Research On Functional Image And Anatomical Image Guided Radiotherapy For Thoracic Carcinoma

PartⅠ. FDG PET/CT guided target definition of esophageal carcinoma1. The value of FDG PET/CT in the length determination of esophageal cancerObjective: To assess the value of FDG PET/CT in the length determination of primary esophageal cancer.Methods: Thirty-two patients underwent esophagoscope, esophagogram and FDG PET/CT scans 1 week before esophagectomy. The location and length of primary tumor were determined by esophagoscope, esophagogram, FDG PET/CT respectively. The length of the abnormality seen on the CT portion of the PET/CT scan was determined independently by two separate investigators from the length of the lesion seen on PET/CT scan. All results were compared with surgical specimens respectively.Results: There were 1 tumor located in the upper thoracic esophagus, 22 in the middle thoracic esophagus, and 9 in the lower thoracic esophagus according to esophagram, CT and PET/CT respectively which were all confirmed by surgical specimen. At same time, the tumor location determined by esophagoscope were not accordance with surgical specimen in 2 patients. The mean length of the primary tumor were 3.797±1.447 cm, 4.140±1.460cm, 4.375±1.596cm, 5.328±1.856cm and 4.734±1.746cm, determined by esophagoscope, esophagogram, CT, PET/CT scan and surgical specimens respectively. Compared with surgical specimen, there was obviously difference, P value was 0.000, 0.007, 0.025, 0.001 respectively. Considered that there were some kind of shrinkage in resected specimens (about 83.59%, as reported by Ma et al), we rectified the length of surgical specimens and compared with others again. It was found that there was no difference between PET/CT and rectified surgical specimens (P=0.082).Conclusions: FDG PET/CT maybe a effective tool for the length determination of primary esophageal cancer, which maybe used to determine the length of resected esophagus for patients eligible for esophagectomy and accurate delineation of the gross tumor volume for patients eligible for radiotherapy.2. Additional value of PET/CT over PET in thoracic esophageal carcinomaObjective: To assess the value of combined FDG PET/CT compared with PET, for detection of lymph node metastases in esophageal carcinoma.Methods: From November 2003 to August 2005, 35 patients with esophageal carcinoma underwent FDG PET/CT before esophagectomy and lymph node dissection. The patients with prior anticancer treatment were excluded, as well as patients with diabetes mellitus, inflammatory lung disease, and inoperability for medical reasons. The results of PET/CT and PET for depiction of lymph node metastases were compared according to pathological proof.Results: There were 29 men and 6 women eligible, who ranged in age from 40 to 72 years (mean age, 57 years). All patients were operated successfully, 3 of them had upper thoracic esophageal cancer, 22 of them had middle thoracic esophageal cancer and 10 of them had tumor located in the lower third. The pathological examination confirmed metastases in 25 patients and 65 out of 313 excised nodal groups. True true positive and true negative interpretations reached 53 and 217 on PET in contrast to 61 and 229 on PET/CT. False negatives were found on PET in 12 lesions and 8 of them were corrected by PET/CT including 1 cervical nodal group, 4 paraesophageal nodal groups, 1 left gastric arterial lymph nodal group, 1 left cardia ventriculi lymph nodal group, 1 arcus minor ventriculi lymph nodal group. False positive interpretations on PET were found in 31 nodal group, 12 of them were corrected by PET/CT including 9 false-positive interpretations for physical tracer uptake (2 in the cervical region and 7 in gastrointestinal tract) and 3 foci with heterogeneous tracer uptake in the primary tumor which were incorrectly considered as local nodal involvement. The sensitivity, specificity and accuracy, respectively, of PET were 81.54% (53of 65 nodes), 87.50% (217of248), and 86.26% (270 of 313) whereas those of PET/CT were 93.85% (61 of 65), 91.24% (229 of 248) and 92.65% (290 of 313). P values were 0.033, 0.074 and 0.009, respectively. The differences of sensitivity and accuracy between PET/CT and PET were statistically significant.Conclusion: PET/CT improves the sensitivity and accuracy of FDG imaging in lymph node metastases of esophageal carcinoma and provides data of diagnostic significance. line profiles were measured for the phantom. Dose measurements were done with a 10 cm long pencil ionisation chamber placed at the isocentre.Results: A variation of less than 3HU is currently being seen across the images in the centre of the phantom and higher variation of 16 HU at the edges in the uniformity modular of phantom. A biggest variation of 22 HU is currently being seen between the images of cone beam CT and the standard value of phantom. This indicates a accurate and stationary correlation between the images of cone beam CT and the real electronic density of target substance. A excellent geometric accuracy is observed with a spatial linearity of pixel size verification less than 0.4% and consistently the 8 th set corresponding to a gap size of 0.63 mm is seen in the high resolution modular. It is possible to detect the circle with a diameter of 4mm at the 1% contrast level and the circle with a diameter in 5mm at the 0.5% contrast level, and a mean dose 2.97±0.19 cGy is measured at the isocenter. Limpid lesion images are available for patients with nasopharyngeal carcinoma and meningothelioma, and unavailable for patients with well differentiated glioma.Conclusions: The images of cone beam CT show a good performance in scan uniformity, electronic density, geometric accuracy, spatial resolution, low irradiation dose and limited low contrast resolution. Tumor and soft images are available by cone beam CT scan. Cone beam CT has been a effective tool for image guided radiotherapy.2.Initial image evaluation of cone beam CT in half fan modeObjective: Cone beam CT mounted on a linear accelerator is available for image guidance applications. In preparation for clinical implementation, phantom and patient imaging studies were conducted to evaluate the image quality of cone beam CT in half fan mode.Methods: This was tested by scanning the phantom and 5 patients with thoracic or abdominal carcinoma using a 2.5 mm slice reconstruction by cone beam CT, and area and line profiles were measured for the phantom. Dose measurements were done with a 10 cm long pencil ionisation chamber placed at the isocentre.Results: A variation of less than 18 Hu is currently being seen across the images in the uniformity modular of phantom, with a mean value of-13.8 Hu and standard deviation (noise) of 2.56 Hu. A biggest variation of 20 Hu is currently being seen across the images of half fan cone beam CT and the nominal value of target material of CT number linearity modular. This indicates a accurate and stationary correlation between the images of half fan cone beam CT and the real electronic density of target material. A excellent geometric accuracy is observed with a spatial linearity of pixel size verification less than 0.6% and consistently the 7th set corresponding to a gap size of 0.71 mm is seen in the high resolution modular. It is possible to detect the circle with a diameter of 7 mm at the 1% contrast level and the circle with a diameter in 9 mm at the 0.5% contrast level, and a mean dose 2.56±0.27 cGy is measured at the isocenter (in-air, in absence of phantom). Limpid lesion images are available for patients with lung carcinoma, and unavailable for patients with mediastinal lymph node metastases.Conclusions: The images of half fan cone beam CT show a good performance in scan uniformity, electronic density, geometric accuracy, spatial resolution, low irradiation dose and limited low contrast resolution. Thoracic or abdominal tumor and soft images are available by half fan cone beam CT scan. Half fan cone beam CT has been a effective tool for image guided radiotherapy of thoracic or abdominal carcinoma. 3. Effect of respiratory motion on KV X-ray cone beam CT image: a phantom studyObjective: To evaluate the effect of respiratory motion on imaging results of KV X-ray cone beam CT(CBCT) equipped on Varian 23EX linear accelerators.Methods: A phantom system to simulate respiratory motion was scanned by KV CBCT and images were analyzed. Results: The images of static phantoms were limpid and undeformed, and its magnitudes were in accordance with real size. The simulated respiratory motion had significant effects on CBCT images of the phantoms in both transverse axis and long axis. Significant decrease in image legibility and increase in deform and magnitude were observed when motion amplitude were increased. No significant effect on image results related to respiratory frequency change were observed In this study, all CBCT images showed accordance with the range of phantom movement (the magnitude of the phantom plus the motion amplitude).Conclusions: This phantom study indicates that CBCT image includes the data of respiratory motion.