Predicting Tumor Recurrence Of Astrocytoma By Ki-67 And Proton Magnetic Resonance Spectra
Jieting Fu,Jiner Shu,Xinfeng Yu,Jiangfeng Pan,Guanghong Lv,Shaobin Zhou,Qiaosheng Jiang,Bilian Zhu,Risheng Yu
2017-01-01
Abstract:Background: Astrocytoma is the most common primary cerebral tumor and prediction of its prognosis is crucial. In the study, we aimed to investigate the correlation between the proton magnetic resonance spectroscopy (MRS) spectra of brain astrocytoma and Ki-67 expression and their clinical value in predicting tumor recurrence in astrocytoma after surgery. Methods: 73 patients with astrocytoma underwent surgery were retrospectively included from January 2012 to January 2015. Ki-67 expression was examined by immunohistochemical staining. The ratios of tumor parenchyma choline (Cho)/N-acetylaspartate (NAA), Cho/creatine (Cr), NAA/Cr, and occurrence rates of lactate (Lac) and lipid (Lip) signals were obtained from MRS and compared with Ki-67 expression. The prognostic value of the metabolite ratios and Ki-67 for astrocytoma was analyzed. Results: Ki-67 expression of astrocytoma was increasing significantly with the ratios of Cho/Cr (r=0.305, P=0.009) and Cho/NAA (r=0.468, P=0.000) but decreasing with NAA/Cr (r=-0.480, P=0.000). Correlation analyses suggested that Ki-67 associated with grade and recurrence, Cho/Cr associated with age and recurrence, Cho/NAA associated with grade and recurrence, NAA/Cr associated with grade, Lac associated with age and recurrence, and Lip associated with grade and recurrence. Multivariate analyses revealed that astrocytoma grade and metabolite ratios Cho/Cr and Lac were independent prognostic factors for recurrence time (Grade, HR=0.337, 95% CI=0.135-0.846; Cho/Cr, HR=0.348, 95% CI=0.145-0.839; Lac, HR=0.457, 95% CI=0.228-0.916). Conclusions: MRS data of astrocytoma were associated with Ki-67 expression and can be used to predict tumor recurrence after tumor surgery, thereby helping clinicians make reasonable treatment decisions for astrocytoma patients.