| Part I: In vivo MR spectroscopic imaging of healthy rat brain and C6 glioma model with a 3.0T clinical human scannerPurpose: To explore the affected factors of MR spectroscopic imaging in rat brain, and the metabolic ratio measured in healthy rat brain and rat C6 glioma model on 3.0T MR unit. Materials and methods: 62 healthy male Sprague-Dawley rats, weighted 250-300g, were random divided into two groups: C6 glioma model group(32 cases) and control group(30 cases). C6 glioma model group had four subgroups(8 cases per subgroup) and all were injected with C6 cells in right candate nucleus. MR Spectroscopic imaging were performed with 2-D multi-voxel PRESS sequence using a GE signa VH/i 3.0T MR unit and rat-special coil, Functool software package was used for post-processing of spectrum. Results: 30 brain gliomas were developed in 32 rats after C6 cell implantation. The mean metabolic ratios in lesions of C6 glioma model were as follow: Cho/Cr 2.090±0.433, Cho/NAA 1.767±0.341, NAA/Cr 1.059±0.189, Lip0.9/Cr 0.761±0.347, IL/Cr 2.700+1.661, mI/Cr 1.506±0.568, and Glx/Cr 0.923±0.382; The mean metabolic ratios in right caudate nucleus of healthy rat brain were as follow: Cho/Cr 0.957±0.179,Cho/NAA 0.833+0.274,NAA/Cr 1.195±0.243, Lip0.9/Cr 0.432±0.170, IL/Cr 0.480±0.260, mI/Cr 1.079±0.314, and Glx/Cr 0.640±0.162. optimize the voxel location, shimming and saturation band is very important for MR spectroscopic imaging. Conclusion: It is feasible to do MR spectroscopic imaging in rat brain with 3.0T MR and the metabolic ratio measured in healthy rat brain and C6 glioma model can be used for reference in future studies.Part II: In vivo MR spectroscopic imaging characteristics of rat C6 glioma model -correlation with quantitative histomorphology in specimenPurpose: To study semiquantitative metabolic changes measured by MR spectroscopic imaging in different zones of C6 glioma model,and investigate the correlations betweenthe semiquantitative metabolic data and the Ki-67 labeling index and quantitative histomorphologic data. Materials and methods: 62 healthy male Sprague-Dawley rats, weighted 250-300g, were random divided into two groups: C6 glioma model group(32 cases) and control group(30 cases). C6 glioma model group had four subgroups(8 cases per subgroup).MR Spectroscopic imaging were performed with 2-D multi-voxel PRESS sequence using a GE signa VH/i 3.0T MR unit and rat-special coil, Functool software package was used for post-processing of spectrum. Voxels were selected at the solitary tumor zone, peritumoral zone and contralateral normal brain for C6 glioma model group, for control group, voxels were located at right caudate nucleus. Nuclear area,nuclear roundness,nuclear density and Ki-67 labelling index were analyzed with Image-Pro PLUS software Using Ki-67 immunostainde paraffin setions. Be assure that MRSI, Ki-67 labelling index and quantitative histomorphologic data were performed in the same areas of C6 glioma model, and examined the relationships of MRSI findings, Ki-67 labelling index and survival. Results: for C6 glioma model group, ①significant difference can be observed in Cho/Cr, Cho/NAA, LL/Cr between solitary tumor zone, peritumoral zone and contralateral normal brain, ②difference can be observed in NAA/Cr between solitary tumor zone and peritumoral zone,contralateral normal brain.③significant difference can be observed in Lip0.9/Cr between solitary tumor zone and peritumoral zone,contralateral normal brain.④significant difference can be observed in Glx/Cr between solitary tumor zone and contralateral normal brain.⑤a positive correlation between Ki-67 labelling index and Cho/Cr,Cho/NAA was obserbed,but no correlation between Ki-67 labeling index and NAA/Cr. No correlation was found between nuclear area, nuclear roundness,nuclear density and metabolic ratios.⑥In Kaplan-Meier survival analysis, significant difference was observed between high (Cho/Cr= 2.0) and low (Cho/Cr<2) Cho/Cr ratio groups. Conclusion: The study verify a significant correlation between MR spectroscopic imaging data acquired at solitary tumor zone of C6 glioma model and Ki-67 labeling index, supporting that preoperative MR spectroscopic imaging could provide additional information on cell proliferation and the prognosis.Part III: Application of MR spectroscopic imaging in radiation therapy of rat C6glioma modelPurpose: To explore the value of MR spectroscopic imaging in radiation therapy ofastrocytic tumors. Materials and methods: 12 healthy male Sprague-Dawley rats, weighted 250-300g, and all were injected with C6 cells in right candate nucleus. MR Spectroscopic imaging were performed with 2-D multi-voxel PRESS sequence using a GE signa VH/i 3.0T MR unit and rat-special coil. The examinations were performed immediately prior to radiation therapy, 4 days,10 days, 14 days,and 20 days after radiation therapy. Functool software package was used for post-processing of spectrum. Voxels were selected at the solitary tumor zone. Results: in general, the metabolic raitos of serial MR spectroscopic imaging after radiation therapy decreases progressively vs. time of therapy. The Cho/Cr ratios immediately prior to radiation therapy differed significantly with that of 10 days, 14 days,and 20 days after radiation therapy;The Cho/NAA ratios immediately prior to radiation therapy differed significantly with that of 14 days and 20 days after radiation therapy; The NAA/Cr ratios immediately prior to radiation therapy differed significantly with that of 20 days after radiation therapy. For results of MR spectroscopic imaging immediately prior to radiation therapy and 4 days after radiation therapy, a positive correlation between changes of tumor volume and changes of Cho/Cr (r = 0.814,P=0.001 ) , LL/Cr (r = 0.720,P=0.008 ) , Glx/Cr ratio was obserbed. Conclusion: MR spectroscopic imaging is a valuable method of radiation therapy monitoring and can help to predict the radiation therapy response of astrocytic tumors. Part IV: Clinical value of pre-treatment MRSI for astrocytic tumors Purpose: To explore the MRSI characteristics of astrocytic tumors and the value of MRSI in preoperative grading and detecting peritumoral zone. Materials and methods: Two groups were included in this study: patient group and control group. MR Spectroscopic imaging were performed with 2-D multi-voxel PRESS sequence using a GE signa VH/i 3.0T MR unit, patient group including 53 cases aged 15-65 years(mean age 41±11 years),31 males and 22 females, of the 53 cases, 1 had pilocytic astrocytoma, 20 had diffuse astrocytomas, 14 had anaplastic astrocytomas and 18 had glioblastomas, all of which were proved by surgery or stereotactic biopsy. Voxels were selected at the solitary tumor zone, peritumoral zone and contralateral normal brain region. Control group including 26 healthy adult volunteers, voxels were located at centrum semiovale and parietal cortex. Results: 1.The Cho/Cr, Cho/NAA and NAA/Cr ratios in the same brain region using PRESS 144 and PRESS35 sequence had no significant difference; 2.althougheight channel phased-array head coil can offer higher SNR, the metabolic changes using eight channel phased-array head coil and quadrature head coil had no difference; 3.paramagnetic contrast agent induced reduce in the choline and creatine signal(no statistical significance), but had no differece to semiquantitative matabolic ratios; 4.significant difference of Cho/Cr,NAA/Cr and Glx/Cr can be obserbed between gray matter and white matter in healthy adult volunteers; 5.astrocytic tumors were characterized by increased Cho signal and decreased NAA signal,while Lipids were present in high-grade astrocytic tumors; 6.Cho/Cr ratio can help to difference between peritumoral zone and solitary tumor region; 7.The Cho/Cr,lip1.3/Cr and Glx/Cr ratios in solitary tumor region are very important markers for preoperative grading of astrocytic tumors. Conclusion: This study demonstrates the potential of MRSI to discriminate different regions of astrocytic tumors,and can help to detect peritumoral zone and preoperative grading of astrocytic tumors.Part V: MR spectroscopic imaging characteristics of astrocytic tumors- correlation with quantitative histomorphology in surgical specimenPurpose: To investigate the correlations between the semiquantitative metabolic data measured by preoperative MR spectroscopic imaging and the Ki-67 labelling index and quantitative histomorphologic data in astrocytic tumors. Materials and methods: 24 patients aged 20～65 years (mean age 44±13years) were studied,16 males and 8 females. MR Spectroscopic imaging were performed with 2-D multi-voxel PRESS sequence using a GE signa VH/i 3.0T MR unit, all cases were proved by navigated surgery or stereotactic biopsy: 9 diffuse astrocytomas, 4 anaplastic astrocytomas and 11 glioblastomas. Nuclear area, nuclear roundness, nuclear density and Ki-67 labelling index were analyzed with Image-Pro PLUS software Using Ki-67 immunostainde paraffin setions. Results: The mean values of Ki-67 labelling index,nuclear roundness, nuclear density and nuclear area were 6.80%±3.16%,1.368±0.164,8.70%±6.84% and 320±58 in diffuse astrocytomas, 9.88%±3.35%,1.465±0.232,12.68%±7.77% and 387±31 in anaplastic astrocytomas, 10.98%±5.94%,1.526±0.306,18.80%±12.39% and 405±88 in glioblastomas. For all the patients with tumor grade 2-4 according to WHO classification, a positive correlation between Ki-67 labelling index, nuclear area and Cho/Cr ratio was obserbed, as well as a positive correlation between nuclear density, nuclear area and Cho/NAA; Lipl.3/Cr ratio was positively correlated with nuclear roundness and nuclear area, this result indicatehigher Lipl.3/Cr ratio with increasing area and irregularity of nuclear. Nuclear density was positively correlated Lip0.9/Cr ration and Glx/Cr ration, while in glioblastoma, a positive correlation between Ki-67 labelling index and Cho/Cr ratio was observed, as well as a positive correlation between nuclear density and Glx/Cr raton and mI/Cr ratio. Conclusion: The study verify a significant correlation between preoperative MR spectroscopic imaging data and Ki-67 labelling index and quantitative histomorphologic data, supporting that preoperative MR spectroscopic imaging could provide accurate information on histological grade and cell proliferation in astrocytic tumors. Part VI: Clinical value of MR Spectroscopic imaging after postoperative radiation therapy of astrocytic tumorsPurpose: To study metabolic changes vs. dose and time of therapy in the acute and early delayed phase of postoperative radiation therapy patients detected with MR Spectroscopic imaging. Materials and methods: Two groups were included in this study: radiation therapy group and control group. MR Spectroscopic imaging were performed with 2-D multi-voxel PRESS sequence using a GE signa VH/i 3.0T MR unit. Radiation therapy group including 18 cases aged 9-67 years(mean age 46±14 years), of the 18 cases, 7 had glioblastomas,4 had anaplastic astrocytomas, 6 had diffuse astrocytomas and 1 had pilocytic astrocytoma, all of which were postoperative(total resected or partial resected) proved. Two voxels were examined, one located at the tumor bed(=60Gy area) and the second lacated at the controlateral normal brain(<40Gy area). All cases were irradiated to 60-68 Gy using fractioned radiation therapy and intensity modulation radiation therapy (IMRT). Control group including 18 healthy adult volunteers, voxel was located at right centrum semiovale. Results: MR Spectroscopic imaging in postoperative patients before radiation therapy revealed a statistically significant decrease in the NAA/Cr ratio and increases in the Cho/Cr, Cho/NAA, Lip0.9/Cr, Lip1.3/Cr, mI/Cr, and Glx/Cr ratios at the tumor bed(=60Gy area), and a statistically significant decrease in the NAA/Cr at the controlateral normal brain(<40Gy area). In the =60Gy area, the Cho/Cr ratio began to decrease at the 60% of total dose, and decreases progressively vs. time of therapy (the ratio decrease from 2.521 before radiation therapy to 1.810 at one month after radiotherpy). while in the <40Gy area, the Cho/Cr ratio rose slightly at the 60% of total dose, then returned to the former level. The NAA/Cr ratio decrease over time for both=60Gy area and <40Gy area, but the magnitude of variation was small(form 1.507 to 1.333). In the =60Gy area, the Lipl.3/Cr ratio began to decrease at the 60% of total dose, and decreases gradually vs. time of therapy (the ratio decrease from 1.398 before radiation therapy to 0.830 at one month after radiotherpy), but still higher than normal value. Conclusion: MR Spectroscopic imaging is able to detect early metabolic changes in the acute and early delayed phase of postoperative radiation therapy patients, and can help to evaluate the early tumor response of astrocytic tumors to radiation therapy.
|
PDF Full Text Request