- Email: [email protected]
Abdominal Organ Motion During Conformal Radiation
Proceedings of the 47th Annual ASTRO Meeting
Conclusions: Overall, patients with gastric cancer treated with radiotherapy showed improved 3 years overall survival, but the magnitude of this benefit was relatively small. In post-operative setting, radiochemotherapy seem to obtain better results respect to radiotherapy alone and criteria are needed to identify patients most likely to benefit from this form of adjuvant treatment. In patients with resectable gastric cancer, preoperative radiotherapy significantly reduced 3 and 5 year mortality rate. A large scale multicentre RCT may prove useful to substantiate the benefit of preoperative radiotherapy on overall survival.
2086
Long Term Follow-Up of Patients of Intrahepatic Malignancies Treated with Iodine-125 Brachytherapy
S. Nag,1 M. DeHaan,1 G.R. Scruggs,1 N. Mayr,1 E.W. Martin2 Radiation Medicine, Ohio State University, Columbus, OH, 2Division of Surgical Oncology, Ohio State University, Columbus, OH 1
Purpose/Objective: The prognosis of patients with unresectable primary and metastatic liver tumors is extremely poor. We investigated the role of intraoperative Iodine-125 (I-125) brachytherapy as a treatment option. Materials/Methods: Between 1989 and 2002, 64 patients with unresectable or residual disease after surgical resection for intrahepatic malignancies (metastatic colorectal cancer: 54, cholangiocarcinoma: 4, and metastatic non-colorectal cancer: 2) underwent permanent with I-125 brachytherapy to deliver 160 Gy (pre-TG 43) to the periphery of the target volume. A volume implant with Mick applicator was used for gross tumors while I-125 impregnated gel foam surface implant was used for microscopic tumor beds. Liver recurrence (defined as a measurable increase of implanted lesions or appearance of new lesions in the liver as determined by CT scan or laparotomy) and survival rates as a function of clinical and treatment variables were examined retrospectively. Results: The median length of follow-up was 13.2 years. There was microscopic residual disease in 59% patients and gross residual disease was present in 41% patients. The overall 1, 3, and 5 -year actuarial intrahepatic local control rates were 44%, 22%, and 22%, respectively with a median time to liver recurrence of 9 months (95% CI 6 –12). The 5-year actuarial intrahepatic control was higher for patients with solitary metastasis (38%) than for those with multiple metastases (6%, p⫽0.04). The 1, 3, and 5-year actuarial overall survival rates were 73%, 23%, and 5%, respectively (median: 20 months; 95% CI: 16 –24; longest survival 7.5 years). Overall survival was higher for patients with smaller-volume implants (p⫽0.003) and for patients without a history of prior liver resection (p⫽0.002). There was no mortality. Radiation-related complications were minimal. Conclusions: Permanent I-125 brachytherapy is a safe and effective adjuvant treatment for unresectable intrahepatic malignancies. It is a simple technique with morbidity and mortality rates comparable to liver resection alone. Patients considered good candidates for I-125 brachytherapy include those with small volume implants, those without prior liver resection, and those with solitary liver metastases. In these select patients for whom curative surgical resection is not an option, I-125 brachytherapy is an important alternative to other locally ablative techniques and can provide long-term local control and increased survival.
2087
Abdominal Organ Motion During Conformal Radiation 1,3
Z. Kassam, J. Brierley,1,3 L. Dawson,1,3 G. Lockwood,2 G. Perkins,1 K. Churcher,1 J. Ringash1,3 Department of Radiation Medicine, The Princess Margaret Hospital, Toronto, ON, Canada, 2Department of Biostatistics, The Princess Margaret Hospital, Toronto, ON, Canada, 3The University of Toronto, Toronto, ON, Canada 1
Purpose/Objective: Radiation has recently been established as an integral modality in the adjuvant therapy of R0 resected adenocarcinoma of the stomach/GE junction (GC). There are no published studies assessing motion of the residual stomach, anastomosis and nodal groups or effects of respiration on target motion. Measurement of translation and deformation due to variable filling of hollow organs, as well as breathing motion, is necessary for accurate delivery of conformal radiotherapy and intensity modulated radiation therapy (IMRT). The current arbitrary PTV margin (1cm) could result in excess normal tissue toxicity or insufficient target coverage. The aims of this study are to determine interfraction motion of abdominal organs during adjuvant 3D conformal radiation for GC, quantify displacement due to respiratory motion, and assess the stability of these organs in voluntary breath hold. Materials/Methods: 20 patients receiving adjuvant chemoradiation for GC will undergo CT scans in free breath (FB), normal voluntary breathhold inhale (I) and exhale (E) immobilized with BodyFix, in week 1, 3 and 5 of therapy. A standard light meal is taken by patients not less that an hour prior to the scans. The serial CT scans are imported into Pinnacle v6.2b and registered to the freebreath planning scan (CT0) using automated bone registration. Volumes of interest (VOIs) are contoured on each scan (stomach, pancreas, kidneys, porta hepatis, celiac axis), by the same physician. Points of interest (POIs) are defined as the centre of mass for each VOI, dome of diaphragm and splenic hilum. Motion during treatment is assessed by measuring differences in position of POIs and in volumes of VOIs in FB scans compared to CT0, in right-left (RL), craniocaudal (CC) and posteroanterior (PA) directions. POIs and VOIs are compared in FB, I and E scans, using week 1 scans as baseline, to determine stability of breath hold. Respiratory movement is determined from I/E scans. Results: 47 CTs from 5 patients have been analyzed. Median absolute displacement for all organs was determined for all scans (see table). Greatest magnitude of displacement in freebreath scans was seen in the CC direction for the kidneys, stomach and splenic hilum, where individual translations were as great as 20 to 50 mm. Breathing motion, determined from I and E scans, showed greatest absolute displacement in CC direction for the diaphragm (median 15mm, range 0 to 50mm) and the porta hepatis (median 12mm, range 2 to 28mm). Displacements were similar when comparing FB, I and E scans, except for greater displacement seen in the CC direction on I scans. Median absolute change in volume (%) compared to week 1 scans in FB/I/E was: stomach 37/ 24.7/ 19.05, pancreas 33.2/ 19.8/ 21.4, kidneys 6.2/ 4/ 5.39. Conclusions: Displacement of the abdominal organs can be substantial, with offsets up to 5cm. This may alter the dose to targets and organs at risk and indicates the potential need for individualised and/or adaptive planning.
S281
S282
I. J. Radiation Oncology
● Biology ● Physics
Volume 63, Number 2, Supplement, 2005
Median displacement over all organs in mm (range of medians)
2088
A Prospective Comparison Study of Liver Tumour Target Definition Based on Triphasic CT and Gadolinium MR
J.J. Voroney,1 K. Brock,3 C. Eccles,3 M. Haider,2 L. Dawson1 Radiation Oncology, Princess Margaret Hospital, Toronto, ON, Canada, 2Radiology, Princess Margaret Hospital, Toronto, ON, Canada, 3Radiation Medicine, Princess Margaret Hospital, Toronto, ON, Canada 1
Purpose/Objective: To quantitatively compare liver cancer gross tumour volume (GTV) defined using MR and triphasic CT. Materials/Methods: Diagnostic quality planning MR scans with and without gadolinium (6 – 8 mm slice thickness) and triphasic planning CT scans (2–5 mm slice thickness) were obtained in twenty-three patients with non-resectable cholangiocarcinoma (n⫽4), hepatoma (n⫽10) and liver metastases (n⫽9) enrolled on a protocol of hypofractionated highly conformal radiotherapy. The liver was immobilized in exhale treatment position, using the active breathing coordinator for CT and voluntary breath hold for MR. On the CT and MR series best demonstrating the tumor, the liver and GTV were contoured with tools from Pinnacle (v6.5) treatment planning software. Anatomical reference points within the liver were identified. Despite careful patient positioning, spatial changes in liver position occurred between CT and MR imaging. A deformable registration method based on finite element modeling accounting for biomechanical forces and resulting positional and volumetric changes was used to register the CT and MR livers and resolve geometric differences between them. Accuracy of registration of anatomical reference points was calculated, and spatial differences between the CT and MR GTVs were then quantified using the same deformable registration technique to register the CT and MR GTVs. Results: The contrast CT best visualizing the GTV was the arterial phase CT for hepatoma (7/11) and the venous phase CT for metastases (7/9) and cholangiocarcinoma (4/4); the MR sequence best visualizing the GTV was gadolinium enhanced T1 weighted MR (22/24) and dual echo MR (2/24). Additional tumour foci were seen with MR in 3 cases. Following deformable registration of the liver volumes to resolve geometric discrepancies of the liver, the average vector magnitude (VM) of the residual registration error based on the displacement of the registered anatomical references points was 4.2 mm (SD ⫽ 1.7 mm), less than the MR slice thickness. The average CT and MR GTVs was 289 cc (range: 9 –1467 cc) and 278 cc (range: 6 –1731 cc) respectively. The average difference in CT and MR GTV was 11 cc (range: 1–263 cc), with CT GTV larger then MR GTV in 3/7 metastases, 4/4 cholangiocarcinomas and 6/11 hepatomas. The average ratio of CT/MR GTV was 1.2 (range: 0.3– 4.2). Deformable registration of the MR GTV to the CT GTV revealed substantial spatial differences. The study population mean of the average distance between the CT and MR GTV surfaces was 2.8, 2.9 and 2.9 mm in the medial-lateral (ML), anterior-posterior (AP) and cranial-caudal (CC) directions, with a standard deviation of 2.0, 2.9 and 2.1 mm. The study population mean of the maximal distance between CT and MR GTV surfaces was 17.6, 17.9 and 16.6, with maximal distances between individual surfaces as high as 39.7, 63.8 and 42.4 in the ML, AP and CC directions. The average percentage of GTV surface area that differed by 3 mm or more, 5 mm or more and 15 mm or more was 56% (max 92%), 35% (max 86%) and 9% (max 56%). Conclusions: MR defined GTVs can be significantly different than CT defined GTVs and this should be considered for high precision liver cancer radiotherapy.
Average (Av) surface area (SA) difference (⌬) statistics between MR and CT defined GTVs by tumour type, hepatoma (H), cholangiocarcinoma (CC) and liver metastases (LM).
Conclusions: Overall, patients with gastric cancer treated with radiotherapy showed improved 3 years overall survival, but the magnitude of this benefit was relatively small. In post-operative setting, radiochemotherapy seem to obtain better results respect to radiotherapy alone and criteria are needed to identify patients most likely to benefit from this form of adjuvant treatment. In patients with resectable gastric cancer, preoperative radiotherapy significantly reduced 3 and 5 year mortality rate. A large scale multicentre RCT may prove useful to substantiate the benefit of preoperative radiotherapy on overall survival.
2086
Long Term Follow-Up of Patients of Intrahepatic Malignancies Treated with Iodine-125 Brachytherapy
S. Nag,1 M. DeHaan,1 G.R. Scruggs,1 N. Mayr,1 E.W. Martin2 Radiation Medicine, Ohio State University, Columbus, OH, 2Division of Surgical Oncology, Ohio State University, Columbus, OH 1
Purpose/Objective: The prognosis of patients with unresectable primary and metastatic liver tumors is extremely poor. We investigated the role of intraoperative Iodine-125 (I-125) brachytherapy as a treatment option. Materials/Methods: Between 1989 and 2002, 64 patients with unresectable or residual disease after surgical resection for intrahepatic malignancies (metastatic colorectal cancer: 54, cholangiocarcinoma: 4, and metastatic non-colorectal cancer: 2) underwent permanent with I-125 brachytherapy to deliver 160 Gy (pre-TG 43) to the periphery of the target volume. A volume implant with Mick applicator was used for gross tumors while I-125 impregnated gel foam surface implant was used for microscopic tumor beds. Liver recurrence (defined as a measurable increase of implanted lesions or appearance of new lesions in the liver as determined by CT scan or laparotomy) and survival rates as a function of clinical and treatment variables were examined retrospectively. Results: The median length of follow-up was 13.2 years. There was microscopic residual disease in 59% patients and gross residual disease was present in 41% patients. The overall 1, 3, and 5 -year actuarial intrahepatic local control rates were 44%, 22%, and 22%, respectively with a median time to liver recurrence of 9 months (95% CI 6 –12). The 5-year actuarial intrahepatic control was higher for patients with solitary metastasis (38%) than for those with multiple metastases (6%, p⫽0.04). The 1, 3, and 5-year actuarial overall survival rates were 73%, 23%, and 5%, respectively (median: 20 months; 95% CI: 16 –24; longest survival 7.5 years). Overall survival was higher for patients with smaller-volume implants (p⫽0.003) and for patients without a history of prior liver resection (p⫽0.002). There was no mortality. Radiation-related complications were minimal. Conclusions: Permanent I-125 brachytherapy is a safe and effective adjuvant treatment for unresectable intrahepatic malignancies. It is a simple technique with morbidity and mortality rates comparable to liver resection alone. Patients considered good candidates for I-125 brachytherapy include those with small volume implants, those without prior liver resection, and those with solitary liver metastases. In these select patients for whom curative surgical resection is not an option, I-125 brachytherapy is an important alternative to other locally ablative techniques and can provide long-term local control and increased survival.
2087
Abdominal Organ Motion During Conformal Radiation 1,3
Z. Kassam, J. Brierley,1,3 L. Dawson,1,3 G. Lockwood,2 G. Perkins,1 K. Churcher,1 J. Ringash1,3 Department of Radiation Medicine, The Princess Margaret Hospital, Toronto, ON, Canada, 2Department of Biostatistics, The Princess Margaret Hospital, Toronto, ON, Canada, 3The University of Toronto, Toronto, ON, Canada 1
Purpose/Objective: Radiation has recently been established as an integral modality in the adjuvant therapy of R0 resected adenocarcinoma of the stomach/GE junction (GC). There are no published studies assessing motion of the residual stomach, anastomosis and nodal groups or effects of respiration on target motion. Measurement of translation and deformation due to variable filling of hollow organs, as well as breathing motion, is necessary for accurate delivery of conformal radiotherapy and intensity modulated radiation therapy (IMRT). The current arbitrary PTV margin (1cm) could result in excess normal tissue toxicity or insufficient target coverage. The aims of this study are to determine interfraction motion of abdominal organs during adjuvant 3D conformal radiation for GC, quantify displacement due to respiratory motion, and assess the stability of these organs in voluntary breath hold. Materials/Methods: 20 patients receiving adjuvant chemoradiation for GC will undergo CT scans in free breath (FB), normal voluntary breathhold inhale (I) and exhale (E) immobilized with BodyFix, in week 1, 3 and 5 of therapy. A standard light meal is taken by patients not less that an hour prior to the scans. The serial CT scans are imported into Pinnacle v6.2b and registered to the freebreath planning scan (CT0) using automated bone registration. Volumes of interest (VOIs) are contoured on each scan (stomach, pancreas, kidneys, porta hepatis, celiac axis), by the same physician. Points of interest (POIs) are defined as the centre of mass for each VOI, dome of diaphragm and splenic hilum. Motion during treatment is assessed by measuring differences in position of POIs and in volumes of VOIs in FB scans compared to CT0, in right-left (RL), craniocaudal (CC) and posteroanterior (PA) directions. POIs and VOIs are compared in FB, I and E scans, using week 1 scans as baseline, to determine stability of breath hold. Respiratory movement is determined from I/E scans. Results: 47 CTs from 5 patients have been analyzed. Median absolute displacement for all organs was determined for all scans (see table). Greatest magnitude of displacement in freebreath scans was seen in the CC direction for the kidneys, stomach and splenic hilum, where individual translations were as great as 20 to 50 mm. Breathing motion, determined from I and E scans, showed greatest absolute displacement in CC direction for the diaphragm (median 15mm, range 0 to 50mm) and the porta hepatis (median 12mm, range 2 to 28mm). Displacements were similar when comparing FB, I and E scans, except for greater displacement seen in the CC direction on I scans. Median absolute change in volume (%) compared to week 1 scans in FB/I/E was: stomach 37/ 24.7/ 19.05, pancreas 33.2/ 19.8/ 21.4, kidneys 6.2/ 4/ 5.39. Conclusions: Displacement of the abdominal organs can be substantial, with offsets up to 5cm. This may alter the dose to targets and organs at risk and indicates the potential need for individualised and/or adaptive planning.
S281
S282
I. J. Radiation Oncology
● Biology ● Physics
Volume 63, Number 2, Supplement, 2005
Median displacement over all organs in mm (range of medians)
2088
A Prospective Comparison Study of Liver Tumour Target Definition Based on Triphasic CT and Gadolinium MR
J.J. Voroney,1 K. Brock,3 C. Eccles,3 M. Haider,2 L. Dawson1 Radiation Oncology, Princess Margaret Hospital, Toronto, ON, Canada, 2Radiology, Princess Margaret Hospital, Toronto, ON, Canada, 3Radiation Medicine, Princess Margaret Hospital, Toronto, ON, Canada 1
Purpose/Objective: To quantitatively compare liver cancer gross tumour volume (GTV) defined using MR and triphasic CT. Materials/Methods: Diagnostic quality planning MR scans with and without gadolinium (6 – 8 mm slice thickness) and triphasic planning CT scans (2–5 mm slice thickness) were obtained in twenty-three patients with non-resectable cholangiocarcinoma (n⫽4), hepatoma (n⫽10) and liver metastases (n⫽9) enrolled on a protocol of hypofractionated highly conformal radiotherapy. The liver was immobilized in exhale treatment position, using the active breathing coordinator for CT and voluntary breath hold for MR. On the CT and MR series best demonstrating the tumor, the liver and GTV were contoured with tools from Pinnacle (v6.5) treatment planning software. Anatomical reference points within the liver were identified. Despite careful patient positioning, spatial changes in liver position occurred between CT and MR imaging. A deformable registration method based on finite element modeling accounting for biomechanical forces and resulting positional and volumetric changes was used to register the CT and MR livers and resolve geometric differences between them. Accuracy of registration of anatomical reference points was calculated, and spatial differences between the CT and MR GTVs were then quantified using the same deformable registration technique to register the CT and MR GTVs. Results: The contrast CT best visualizing the GTV was the arterial phase CT for hepatoma (7/11) and the venous phase CT for metastases (7/9) and cholangiocarcinoma (4/4); the MR sequence best visualizing the GTV was gadolinium enhanced T1 weighted MR (22/24) and dual echo MR (2/24). Additional tumour foci were seen with MR in 3 cases. Following deformable registration of the liver volumes to resolve geometric discrepancies of the liver, the average vector magnitude (VM) of the residual registration error based on the displacement of the registered anatomical references points was 4.2 mm (SD ⫽ 1.7 mm), less than the MR slice thickness. The average CT and MR GTVs was 289 cc (range: 9 –1467 cc) and 278 cc (range: 6 –1731 cc) respectively. The average difference in CT and MR GTV was 11 cc (range: 1–263 cc), with CT GTV larger then MR GTV in 3/7 metastases, 4/4 cholangiocarcinomas and 6/11 hepatomas. The average ratio of CT/MR GTV was 1.2 (range: 0.3– 4.2). Deformable registration of the MR GTV to the CT GTV revealed substantial spatial differences. The study population mean of the average distance between the CT and MR GTV surfaces was 2.8, 2.9 and 2.9 mm in the medial-lateral (ML), anterior-posterior (AP) and cranial-caudal (CC) directions, with a standard deviation of 2.0, 2.9 and 2.1 mm. The study population mean of the maximal distance between CT and MR GTV surfaces was 17.6, 17.9 and 16.6, with maximal distances between individual surfaces as high as 39.7, 63.8 and 42.4 in the ML, AP and CC directions. The average percentage of GTV surface area that differed by 3 mm or more, 5 mm or more and 15 mm or more was 56% (max 92%), 35% (max 86%) and 9% (max 56%). Conclusions: MR defined GTVs can be significantly different than CT defined GTVs and this should be considered for high precision liver cancer radiotherapy.
Average (Av) surface area (SA) difference (⌬) statistics between MR and CT defined GTVs by tumour type, hepatoma (H), cholangiocarcinoma (CC) and liver metastases (LM).