Cortical Morphological Alterations And Gray Matter Volume Structural Covariant Networks Topology Changes In Children Diagnosed With Acute Lymphoblastic Leukemia After Chemotherapy

Objective:This study aimed to elucidate the changes in cortical morphological characteristics（including cortical thickness,sulcus depth,gyrification index and fractal dimension）of chemotherapy in children diagnosed with acute lymphoblastic leukemia（ALL）by surface-based morphometry（SBM）at the early stage of chemotherapy.And the correlation between cortical thickness changes and drugs dosage was evaluated.Graph theory analysis was used to evaluate the recombination of structural covariance network（SCN）in children diagnosed with acute lymphoblastic leukemia in complete remission on days 46-52 of chemotherapy at the network level by using cortex volume index in terms of both global and local network properties.Methods:Twenty-six children（7.07±3.09 years old,15 males and 11 females）with ALL diagnosed by bone marrow biopsy,respectively had once magnetic resonance imaging（MRI）scans（two times in total）of the head before the start of chemotherapy and after achieving complete remission as assessed on days 46-52 of chemotherapy from September 2020 to December 2022.The treatment plan for all patients followed The Chinese Children’s Cancer Group study ALL 2015/2020（CCCG-ALL-2015/2020）.The oral cumulative doses of dexamethasone（DXM）and intrathecal injection of DXM,methotrexate（MTX）and cytosine arabinoside（Ara-c）and the number of intrathecal injection treatments were collected.1.SBM analysis was carried out by CAT12 toolkit.Using the 3D-MPRAGE sequence,and using the CAT12 toolkit of SPM12 in MATLAB software,the cerebral hemisphere was segmented into 90 regions based on the automated anatomical labeling（AAL）90 template according to the DARTEL algorithm.The changes in cerebral cortex thickness,sulcus depth,gyrification index and fractal dimension in children diagnosed with ALL on the 46-52 days of complete remission were analyzed.Paired T-test（0.05）was used to statistically the cortical morphological changes with Family Wise Error（FWE）correction.The changes in cortex thickness（Δ）before and after chemotherapy in 90 brain regions were calculated.Spearman correlation analysis was used to calculate the correlation between theΔcortical thickness of each brain region and the oral DXM cumulative doses and the intrathecal injection of MTX,DXM and Ara-c cumulative doses.2.The gray matter volume of each 90 region was calculated.The correlation coefficient of the gray matter volume of each brain region was calculated by Pearson correlation analysis,and the correlation matrix was obtained.The global SCN parameters（including small-word topology,clustering coefficient,characteristic path length,global efficiency,local efficiency,assortativity,transitivity,and modularity）and local SCN parameters（including degree,degree distribution,betweenness and hubs）of the brain covariance network were calculated by graph theory.Statistical analysis was performed by 1000nonparametric replacement tests.0.05（two-tailed）was considered statistically significant,and a false discovery rate（FDR）was used to correct the local network attributes.Results:1.When complete remission was achieved on days 46-52 of chemotherapy,the cortical thickness of multiple brain regions in children diagnosed with ALL became extensively thinner,mainly in bilateral frontal and parietal lobes.Shallow sulcus depth appeared in multiple brain regions,especially in the left frontal lobe,right parietal lobe and right insula lobe.The gyrification index was reduced in several brain regions,especially in the bilateral frontal lobe and right insula.The reduction of fractal dimension was found in the left inferior medial frontal gyrus.In this study,there were no statistically significant brain regions with increased cortical thickness,deepened sulci,increased gyrification index and increased fractal dimension after chemotherapy in children diagnosed with ALL.The changes in cortical thickness（Δ）of the right medial orbitofrontal cortex（r₁=0.61,₁=0.0010）,the right precuneus（r₂=0.56,₂=0.0028）,the right cuneus（r₃=0.55,₃=0.0038）,the right superior temporal gyrus（r₄=0.52,₄=0.0060）,the left precentral gyrus（r₅=0.45,₅=0.0215）and the right insula（r₆=0.41,₆=0.0398）in children diagnosed with ALL was positively linear correlated with the oral cumulative doses of DXM.TheΔcortical thickness of the right transversal temporal gyrus was positively correlated with the cumulative doses of MTX intrathecal injection（r₇=0.42,₇=0.0327）.2.On the 46-52 days after chemotherapy for children diagnosed with ALL in complete remission,the nodal degree of the local network parameter increased in the left supplementary motor area.And the nodal betweenness decreased in the left precuneus.Small-world topology existed in children diagnosed with ALL both before and after chemotherapy,and no statistically significant changes were found in small-world properties parameters and other global network parameters（including clustering coefficient,characteristic path length,global efficiency,local efficiency,assortativity,transitivity,and modularity）.Conclusion:On the 46-52 days of complete remission after chemotherapy,morphological changes in the cerebral cortex were already present in children diagnosed with ALL.Bilateral frontal and parietal lobes and the right insula may be the most susceptible brain regions.The oral cumulative doses of DXM and the cumulative doses of MTX intrathecal injection were positively correlated with the changes in cortical thickness in multiple brain regions.Higher doses of DXM and MTX may be one of the risk factors for cognitive impairment in children diagnosed with ALL.The nodal degree,betweenness and hubs of the SCN based on gray matter volume changed in children diagnosed with ALL on days 46-52 of chemotherapy in complete remission,which indicated that the topological properties of these children were reorganized and compensated after chemotherapy.That may reduce the communication efficiency of brain regions and change the information transmission path.The changes in network topological properties may be the potential network mechanism of cognitive impairment in children diagnosed with ALL after chemotherapy.The changes in cerebral cortex morphological and network topological attributes are expected to provide evidence for early diagnosis and intervention of chemotherapy induced cognitive impairment in children diagnosed with ALL.