The Application Study Of PET-CT Image Fusion In Clinical Stage And Precision Radiotherapy Of Esophageal Carcinoma

Objective: To investigate the influence of PET-CT image in esophageal carcinoma staging and precision radiotherapy.Methods: Twenty patients with potentially radical radiotherapy and biopsy-proven esophageal cancer from September 2007 to November 2008 were included in this study that underwent PET-CT. All patients were taken esophagogram and esophagoscope. Fixed by a heat plastic device, each patient was examined by a combine PET-CT scan as same as the radiotherapy treatment position. The obtained images were transmitted to treatment planning system by local network and the image fusion was made by Syntegra software. The PET, CT and coregistered PET-CT images were then reviewed on a workstation (Pinnacle 8.0c, Syntegra Viewer). (1)The tumor lengths were compared measured by esophagogram, esophagoscope, CT and PET-CT image showed at 40% of the maximum standard uptake value (SUVmax) and standard uptake value cut-off method 2.5. Also the tumor transverse diameters were compared on CT and PET-CT image at 40% SUVmax and SUV2.5. (2) One of the aims was to oberserve the difference of conformed mediastinum lymph node and the distant metastasis on CT and PET-CT and restaging. (3) For each patient, three separate GTVs were defined, one based on CT data alone (GTV-CT) and the others based on combined PET-CT data at SUV2.5 and 40%SUVmax (GTV-2.5and GTV-40), respectively. With the same margin, CTV and PTV were generated and esophagoscopy and esophagogram were taken into account when defining the GTVs. All treatment plans were desioned by experienced physicist with the same conditions including beams, gantry angles and prescription dose but the MLC settings readjusted according to the prescription dose and clinical request optimization design. The three treatment plans for each patient were compared with dose volume histogram (DVH). The dose limiting normal tissues including lung, spinal cord and heart was also evaluated with DVH parameters. The assessed parameters about normal tissues included were the volume of lung receiving≥20Gy (V20), maximum spinal cord dose, and the mean heart dose (MHD), etal. Statistical analysis was performed using an SPSS statistical package (Version 13.0).Results: (1) Tumor lengths of esophagoscope, esophagogram , CT and PET-CT data at SUV2.5 and 40%SUVmax were: 5.33㎝, 5.38cm, 6.88㎝, 5.50㎝ and 5.00㎝ respectively. The lengths between CT and esophagoscope, CT and esophagogram, CT and 40%SUVmax had statistical significance, P=0.003 , 0.007 and 0.002 respectively. The diameters of tumor shown on CT, SUV2.5 and 40%SUVmax were4.05㎝, 3.38cm, 2.95㎝, respectively. The transverse diameters between CT and 40%SUVmax had statistical significance, P=0.002. (2) Thirty-one lymph nodes were assessed positive on CT image, 21 on the PET-CT image and 14 on both. PET-CT image excluded 17 positive lymph nodes assessed by CT; also confirmed 7 negative lymph nodes excluded by CT (5 of them located in supraclavicular lymph region and superior mediastinum). The addition of PET information altered the clinical stage in 6 patients. Five patients had distant metastatic disease negative on CT and one patient had supraclavicular nodal metastatic diagnosed on CT image but exclude by PET-CT image. (3) There were not significant difference between the dose of GTV, CTV and PTV in three separate treatment plans of CT, SUV2.5 and 40%SUVmax. The mean lung doses (MLD) were 1340.01cGy, 1244.70 cGy, and 1098.08cGy respectively. The V10 and V15 of the three plans were 45.48 %, 41.60%, 36.40% and 33.51 %, 30.49%, 27.19%, respectively. The NTCP of lung were 1.75%, 0.85%, 0.35%, respectively. MHD and the maximum dose of spinal cord were 2201.20cGy, 2188.50cGy, 1728.10cGy and 4284.86 cGy, 4179.33 cGy, 4181.58cGy, respectively. The results of ANOVA and Krustal Wallis H test showed that MLD,V10,V15 and NTCP had statistical significance, P=0.01,0.037,0.037,0.028, respectively, there were not significant difference between V20, V30, the maximum dose of spinal cord and MHD. Compared with GTV-ct and GTV-2.5, 5 patients increased and 13 patients decreased. The V20 and V30 of two groups were 24.91%, 22.78% and 14.10%, 12.34% respectively and mean lung dose were 1393.62cGy and 1326.62cGy, respectively.Conclusion: Tumor lengths measured on PET scan by a fixed SUV value of 2.5 and esophagogram had the optimal approximation. By the additional PET information, the clinical stage and treatment aims were alerted in some patients of esophageal carcinoma; Treatment plans based on PET image could decrease the MLD,V10,V15 and NTCP, but there were not significant difference between V20,V30,MHD and the maximum dose of spinal cord. The alteration of irradiation field led to the change of normal tissues'dosage.