Intensity Modulated Radiation Therapy Implementation Process Of Quality Assurance Studies

1 Study of set up errors in intensity-modulated radiation therapy (IMRT)for nasopharyngeal carcinoma1.1 Comparison of setup accuracy of head mask and head and shoulder mask in IMRT for nasopharyngeal carcinomaPurpose: IMRT for nasopharyngeal carcinoma requires a reliable immobilization for an accurate and consistent treatment setup. This study was done to compare the reproducibility of two types of thermoplastic face masks: head mask and head and shoulder mask.Methods and Materials: The positioning data of 20 patients treated with IMRT for nasopharyngeal carcinoma were analyzed. 20 patients were consecutively assigned to head mask or head and shoulder mask. Once a week, during the session with a 6MV linac (Elekta 5954), orthogonal verification films were acquired. Field shape and an anatomy template consisting of contours of relevant bony structures on DRR and portal films were drawn and compared using the iVewGT software, and displacements in antero-posterior(AP), cranio-caudal(CC) and medio-lateral (ML) directions were calculated. From these displacements, three dimensional errors were also calculated. The frequency of repositioning of head mask and head and shoulder mask were compared.Results: For patients with head and shoulder mask, the systematic error is 1.2mm, 1. 1mm, 1.1mm and the random set-up component was 1. 4mm, 1. 2mm, 1. 4mm in the AP, CC and ML direction, respectively. For patients with head mask , the systematic error is 1. 6mm, 1. 5mm, 1. 5mm and the random set-up component was 1. 6mm, 1. 6mm, 1. 5mm in the AP, CC and ML direction, respectively. Compare the frequency of repositioning of head mask and head and shoulder mask and there was significant difference in the AP, CC and ML direction, with p value 0. 015, 0. 032, 0. 015 respectively.Conclusion: Head mask and head and shoulder mask provide an accurate patient immobilization in IMRT for nasopharyngeal carcinoma. However, the head and shoulder mask had a higher reproducibility of the setup. 1. 2 The analysis of set up errors during the IMRT for nasopharyngeal carcinomaPurpose: Intensity Modulated Radiation Therapy (IMRT) is now widely used in the radiation therapy of nasopharyngeal carcinoma. Higher positioning precision is required in IMRT, compared to conventional radiotherapy. High precision radiotherapy makes it important to research set-up errors. Because of different equipments and technologists' experience, every institute should have its own data. In this paper, we investigated the set up errors with portal film for nasopharyngeal carcinoma with IMRT and defined margins for planning target volume definition, accounting for systematic and random set up uncertainties. At the same time, we explored the time trend of setup errors and the effect of weight loss on setup reproducibility.Method and materials: The positioning data of 30 patients treated with IMRT for nasopharyngeal carcinoma were analyzed. All patients were immobilized with head and shoulder mask. Orthogonal verification films were taken once a week before treatment. Field shape and an anatomy template consisting of contours of relevant bony structures on DRR and portal films were drawn and compared to obtain the positioning data. For all three directions(x, y, z), random and systematic set-up deviations were measured. Margins for planning target volume were defined accounting for systematic and random set up uncertainties. Weight data was recorded when patients were examined every week.Results: Three hundred and sixty images were obtained. Eleven images were excluded due to poor quality to identify two or more relevant bony structures. For each patient, the systematic and random errors for the directions were measured separately. The systematic error showed 1. 2mm, 1.1mm, 1.1mm, along the medio-lateral (ML), cranio-caudal (CC) and antero-posterior (AP) direction and the random set-up component was 1. 4mm, 1. 5mm, 1. 5mm along ML, CC and AP direction, respectively. The overall mean deviation was -0.3mm,-0.3mm,-0.3mm along ML, CC and AP direction, respectively. To compare setup errors in first three weeks and last three weeks, increase of both systematic and random set-up error in three directions was found in last three weeks. The displacements increased with treatment progress. Media weight change was 6. 1%, from to -14. 6% to 5. 5%. For patients with weight loss≤6. 1%, the systematic error was 0.8mm, 0.8mm, 0.8mm along the ML, CC and AP direction and the random set-up component was 1. 4mm, 1.3mm, 1. 3mm along ML, CC and AP direction, respectively without significant time trend. While for patients with weight loss >6. 1%, the systematic error was 1. 5mm, 1. 3mm, 1. 4mm along the ML, CC and AP direction and the random set-up component was 1. 4mm, 1. 3mm, 1. 3mm along the ML, CC and AP direction, respectively with significant time trend. Calculation of CTV-PTV margin was based on the recipe of Stroom. CTV-PTV margin was 3. 4mm, 3. 3mm, 3. 4mm along ML, CC and AP direction respectively.Conclusion: Head and shoulder mask provided an accurate patient immobilization in IMRT for nasopharyngeal carcinoma. Application of proposed recipes for CTV-PTV margins indicates that 4mm was feasible. There was a significant increase in systematic and random set-up errors in last three weeks of treatment. The displacements increased with treatment progress. Weight loss during treatment compromised the reproducibility of setup. 1. 3 Systematic setup errors for IMRT in the nasopharyngeal carcinoma: effect on dose distributionPurpose: The purpose of this study is to determine dose delivery errors that could result from systematic setup errors for IMRT in the nasopharyngeal carcinoma and compare the effects on the dose distribution with the effects of DCRT treatments according to simulating systematic setup errors by moving the isocenter.Method and materials: Eight patients with IMRT and DCRT plan were evaluated to assess the impact of systematic setup errors. The dosimetric effect of systematic error was simulated by randomly shifting the patient isocenter along each of the three Cartesian axes, with each shift selected from a normal distribution. Systematic setup error of 2mm and 5mm were simulated. For each dose calculation, analyze the ratio of each dose evaluation index to the normal plan used for treatment to compare the sensitivity of DCRT and IMRT to setup errors.Results: Frequency distribution showed that simulating systematic setup error with 5mm resulted in 4. 2% GTV98 and 12. 5% CTV95 having more than a 3% dose error, whereas simulating systematic setup error with 2mm resulted in having no more than 3% dose error in IMRT. For DCRT, neither simulated systematic error with 5mm or with 2mm resulted in having more than 3% dose error. With a 5mm systematic setup error, the dose of critical organs increased with 10.4%spinal cord, 14.6% brain stem, 31.2%left parotid and 25% right parotid having more than 10% dose error and with a 2mm systematic setup error, the dose of critical organs increased only with 6. 3%spinal cord, 4. 2% brain stem, having more than 5% dose error and 12. 5%left parotid and 8. 3% right parotid having more than 10% dose error in IMRT. However in DCRT, with a 5mm systematic setup error, only 10. 4%spinal cord, 6. 3% brain stem had more than 5% dose error but there was no significant increase in the dose of left parotid and right parotid. For both targets and critical organs, comparing IMRT with DCRT, the mean percent dose change of the simulated structure dose with respect to the normal plan was more sensitive to IMRT (p<0.05) . The systematic setup error was bigger, the dose error was more significant (p<0. 05) .Conclusion: Bigger systematic setup error induced more significant dose error. For both targets and critical organs, IMRT was more sensitive tosystematic setup error than DCRT. For IMRT, GTV was more sensitive topatient positioning than CTV and parotids were more sensitive to patientpositioning than spinal cord and brain stem. 2 Repeat CT imaing and replanning during the couse of IMRT for nasopharyngeal carcinomaPurpose: Many patients receiving IMRT for nasopharyngeal carcinoma have marked anatomic change including tumor shrinkage and/or weight loss during the course of radiotherapy. The aim of this study was to quantify the magnitude of these anatomic changes and determine the dosimetric effects of repeat CT imaging and replanning on both normal and target volumes.Method and materials: Fifteen assessable patients with local advanced nasopharyngeal carcinoma were enrolled in this pilot study, having repeat CT imaging after 18 fractions with same position. CT-CT fusion was utilized to correct patient positioning differences between the scans. Both targets and normal tissues were manually contoured on every axial slice. Attempts were made to maintain the size of GTV, CTV in the second CT without extending it beyond the skin contour. Two central body reference (the centers of mass of the C2 vertebral body and odontoid process) were used to determine the volumetric and positional changes of parotids and external contour. The first plan for each patient was generated based on the original planning CT scan. By utilizing a inverse IMRT planning system, a hybrid IMRT plan (including the intensity profile of each beam) to the anatomy of the second CT scan. The dose of the first plan and hybrid plans were compared. New plan was made at the second CT with the same objective function. The dose of the new plan and hybrid plans were also compared.Results: Fifteen patient with local advanced nasopharyngeal carcinoma were enrolled, including 10 patients with III and 5 patients with IV disease. All patients lost weight throughout their course of treatment. The median weight change was 8.0% (2.7-11.8%). Parotid glands decreased in volume and generally shifted medially. The mean volume of left parotid and right parotid decreased 6. 19c^ 6. 44cc and the media displacement was 4.8mm,4.3mm, respectively. The external diameter contracted (P<0. 01) with the mean contraction of 7.7mm(-1.3-14.0mm), while the internal had no significant change. The external contours of the upper bound of odontoid process , the center of odontoid process and the center of the C2 vertebral body slices contracted (P<0. 01) with the mean contraction of 8.2±4.1mm , 9.4±5.0mm, 7.6±3.8mm. Comparing the hybrid plan to treatment plan, the coverage of target was maintained and dose to V105 and V110 was increased (P<0.05). The maximum dose to the brain stem in 80% patients increased by 0.08-6.51 Gy and the maximum dose to the spinal cord in 86. 67% patients increased by 0. 05-7. 8 Gy . All dosimetric points of parotids increased :the mean dose of left parotid increased by 2. 97Gy (—0.13-7.04Gy) and the mean dose of left parotid increased by 2. 57 Gy (0.04-6.24Gy) , P<0. 05. A new plan made on the second CT maintained the coverage of target and reduced the dose of critical organs including spinal cord, brain stem and parotids. The percentage of weight loss correlated highly with the change of external contours on three slices and the decreased volume and percentage of parotids. In addition, the extent of weight loss had a high correlation with the change of the mean dose to parotids and the maximum dose to spinal cord and had no significant correlation with the maximum dose to brain stem. The analysis of the extent of weight loss and target dosimetric points showed that only PTV-G V110 has a statistics significant correlation with the extent of weight loss. Conclusion: Measurable anatomic changes occurred during the IMRT for local advanced nasopharyngeal carcinoma including external contour contraction, parotid volume reduction and displacement to media line. These anatomic changes maintained the coverage of target but increased the dose to critical organs in some patients. These patients can get benefit from repeat CT imaging and replanning. 3 The in vivo study on the effect of prolonged delivery time to tumor control in C57BL mice implanted with Lewis lung cancer Purpose: Research the effect of different delivery time with same dose on the tumor growth delay and survival in C57BL mice implanted with Lewis lung cancer to determine whether prolonged delivery time would decrease the tumor response to radiation or not.Method and materials: 48 mice implanted with Lewis lung cancer in the back legs were involved and the experiment started when the transplanted tumor diameter reached 0. 8 to 1cm. 18Gy was divided into different subfractions, simulating the delivery of a complicated treatment, so 48 mice were randomized into 6 groups: the normal control group, the single fraction with 18 Gy group, the two subfractions with 30min interval group, the seven subfractions with 5min interval group, the two subfractions with 60min interval group, the seven subfractions with 10min interval group. Record the maximal and minimal diameters of the tumor to observe the tumor growth tendency, the tumor growth delay and the mice survival time.Results: The tumor grow delay of groups with prolonged delivery time was shorter than the group with single fraction of 18 Gy (P<0. 05). The tumor grow delay of groups with prolonged delivery time 30min was longer than that of groups with prolonged delivery time 60min P<0.05). There was no significant difference between groups with same delivery time(P>0. 05). Compared to group with single fraction of 18 Gy, the groups with prolonged delivery time shorten the mice survival time while there was no significant difference between the group with prolonged delivery time 30min and the group with prolonged delivery time 60min.Conclusion: The prolonged delivery time with same radiation dose shorten the tumor growth delay and survival time in the mice implanted with Lewis lung cancer. Longer the delivery time prolonged, the effect on tumor control was more significant. However the different prolonged delivery time 30min and 60min had no significant difference of the effect on mice survival time.