In this study, we assessed the importance of the identification of imaging necrosis in the preoperative evaluation of glioma. We found strong agreement between observations of Imnecrosis and Panecrosis. Moreover, Imnecrosis was found to be significantly related with glioma-related key gene mutations, such as 1p19q non-codeletion and CDKN2A/B homozygous deletion. And it is an independent imaging marker for predicting tumor prognosis.
Our study indicated that there was strong agreement between the inter-observer agreement of Imnecrosis and Panecrosis. And during the analysis, we found that the regions with an absence or marked decrease of enhancement inside the intensified areas was easily mistook as Imnecrosis. To be mentioned that there was one WHO grade 2 IDH-mutant astrocytomas with an very short OS (5 months) who died from glioma-related causes. When we reviewed the raw data and identified that this patient had a small extent of Imnecrosis, which indicated a high grade gliomas. And we deduced that Panecrosis was unluckily missed due to the limited tumor specimen. Considering that pathological samples were partial while imaging observation can capture full tumors, above mentioned situation can be avoided if to made a judgement of Imnecrosis, which was exactly one unique advantage for radiographic examination. In comparison, we also found seven patients with Imnecrosis were diagnosed as WHO grade 2 or 3 oligodendrogliomas, indicated that necrosis plays a limited prognostic value in oligodendrogliomas. Hence, if there is evidence of oligodendroglioma, such as calcification and filiform or localized internal homogeneous enhancement, presence of imnecrosis did not indicated a high-grade tumor.
Previous studies have highlighted that Imnecrosis shown by pre-operative MRI is an independent unfavorable prognosis factor [5, 6, 10, 19–21]. Our results are in accordance, and they showed that patients with Imnecrosis had a poorer prognosis than those without Imnecrosis. Beside, the latest WHO CNS classification emphasizes the role of molecular markers in the diagnosis and prognosis of gliomas, such as IDH, 1p19q, CDKN2A/B, 7+/10-, and EGFR [4]. Consistent with previous findings [16–18] and our results demonstrated that Imnecrosis is less likely to occur in gliomas with IDH-mutation, 1p19q codeletion, or CDKN2A/B homozygous deletion; thus, Imnecrosis can be an imaging marker helpful for gene prediction. From this prospect, Imnecrosis might be more important than Panecrosis, since it can be non-invasively obtained before operation. However, there was no significant difference between the expressions of 7+/10- cytogenetic signature or EGFR amplification and the presence of Imnecrosis. This negative result might also due to the small sample size and insufficient number of events.
In this study, we also seeked for a quantitative metrics for indicating tumor necrosis. Our results revealed that, compared with tumor without Imnecrosis/Panecrosis, DCE-rerived parameters in tumor parenchyma were significantly higher in gliomas with Imnecrosis/Panecrosis. And ve in tumor parenchyma demonstrated the highest diagnostic efficiency in identifying tumor necrosis with high sensitivity and specificity. This may be attributed to the fact that gliomas grow uncontrollably fast, resulting in severe hypoxia and necrosis, and thus an extensively hyperpermeable vasculature is generated resulting in inadequate oxygen and supplements. The greater the levels of perfusion and permeability in the tumor tissue, the higher the ktrans and ve values, and the higher the degree of tumor malignancy [22–24]. Hence, DCE-MRI metrics, especially ve in tumor parenchyma (cut-off value: 0.17), might be used as a supplementary metric to the morphological observation for delinating of tumor necrosis.
The current study has some limitations. First, since evidence of pathological necrosis was obtained from pathology reports of our hospital, there may be a detection bias. However, this study based on clinical, real-world evidence can exactly address the current clinical deficits. Second, this is a single-center study, subgroups analysis (such as the + 7/–10 cytogenetic signature subgroup) had a small sample, which might resulting in insufficient power to reach definite conclusions. Further multicenter studies with large sample sizes will help improve the efficacy of Imnecrosis in predicting the expression of molecular markers and prognosis.