RESUMO
OBJECTIVES: To determine whether radiomics features derived from diffusion-weighted imaging (DWI) and arterial spin labeling (ASL) can improve the differentiation between radiation-induced brain injury (RIBI) and tumor recurrence (TR) in glioma patients. METHODS: A total of 4199 radiomics features were extracted from conventional MRI, apparent diffusion coefficient (ADC), and cerebral blood flow (CBF) maps, obtained from 96 pathologically confirmed WHO grade 2~4 gliomas with enhancement after standard treatment. The intraclass correlation coefficient (ICC) was used to test segmentation stability between two doctors. Radiomics features were selected using the Mann-Whitney U test, LASSO regression, and RFE algorithms. Four machine learning classifiers were adopted to establish radiomics models. The diagnostic performance of multiparameter, conventional, and single-parameter MRI radiomics models was compared using the area under the curve (AUC). The models were evaluated in the subsequent independent validation set (n = 30). RESULTS: Eight important radiomics features (3 from conventional MRI, 1 from ADC, and 4 from CBF) were selected. Support vector machine (SVM) was chosen as the optimal classifier. The diagnostic performance of the multiparameter MRI radiomics model (AUC 0.96) was higher than that of the conventional MRI (AUC 0.88), ADC (AUC 0.91), and CBF (AUC 0.95) radiomics models. For subgroup analysis, the multiparameter MRI radiomics model showed similar performance, with AUCs of 0.98 in WHO grade 2~3 and 0.96 in WHO grade 4. CONCLUSION: The incorporation of noninvasive DWI and ASL into the MRI radiomics model improved the diagnostic performance in differentiating RIBI from TR; ASL, especially, played a significant role. KEY POINTS: ⢠The multiparameter MRI radiomics model was superior to the conventional MRI radiomics model in differentiating glioma recurrence from radiation-induced brain injury. ⢠Diffusion and perfusion MRI could improve the ability of the radiomics model in predicting the progression in patients with glioma. ⢠Arterial spin labeling played an important role in predicting glioma progression using radiomics models.
Assuntos
Lesões Encefálicas , Neoplasias Encefálicas , Glioma , Humanos , Neoplasias Encefálicas/diagnóstico por imagem , Neoplasias Encefálicas/radioterapia , Neoplasias Encefálicas/patologia , Marcadores de Spin , Recidiva Local de Neoplasia/diagnóstico por imagem , Glioma/diagnóstico por imagem , Glioma/patologia , Imagem de Difusão por Ressonância Magnética/métodos , Imageamento por Ressonância Magnética , Estudos RetrospectivosRESUMO
Background: Tumor recurrence and pseudoprogression (PsP) have similar imaging manifestations in conventional magnetic resonance imaging (MRI), although the subsequent treatments are completely different. This study aimed to evaluate the value of perfusion-weighted imaging (PWI) in differentiating PsP from glioma recurrence. Methods: A comprehensive literature search was performed to evaluate clinical studies focused on differentiating recurrent glioma from PsP using PWI, including dynamic susceptibility contrast MRI (DSC-MRI), dynamic contrast enhanced MRI (DCE-MRI), and arterial spin labeling (ASL). Study selection and data extraction were independently completed by two reviewers. The Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tool was applied to evaluate the quality of the included studies. The software Stata 16.0 and Meta-Disc 1.4 were used for the meta-analysis. Meta-regression and subgroup analyses were applied to identify the sources of heterogeneity in the studies. This study was registered in the International Prospective Register of Systematic Reviews (PROSPERO) prior to initiation (CRD42022304404). Results: A total of 40 studies were included, including 27 English studies and 13 Chinese studies. There were 1,341 patients with glioma recurrence and 876 patients with PsP. The pooled sensitivity and specificity of DSC-MRI for differentiating glioma recurrence from PsP were 0.82 [95% confidence interval (CI): 0.78 to 0.86] and 0.87 (95% CI: 0.80 to 0.92), respectively. The pooled sensitivity and specificity of DCE-MRI were 0.83 (95% CI: 0.76 to 0.89) and 0.83 (95% CI: 0.78 to 0.87), respectively. The pooled sensitivity and specificity of ASL were 0.80 (95% CI: 0.73 to 0.86) and 0.86 (95% CI: 0.76 to 0.92), respectively. Discussion: The DSC-MRI, DCE-MRI, and ASL perfusion techniques displayed high accuracy in distinguishing glioma recurrence from PsP, and DSC-MRI had a higher diagnostic performance than the other two techniques. However, due to the diversity of the parameters and threshold differences, further investigation and standardization are needed.
RESUMO
BACKGROUND: Diffuse midline glioma (DMG), H3 K27M mutant, occurs in both adult and pediatric populations. The characteristics of the 2 DMG groups were systematically explored in this study. METHODS: H3 K27M-mutant DMG was diagnosed in 116 patients at Beijing Tiantan Hospital from May 2016 to December 2018 who were included in our study. Patients were classified into an adult group (n = 57; 49.1%) and a pediatric group (n = 59; 50.9%). Clinical, radiological, and molecular features were compared between the groups. Univariate and multivariate analyses were performed to identify prognostic factors. RESULTS: Compared with the adult group, pediatric patients had a younger age (8.9 ± 4.1 y vs 35.1 ± 11.8 y, P < 0.001), a lower preoperative Karnofsky performance scale score (62.9 ± 15.5 vs 72.1 ± 16.5, P = 0.004), a lower rate of total resection (5.7% vs 26.8%, P = 0.009), a larger tumor size (4.4 ± 0.9 vs 3.9 ± 1.5 cm, P = 0.045), a higher Ki-67 index (63.0% vs 37.8%, P = 0.047), and higher rates of postoperative cranial nerve palsy (61.0% vs 36.8%, P = 0.009) and ataxia (45.8% vs 26.3%, P = 0.029). Adult DMG was located predominantly in the thalamus, while the predilection site for pediatric DMG was brainstem (P < 0.001). Kaplan-Meier plot showed that the median survival of adult and pediatric DMG was 16.0 (9.7-22.3) months and 10.0 (8.3-11.7) months, respectively, which imparted a significant difference (P = 0.008). Age at diagnosis, radiotherapy, and motor deficit were confirmed as independent prognostic factors according to the multivariate analysis (P < 0.05). CONCLUSION: Compared with adult patients, children with H3 K27M-mutant DMG confer distinct clinical, radiological, and molecular characteristics and have a dismal prognosis. Radiotherapy is an independent factor associated with prolonged survival.