
N (%) 22 38 50 51
(14) (23) (31) (32)
Income (I) Below $30,000 $30,000-$75,000 Above $75,000 Unknown
N (%) 24 65 59 13
(15) (40) (37) (8)
Author Disclosure: M. Akbari, None; S.A. Russo, None; J.S. Jacobson, None; F.R. Schnabel, None; M.B. El-Tamer, None; K.A. Joseph, None; L.A. Klein, None; C. Chen, None; V.R. Grann, None.
2036
APBI Using Respiratory Gated IMRT - Review of First 24 Patients
A. Lewin Baptist Hospital of Miami, Miami, FL Purpose/Objective(s): Intensity Modulated Radiation Therapy (IMRT) with respiratory gating is used to deliver Accelerated Partial Breast Irradiation (APBI) in selected patients with early breast cancer following breast conserving surgery. APBI can be delivered using conformal external beam or brachytherapy, such as MammositeTM or interstitial implant. IMRT has not been widely used due to the intra-fraction breast motion associated with the respiratory cycle. However, respiratory gating provides a useful tool to minimize the impact of breast motion relative to the MLC segments. IMRT enables delivery of highly conformal dose to the surgical cavity. Materials/Methods: 24 postmenopausal women with Stage I and II breast cancer elected to receive APBI using IMRT with respiratory gating following lumpectomy. Patients were enrolled in an IRB approved protocol to evaluate tumor bed irradiation after lumpectomy. 13 out of 24 patients had right-sided breast tumors while 11 had tumors located in the left breast. Patients were treated in the supine position, immobilized on a breast board with both arms up. Respiratory gating was used at the time of simulation and treatment. Only exhalation part of the cycle was used. This reduced the breast excursion in the AP direction from 8 –10 mm to about 2 mm. Verification CT was repeated in 8 out of the 24 patients towards the end of the course of radiation to evaluate reproducibility. EPID was used for set up verification before each treatment delivery. In all patients the lumpectomy cavity was clearly defined on CT scan. 12 patients had surgical clips outlining the surgical cavity. The CTV consisted of the lumpectomy cavity plus a 10 –15 mm margin. No further margin was added to define the PTV. The CTV was treated BID, 380 cGy per fraction for 5 days to a total dose of 3800 cGy. Results: The median age of the patients was 70. The median CTV treated is 58 cm3. In this group of patients, the CTV received at least 95% of the prescribed dose. Dose to the skin and chest wall musculature was limited to 30 Gy. The mean dose to the CTV was 3885 cGy. Twelve out of 24 patients received dose greater than 114 cGy (3%) to the contra lateral breast while dose to the heart was greater than 190 cGy (5%) for 3 out of 24 patients. Of these three patients, two had left breast tumors while one patient had a right breast tumor. Other than the contra lateral breast and heart, results show that IMRT-based APBI objectives were achieved when compared with the currently accruing NSABP/RTOG study, while dose conformity to the CTV was excellent. Acute toxicity was limited to Grade I erythema (13 out of 24 patients), Grade I myositis (5 out of 24 patients) and Grade I hyper pigmentation (8 out of 24 patients). With a median follow up of 13 months (range 3–26 months), only 1 patient developed late toxicity (grade II telangentasia) and no patient developed local recurrence. No skin changes greater than Grade I erythema were noted during treatment. Cosmetic results were excellent in 22 patients, good in 2 patients and fair in one patient. Conclusions: 1. Results are similar to previous published techniques to deliver APBI. 2. Respiratory gating and EPID provide a reproducible patient set up for APBI. 3. The radiation dose to the rest of the ipsilateral breast is less when compared to 3D conformal treatment and more homogenous than the other APBI techniques. 4. With short term follow up this approach is well tolerated, with excellent cosmetic results in the majority of patients. Author Disclosure: A. Lewin, None.
I. J. Radiation Oncology
● Biology ● Physics
Volume 66, Number 3, Supplement, 2006
Results: Coronal tumor bed area was verified to correlate strongly with tumor bed volume, correlation coefficient 0.818 (p⬍0.001). The mean distance between the geometric centers of the seroma and clip defined abnormalities was 6.1 mm. On average, the seroma extends beyond the clips in medial/lateral/superior/inferior directions by 3 mm. Tumor bed areas defined by the two methods correlate strongly (correlation coefficient 0.758, p⬍0.001). The tumor bed area defined by the seroma was significantly larger than the tumor bed area defined by clips, with a mean difference of 9.9 cm2 (p⫽0.003). Conclusions: Tumor bed definition post lumpectomy, as defined by surgical clips and seroma formation, varies significantly. Radiation oncologists should incorporate the seroma and surgical clips in defining the tumor bed. Future studies will be required to define volumetric changes over time. Author Disclosure: Z. Gabos, None; S. Chafe, None; J. Hanson, None.
2034
Pilot Study of a Real-Time Stereovision-Based Image-Guided Radiotherapy for Breast Cancer
S. Li, E. Walker, D. Liu, S. Andrews, K. Aldridge, J. Kim, C. Fraser, J. Dragovic, I. Aref, B. Movsas Henry Ford Health System, Detroit, MI Purpose/Objective(s): To validate a real-time stereovision-based IGRT for breast cancer (BC) patients undergoing IMRT. Materials/Methods: Ten BC patients were accrued over the last six months on an IRB-approved study. The PTV defined by the physician includes the ipsilateral breast tissue with a skin sparing margin of 5 mm that was encompassed by the prescribed isodose in an IMRT plan and the surgical bed was boosted using an electron field. Our in-house IGRT program automatically loaded the approved plan (included CT image and structures as well as beam fluence) and created a reference surface with critical structures and beams attached. After marker-based daily patient setup, a 3D surface image was snapshot captured using a 3D camera mounted on the ceiling of the LINAC vault. The 3D surface image was semi-automatically matched with the reference image and the table-beam corrections along with the PTV coverage and the lung intersection shown on the screen. For this pilot study, only 3D isocenter shifts were implemented. A second surface image was captured to verify the correction. During the tangential field irradiation, two additional images were captured to monitor the intra-fraction impact of thoracic respiration or patient motion. All results were automatically stored in the IGRT computer for individual patients. X-ray images were taken weekly and at any shift ⬎10 mm. Results: Table 1 shows that all patients excepting for the one under treatment have at least 21 days of images. The 3D isocenter setup errors have been improved from 6.3 ⫾ 6.2 mm to 1.7 ⫾ 3.3 mm after the application of IGRT. Isocenter shifts ⬎ 10 mm caused by either large breast (pts 6 and 7) or setup-marking error (pt 10) were detected and corrected. All major changes were confirmed with portal images. Less change for the isocenter shift on the first pt and for the rotational errors on all pts agrees with no correction at the early trial. Importantly, our data shows that respiration only causes 0.8 ⫾ 2.0 mm isocenter shift and 0.3o ⫾ 1.1o orientation changes indicating achievable accuracy with an IGRT of BC. The system works equally well for White and African-American patients and the entire procedure (including the table shifts) takes ⬍ 5 minutes per day. Conclusions: A real-time stereovision-based IGRT system has been successfully implemented at our institution. The clinical results demonstrate significant improvement for breast target positioning. The test is completely performed by the treatment therapists. Thus, this IGRT for BC is clinically feasible, efficient, accurate, precise, and cost-efficient, and involves no additional radiation exposure.
Author Disclosure: S. Li, inventor, E. Ownership Interest; E. Walker, None; D. Liu, co-inventor, E. Ownership Interest; S. Andrews, None; K. Aldridge, None; J. Kim, None; C. Fraser, None; J. Dragovic, None; I. Aref, None; B. Movsas, None.
2035
Association of Women’s Local Treatment Decisions With Local Recurrence Risk and Life Expectancy 1
M. Akbari , S. A. Russo2, J. S. Jacobson3, F. R. Schnabel4, M. B. El-Tamer4, K. A. Joseph4, L. A. Klein4, C. Chen5, V. R. Grann6 1 Columbia University Medical Center, New York, NY, 2Columbia University Medical Center, Department of Radiation Oncology, New York, NY, 3Columbia University Medical Center, Department of Epidemiology, New York, NY, 4Columbia University Medical Center, Department of Surgery, New York, NY, 5Columbia University Medical Center, Department of Biostatistics, New York, NY, 6Columbia University Medical Center, Department of Medicine and Epidemiology, New York, NY Purpose/Objective(s): Women diagnosed with early stage breast cancer (BC) have several local treatment options including mastectomy (M), conservative surgery (CS) ⫹ whole breast radiation therapy (WRT), CS ⫹ accelerated partial breast irradiation (APBI), or randomization in the RTOG 0413 trial. Depending on their age at diagnosis and receptor status, they may be offered oral adjuvant systemic treatments, such as tamoxifen (T), with or without local breast irradiation. The purpose of this study was to evaluate how the effects of age (A), race/ethnicity (RE), education (E), income (I), risks of local recurrence (LR), and life expectancy influence a women’s local treatment preferences.
Proceedings of the 48th Annual ASTRO Meeting
S229
Materials/Methods: Between 6/05 and 9/05, we asked 161 women (106 without a diagnosis of BC and 55 with a diagnosis of BC) to participate in an IRB-approved questionnaire. Women answering the survey were between the ages of 25 and 65. The questionnaire presented women with clinical vignettes that included information about age at diagnosis (50 versus 70), local treatment choices (M, CS ⫹/- {WRT versus APBI} ⫹/- T, participation in RTOG 0413 trial) and risks of LR. It then asked women to choose among these treatment options and to rate them, using a time trade-off method given different local treatments and LR risks. Results: The distribution of all respondents by A, RE, E and I is shown in the table below. In the total population of women surveyed, A and RE were associated with a women’s choice of CS⫹WRT⫹T to decrease her risk of LR (p⬍ 0.009 and p⬍ 0.056). 28.5% of all women surveyed answered they would be willing to participate in the RTOG 0413 trial; less E and lower I were associated with the decision to participate (p⬍0.009 and p⬍0.03). When asked to choose M, CS⫹WRT or CS⫹APBI given related risks of LR, BC patients most often chose CS⫹WRT (64%) with only 9% of women choosing CS⫹APBI; RE and E influenced this choice (p ⬍ 0.04 and p ⬍ 0.05). Stratified analysis showed an association between CS⫹WRT⫹T and E (p⬍0.006) in women with BC. Women with BC were less likely than controls to trade time to receive CS⫹WRT instead of M and were also less likely to give up years of life to receive CS⫹APBI instead of CS⫹WRT. Conclusions: When offered local treatment choices and information on their risks of LR, women chose local therapies that provided the best chance of local control. Local treatment choices were influenced by A, RE, E and I. Women diagnosed with BC were less likely to trade years of life for less aggressive local treatments. Table 1: Age, race/ethnicity, education, income of respondents
Age (A) 25-34 35-45 46-55 56-65
N (%) 43 40 41 37
(27) (25) (25) (23)
Race/Ethnicity (RE) Asian Black Hispanic White Other
N (%) 11 21 47 75 7
(7) (13) (29) (47) (4)
Education (E) High School (HS) or below Above HS, some college Bachelors degree Master’s or higher
N (%) 22 38 50 51
(14) (23) (31) (32)
Income (I) Below $30,000 $30,000-$75,000 Above $75,000 Unknown
N (%) 24 65 59 13
(15) (40) (37) (8)
Author Disclosure: M. Akbari, None; S.A. Russo, None; J.S. Jacobson, None; F.R. Schnabel, None; M.B. El-Tamer, None; K.A. Joseph, None; L.A. Klein, None; C. Chen, None; V.R. Grann, None.
2036
APBI Using Respiratory Gated IMRT - Review of First 24 Patients
A. Lewin Baptist Hospital of Miami, Miami, FL Purpose/Objective(s): Intensity Modulated Radiation Therapy (IMRT) with respiratory gating is used to deliver Accelerated Partial Breast Irradiation (APBI) in selected patients with early breast cancer following breast conserving surgery. APBI can be delivered using conformal external beam or brachytherapy, such as MammositeTM or interstitial implant. IMRT has not been widely used due to the intra-fraction breast motion associated with the respiratory cycle. However, respiratory gating provides a useful tool to minimize the impact of breast motion relative to the MLC segments. IMRT enables delivery of highly conformal dose to the surgical cavity. Materials/Methods: 24 postmenopausal women with Stage I and II breast cancer elected to receive APBI using IMRT with respiratory gating following lumpectomy. Patients were enrolled in an IRB approved protocol to evaluate tumor bed irradiation after lumpectomy. 13 out of 24 patients had right-sided breast tumors while 11 had tumors located in the left breast. Patients were treated in the supine position, immobilized on a breast board with both arms up. Respiratory gating was used at the time of simulation and treatment. Only exhalation part of the cycle was used. This reduced the breast excursion in the AP direction from 8 –10 mm to about 2 mm. Verification CT was repeated in 8 out of the 24 patients towards the end of the course of radiation to evaluate reproducibility. EPID was used for set up verification before each treatment delivery. In all patients the lumpectomy cavity was clearly defined on CT scan. 12 patients had surgical clips outlining the surgical cavity. The CTV consisted of the lumpectomy cavity plus a 10 –15 mm margin. No further margin was added to define the PTV. The CTV was treated BID, 380 cGy per fraction for 5 days to a total dose of 3800 cGy. Results: The median age of the patients was 70. The median CTV treated is 58 cm3. In this group of patients, the CTV received at least 95% of the prescribed dose. Dose to the skin and chest wall musculature was limited to 30 Gy. The mean dose to the CTV was 3885 cGy. Twelve out of 24 patients received dose greater than 114 cGy (3%) to the contra lateral breast while dose to the heart was greater than 190 cGy (5%) for 3 out of 24 patients. Of these three patients, two had left breast tumors while one patient had a right breast tumor. Other than the contra lateral breast and heart, results show that IMRT-based APBI objectives were achieved when compared with the currently accruing NSABP/RTOG study, while dose conformity to the CTV was excellent. Acute toxicity was limited to Grade I erythema (13 out of 24 patients), Grade I myositis (5 out of 24 patients) and Grade I hyper pigmentation (8 out of 24 patients). With a median follow up of 13 months (range 3–26 months), only 1 patient developed late toxicity (grade II telangentasia) and no patient developed local recurrence. No skin changes greater than Grade I erythema were noted during treatment. Cosmetic results were excellent in 22 patients, good in 2 patients and fair in one patient. Conclusions: 1. Results are similar to previous published techniques to deliver APBI. 2. Respiratory gating and EPID provide a reproducible patient set up for APBI. 3. The radiation dose to the rest of the ipsilateral breast is less when compared to 3D conformal treatment and more homogenous than the other APBI techniques. 4. With short term follow up this approach is well tolerated, with excellent cosmetic results in the majority of patients. Author Disclosure: A. Lewin, None.