- Email: [email protected]
Does 120 MLC-A Give an Advantage Over 120 MLC-B in the Treatment of Head-and-Neck Cancer (HNC) Patients?
Volume 84 Number 3S Supplement 2012 Author Disclosure: E.E. Ahunbay: None. B.H. Kimura: None. C. Schultz: None. X.A. Li: None.
3559 Comparison of Cone Beam CT and kV/kV Imaging Using Fiducial Marker for PTV margin Definition in IGRT of Localized Prostate Cancer C. Oehler,1,2 S. Lang,2 P. Dimmerling,2 C. Bolesch,2 A. Tini,2 C. Glanzmann,2 U.M. Luetolf,3 Y. Najafi,2 G. Studer,2 and D.R. Zwahlen1,2; 1Hospital Graubunden, Chur, Switzerland, 2 University Hospital Zurich, Zurich, Switzerland, 3University Hospital Zurich, Zurich, Switzerland Purpose/Objective(s): Image-guided radiation therapy (IGRT) using fiducial markers (FM) or cone beam CT (CBCT) is the preferred method in delivering RT for prostate cancer (PC) enabling precise targeting and dose escalation. These techniques should enable reduced CTV to PTV margins resulting in lower toxicity. We applied the Van Herk formula to calculate the margin for PTV of prostate using CBCT or FM. Materials/Methods: Twenty patients with low, intermediate or high risk PC received IGRT (VMAT) to a dose of 74Gy/37f, 60Gy/20f or 57Gy/ 19f. Three gold FM were implanted prior to treatment. Patients were planned and treated with rectal balloon (RB). CBCT was performed on day 1 to 3, then once every week. KV-image pairs (kV/kV) were used on all other days. For 10 patients CBCT was performed after each fraction to assess intra-fractional motion. Contouring of the prostate (P) for low risk and of the prostate/seminal vesicles (PSV) for intermediate/high risk PC was performed independently by 3 clinicians and mean differences in vertical, lateral and longitudinal directions between the intersections and the unions of the three contours were assessed. The Van Herk formula M Z 2.5S + 0.7s3 was used for PTV margin calculation, where S represents the systematic and s the random error. We calculated the systematic errors for intra-fractional motion, contouring as well as setup using CBCT, kV/kV and FM referring to prostate position in the CBCT and random errors for intra-fractional motion and patient setup. Results: Calculated PTV margins were 6.1 mm vertically, 7.1 mm longitudinally and 4.4 mm laterally for contouring and intra-fractional but not setup errors. Systematic error for contouring was more influential than intra-fractional motion (0.9-1.4mm) and more pronounced in intermediate/high risk (2.5-3.2mm) than in low risk PC (1.3-2.1mm). Including setup errors, enlargement was least for CBCT and kV/kVimaging using FM: +0.4-0.6mm (P) / +2.0-2.3mm (PSV) vertically, +0.1-0.7mm (P) / +2.5-2.9mm (PSV) longitudinally and +0.3mm (P) / +3.0-3.1mm (PSV) laterally. Higher increases were noticed for bone matching: +1.8-2.0mm (P) / +2.2-2.4mm (PSV) vertically, +1.4-1.5mm (P) / +3.6mm (PSV) longitudinally and +0.4-0.7mm (P) / +3.0-3.3mm (PSV) laterally. Largest increases were found for rectal balloon matching: +1.3-1.6mm (P) / +2.8-3.0mm (PSV) vertically, +0.9-4.5mm (P) / +3.1-6.2mm (PSV) longitudinally and +0.5-2.0mm (P) / +3.1-4.1mm (PSV) laterally. Conclusion: Calculated PTV margins are comparable using either CBCT or kV/kV-imaging FM matching. The systematic error for contouring was more influential than intra-fractional motion. Margins for intermediate/ high risk PC are notably larger than those for low risk PC due to seminal vesicles. Author Disclosure: C. Oehler: None. S. Lang: None. P. Dimmerling: None. C. Bolesch: None. A. Tini: None. C. Glanzmann: None. U.M. Luetolf: None. Y. Najafi: None. G. Studer: None. D.R. Zwahlen: None.
3560 A Predictive Model of Correlation Between Site and Diameter of Brain Metastases and Hippocampal Sparing Using Rotational Techniques S. Chiesa, L. Azario, F. Cellini, G.C. Mattiucci, S. Gaudino, L. De Filippo, M. Ferro, V. Frascino, S. Silipigni, and M. Balducci; Catholic University of Sacred Heart, Rome, Italy
Poster Viewing Abstracts S805 Purpose/Objective(s): Whole brain radiation therapy (WBRT), resection, stereotactic radiosurgery (SRS), and chemotherapy are the current treatment in brain metastases and it is known that metastases are localized in less than 4% within 5 mm of hippocampal. It is interesting because hippocampal irradiation during WBRT correlates with a neurocognitive toxicity and modern IMRT can reduce the dose to critical structures as hippocampus, but it is time consuming during treatment planning. The aim of our investigation is to evaluate if site and diameter of brain metastases can influence the sparing of hippocampus using volumetric modulated arc therapy (VMAT-RA) to decide in advance when using this technique is better than using 3D conformal radiation therapy. Materials/Methods: We contoured concentric brain metastases of 0.5, 1, 2, 3 cm of diameter. Their center was the gravity center of each lobe. The lobes, the deep brain areas and the hippocampus were contoured manually by the collaboration with a neuroradiologist using fused CT-MRI axial images. A VMAT treatment plan was generated for each metastases with only one arc (from 179 to 181, CCW). The whole brain dose was 32.25 Gy in 15 fractions, while total dose to brain metastases was modulated according to the diameter, 63 Gy if 2.0 cm, 70.8 Gy if <2.0 cm. Optimization aims to minimizing hippocampal dose, conforming target coverage. Results: We contoured frontal, parietal, occipital, temporal lobe and deep brain structures, therefore we obtained twenty brain lesions and we generated twenty RA treatment plans. Hippocampal volume was 4,09 cm3. According to the site we observed following mean of mean hippocampal dose: 11 Gy2 in frontal lobe, 11.92 Gy2 in parietal lobe, 9.66 Gy2 in occipital lobe, 13.47 Gy2 in deep brain areas, 44.38 Gy2 in temporal lobe, where hippocampal is localized. According to the diameter we reported following mean of mean hippocampal dose: 13.82 Gy2 in case of 1 cm of diameter, 16.32 Gy2 if 2 cm, 16.65 Gy2 if 3 cm, while it was 19.46 Gy2 in diameter < 1cm (0,5 cm). Mean target coverage and homogeneity index of whole brain were 0.9 0.04 and 0,65 0.18 respectively, while mean target coverage was 0.54 0.10 for the brain lesions. Conclusion: Size of brain metastases 1 cm seems to improve the feasibility of sparing hippocampal; it is more difficult if the metastases are in the temporal lobe. Further investigation are warranted to obtain definitive considerations. Author Disclosure: S. Chiesa: None. L. Azario: None. F. Cellini: None. G.C. Mattiucci: None. S. Gaudino: None. L. De Filippo: None. M. Ferro: None. V. Frascino: None. S. Silipigni: None. M. Balducci: None.
3561 Does 120 MLC-A Give an Advantage Over 120 MLC-B in the Treatment of Head-and-Neck Cancer (HNC) Patients? S. Ng, A.L. Damato, R.B. Tishler, and R.A. Cormack; Brigham and Women’s Hospital & Dana Faber Cancer Institute, Harvard Medical School, Boston, MA Purpose/Objective(s): To evaluate dosimetric differences associated with the use of 120 MLCs in the IMRT treatment of HNC patients. Materials/Methods: The records of 11 HNC patients recently treated in our clinic were retrospectively analyzed with an IRB-approved protocol. Two clinical plans were generated for each patient in this study: the plan delivered with HD (P_HD) and the plan delivered with ML (P_ML). P_ML was obtained from P_HD, where the planners were free to change the fluence map and renormalize the plan. An additional plan (P_FL) was created with ML leaf motion generated from the fluence map associated to P_HD, without any other modification. Target coverage of V98, conformity index (CI Z ratio of volume receiving 100% prescription dose to planning target volume), homogeneity index (HI Z (D_max - D_min)/ D_mean) and mean dose to left and right parotid were calculated. D_max, D_min and D_mean are maximum, minimum and mean dose to the PTV. Means and standard deviations (SD) of all metrics across plans for each patient and across patients for each plan were calculated. All plans underwent routine IMRT QA.
International Journal of Radiation Oncology Biology Physics
S806 Poster Viewing Abstract 3561; Table Comparison of CI, HI, and target coverage of plans delivered with HD120 and Millennium MLC CI for all patients
HI for all patients
V98 for all patients
P_HD P_ML P_FL P_HD P_ML P_FL P_HD P_ML Mean SD
1.39 0.44
1.54 0.55
1.46 0.56
Mean SD
CI across different MLC 1.48 0.50
0.19 0.06
0.20 0.06
0.19 0.07
HI across different MLC 0.18 0.01
P_FL
97.0% 97.9% 97.5% 3.6% 3.1% 3.9% V98 across different MLC 97.5% 3.4%
Results: While 120 MLC-A plans achieved better conformity in the comparison of CI across patients, the SD across plans was small (mean CI Z 1.48+/- 0.5). Comparison of target dose homogeneity across all plans and all patients showed little difference (mean HI Z 0.18 +/- 0.01). Difference of mean dose to left and right parotid delivered using P_HD and P_ML was 0.6+/- 0.6 Gy and 1.0 +/- 1.1 Gy, respectively. Mean target coverage, V98 across all plans was 97.5+/- 3.4 %. Results are summarized in Table. All plans passed IMRT QA according to the standard at this institute. Conclusion: No clinically meaningful differences were observed between P_HD, P_ML and P_FL. These results suggest that 120 MLC-A (leaf width Z 2.5mm) and 120 MLC-B (leaf width Z 5.0mm) are equivalent for the treatment of HNC patients. Moreover the data suggest that changing the treatment of a HNC patient from an HD equipped machine to a matched ML equipped machine, while clearly requiring IMRT QA, does not require re-planning. Author Disclosure: S. Ng: None. A.L. Damato: None. R.B. Tishler: None. R.A. Cormack: None.
3562 Gamma Analysis of Normalized and Un-normalized Dose Distributions S. Stojadinovic, L. Ouyang, Q. Bao, A. Pompos, X. Gu, and T.D. Solberg; University of Texas Southwestern Medical Center, Dallas, TX Purpose/Objective(s): The gamma index method, as currently implemented in a typical IMRT QA analysis, calls for selection of a normalization point to evaluate agreement between two dose distributions. The implication of this is that there is an infinite number of possible solutions! Which one to pick? A unique and more relevant solution is obtained only if no normalization point is used. Materials/Methods: The set of test cases suggested by the AAPM TG119 were planned and delivered on a linac for 6 and 18 MV photons. The recommended point and planar dose measurements were obtained using an ion chamber, EDR2 film and detector. The gamma index method using typical 3%, 3 mm criteria with and without a normalization point was used to assess the agreement between calculated and delivered planar dose distributions. The analysis was extended to a set of data for clinically treated patients. Results: The comparison with the TG119 benchmark data showed that all point dose and planar measurements for 6 MV were within the published range. Similar results, although without published data to compare with, were obtained for 18 MV as well. For all complex tests, the percentage of points passing the gamma criteria of 3%, 3 mm was (95.81.6)% and (95.61.0)% for 6 MV and 18 MV, respectively. Without a normalization point, however, the same gamma analysis fell to (20.76.7)% and (13.94.0)% for 6 MV and 18 MV, respectively. The clinical data set showed the same trend, with the gamma passing rate declining from (98.90.7)% to (33.413.1)%. Conclusion: The gamma index method provides a unique answer for gamma passing rate only without normalizing dose distributions to any particular point. The common gamma criteria of 3%, 3 mm, however, is a very poor metric in that case. Author Disclosure: S. Stojadinovic: None. L. Ouyang: None. Q. Bao: None. A. Pompos: None. X. Gu: None. T.D. Solberg: None.
3563 Implementation of Intensity Modulated Radiation Therapy in a Developing Country: Lessons for Everyone I. Mohamad, A. Salem, R. Abu-Hijlih, A. Dayyat, S. Ramahi, J. Khader, I. Jaradat, and A. Almousa; King Hussein Cancer Center, Amman, Jordan Purpose/Objective(s): Review our experience with the introduction of IMRT at our center in an attempt to present pertinent lessons applicable for radiation oncology departments in developed and developing nations alike. Materials/Methods: Between December 2010 and February 2012, 62 patients treated by IMRT (28 head and neck, 27 prostate, 4 gynecological, and 3 metastatic brain lesions) were retrospectively reviewed. Extensive analysis of dosimetric and quality assurance (QA) data was performed. The time from CT simulation to completion of contouring was utilized to assess the learning curve for radiation oncologists, while the conformity index (CI) was utilized to assess the learning curve for physicists/ dosimetrists. Results: In head and neck cancer, IMRT was attempted in 16 patients with the sole intention of sparing the brain stem, spinal cord, and/or ocular apparatus due to huge primary tumors. In these cases, the huge target volumes lead to dosimetric difficulties in order to balance tumor coverage while maintaining tolerance of near-by normal tissues. Nevertheless, at least one parotid gland was successfully spared in all patients (even in patients with level II nodal involvement). To ensure adequate tumor coverage in these cases, co-registration with MRI and/or PET-CT was performed. Due to numerous financial difficulties, deformable registration is not currently available at our institution. To overcome the challenge of fusing images in cases where the patient’s anatomy has significantly changed due to weight loss or following induction chemotherapy, we proposed a “manual deformation” regimen in which we manually divide the neck into multiple levels (up to 7) and perform individual registration for each level independently of the next. Although labor intensive, this obviates the need to purchase expensive deformable registration software. In addition, we have demonstrated a learning curve for radiation oncologists and physicists/dosimetrists after the first 14 cases for both prostate and head and neck cancer sites. Gynecological cancer cases were treated utilizing RTOG contouring guidelines for nodal delineation. However, we opted to include the obturator lymph nodes as a treatment target in variance to RTOG protocols. IMRT was used for solitary metastatic brain lesions following whole brain radiation therapy as an alternative to stereotactic radiosurgery which is currently unavailable at our center. We are currently implementing an intense physics QA program consisting of fluence and MU checks for each individual treatment beam. This is in variance to less labor intensive QA programs implemented in more advanced centers worldwide. Conclusions: This review represents a detailed report addressing important lessons learned from the successful establishment of IMRT in developing countries. Author Disclosure: I. Mohamad: None. A. Salem: None. R. Abu-Hijlih: None. A. Dayyat: None. S. Ramahi: None. J. Khader: None. I. Jaradat: None. A. Almousa: None.
3564 Dosimetric Benefits of Online Adaptive Replanning for Radiation Therapy of Prostate Plus Seminal Vesicle: Evaluation Based on Cumulative Dose F. Liu, E. Ahunbay, E. Chang, C. Lawton, and X. Li; Medical College of Wisconsin, Milwaukee, WI Purpose/Objective(s): To investigate cumulative dose calculation and to compare plans generated with IGRT repositioning and online replanning for radiation therapy of prostate plus seminal vesicle. Materials/Methods: Daily CT datasets acquired during IGRT using an inroom CT for 5 prostate cancer patients with prostate and seminal vesicle irradiated were analyzed. Target and organs at risk (OARs) were delineated on each set of daily CT by populating the planning contours to the daily CT using an auto-segmentation tool with manual editing. For each case,
3559 Comparison of Cone Beam CT and kV/kV Imaging Using Fiducial Marker for PTV margin Definition in IGRT of Localized Prostate Cancer C. Oehler,1,2 S. Lang,2 P. Dimmerling,2 C. Bolesch,2 A. Tini,2 C. Glanzmann,2 U.M. Luetolf,3 Y. Najafi,2 G. Studer,2 and D.R. Zwahlen1,2; 1Hospital Graubunden, Chur, Switzerland, 2 University Hospital Zurich, Zurich, Switzerland, 3University Hospital Zurich, Zurich, Switzerland Purpose/Objective(s): Image-guided radiation therapy (IGRT) using fiducial markers (FM) or cone beam CT (CBCT) is the preferred method in delivering RT for prostate cancer (PC) enabling precise targeting and dose escalation. These techniques should enable reduced CTV to PTV margins resulting in lower toxicity. We applied the Van Herk formula to calculate the margin for PTV of prostate using CBCT or FM. Materials/Methods: Twenty patients with low, intermediate or high risk PC received IGRT (VMAT) to a dose of 74Gy/37f, 60Gy/20f or 57Gy/ 19f. Three gold FM were implanted prior to treatment. Patients were planned and treated with rectal balloon (RB). CBCT was performed on day 1 to 3, then once every week. KV-image pairs (kV/kV) were used on all other days. For 10 patients CBCT was performed after each fraction to assess intra-fractional motion. Contouring of the prostate (P) for low risk and of the prostate/seminal vesicles (PSV) for intermediate/high risk PC was performed independently by 3 clinicians and mean differences in vertical, lateral and longitudinal directions between the intersections and the unions of the three contours were assessed. The Van Herk formula M Z 2.5S + 0.7s3 was used for PTV margin calculation, where S represents the systematic and s the random error. We calculated the systematic errors for intra-fractional motion, contouring as well as setup using CBCT, kV/kV and FM referring to prostate position in the CBCT and random errors for intra-fractional motion and patient setup. Results: Calculated PTV margins were 6.1 mm vertically, 7.1 mm longitudinally and 4.4 mm laterally for contouring and intra-fractional but not setup errors. Systematic error for contouring was more influential than intra-fractional motion (0.9-1.4mm) and more pronounced in intermediate/high risk (2.5-3.2mm) than in low risk PC (1.3-2.1mm). Including setup errors, enlargement was least for CBCT and kV/kVimaging using FM: +0.4-0.6mm (P) / +2.0-2.3mm (PSV) vertically, +0.1-0.7mm (P) / +2.5-2.9mm (PSV) longitudinally and +0.3mm (P) / +3.0-3.1mm (PSV) laterally. Higher increases were noticed for bone matching: +1.8-2.0mm (P) / +2.2-2.4mm (PSV) vertically, +1.4-1.5mm (P) / +3.6mm (PSV) longitudinally and +0.4-0.7mm (P) / +3.0-3.3mm (PSV) laterally. Largest increases were found for rectal balloon matching: +1.3-1.6mm (P) / +2.8-3.0mm (PSV) vertically, +0.9-4.5mm (P) / +3.1-6.2mm (PSV) longitudinally and +0.5-2.0mm (P) / +3.1-4.1mm (PSV) laterally. Conclusion: Calculated PTV margins are comparable using either CBCT or kV/kV-imaging FM matching. The systematic error for contouring was more influential than intra-fractional motion. Margins for intermediate/ high risk PC are notably larger than those for low risk PC due to seminal vesicles. Author Disclosure: C. Oehler: None. S. Lang: None. P. Dimmerling: None. C. Bolesch: None. A. Tini: None. C. Glanzmann: None. U.M. Luetolf: None. Y. Najafi: None. G. Studer: None. D.R. Zwahlen: None.
3560 A Predictive Model of Correlation Between Site and Diameter of Brain Metastases and Hippocampal Sparing Using Rotational Techniques S. Chiesa, L. Azario, F. Cellini, G.C. Mattiucci, S. Gaudino, L. De Filippo, M. Ferro, V. Frascino, S. Silipigni, and M. Balducci; Catholic University of Sacred Heart, Rome, Italy
Poster Viewing Abstracts S805 Purpose/Objective(s): Whole brain radiation therapy (WBRT), resection, stereotactic radiosurgery (SRS), and chemotherapy are the current treatment in brain metastases and it is known that metastases are localized in less than 4% within 5 mm of hippocampal. It is interesting because hippocampal irradiation during WBRT correlates with a neurocognitive toxicity and modern IMRT can reduce the dose to critical structures as hippocampus, but it is time consuming during treatment planning. The aim of our investigation is to evaluate if site and diameter of brain metastases can influence the sparing of hippocampus using volumetric modulated arc therapy (VMAT-RA) to decide in advance when using this technique is better than using 3D conformal radiation therapy. Materials/Methods: We contoured concentric brain metastases of 0.5, 1, 2, 3 cm of diameter. Their center was the gravity center of each lobe. The lobes, the deep brain areas and the hippocampus were contoured manually by the collaboration with a neuroradiologist using fused CT-MRI axial images. A VMAT treatment plan was generated for each metastases with only one arc (from 179 to 181, CCW). The whole brain dose was 32.25 Gy in 15 fractions, while total dose to brain metastases was modulated according to the diameter, 63 Gy if 2.0 cm, 70.8 Gy if <2.0 cm. Optimization aims to minimizing hippocampal dose, conforming target coverage. Results: We contoured frontal, parietal, occipital, temporal lobe and deep brain structures, therefore we obtained twenty brain lesions and we generated twenty RA treatment plans. Hippocampal volume was 4,09 cm3. According to the site we observed following mean of mean hippocampal dose: 11 Gy2 in frontal lobe, 11.92 Gy2 in parietal lobe, 9.66 Gy2 in occipital lobe, 13.47 Gy2 in deep brain areas, 44.38 Gy2 in temporal lobe, where hippocampal is localized. According to the diameter we reported following mean of mean hippocampal dose: 13.82 Gy2 in case of 1 cm of diameter, 16.32 Gy2 if 2 cm, 16.65 Gy2 if 3 cm, while it was 19.46 Gy2 in diameter < 1cm (0,5 cm). Mean target coverage and homogeneity index of whole brain were 0.9 0.04 and 0,65 0.18 respectively, while mean target coverage was 0.54 0.10 for the brain lesions. Conclusion: Size of brain metastases 1 cm seems to improve the feasibility of sparing hippocampal; it is more difficult if the metastases are in the temporal lobe. Further investigation are warranted to obtain definitive considerations. Author Disclosure: S. Chiesa: None. L. Azario: None. F. Cellini: None. G.C. Mattiucci: None. S. Gaudino: None. L. De Filippo: None. M. Ferro: None. V. Frascino: None. S. Silipigni: None. M. Balducci: None.
3561 Does 120 MLC-A Give an Advantage Over 120 MLC-B in the Treatment of Head-and-Neck Cancer (HNC) Patients? S. Ng, A.L. Damato, R.B. Tishler, and R.A. Cormack; Brigham and Women’s Hospital & Dana Faber Cancer Institute, Harvard Medical School, Boston, MA Purpose/Objective(s): To evaluate dosimetric differences associated with the use of 120 MLCs in the IMRT treatment of HNC patients. Materials/Methods: The records of 11 HNC patients recently treated in our clinic were retrospectively analyzed with an IRB-approved protocol. Two clinical plans were generated for each patient in this study: the plan delivered with HD (P_HD) and the plan delivered with ML (P_ML). P_ML was obtained from P_HD, where the planners were free to change the fluence map and renormalize the plan. An additional plan (P_FL) was created with ML leaf motion generated from the fluence map associated to P_HD, without any other modification. Target coverage of V98, conformity index (CI Z ratio of volume receiving 100% prescription dose to planning target volume), homogeneity index (HI Z (D_max - D_min)/ D_mean) and mean dose to left and right parotid were calculated. D_max, D_min and D_mean are maximum, minimum and mean dose to the PTV. Means and standard deviations (SD) of all metrics across plans for each patient and across patients for each plan were calculated. All plans underwent routine IMRT QA.
International Journal of Radiation Oncology Biology Physics
S806 Poster Viewing Abstract 3561; Table Comparison of CI, HI, and target coverage of plans delivered with HD120 and Millennium MLC CI for all patients
HI for all patients
V98 for all patients
P_HD P_ML P_FL P_HD P_ML P_FL P_HD P_ML Mean SD
1.39 0.44
1.54 0.55
1.46 0.56
Mean SD
CI across different MLC 1.48 0.50
0.19 0.06
0.20 0.06
0.19 0.07
HI across different MLC 0.18 0.01
P_FL
97.0% 97.9% 97.5% 3.6% 3.1% 3.9% V98 across different MLC 97.5% 3.4%
Results: While 120 MLC-A plans achieved better conformity in the comparison of CI across patients, the SD across plans was small (mean CI Z 1.48+/- 0.5). Comparison of target dose homogeneity across all plans and all patients showed little difference (mean HI Z 0.18 +/- 0.01). Difference of mean dose to left and right parotid delivered using P_HD and P_ML was 0.6+/- 0.6 Gy and 1.0 +/- 1.1 Gy, respectively. Mean target coverage, V98 across all plans was 97.5+/- 3.4 %. Results are summarized in Table. All plans passed IMRT QA according to the standard at this institute. Conclusion: No clinically meaningful differences were observed between P_HD, P_ML and P_FL. These results suggest that 120 MLC-A (leaf width Z 2.5mm) and 120 MLC-B (leaf width Z 5.0mm) are equivalent for the treatment of HNC patients. Moreover the data suggest that changing the treatment of a HNC patient from an HD equipped machine to a matched ML equipped machine, while clearly requiring IMRT QA, does not require re-planning. Author Disclosure: S. Ng: None. A.L. Damato: None. R.B. Tishler: None. R.A. Cormack: None.
3562 Gamma Analysis of Normalized and Un-normalized Dose Distributions S. Stojadinovic, L. Ouyang, Q. Bao, A. Pompos, X. Gu, and T.D. Solberg; University of Texas Southwestern Medical Center, Dallas, TX Purpose/Objective(s): The gamma index method, as currently implemented in a typical IMRT QA analysis, calls for selection of a normalization point to evaluate agreement between two dose distributions. The implication of this is that there is an infinite number of possible solutions! Which one to pick? A unique and more relevant solution is obtained only if no normalization point is used. Materials/Methods: The set of test cases suggested by the AAPM TG119 were planned and delivered on a linac for 6 and 18 MV photons. The recommended point and planar dose measurements were obtained using an ion chamber, EDR2 film and detector. The gamma index method using typical 3%, 3 mm criteria with and without a normalization point was used to assess the agreement between calculated and delivered planar dose distributions. The analysis was extended to a set of data for clinically treated patients. Results: The comparison with the TG119 benchmark data showed that all point dose and planar measurements for 6 MV were within the published range. Similar results, although without published data to compare with, were obtained for 18 MV as well. For all complex tests, the percentage of points passing the gamma criteria of 3%, 3 mm was (95.81.6)% and (95.61.0)% for 6 MV and 18 MV, respectively. Without a normalization point, however, the same gamma analysis fell to (20.76.7)% and (13.94.0)% for 6 MV and 18 MV, respectively. The clinical data set showed the same trend, with the gamma passing rate declining from (98.90.7)% to (33.413.1)%. Conclusion: The gamma index method provides a unique answer for gamma passing rate only without normalizing dose distributions to any particular point. The common gamma criteria of 3%, 3 mm, however, is a very poor metric in that case. Author Disclosure: S. Stojadinovic: None. L. Ouyang: None. Q. Bao: None. A. Pompos: None. X. Gu: None. T.D. Solberg: None.
3563 Implementation of Intensity Modulated Radiation Therapy in a Developing Country: Lessons for Everyone I. Mohamad, A. Salem, R. Abu-Hijlih, A. Dayyat, S. Ramahi, J. Khader, I. Jaradat, and A. Almousa; King Hussein Cancer Center, Amman, Jordan Purpose/Objective(s): Review our experience with the introduction of IMRT at our center in an attempt to present pertinent lessons applicable for radiation oncology departments in developed and developing nations alike. Materials/Methods: Between December 2010 and February 2012, 62 patients treated by IMRT (28 head and neck, 27 prostate, 4 gynecological, and 3 metastatic brain lesions) were retrospectively reviewed. Extensive analysis of dosimetric and quality assurance (QA) data was performed. The time from CT simulation to completion of contouring was utilized to assess the learning curve for radiation oncologists, while the conformity index (CI) was utilized to assess the learning curve for physicists/ dosimetrists. Results: In head and neck cancer, IMRT was attempted in 16 patients with the sole intention of sparing the brain stem, spinal cord, and/or ocular apparatus due to huge primary tumors. In these cases, the huge target volumes lead to dosimetric difficulties in order to balance tumor coverage while maintaining tolerance of near-by normal tissues. Nevertheless, at least one parotid gland was successfully spared in all patients (even in patients with level II nodal involvement). To ensure adequate tumor coverage in these cases, co-registration with MRI and/or PET-CT was performed. Due to numerous financial difficulties, deformable registration is not currently available at our institution. To overcome the challenge of fusing images in cases where the patient’s anatomy has significantly changed due to weight loss or following induction chemotherapy, we proposed a “manual deformation” regimen in which we manually divide the neck into multiple levels (up to 7) and perform individual registration for each level independently of the next. Although labor intensive, this obviates the need to purchase expensive deformable registration software. In addition, we have demonstrated a learning curve for radiation oncologists and physicists/dosimetrists after the first 14 cases for both prostate and head and neck cancer sites. Gynecological cancer cases were treated utilizing RTOG contouring guidelines for nodal delineation. However, we opted to include the obturator lymph nodes as a treatment target in variance to RTOG protocols. IMRT was used for solitary metastatic brain lesions following whole brain radiation therapy as an alternative to stereotactic radiosurgery which is currently unavailable at our center. We are currently implementing an intense physics QA program consisting of fluence and MU checks for each individual treatment beam. This is in variance to less labor intensive QA programs implemented in more advanced centers worldwide. Conclusions: This review represents a detailed report addressing important lessons learned from the successful establishment of IMRT in developing countries. Author Disclosure: I. Mohamad: None. A. Salem: None. R. Abu-Hijlih: None. A. Dayyat: None. S. Ramahi: None. J. Khader: None. I. Jaradat: None. A. Almousa: None.
3564 Dosimetric Benefits of Online Adaptive Replanning for Radiation Therapy of Prostate Plus Seminal Vesicle: Evaluation Based on Cumulative Dose F. Liu, E. Ahunbay, E. Chang, C. Lawton, and X. Li; Medical College of Wisconsin, Milwaukee, WI Purpose/Objective(s): To investigate cumulative dose calculation and to compare plans generated with IGRT repositioning and online replanning for radiation therapy of prostate plus seminal vesicle. Materials/Methods: Daily CT datasets acquired during IGRT using an inroom CT for 5 prostate cancer patients with prostate and seminal vesicle irradiated were analyzed. Target and organs at risk (OARs) were delineated on each set of daily CT by populating the planning contours to the daily CT using an auto-segmentation tool with manual editing. For each case,