Brain alterations detected by multiplex magnetic resonance imaging in nasopharyngeal carcinoma patients after induction chemotherapy: a longitudinal study
Dan Liu, Kai Yang, Panpan Dong, Zien Chen, Shuangquan Ai, Xiaopeng Song, Tao You, Jianping Bi, Zilong Yuan, Xiaofang Guo, Yulin LiuBackground
Alterations in brain regions and cognitive changes in patients with nasopharyngeal carcinoma (NPC) after induction chemotherapy (ICT) remain overlooked. This study aimed to investigate brain alterations and their cognitive correlations using a novel Multiplex (MTP) magnetic resonance imaging (MRI) sequence.
Methods
Thirty-nine patients underwent MTP imaging and Montreal Cognitive Assessment (MoCA) test before (pre-ICT) and after (post-ICT) ICT. Paired-samples permutation test determined the differences in quantitative MRI parameters of brain regions between pre- and post-ICT assessments. Multiple linear regression analysis, adjusted for age, sex, and education, examined the relationship between longitudinal changes in regional parameters and MoCA scores. Analysis of covariance, controlling for baseline MoCA scores and demographic variables, compared the parameters of longitudinally altered brain regions between MoCA-decreased (MoCA-dec) and MoCA-non-decreased (MoCA-nd) groups. Multivariate logistic regression and receiver operating characteristic curve analysis evaluated the combined performance of clinical variables and brain regional parameters in predicting cognitive impairment.
Results
Post-ICT patients exhibited decreased volume and R2*, as well as increased T2* and quantitative susceptibility mapping values, predominantly in the default mode network, occipital lobe, temporal lobe, and subcortical regions. Baseline R2* values in the right superior frontal gyrus, left cuneus, and left lingual gyrus showed significant false discovery rate (FDR)-corrected negative associations with post-ICT naming subscores. The MoCA-dec group exhibited significantly larger right amygdala volumes both before and after ICT, and smaller left inferior temporal gyrus (ITG.L) volumes after ICT compared with the MoCA-nd group. A combined model of sex, baseline MoCA score, and post-ICT ITG.L volume achieved a preliminary and exploratory area under the curve (AUC) of 0.706 for predicting cognitive decline.
Conclusions
MTP demonstrates potential for evaluating brain alterations in NPC following ICT, and these multiparametric indices showed preliminary associations with cognitive decline. However, these findings should be interpreted as exploratory, as they were derived from a cohort without a non-ICT control group. In addition, the single post-ICT time point cannot determine whether the observed alterations are transient, persistent, or progressive. Consequently, this exploratory study provides neurobiological insights into ICT-related cognitive changes and generates hypotheses regarding the role of MTP-derived parameters as imaging biomarkers in future studies.