Diagnostic Accuracy of Dynamic Contrast-Enhanced Magnetic Resonance Imaging for Distinguishing Pseudoprogression from Glioma Recurrence: A Meta-Analysis
Introduction
Gliomas represent the most prevalent primary brain tumors, constituting approximately 80% of all malignant brain tumors and 30% of all central nervous system tumors. The standard treatment for gliomas typically involves surgical intervention, such as gross total or subtotal excision, followed by concomitant chemoradiotherapy and temozolomide adjuvant chemotherapy. However, this treatment regimen can lead to radiation-induced damage to brain tissue, increasing the risk of recurrence. Pseudoprogression is a sub-acute clinical phenomenon characterized by the expansion of existing lesions or the appearance of new lesions within 12 weeks after radiation therapy. Unlike true tumor progression, pseudoprogression lesions stabilize or shrink without further treatment. The occurrence rate of pseudoprogression in glioma patients ranges from 15% to 60%. Pseudoprogression lesions often exhibit contrast enhancement on MRI or CT, similar to tumor progression, making differentiation challenging.
Given the entirely different prognoses and treatment strategies for glioma recurrence and pseudoprogression, accurate differentiation is crucial. This study aims to assess the diagnostic accuracy of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) in distinguishing glioma recurrence from pseudoprogression.
Methods
A comprehensive literature search was conducted using PubMed, Embase, Cochrane Library, and Chinese biomedical databases up to May 1, 2019. The search strategy included terms related to glioma, DCE-MRI, and pseudoprogression. Only English-language articles were included. The quality of the included studies was evaluated using the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) tool. Data extraction included study characteristics, patient demographics, tumor treatment details, and DCE-MRI parameters. Statistical analysis was performed using the “mada” package in R, with heterogeneity assessed using Cochran Q-statistic and I² statistic. Pooled sensitivity, specificity, and diagnostic odds ratio were calculated, and summary receiver operating characteristic (SROC) curves were generated. Publication bias was evaluated using funnel plots and Deek’s test.
Results
The meta-analysis included 11 studies involving 616 patients. The pooled sensitivity and specificity of DCE-MRI for distinguishing glioma recurrence from pseudoprogression were 0.792 (95% CI 0.707–0.857) and 0.779 (95% CI 0.715–0.832), respectively. The diagnostic odds ratio was 16.219 (97.5% CI 9.123–28.833), indicating that DCE-MRI significantly improves the likelihood of accurate differentiation. The area under the SROC curve was 0.846, suggesting high diagnostic efficiency. Most studies showed high sensitivities (>0.6) and low false positive rates (<0.5). However, significant heterogeneity (I² = 77.5%) and publication bias were observed.
Discussion
DCE-MRI has emerged as a valuable tool in tumor diagnosis, offering insights into tissue microcirculation through pharmacokinetic parameters such as the extravascular extracellular space per unit volume of tissue (Ve), the rate transfer constant (Kep), the blood plasma volume per unit volume of tissue (Vp), and the volume transfer constant (Ktrans). Previous studies have indicated that these parameters differ significantly between pseudoprogression and true tumor progression, providing a basis for differentiation.
This meta-analysis demonstrates that DCE-MRI can effectively differentiate glioma recurrence from pseudoprogression, with high sensitivity and specificity. The diagnostic odds ratio of 16.219 underscores the clinical utility of DCE-MRI in this context. The SROC curve further supports the diagnostic accuracy, with an AUC value of 0.846 indicating strong diagnostic performance.
Despite these promising results, the study has several limitations. The heterogeneity among the included studies was significant, which may affect the reliability of the findings. Most studies were retrospective, and the sample sizes were relatively small, potentially introducing bias. Additionally, variations in patient characteristics, such as age and treatment regimens, could contribute to heterogeneity. The exclusion of non-English publications may also have led to publication bias.
Future research should focus on larger, prospective studies to validate these findings. Combining DCE-MRI with other imaging modalities, such as dynamic susceptibility contrast (DSC)-MRI, may further enhance diagnostic accuracy. Advanced MRI techniques, including diffusion-weighted imaging (DWI) and magnetic resonance spectroscopy (MRS), could also be integrated into a comprehensive diagnostic approach.
Conclusion
DCE-MRI shows significant potential in improving the diagnostic accuracy of distinguishing glioma recurrence from pseudoprogression. While the technique is not without limitations, it offers valuable insights that can guide clinical decision-making. Further research, particularly larger prospective studies and the integration of multiple imaging modalities, is essential to establish a robust diagnostic framework for glioma patients.
doi.org/10.1097/CM9.0000000000001445
Was this helpful?
0 / 0