Catherine Daumas-Duport

Stereotactic Histologic Correlations of Computed Tomography- and Magnetic Resonance Imaging-Defined Abnormalities in Patients With Glial Neoplasms

In 39 patients who harbored previously untreated astrocytomas (21 patients), oligoastrocytomas (9 patients), or oligodendrogliomas (9 patients), computed tomographic (CT) and magnetic resonance imaging (MRI) findings were correlated with stereotactic serial biopsy findings. The 39 patients were classified as having one of three types of tumor: type I (1 patient), which consisted only of circumscribed tumor tissue; type II (26 patients), which consisted of tumor tissue and isolated tumor cells; or type III (11 patients), which consisted of intact parenchyma infiltrated by isolated tumor cells. (In one patient, the biopsy sampling was inadequate for determining the type of tumor.) In high-grade lesions, tumor tissue was obtained from CT contrast-enhancing regions, and the area of enhancement accurately defined the tumor tissue volume. In low-grade lesions, tumor tissue was hypodense and indistinguishable from parenchyma infiltrated by isolated tumor cells on both CT and MRI. Isolated tumor cells usually extended as far as the prolongation of T2 on T2-weighted MRI of high-grade and low-grade tumors. CT and MRI detection of boundaries and stereotactic serial biopsies are necessary for the demarcation of glial neoplasms into tumor tissue and isolated tumor cell volumes as well as for the determination of the spatial extent of each component. This information is important for determining appropriate treatment.

Dysembryoplastic neuroepithelial tumours.

Dysembryoplastic neuroepithelial tumours (DNTs) are a group of supratentorial cortical benignant lesions that superficially resemble mixed oligoastrocytomas, oligodendrogliomas or astrocytomas. Clinically these tumours are associated with partial seizures beginning before the age of 20 years, with no neurologic deficit and no stigmata of phacomatosis. In the revised WHO classification, DNTs have been incorporated among the category of neuronal and mixed neuronoglial tumours. This classification describes a histologic variant characterized by the following criteria: cortical location, multinodular architecture ‐ the nodule being made of multiple variants looking like astrocytomas, oligodendrogliomas or oligo‐astrocytomas, foci of dysplastic cortical disorganization and the presence of a glioneuronal element showing a columnar structure perpendicular to the cortical surface. A study of 14 cases for which only a specific glioneuronal element could be identified demonstrated that this specific element is sufficient for diagnosing DNTs and that the spectrum of DNTs includes a simple form with a unique glioneuronal element. Preoperative imaging follow‐up data, in the series of 23 simple and complex forms, indicated that DNTs are perfectly stable. However, these tumours may show a high MIB1 labeling index.

ABC transporters, cytochromes P450 and their main transcription factors: expression at the human blood–brain barrier

We have established the expression patterns of the genes encoding ATP‐binding cassette (ABC) transporters and cytochromes P450 (CYPs) at the adult human blood–brain barrier (BBB) using isolated brain microvessels and cortex biopsies from patients with epilepsia or glioma. Microves synaptophysin (neurons) and neuron‐glial antigen 2 (NG2) (pericytes). ABCG2 [breast cancer resistance protein (BCRP)] and ABCB1 (MDR1) were the main ABC transporter genes expressed in microvessels, with 20 times more ABCG2 and 25 times more ABCB1 in microvessels than in the cortex. The CYP1B1 isoform represented over 80% of all the CYPs genes detected in microvessels. There were 14 times more CYP1B1 in microvessels than in the cortex, showing that CYP1B1 is mainly expressed at the BBB. p‐glycoprotein (ABCB1), BCRP (ABCG2) and CYP1B1 proteins were found in microvessels by western blotting. The expression of genes encoding three transcription factors [pregnane xenobiotic receptor (PXR), constitutive androstane receptor (CAR), aryl hydrocarbon receptor (AhR)] was also investigated. The AhR gene, involved in the regulation of CYP1B1 expression, was highly expressed in brain microvessels, whereas PXR and CAR genes were almost undetected. This detailed pattern of ABC and CYPs gene expression at the human BBB provides useful information for understanding how their substrates enter the brain.

Oligodendrogliomas. Part I : Patterns of growth, histological diagnosis, clinical and imaging correlations : A study of 153 cases

The present study has attempted to demonstrate that the morphological spectrum of oligodendrogliomas includes tumors which are traditionally misinterpreted as ‘diffuse fibrillary astrocytoma’. We have shown that these tumors are in fact made of isolated neoplastic oligodendrocytes which are entrapped in a fibrillary background composed of axons and fibrillary reactive gliosis. Analysis in a series of 153 ‘pure’ supratentorial oligodendrogliomas composed of ‘classical’ or pseudo ‘diffuse fibrillary oligodendrogliomas’ diagnosed by imaging-based serial stereotactic biopsies showed that 2/3 of the tumors were exclusively made of isolated tumor cells (ITCs) (structure type III) and that only 1/3 of them exhibited both ITCs and solid tumor tissue components (structure type II). The tumor tissue destroys the brain parenchyma and contains new formed microblood vessels whereas ITCs do not destroy the parenchyma and are not associated with microangiogenesis. These fundamentally opposite morphological characteristics were reflected by the following findings: 1) contrast enhancement was observed in 64% of structure type II but was never seen in structure type III oligodendrogliomas. 2) a neurological deficit occurred in 57% of structure type II but in only 8% of structure type III oligodendrogliomas. 3) using the new grading system described in the companion paper to this study, we found that the biological behavior of oligodendrogliomas was also closely related to the patterns of tumor growth. From a synthesis of data gathered in this study it is suggested that emergence of microangiogenesis within a tumor which at first grows slowly with a structure type III pattern is a crutial event toward more aggressive behavior.

Transcriptomic and Quantitative Proteomic Analysis of Transporters and Drug Metabolizing Enzymes in Freshly Isolated Human Brain Microvessels

We have investigated the transcriptomic and/or proteomic patterns of 71 solute carrier (SLC) and organic solute (OST) transporters, 34 ATP-binding cassette (ABC) transporters, and 51 metabolizing enzymes in human brain microvessels. We used quantitative RT-PCR and LC-MS/MS to examine isolated brain microvessels and cortex biopsies from 12 patients with epilepsia or glioma. SLC2A1/GLUT1, SLC1A3/EAAT1, and SLC1A2/EAAT2 were the main SLC proteins whereas ABCG2/BCRP, ABCB1/MDR1, ABCA2 and ABCA8 were the main ABC quantified in isolated brain microvessels; ABCG2/BCRP was 1.6-fold more expressed than ABCB1/MDR1, and ABCC4/MRP4 was 10 times less abundant than ABCB1/MDR1. CYP1B1 and CYP2U1 were the only quantifiable CYPs. Finally, GSTP1, COMT, GSTM3, GSTO1 and GSTM2 proteins were the main phase II enzymes quantified; UGTs and NATs were not detected. Our extensive investigation of gene and protein patterns of transporters and metabolizing enzymes provides new molecular information for understanding drug entry and metabolism in the human blood-brain barrier.

Dysembryoplastic neuroepithelial tumors: nonspecific histological forms -- a study of 40 cases.

Objective. To demonstrate that DNTs include a large morphological spectrum of tumors that cannot be histologically distinguished from conventional categories of gliomas.Methods. All tumors from patients who underwent epilepsy surgery in Sainte-Anne hospital (Paris) that histologically resembled gliomas and did not conform to current histological criteria for DNTs or gangliogliomas were entered in the study.Results. According to the WHO histological classification, the 40 tumors resembled: pilocytic astrocytomas (4 cases), astrocytomas (16 cases), anaplastic astrocytoma (1 case), oligodendrogliomas (10 cases), oligo-astrocytomas (8 cases) or anaplastic oligo-astrocytomas (1 case). However foci of cortical dysplasia could be observed in 47% of the cases. Clinical presentation and imaging features were strikingly similar to that observed in typical DNTs. Although surgical removal was incomplete in 28% of the cases and none of the patients underwent chemo or radiotherapy, none of the tumors recurred (mean follow-up: 7 years). Moreover, serial preoperative imaging in 26 patients (mean follow-up: 4.5 years) demonstrated that these lesions were perfectly stable.Conclusions. Whatever the histological appearance of a glial tumor, the diagnosis of DNT must be considered when all the following criteria are associated: (1) partial seizures, with or without secondary generalization, beginning before the age 20 years, (2) no neurological deficit or stable congenital deficit, (3) cortical topography of the lesion as better demonstrated by MRI and (4) no mass effect on imaging.

Oligodendrogliomas. Part II: A new grading system based on morphological and imaging criteria

This second part of our study of ‘pure’ oligodendrogliomas focuses on survival data analysis. In order to identify potentially useful prognostic factors and to assess the effectiveness of a new grading system, the 79 patients in the previously analyzed series for whom adequate follow-up could be obtained (52%) were entered in the present analysis. Statistical analysis demonstrated that contrast enhancement and endothelial hyperplasia had powerful and similar influence on survival. Median survival with and without contrast enhancement were: 3 versus 11 years, and with or without endothelial hyperplasia were: 3.5 versus 11 years. Conversely, the degree of nuclear atypia and presence or absence of mitosis or necrosis were not correlated with survival. These findings allowed us to devise a simple grading system which discriminates two malignancy grades as follows: absence of endothelial hyperplasia and of contrast enhancement=Grade A, presence of endothelial hyperplasia and/or of contrast enhancement=Grade B. Of the 79 oligodendrogliomas in this study, 59 tumors were categorized as grade A and 20 as grade B. Median survival were: 11 years in grade A and 3.5 years in grade B. Five-year and 8-year survival rates were: 89% and 60% in grade A and: 33% and 15% in grade B. Double blind grading between two independent observers was concordant in 96% of the cases. Application of this simple efficient and reproducible grading scheme should permit reliable comparison of retrospective or prospective therapeutic data emanating from various institutions.

A histologic and cytologic method for the spatial definition of gliomas.

Because of the increasing application of stereotactic neurosurgical techniques not only in the diagnosis but also in the treatment of brain tumors, a demand for data regarding the spatial delimitation of gliomas is being created. Traditional histologic criteria were not established for detecting isolated tumor cells located at the boundaries of gliomas. Standard histologic techniques used for processing surgical specimens result in preparations of insufficient quality to allow the detection of minimal tumor infiltrates within normal brain parenchyma. Consequently, histologic examination has been considered an unreliable method of determining the boundaries of gliomas. In this study, we describe in detail an improved technique of biopsy specimen preparation that involves the use of glutaraldehyde-fixed and hemalum-phloxine-stained paraffin sections as well as alcohol-fixed and hemalum-phloxine-stained smears. In concert, these methods allow visualization of delicate cytologic details and are complementary in assessing the presence of isolated tumor cells at the periphery of gliomas. We define morphologic criteria for the accurate identification of isolated neoplastic cells and for their distinction from normal or reactive glial elements. These morphologic criteria include classic cytologic features of malignant lesions (such as nuclear pleomorphism and hyperchromasia) as well as the assessment of tumor architecture and cellular spatial relationships.

FDG-PET improves surgical outcome in negative MRI Taylor-type focal cortical dysplasias

Objective: To determine the diagnostic accuracy and prognostic value of 18FDG-PET in a recent series of patients operated for intractable partial epilepsy associated with histologically proven Taylor-type focal cortical dysplasia (TTFCD) and negative MRI. Methods: Of 23 consecutive patients (12 male, 7–38 years old) with negative 1.5-Tesla MRI, 10 exhibited subtle nonspecific abnormalities (e.g., unusual sulcus depth or gyral pattern) and the 13 others had strictly normal MRI. FDG-PET was analyzed both visually after coregistration on MRI and using SPM5 software. Metabolic data were compared with the epileptogenic zone (EZ) determined by stereo-EEG (SEEG) and surgical outcome. Results: Visual PET analysis disclosed a focal or regional hypometabolism in 18 cases (78%) corresponding to a single gyrus (n = 9) or a larger cortical region (n = 9). PET/MRI coregistration detected a partially hypometabolic gyrus in 4 additional cases. SPM5 PET analysis (n = 18) was concordant with visual analysis in 13 cases. Location of PET abnormalities was extratemporal in all cases, involving eloquent cortex in 15 (65%). Correlations between SEEG, PET/MRI, and histologic findings (n = 20) demonstrated that single hypometabolic gyri (n = 11) corresponded to EZ and TTFCD, which was localized at the bottom of the sulcus. Larger hypometabolic areas (n = 9) also included the EZ and the dysplastic cortex but were more extensive. Following limited cortical resection (mean follow-up 4 years), seizure freedom without permanent motor deficit was obtained in 20/23 patients (87%). Conclusions: 18FDG-PET coregistered with MRI is highly sensitive to detect TTFCD and greatly improves diagnosis and surgical prognosis of patients with negative MRI.

Diffuse low-grade oligodendrogliomas extend beyond MRI-defined abnormalities

Background: Imaging determinations of the spatial extent of diffuse low-grade gliomas (DLGGs) are of paramount importance in evaluating the risk-to-benefit ratio of surgical resection. However, it is not clear how accurately preoperative conventional MRI can delineate DLGGs. Methods: We report a retrospective histologic and imaging correlation study in 16 adult patients who underwent serial stereotactic biopsies for the diagnosis of untreated supratentorial well-defined and non–contrast-enhanced DLGG, in whom biopsy samples were taken within and beyond (OutBSs) MRI-defined abnormalities. Results: Thirty-seven OutBSs that extended from 10 to 26 mm beyond MRI-defined abnormalities were studied. Immunostaining revealed MIB-1–positive cells (i.e., cycling cells) in all but 2 of the OutBSs. None of the MIB-1–positive cells coexpressed glial fibrillary acidic protein, and all of them coexpressed OLIG2. MIB-1–positive cells were cycling isolated tumor cells, because 1) their morphologic characteristics reflected those of tumor cells, 2) the number of MIB-1–positive cells per square centimeter was significantly higher than that of controls, 3) the number of MIB-1–positive cells per square centimeter was positively correlated with the tumor growth fraction (p = 0.012), and 4) the number of MIB-1–positive cells per square centimeter in OutBSs decreased with distance from the tumor (p = 0.003). Conclusions: This study demonstrates, using a multiscale correlative approach, that conventional MRI underestimates the actual spatial extent of diffuse low-grade gliomas (DLGGs), even when they are well delineated. These results suggest that an extended resection of a margin beyond MRI-defined abnormalities, whenever feasible in noneloquent brain areas, might improve the outcome of DLGGs.