Search

Purpose: To understand the tumor immune microenvironment precisely, it is important to secure the quantified data of tumor-infiltrating immune cells, since the immune cells are true working unit. We analyzed unit immune cell number per unit volume of core tumor tissue of high-grade gliomas (HGG) to correlate their immune microenvironment characteristics with clinical prognosis and radiomic signatures., Methods: The number of tumor-infiltrating immune cells from 64 HGG core tissue were analyzed using flow cytometry and standardized. After sorting out patient groups according to diverse immune characteristics, the groups were tested if they have any clinical prognostic relevance and specific radiomic signature relationships. Sparse partial least square with discriminant analysis using multimodal magnetic resonance images was employed for all radiomic classifications., Results: The median number of CD45 + cells per one gram of HGG core tissue counted 865,770 cells which was equivalent to 8.0% of total cells including tumor cells. There was heterogeneity in the distribution of immune cell subpopulations among patients. Overall survival was significantly better in T cell-deficient group than T cell-enriched group (p = 0.019), and T8 dominant group than T4 dominant group (p = 0.023). The number of tumor-associated macrophages (TAM) and M2-TAM was significantly decreased in isocitrate dehydrogenase mutated HGG. Radiomic signature classification showed good performance in predicting immune phenotypes especially with features extracted from apparent diffusion coefficient maps., Conclusions: Absolute quantification of tumor-infiltrating immune cells confirmed the heterogeneity of immune microenvironment in HGG which harbors prognostic impact. This immune microenvironment could be predicted by radiomic signatures non-invasively.

Glioblastoma remains the most devastating brain tumor despite optimal treatment, because of the high rate of recurrence. Distant recurrence has distinct genomic alterations compared to local recurrence, which requires different treatment planning both in clinical practice and trials. To date, perfusion-weighted MRI has revealed that perfusional characteristics of tumor are associated with prognosis. However, not much research has focused on recurrence patterns in glioblastoma: namely, local and distant recurrence. Here, we propose two different neural network models to predict the recurrence patterns in glioblastoma that utilizes high-dimensional radiomic profiles based on perfusion MRI: area under the curve (AUC) (95% confidence interval), 0.969 (0.903-1.000) for local recurrence; 0.864 (0.726-0.976) for distant recurrence for each patient in the validation set. This creates an opportunity to provide personalized medicine in contrast to studies investigating only group differences. Moreover, interpretable deep learning identified that salient radiomic features for each recurrence pattern are related to perfusional intratumoral heterogeneity. We also demonstrated that the combined salient radiomic features, or "radiomic risk score", increased risk of recurrence/progression (hazard ratio, 1.61; p = 0.03) in multivariate Cox regression on progression-free survival.

Adult-type diffuse glioma (grade 4) has infiltrating nature, and therefore local progression is likely to occur within surrounding non-enhancing T2 hyperintense areas even after gross total resection of contrast-enhancing lesions. Cerebral blood volume (CBV) obtained from dynamic susceptibility contrast perfusion-weighted imaging (DSC-PWI) is a parameter that is well-known to be a surrogate marker of both histologic and angiographic vascularity in tumors. We built two nnU-Net deep learning models for prediction of early local progression in adult-type diffuse glioma (grade 4), one using conventional MRI alone and one using multiparametric MRI, including conventional MRI and DSC-PWI. Local progression areas were annotated in a non-enhancing T2 hyperintense lesion on preoperative T2 FLAIR images, using the follow-up contrast-enhanced (CE) T1-weighted (T1W) images as the reference standard. The sensitivity was doubled with the addition of nCBV (80% vs. 40%, P = 0.02) while the specificity was decreased nonsignificantly (29% vs. 48%, P = 0.39), suggesting that fewer cases of early local progression would be missed with the addition of nCBV. While the diagnostic performance of CBV model is still poor and needs improving, the multiparametric deep learning model, which presumably learned from the subtle difference in vascularity between early local progression and non-progression voxels within perilesional T2 hyperintensity, may facilitate risk-adapted radiotherapy planning in adult-type diffuse glioma (grade 4) patients. [ABSTRACT FROM AUTHOR]