Extension of diffuse low-grade gliomas beyond radiological borders as shown by the coregistration of histopathological and magnetic resonance imaging data

Full access

OBJECTIVE

Magnetic resonance imaging tends to underestimate the extent of diffuse low-grade gliomas (DLGGs). With the aim of studying the presence of tumor cells outside the radiological border, the authors developed a method of correlating MRI findings with histological data in patients with suspected DLGGs in whom en bloc resections were performed.

METHODS

Five patients with suspected DLGG suitable for en bloc resection were recruited from an ongoing prospective study. Sections of the entire tumor were immunostained with antibodies against mutated IDH1 protein (IDH1-R132H). Magnetic resonance images were coregistered with corresponding IDH1 images. The growth pattern of tumor cells in white and gray matter was assessed in comparison with signal changes on corresponding MRI slices.

RESULTS

Neuropathological assessment revealed DLGG in 4 patients and progression to WHO Grade III glioma in 1 patient. The tumor core consisted of a high density of IDH1-R132H–positive tumor cells and was located in both gray and white matter. Tumor cells infiltrated along the peripheral fibers of the white matter tracts. In all cases, tumor cells were found outside the radiological tumor border delineated on T2-FLAIR MRI sequences.

CONCLUSIONS

The authors present a new method for the coregistration of histological and radiological characteristics of en bloc–removed infiltrative brain tumors that discloses tumor invasion at the radiological tumor borders. This technique can be applied to evaluate the sensitivity of alternative imaging methods to detect scattered tumor cells at tumor borders. Accurate methods for detection of infiltrative tumor cells will improve the possibility of performing radical tumor resection. In future studies, the method could also be used for in vivo studies of tumor invasion.

ABBREVIATIONSDLGG = diffuse low-grade glioma; DTI = diffusion tensor imaging; FLAIR = fluid-attenuated inversion recovery; PACS = picture archiving and communication system; SE = spin echo.

Abstract

OBJECTIVE

Magnetic resonance imaging tends to underestimate the extent of diffuse low-grade gliomas (DLGGs). With the aim of studying the presence of tumor cells outside the radiological border, the authors developed a method of correlating MRI findings with histological data in patients with suspected DLGGs in whom en bloc resections were performed.

METHODS

Five patients with suspected DLGG suitable for en bloc resection were recruited from an ongoing prospective study. Sections of the entire tumor were immunostained with antibodies against mutated IDH1 protein (IDH1-R132H). Magnetic resonance images were coregistered with corresponding IDH1 images. The growth pattern of tumor cells in white and gray matter was assessed in comparison with signal changes on corresponding MRI slices.

RESULTS

Neuropathological assessment revealed DLGG in 4 patients and progression to WHO Grade III glioma in 1 patient. The tumor core consisted of a high density of IDH1-R132H–positive tumor cells and was located in both gray and white matter. Tumor cells infiltrated along the peripheral fibers of the white matter tracts. In all cases, tumor cells were found outside the radiological tumor border delineated on T2-FLAIR MRI sequences.

CONCLUSIONS

The authors present a new method for the coregistration of histological and radiological characteristics of en bloc–removed infiltrative brain tumors that discloses tumor invasion at the radiological tumor borders. This technique can be applied to evaluate the sensitivity of alternative imaging methods to detect scattered tumor cells at tumor borders. Accurate methods for detection of infiltrative tumor cells will improve the possibility of performing radical tumor resection. In future studies, the method could also be used for in vivo studies of tumor invasion.

Diffuse low-grade glioma (DLGG) is a Grade II glioma according to the WHO classification of tumors of the central nervous system40 and represents a chronic infiltrative disease of the brain. It affects mainly adults with a mean age of 39 years at the time of diagnosis.7 Median patient survival is around 5 years.7,57 Its incidence is 1 case (0.5–1.3 cases) per 100,000 inhabitants.2

Today, there is strong evidence of a positive correlation between the extent of tumor resection and survival in patients with DLGG.4,14,46,60,63,65 Since DLGG, after varying periods of time, undergoes malignant transformation,45,52 tumor resection should be performed early to prevent uncontrollable disease.31 In addition, some data support the notion that surgery for DLGG should aim for supratotal resection, which not only improves patient outcome, but also inhibits the natural progression of the disease.67

Morphological MRI tends to underestimate the extent of DLGG. Previous studies using multiple biopsies have identified tumor cells outside the radiological border on fluid-attenuated inversion recovery (FLAIR) or T2-weighted MRI sequences.33,53 Because of the slow growth rate of DLGG,20 proliferation markers are not suitable for histological detection of tumor cells beyond the radiological border. Furthermore, some tumors cells invading the peritumoral areas may escape detection because the invasive properties of gliomas appear to occur at the cost of proliferation.22

The majority of tumor cells in DLGG are characterized by a mutation in the isocitrate dehydrogenase gene (IDH1) at position 132 (R132H). A mutation-specific monoclonal antibody recognizing tumor cells carrying the IDH1-R132H mutated protein has been generated.5 Since the complete population of tumor cells in an IDH1-positive tumor carries the mutation, this antibody is useful in the detection of infiltrating tumor cells.

Only a few studies have correlated imaging techniques with histological tumor characteristics.33,53,58,66 More accurate noninvasive neuroimaging techniques are warranted to detect the infiltration of tumor cells in the brain.9 These issues are directly related to the unsolved neurosurgical questions regarding radical tumor resection and the sensitivity of MRI to detect the scattered tumor cells surrounding the tumor.

The aim of this study was to disclose the presence of tumor cells outside the radiological border. For this purpose, we developed a new method to compare MRI findings with histological data in patients in whom en bloc tumor resection was performed. We describe the applications of this method in 5 patients who underwent macroscopically complete resection of the tumor plus a margin outside the MRI-defined tumor border. Consecutive T2-weighted FLAIR images through the whole tumor were matched with the corresponding images of IDH1-R132H–labeled tissue sections, and the histological findings at the radiological borders are reported.

Methods

Patients

Five patients were recruited from a longitudinal followup study of adults with DLGGs at our hospital.3 The institutional review board approved the protocol for the present analysis, and written informed consent was obtained prior to patient participation. Among the specific inclusion criteria was suspected DLGG suitable for complete en bloc tumor resection that included a margin of normal-appearing brain located outside the radiological border. Tumor locations considered suitable for en bloc resection were noneloquent areas (those outside the cortical primary motor, somatosensory, speech, and language areas and the insula).29 In contrast, tumors located in or in the immediate vicinity of critical white matter tracts29 that required close monitoring of function and tumors located in central gray nuclei were not considered suitable. Tumors that were better approached by piecemeal removal were also not included. The ideal tumor location for en bloc resection was the right anterior frontal lobe, an area with a high resection probability in accordance with the “resection probability map” presented by Sarubbo et al.61 In the present study, left-sided tumors in the anterior frontal pole were also considered suitable for inclusion.

Preoperative MRI

Prior to surgery, patients were examined via 3-T MRI (Philips Achieva) according to the original study protocol.3 Morphological MRI sequences included sagittal and axial T2-weighted turbo spin echo (SE), coronal and axial T2-FLAIR, axial T1-weighted SE before and after contrast injection, and sagittal T1-weighted 3D turbo field echo after contrast injection. Perfusion and diffusion MRI sequences were also obtained3 but were not used in the present study.

A neuroradiologist manually delineated tumor borders on T2-FLAIR sequences by using a picture archiving and communication system (PACS; Vue PACS, Carestream Health Inc.). The borders were drawn at the transition between high signal intensity and normal parenchymal signal intensity based on visual evaluation. In ambiguous cases the signal intensity was measured in regions adjacent to the tumor and compared with the normal parenchymal signal at a distance from the tumor.

Surgery

Tumor borders with possible peritumoral margins were delineated on preoperative T2-FLAIR MRI sequences. The aim of tumor resection was to include the complete tumor volume recorded on T2-FLAIR MRI plus a margin located outside the radiological border and extending into the normal-appearing brain tissue. Magnetic resonance images were loaded onto the neuronavigation system (Curve and Kolibri, Brainlab). Intraoperatively, tumor margins were estimated with the guidance of neuronavigation and confirmed by perioperative ultrasound (Flex Focus 700, BK Medical). Microsurgical en bloc resections were performed. Ultrasound was used to confirm the extent of tumor resection.

Preparation of Tumor Samples

The resection surfaces (anterior, posterior, medial, and lateral) were marked with tissue marking dye (Triangle Biomedical Sciences Inc.) for spatial orientation. The en bloc tissue sample was fixed in formalin for at least 2 days and subsequently cut into 8–12 consecutive tissue slices approximately 6- to 8-mm thick. The slices were photographed using a digital camera (Olympus Corp.). Tissue slices were fixed in formalin for an additional 24 hours and then embedded in paraffin wax. Subsequently, serial paraffin-embedded 4-μm-thick sections were placed on Super-Frost Plus slides (Gerhard Menzel GmbH), rehydrated, and stained. For diagnosis, all sections were stained with H & E. Immunohistochemical staining was applied to the slide containing the most representative tumor section using the Dako Autostainer Plus (DakoCytomation) and Dako EnVision FLEX detection system (DakoCytomation). Antibodies selected for diagnostics included glial fibrillary acidic protein (DakoCytomation), microtubule-associated protein 2 (Sigma-Aldrich), oligodendrocyte transcription factor 2 (Abnova Corp.), tumor protein 53 (DakoCytomatics), IDH1-R132H (Dianova), and Ki 67 (DakoCytomation). All staining was performed following the manufacturer's recommendations. A neuropathologist (I.A.) determined all diagnoses according to the 2007 WHO classification of tumors of the central nervous system.40

In addition to the diagnostics, immunostaining with IDH1-R132H was performed on all histological sections. The IDH1-R132H–stained sections were scanned to obtain digital images using an Aperio slide scanner.

Postoperative MRI

Postoperative MRI was performed within 48 hours after surgery. Tumor resection was considered complete if all signal changes that had been preoperatively evaluated as tumor on T2-FLAIR MRI were removed.

Volume Calculations

The Vue PACS software (version 11.1.0, Carestream Health Inc.) was used to segment the lesions with the aid of a semiautomatic method (Livewire Algorithm). The software is supported by an algorithm that uses an active contour model to define and segment the lesions on each slice. The total tumor volume was calculated on preoperative axial T2-FLAIR sequences, and the volume of the postoperative cavity was calculated using postoperative axial T2-FLAIR and/or T2 sequences.

Coregistration of Histological Data and MRI Findings

The stepwise procedure that we used for coregistration is described in Table 1. In summary, pre- and postoperative coronal MR images were reoriented perpendicular to the longitudinal orientation of sagittal MR images of the postoperative defect in PACS so that coronal MR images were oriented in the same direction as the histological sections (Fig. 1).18 Scanned IDH1 images were overlaid on coronal radiological images using Photoshop CC (Adobe Systems Software). The best-fitting preoperative MR image was identified and matched with the IDH1 image. The IDH1 images were subsequently compared with more anterior and posterior MR images to find the best match, similar to the method used by Eriksson et al.18 The gyral branching patterns seen in IDH1 images were used in combination with color codes to identify the corresponding orientation, location, and size of the resected tumor. After matching, the IDH1 image volumes were correlated using linear registration that included translation, rotation, and scaling.6 Postoperative MRI was used to validate the size of the resected tumor volume. Photoshop CC was used to manually mark the radiological tumor border on the corresponding IDH1 image.

TABLE 1.

The coregistration procedure

ImageDescription
Radiological
  1. Prior to surgery, morphological MRI sequences, including sagittal & axial T2-weighted turbo SE, coronal & axial T2-weighted FLAIR, axial T1-weighted SE before & after contrast injection, & sagittal T1-weighted 3D turbo field echo after contrast injection were obtained using a 3-T scanner (Philips Achieva).

  2. The pre- & postoperative images were stored in a PACS (Carestream Health).

  3. Using the PACS, pre- & postoperative coronal & sagittal FLAIR images were linked to each other & aligned in a presentation.

  4. Pre- & postoperative coronal MR images were reoriented perpendicular to the longitudinal orientation of sagittal MR images of the postoperative defect using the PACS software.

  5. Preoperative coronal FLAIR images (6-mm thickness, 3-mm spacing) of the tumor were saved as TIFF files.

Histological
  1. After resection, the surfaces (anterior, posterior, medial, & lateral) were marked with different colors for spatial orientation.

  2. The tumor was cut into 8–12 consecutive tissue slices 6 mm thick.

  3. Slices were formalin-fixed & paraffin-embedded. Several thin (5 μm) histological sections were cut from each paraffin block.

  4. Immunostaining with IDH1-R132H antibodies was performed on histological sections from each tumor slice, which were put on object slides for light microscopy.

  5. Microscans were obtained using an Aperio microscope scanner & saved as TIFF files (using several percentages of quality for reducing image size).

PhotoshopBoth radiological & histological images were manipulated using Photoshop CC (Adobe Systems Software).
  1. Radiological TIFF files (480 kb) were enlarged by using the resolution tool. In this way voxels are added to the file to make scaling possible without losing quality.

  2. Histological TIFF files (2–4 Gb) were adjusted using the Magic Wand tool to select the resection & were reduced in scale while maintaining the aspect ratio.

  3. The best fitting preoperative MR image was identified & matched for each histological section separately.

  4. The gyral branching patterns of the histological sections were used in combination with color codes to identify corresponding orientation, location, & size of the resected tumor & radiological image.

  5. The histological file was overlaid on the radiological file as a separate layer to enable accurate adjustments.

  6. The histological sections were subsequently compared with more anterior & posterior MR images to find the best match.

  7. After matching, the histological & radiological tumor volumes were correlated using linear registration including translation, rotation, & scaling.

  8. Postoperative MRI was used to validate the size of the resected tumor volume.

  9. Photoshop CC was used to manually mark the radiological tumor border on the corresponding histological sections.

FIG. 1.
FIG. 1.

Case 1. Preoperative coronal T2-FLAIR (A) and sagittal T2-weighted (B) MR images showing a nonenhancing tumor in the right frontal pole. Postoperative coronal T1-weighted (C) and sagittal T2-weighted (D) MR images showing complete resection of the tumor.

Results

The tumor core (area with the highest density of tumor cells) was located in both gray and white matter. Infiltration of tumor cells occurred along the white matter tracts, with tumor cells occupying the peripheral fibers of the tract. In all 5 cases, tumor cells extended beyond the radiological border and IDH1-immunolabeled tumor cells were detected at a maximum distance of 1.4 cm from the radiological border. Clinical data, MRI results, and tumor characteristics of the 5 patients included in this study are summarized in Table 2. The separate cases are described in more detail below.

TABLE 2.

Clinical, MRI, and tumor characteristics in 5 patients with DLGG

CaseAge (yrs), SexPreop SymptomPreop/Postop Neurological FunctionTumor LocationMRI Signal on T1/T2-FLAIRTumor Vol on FLAIR (ml)Volume of Postop Cavity (ml)Tissue Weight (g)*DiagnosisLOH 1p19q/IDH1/Ki 67 (%)
133, MUnrelated vertigoN/NFrontal pole, rtLow/high285867OII+/+/<5
239, MHeadacheN/NFrontal pole, ltLow/high688387OII−/+/<5
360, MSeizures, postictal hemiparesisN/SMA hemiparesisPremotor frontal, rtLow/high203644AIINP/+/<5
450, FHeadache, vertigoN/transient dysphasiaFrontal pole, ltLow/high455638OII+/+/<5
553, MSeizuresNFrontal pole, ltLow/high848582OIII+/+/25

AII = astrocytoma WHO Grade II; LOH 1p/19q = loss of heterozygosity at 1p/19q; N = normal; NP = not performed; OII = oligodendroglioma WHO Grade II; OIII = oligodendroglioma WHO Grade III; SMA = supplementary motor area.

Tumor weight after fixation.

Case 1

Summary

This 33-year-old man was examined because of vertigo. Magnetic resonance imaging showed an incidental, nonenhancing heterogeneous tumor in the right frontal pole (Table 2 and Fig. 1A and B). A right frontal lobectomy was performed as described above, with en bloc removal of the tumor. Postoperative MRI confirmed that the tumor had been completely resected (Fig. 1C and D). Neuropathological examination showed a WHO Grade II oligodendroglioma. Figure 2 shows MR images of the preoperatively delineated tumor (Fig. 2A), the postoperative cavity (Fig. 2B), the overlay of preoperative delineated tumor and postoperative cavity illustrating the resected peritumoral margins (Fig. 2C), and the overlay of the IDH1 image with the T2-FLAIR image (Fig. 2D).

FIG. 2.
FIG. 2.

Case 1. A: Preoperative delineated tumor on coronal T2-FLAIR MRI. B: Postoperative delineated cavity on coronal T1-weighted MRI. C: Overlay of the preoperative delineated tumor and postoperative delineated cavity showing the resected peritumoral margins on coronal T2-FLAIR MRI. D: Overlay of IDH1 image on coronal T2-FLAIR MR image. Figure is available in color online only.

Infiltration Pattern

White Matter. The infiltrating tumor cells followed the white matter tracts. Tumor cells were concentrated in the periphery of the tracts, with the highest density in the most peripheral parts. The white matter tract of the peritumoral margin was occupied by a “single lane” of tumor cells (Fig. 3).

FIG. 3.
FIG. 3.

Case 1. An IDH1 image from the peripheral part of the tumor shows the brown-stained, infiltrating tumor cells following the white matter tracts. The peripheral parts of the white matter tracts show the highest tumor cell density. At the tumor margins, the white matter is occupied by a “single lane” of tumor cells. The tumor cells are visible outside the radiological border (inset). Figure is available in color online only.

Gray Matter. In the tumor core, tumor cells occurred at a high density in the gray matter. Figure 4 left illustrates the core of the tumor, whereas Figs. 3 and 5 show the peripheral part of the tumor.

FIG. 4.
FIG. 4.

Case 1. An IDH1 image and corresponding coronal T2-FLAIR MR image from the core of the tumor. Left: In the tumor core, tumor cells are found at a high density in the gray matter. One area has a central part with relatively few tumor cells and vacuoles surrounded by a rim of densely packed tumor cells. Inset: The central part of the core consists of tumor cells and vacuoles. Right: The central part with tumor cells and vacuoles corresponds to low signal intensity on T2-FLAIR MRI, whereas the densely packed tumor cells correspond to a ring of high signal intensity on a T2-FLAIR sequence. Figure is available in color online only.

FIG. 5.
FIG. 5.

Case 1. An area with a high density of tumor cells was detected outside the radiological border. Left: An IDH1 image with the radiological border delineated by a solid line and the tumor cells outside the radiological border indicated by a dashed line. Right: The corresponding preoperative T2-FLAIR image with the radiological tumor border delineated by a solid line and the area with tumor cells missed on MRI delineated by a dashed line. Figure is available in color online only.

On T2-FLAIR sequences, we noted a round area in the central part of the tumor with relatively low signal intensity surrounded by a ring of high signal intensity. These radiological areas corresponded to histological areas with a low density of tumor cells and vacuoles (low FLAIR signal, central part) and a high density of tumor cells (high FLAIR signal, peripheral ring; Fig. 4). The vacuoles within the central part were considered to represent tumor edema, a phenomenon in DLGGs that has been described by Gerin et al.21

Tumor cells did not cross the pia mater in the sulcus, resulting in a notable difference in tumor cell density in the gray matter of 2 adjacent gyri.

Radiological Versus Histological Tumor Border

The maximum radiological peritumoral margin (that is, tissue outside the radiological border on T2-FLAIR sequences) measured 2 cm. Tumor cells were found outside the radiological border (Figs. 3 and 5) at a maximum distance of 1.4 cm from the radiological border.

An area with high tumor cell density and low signal on T2-FLAIR MRI was found outside the radiological border in the white matter (Fig. 5 left), which had not been detected on preoperative FLAIR sequences (Fig. 5 right).

Case 2

Summary

The MRI studies in this 39-year-old man, who had presented with headache, showed a nonenhancing, left frontal, slightly heterogeneous tumor. Left frontal en bloc lobectomy was performed, and postoperative MRI confirmed complete tumor resection. Neuropathological examination showed a WHO Grade II oligodendroglioma (Table 2).

Infiltration Pattern

White Matter. Similar to Case 1, tumor cells followed the white matter tracts and were slightly more concentrated in the peripheral parts of the tracts (Fig. 6A).

FIG. 6.
FIG. 6.

Case 2. IDH1 images from the peripheral part and core of the tumor. Black lines delineate the radiological border. A: An IDH1 image from a peripheral part of the tumor illustrating tumor cells that follow the white matter tracts. B: In the tumor core, tumor cells are found at high density in both the gray and white matter. Tumor cells are present outside the radiological border. There are fewer tumor cells at the distant side of the sulcus (right side of inset C) compared with the proximal side (left side of inset C). Tumor cells infiltrated along the white mater tracts, resulting in a higher density of tumor cells at the bottom of the sulcus (inset D), compared with the more rostral part of the gyrus (inset E). Figure is available in color online only.

Gray Matter. In the tumor core, tumor cells were found at high density in the gray matter (Fig. 6B). Farther away from the core, fewer tumor cells were present in the gray matter (Fig. 6A).

The density of tumor cells was lower in the gray matter of the adjacent gyrus, which was most pronounced at the tumor borders. Thus, tumor cells followed the route depicted by the white matter tracts at the bottom of the sulcus (Fig. 6C), resulting in a higher density of tumor cells in the bottom of the sulcus (Fig. 6D), compared with the more rostral part of the gyrus (Fig. 6E).

Radiological Versus Histological Tumor Border

The tumor volume was large (Table 2), and some tissue corresponding to the medial part of the resection was lost. The other tumor portions were well preserved, enabling reliable coregistration with the radiological data.

The maximum radiological margin on T2-FLAIR sequences measured 2 cm. In the lateral and posterior portion of the tumor, in the white matter as well as the gray, tumor cells were identified outside the radiological border (Fig. 7). In fact, tumor cells were detected 1 cm from the radiological border. However, the opposite situation (radiological tumor border outside the histological tumor border) was also encountered. Thus, it was difficult to appreciate the correct tumor border on T2-FLAIR sequences in the gray matter because tumor cells were seen outside the radiological border and the radiological border extended into tumor-free tissue.

FIG. 7.
FIG. 7.

Case 2. Tumor cells are present outside the radiological border, which is indicated by a black line. The area marked with an asterisk is outside the radiological border. Fewer tumor cells occur farther away from the border. Tumor cells outside the radiological border occur more frequently closer to the border (lower inset). Farther from the border no tumor cells are found (upper inset). Figure is available in color online only.

Case 3

Summary

This 60-year-old man had partial seizures followed by a transitory postictal left-sided hemiparesis. Magnetic resonance imaging showed a right-sided, homogeneous nonenhancing tumor in the premotor area (Fig. 8A and B). En bloc tumor resection was performed, and postoperative MRI confirmed complete resection (Fig. 8C). After surgery, the patient suffered from an inability to initiate voluntary movements, most likely due to resection of the supplementary motor area,34 but he made a full recovery within 1 month.

FIG. 8.
FIG. 8.

Case 3. Preoperative coronal T2-FLAIR (A) and sagittal T1-weighted (B) MR images showing a right-sided nonenhancing tumor in the premotor area. Postoperative coronal T1-weighted MR image (C) showing complete tumor resection.

During tissue preparation, some technical difficulties resulted in some tissue loss and distortion of the en bloc sample at the lateral tumor side. Neuropathological examination revealed WHO Grade II astrocytoma (Table 2).

Infiltration Pattern: White and Gray Matter

The tumor core was located in gray matter as well as white and showed a high density of tumor cells (Fig. 9A). At the periphery of the tumor, we saw a pattern similar to the one described before, with tumor cells primarily infiltrating the white matter tracts, at a somewhat higher cell density at the border of the gray matter, and partly infiltrating the white matter tracts (Fig. 9E).

FIG. 9.
FIG. 9.

Case 3. IDH1 images from the core and the periphery of the tumor. A: In the tumor core, the tumor cells occur at a high density in both gray and white matter. B: There is a high tumor cell density in the proximal gray matter in connection to the pia overlaying the sulcus. In the adjacent gyrus, tumor cells are scarce, with a declining number of tumor cells farther from the bottom of the sulcus (that is, from left to right in the figure, magnified in insets C and D, respectively). C: More tumor cells are found close to the bottom of the sulcus. D: No tumor cells are found farther from the bottom of the sulcus. E: In the tumor periphery, tumor cells primarily follow the white matter tracts, at a slightly higher density at the border of the gray matter, and partly infiltrate the white matter tracts.

Figure 9B–D illustrates the difference in tumor cell density in the gray matter of 2 adjacent gyri: There is a high cell density in the proximal gray matter, in connection with the pia overlaying the sulcus (Fig. 9B). In the adjacent gyrus, there are only a few scattered tumor cells, declining in number farther from the bottom of the sulcus (Fig. 9C and D).

Radiological Versus Histological Tumor Border

The radiological tumor-free zone included in the en bloc resection measured 2.3 cm. Tumor cells were found at a maximal distance of 1.2 cm from the radiological border.

Case 4

This 50-year-old woman had a tumor in the left frontal pole with heterogeneous signal characteristics and cystic areas on MRI (Table 2). The tumor had been diagnosed 10 years earlier, and she was referred to the neurosurgical department because of a slow progression in tumor size. The tumor was completely removed en bloc. Postoperatively she suffered from dysphasia that spontaneously resolved within a few days. Neuropathological examination showed WHO Grade II oligodendroglioma with densely packed IDH1-labeled tumor cells mainly in the gray matter. A relatively high concentration of tumor cells was also found at the border between gray and white matter in the peripheral part of the tumor (not shown). Tumor cells were identified at a distance of 1.1 cm from the radiological border, which corresponded to the maximum tissue margin (not shown).

Case 5

This 53-year-old man had a left frontal pole tumor with minimal contrast enhancement that was compatible with a DLGG on initial MRI (Table 2). However, MRI 1 month later showed signs of tumor progression with some hemorrhagic transformation and the appearance of atypical gyral contrast enhancement in the medial part of the tumor. An en bloc resection was performed, and the patient was neurologically intact after surgery. The tumor volume was large, and there was some loss of white matter tissue on the medial/inferior side of the tumor after en bloc resection. Neuropathological examination verified focal progression to WHO Grade III oligodendroglioma. The tumor core was situated in both gray and white matter, and as in the previous cases, tumor cell concentration was high at the border between gray and white matter. Tumor cells were identified at a distance of 1.3 cm from the radiological border, corresponding to the maximal resected peritumoral margin (not shown).

Discussion

In this study we developed and applied a new method for the coregistration of MRI and histological data in DLGG. Our findings show that this method is a useful aid in evaluating the growth pattern of these tumors in relation to radiological findings because it illustrates the characteristic tumor infiltration of the white matter tracts and the extension of tumor cells outside the radiological border. Application of this new technique in larger series of patients will enable a more reliable assessment of tumor invasion in the brain. A better understanding of the behavior of tumor cells beyond the radiological border will enhance the possibility of performing a radical tumor resection, which is crucial for optimizing the outcome of patients with DLGG.4,11,14,46,60,63,62

There are no previous reports on the coregistration of histological data with MRI of en bloc–removed DLGGs as described here. However, the correlation of tumor characteristics on MRI with histological findings of DLGGs has been the focus of previous studies.30,33,41,53 In these studies tumor cells have been demonstrated outside the radiological tumor borders, that is, in normal-appearing brain on FLAIR or T2-weighted images. In a postmortem study of 4 patients with high-grade glioma, IDH1-positive tumor cells were found to infiltrate diffusively in the brain at a great distance from the primary tumor location, supporting the notion that a high-grade tumor is a surgically untreatable systemic brain disease.59 The goal when treating DLGG is to postpone or, ideally, to prevent the ominous progression from low-grade tumor to high-grade tumor.

In a stereotactic biopsy study by Kelly et al., samples were obtained from areas that appeared normal on T1- and T2-weighted MR images in 5 of 17 DLGGs.33 In 3 of the 5 tumors, infiltrating tumor cells were found in normal-appearing T1 and T2 areas. Watanabe et al. performed en bloc resections of 8 DLGGs and found isolated tumor cells extending beyond the hyperintense areas on T2-weighted images for 5 of the DLGGs.66 In a stereotactic biopsy study including 16 DLGGs, Pallud et al. showed tumor cells as far as 26 mm beyond MRI-defined abnormalities.53

Thus, these study data together with our findings indicate that the extension of surgery for DLGGs should not be limited to the radiological border as delineated on FLAIR MRI sequences and support the concept of supratotal resection.16 Because the extent of resection positively influences the time to anaplastic transformation4,14,46,60,63,65 and tumor cells infiltrate beyond the radiological border,33,53 it is likely that a supratotal resection—that is, resection of the tumor together with a margin outside the radiological border—will improve the prognosis of patients with DLGG.51 This was actually shown by Yordanova et al.67 and Duffau.11 Anaplastic transformation was avoided in patients with DLGG who had undergone supratotal resection, a significantly better result compared with that in their control group in which only complete resection was performed.67,1 The optimal treatment of DLGG is probably resection to the functional borders during awake surgery, a strategy well described by Duffau and colleagues.10,15,16 Based on their extensive experience with awake surgery and direct electrical stimulation, probability maps on the resectability of different cortical and subcortical regions have been created and most likely represent what is possible to achieve in terms of complete resections of DLGGs.29,61

It is well accepted that DLGGs prefer to infiltrate along white matter tracts.22,23,44,62 Indeed, infiltrating tumor cells were detected in the white matter tracts in all cases in the present study. Interestingly, in 2 of the cases the white matter tracts were only partly infiltrated. In 4 cases tumor cells were concentrated in the peripheral fibers of the tracts. It is likely that this limited infiltration of the white matter reflects an early stage of tumor invasion. To spread to the adjacent gyrus, tumor cells followed the white matter tracts. They reached the molecular layer of cortex but did not infiltrate the pia mater. The short and intracortical U-fibers are the most probable way of diffusion for tumor cells between 2 adjacent gyri.42,49,54

Differences in tumor genotype and phenotype are associated with different invasion patterns in DLGG.25 In addition, the site of origin is also thought to affect the growth properties of the tumor.24,44,68

The migration properties of oligodendrocyte precursors during early development are influenced by their specific origin in the brain,50 and different proteins can guide the migration of cells.8,25,48 In a transgenic mouse model, oligodendrogliomas showed a preference for the white matter regions, while low-grade astrocytomas were more often associated with the lateral ventricles.56 Interestingly, metalloproteinases, expressed by the tumor cells, may enhance the spread of glioma cells in the white matter.19,47 In a study using tumor cell lines, low-grade astrocytomas were sensitive to the inhibitory proteins present in myelin, whereas oligodendrogliomas were able to overcome the inhibitory effects.1 Similar factors may be involved in tumor cell migration and tumor spread.25 By obtaining large resections including tumor margins, as described here, we can study the expression of proteins that may be involved in tumor growth and spread in the brain.

Our method allowed a detailed and reliable comparison between the histological properties of specific intra- and peritumoral regions and the corresponding signal changes on MRI. Tumor cells were identified outside the radiological border in all cases. In Case 1, we found a relatively large area of high-density tumor cells outside the radiological border that had appeared normal on preoperative T2-weighted FLAIR images and was therefore interpreted as normal tissue (Fig. 5).

The tumor is expected to be visible on FLAIR MRI if the tumor cell density is above a certain value, that is, the “visibility threshold.” In addition to the cell density, the MRI signal is influenced by several other factors, including the water content of the tissue.43 Data regarding the threshold for detecting tumor cells on FLAIR MRI43 or diffusion tensor imaging (DTI) are scarce.17 Tumor cell concentrations and edema fraction were correlated in biopsy samples derived from inside and outside T2-defined abnormalities in a recent study.21 The tumor cell density is inversely proportional to the MRI apparent diffusion coefficient value.17,26,36 In a DTI study of gliomas, the isotropic and the anisotropic diffusion components could differentiate between normal and tumor-infiltrated brain.58

Our results showing tumor cells outside the radiological border on FLAIR MRI support the notion that FLAIR sequences are not sensitive enough to accurately detect infiltration of tumor cells. Our method provides a new tool for further studies on how the signal changes on different MRI sequences are affected by the concentration of tumor cells, the water content of the tumor, and other factors present in the tissue. Moreover, the sensitivity of diffusion MRI with regard to detecting infiltrating tumor cells can be further evaluated.

Positron emission tomography with amino acid tracers is another potential tool for identifying the infiltrative “penumbra” of tumor cells. Because of the low background activity of 11C-l-methionine and 18F-fluoroethyl-l-tyrosine, these tracers are considered better than FDG in delineating gliomas.32,64 A threshold for the detection of tumor cells in the infiltration area has been discussed,64 but a common cutoff value has been difficult to define because of tumor heterogeneity.27,35,37 Coregistration of PET and histological data in the way described here enables an evaluation of the potential of these PET tracers to identify scattered tumor cells within the infiltrative tumor zone.

Finally, large tumor resections provide more reliable representative samples for histopathological tumor diagnosis. When tissue loss is minimized and the whole tumor can be analyzed, the probability of a correct diagnosis with respect to tumor grade and focal signs of anaplastic transformation is augmented. The importance of intact tumor samples was illustrated in a recent study in which a subset of WHO Grade II gliomas showed microfoci of high cellularity and with molecular and genetic aberrations, indicating early anaplastic transformation. The presence of these microfoci was associated with a shorter survival than in patients with homogeneous DLGGs.55

Limitations of the Study

The main shortcoming of this study is the limited number of cases included. Therefore, undue extrapolation of our findings should be avoided, and our results should be confirmed by larger series. Another limitation of our method is that tumor localization permitting an en bloc resection technique is mandatory; however, this technique cannot be used for all tumor locations. Tumors in brain areas that are better approached by piecemeal removal are not appropriate, and tumors in or in close connection to eloquent areas with extended subcortical involvement that require close monitoring of motor or speech function are not suitable for en bloc removal. As tumor characteristics can be expected to vary for tumors with different localizations,13,24,25,38,44 this bias by tumor location is important to bear in mind.

For optimal assessment of infiltrated brain outside the radiological border, a larger margin of peritumoral tissue is desirable but must be weighed against the risk of introducing neurological deficits in the event that eloquent areas are breached.

We also encountered a few technical obstacles. To achieve an optimal fit between the coronal MRI and the histological slices, a correct angle (perpendicular to the horizontal plane when cutting the tissue) must be achieved. The technical problems encountered in Case 3 were probably attributable to a mismatch between the cutting angle and the angle of the coronal MRI slices. There is the risk of tissue loss during tumor resection and/or preparation of the slices, which was the case at the medial resection border in the patients in Cases 2 and 5. The tissue can also be distorted during surgery or preparation, and great care is warranted to avoid damage to vulnerable brain tissue.

Finally, detection of tumor cells with the IDH1-R132H antibody is only possible in IDH1-mutated tumors. It has been shown that 68%–94% of all Grade II oligodendrogliomas and 59%–72% of all Grade II astrocytomas have IDH1 mutations.28,39 For the remaining IDH1-negative tumors, another method for detecting tumor cells will be necessary. To restrict the tumor population to only IDH1-positive tumors also introduces a risk of “selection bias.”

Conclusions

In summary, we present a method for the coregistration of histological and radiological tumor characteristics. This method provides a valuable tool to study the invasive growth of DLGGs and to detect infiltrating tumor cells at the radiological border. Our results are encouraging and will contribute to a deeper understanding of the biological behavior of DLGGs and the intrinsic limitations of MRI.

Application of the method on a larger scale offers the possibility to evaluate the sensitivity of alternative imaging methods in detecting scattered tumor cells in the infiltration zone. The method also allows in vivo studies of factors with a potential role in regulating tumor growth and tumor spread in the brain.

Acknowledgments

This work was supported by research funds from the Erik, Karin och Gösta Selanders Stiftelse (A.S.), Hanna Eklund Stiftelsen (A.S.), and the Uppsala County Council (M.Z., A.S.).

References

  • 1

    Amberger VRHensel TOgata NSchwab ME: Spreading and migration of human glioma and rat C6 cells on central nervous system myelin in vitro is correlated with tumor malignancy and involves a metalloproteolytic activity. Cancer Res 58:1491581998

  • 2

    Bauchet LEpidemiology of diffuse low-grade gliomas. Duffau H: Diffuse Low-Grade Gliomas in Adults LondonSpringer2013. 15

  • 3

    Berntsson SGFalk ASavitcheva IGodau AZetterling MHesselager G: Perfusion and diffusion MRI combined with 11C-methionine PET in the preoperative evaluation of suspected adult low-grade gliomas. J Neurooncol 114:2412492013

  • 4

    Capelle LFontaine DMandonnet ETaillandier LGolmard JLBauchet L: Spontaneous and therapeutic prognostic factors in adult hemispheric World Health Organization Grade II gliomas: a series of 1097 cases: clinical article. J Neurosurg 118:115711682013

  • 5

    Capper DZentgraf HBalss JHartmann Cvon Deimling A: Monoclonal antibody specific for IDH1 R132H mutation. Acta Neuropathol 118:5996012009

  • 6

    Choe ASGao YLi XCompton KBStepniewska IAnderson AW: Accuracy of image registration between MRI and light microscopy in the ex vivo brain. Magn Reson Imaging 29:6836922011

  • 7

    Claus EBBlack PM: Survival rates and patterns of care for patients diagnosed with supratentorial low-grade gliomas: data from the SEER program, 1973–2001. Cancer 106:135813632006

  • 8

    de Castro FBribián A: The molecular orchestra of the migration of oligodendrocyte precursors during development. Brain Res Brain Res Rev 49:2272412005

  • 9

    Dhermain FGHau PLanfermann HJacobs AHvan den Bent MJ: Advanced MRI and PET imaging for assessment of treatment response in patients with gliomas. Lancet Neurol 9:9069202010

  • 10

    Duffau H: The challenge to remove diffuse low-grade gliomas while preserving brain functions. Acta Neurochir (Wien) 154:5695742012

  • 11

    Duffau H: Long-term outcomes after supratotal resection of diffuse low-grade gliomas: a consecutive series with 11-year follow-up. Acta Neurochir (Wien) 158:51582016

  • 12

    Duffau H: Surgery of low-grade gliomas: towards a ‘functional neurooncology’. Curr Opin Oncol 21:5435492009

  • 13

    Duffau HCapelle L: Preferential brain locations of low-grade gliomas. Cancer 100:262226262004

  • 14

    Duffau HLopes MArthuis FBitar ASichez JPVan Effenterre R: Contribution of intraoperative electrical stimulations in surgery of low grade gliomas: a comparative study between two series without (1985–96) and with (1996–2003) functional mapping in the same institution. J Neurol Neurosurg Psychiatry 76:8458512005

  • 15

    Duffau HMandonnet E: The “oncofunctional balance” in surgery for diffuse low-grade glioma: integrating the extent of resection with quality of life. Acta Neurochir (Wien) 155:9519572013

  • 16

    Duffau HTaillandier L: New concepts in the management of diffuse low-grade glioma: Proposal of a multistage and individualized therapeutic approach. Neuro Oncol 17:3323422015

  • 17

    Ellingson BMMalkin MGRand SDConnelly JMQuinsey CLaViolette PS: Validation of functional diffusion maps (fDMs) as a biomarker for human glioma cellularity. J Magn Reson Imaging 31:5385482010

  • 18

    Eriksson SHFree SLThom MHarkness WSisodiya SMDuncan JS: Reliable registration of preoperative MRI with histopathology after temporal lobe resections. Epilepsia 46:164616532005

  • 19

    Fillmore HLVanMeter TEBroaddus WC: Membrane-type matrix metalloproteinases (MT-MMPs): expression and function during glioma invasion. J Neurooncol 53:1872022001

  • 20

    Fisher BJNaumova ELeighton CCNaumov GNKerklviet NFortin D: Ki-67: a prognostic factor for low-grade glioma?. Int J Radiat Oncol Biol Phys 52:99610012002

  • 21

    Gerin CPallud JDeroulers CVarlet POppenheim CRoux FX: Quantitative characterization of the imaging limits of diffuse low-grade oligodendrogliomas. Neuro Oncol 15:137913882013

  • 22

    Giese ABjerkvig RBerens MEWestphal M: Cost of migration: invasion of malignant gliomas and implications for treatment. J Clin Oncol 21:162416362003

  • 23

    Giese AWestphal M: Glioma invasion in the central nervous system. Neurosurgery 39:2352521996

  • 24

    Guichet POHugnot JPCellular origin of Grade II gliomas. Duffau H: Diffuse Low-Grade Gliomas in Adults LondonSpringer2013. 7590

  • 25

    Hassani ZHugnot JPDeciphering the molecular and cellular basis for dissemination of diffuse low-grade gliomas. Duffau H: Diffuse Low-Grade Gliomas in Adults LondonSpringer2013. 9198

  • 26

    Hayashida YHirai TMorishita SKitajima MMurakami RKorogi Y: Diffusion-weighted imaging of metastatic brain tumors: comparison with histologic type and tumor cellularity. AJNR Am J Neuroradiol 27:141914252006

  • 27

    Herholz KHölzer TBauer BSchröder RVoges JErnestus RI: 11C-methionine PET for differential diagnosis of low-grade gliomas. Neurology 50:131613221998

  • 28

    Ichimura KPearson DMKocialkowski SBäcklund LMChan RJones DT: IDH1 mutations are present in the majority of common adult gliomas but rare in primary glioblastomas. Neuro Oncol 11:3413472009

  • 29

    Ius TAngelini EThiebaut de Schotten MMandonnet EDuffau H: Evidence for potentials and limitations of brain plasticity using an atlas of functional resectability of WHO grade II gliomas: towards a “minimal common brain”. Neuroimage 56:99210002011

  • 30

    Iwama TYamada HSakai NAndoh TNakashima THirata T: Correlation between magnetic resonance imaging and histopathology of intracranial glioma. Neurol Res 13:48541991

  • 31

    Jakola ASMyrmel KSKloster RTorp SHLindal SUnsgård G: Comparison of a strategy favoring early surgical resection vs a strategy favoring watchful waiting in low-grade gliomas. JAMA 308:188118882012

  • 32

    Kaschten BStevenaert ASadzot BDeprez MDegueldre CDel Fiore G: Preoperative evaluation of 54 gliomas by PET with fluorine-18-fluorodeoxyglucose and/or carbon-11-methionine. J Nucl Med 39:7787851998

  • 33

    Kelly PJDaumas-Duport CKispert DBKall BAScheithauer BWIllig JJ: Imaging-based stereotaxic serial biopsies in untreated intracranial glial neoplasms. J Neurosurg 66:8658741987

  • 34

    Kim YHKim CHKim JSLee SKHan JHKim CY: Risk factor analysis of the development of new neurological deficits following supplementary motor area resection. J Neurosurg 119:7142013

  • 35

    Kinoshita MHashimoto NGoto TYanagisawa TOkita YKagawa N: Use of fractional anisotropy for determination of the cut-off value in 11C-methionine positron emission tomography for glioma. Neuroimage 45:3123182009

  • 36

    Kono KInoue YNakayama KShakudo MMorino MOhata K: The role of diffusion-weighted imaging in patients with brain tumors. AJNR Am J Neuroradiol 22:108110882001

  • 37

    Kracht LWMiletic HBusch SJacobs AHVoges JHoevels M: Delineation of brain tumor extent with [11C]l-methionine positron emission tomography: local comparison with stereotactic histopathology. Clin Cancer Res 10:716371702004

  • 38

    Laigle-Donadey FMartin-Duverneuil NLejeune JCrinière ECapelle LDuffau H: Correlations between molecular profile and radiologic pattern in oligodendroglial tumors. Neurology 63:236023622004

  • 39

    Leu Svon Felten SFrank SVassella EVajtai ITaylor E: IDH/MGMT-driven molecular classification of low-grade glioma is a strong predictor for long-term survival. Neuro Oncol 15:4694792013

  • 40

    Louis DNOhgaki HWiestler ODCavenee WKBurger PCJouvet A: The 2007 WHO classification of tumours of the central nervous system. Acta Neuropathol 114:971092007

  • 41

    Lunsford LDMartinez AJLatchaw RE: Magnetic resonance imaging does not define tumor boundaries. Acta Radiol Suppl 369:1541561986

  • 42

    Magnain CAugustinack JCReuter MWachinger CFrosch MPRagan T: Blockface histology with optical coherence tomography: a comparison with Nissl staining. Neuroimage 84:5245332014

  • 43

    Mandonnet EBiomathematical modeling of DLGG behavior. Duffau H: Diffuse Low-Grade Gliomas in Adults LondonSpringer2013. 447455

  • 44

    Mandonnet ECapelle LDuffau H: Extension of paralimbic low grade gliomas: toward an anatomical classification based on white matter invasion patterns. J Neurooncol 78:1791852006

  • 45

    Mandonnet EDelattre JYTanguy MLSwanson KRCarpentier AFDuffau H: Continuous growth of mean tumor diameter in a subset of grade II gliomas. Ann Neurol 53:5245282003

  • 46

    McGirt MJChaichana KLAttenello FJWeingart JDThan KBurger PC: Extent of surgical resection is independently associated with survival in patients with hemispheric infiltrating low-grade gliomas. Neurosurgery 63:7007082008

  • 47

    Mentlein RHattermann KHeld-Feindt J: Lost in disruption: role of proteases in glioma invasion and progression. Biochim Biophys Acta 1825:1781852012

  • 48

    Miyamoto YYamauchi JTanoue A: Cdk5 phosphorylation of WAVE2 regulates oligodendrocyte precursor cell migration through nonreceptor tyrosine kinase Fyn. J Neurosci 28:832683372008

  • 49

    Oishi KZilles KAmunts KFaria AJiang HLi X: Human brain white matter atlas: identification and assignment of common anatomical structures in superficial white matter. Neuroimage 43:4474572008

  • 50

    Olivier CCobos IPerez Villegas EMSpassky NZalc BMartinez S: Monofocal origin of telencephalic oligodendrocytes in the anterior entopeduncular area of the chick embryo. Development 128:175717692001

  • 51

    Pallud JDiffuse low-grade gliomas: What does “complete resection” mean?. Hayat MA: Tumors of the Central Nervous System DordrechtSpringer2011. 2:153161

  • 52

    Pallud JMandonnet EDuffau HKujas MGuillevin RGalanaud D: Prognostic value of initial magnetic resonance imaging growth rates for World Health Organization grade II gliomas. Ann Neurol 60:3803832006

  • 53

    Pallud JVarlet PDevaux BGeha SBadoual MDeroulers C: Diffuse low-grade oligodendrogliomas extend beyond MRI-defined abnormalities. Neurology 74:172417312010

  • 54

    Palm CAxer MGräßel DDammers JLindemeyer JZilles K: Towards ultra-high resolution fibre tract mapping of the human brain – registration of polarised light images and reorientation of fibre vectors. Front Hum Neurosci 4:92010

  • 55

    Pedeutour-Braccini ZBurel-Vandenbos FGozé CRoger CBazin ACostes-Martineau V: Microfoci of malignant progression in diffuse low-grade gliomas: towards the creation of an intermediate grade in glioma classification?. Virchows Arch 466:4334442015

  • 56

    Persson AIPetritsch CSwartling FJItsara MSim FJAuvergne R: Non-stem cell origin for oligodendroglioma. Cancer Cell 18:6696822010

  • 57

    Pignatti Fvan den Bent MCurran DDebruyne CSylvester RTherasse P: Prognostic factors for survival in adult patients with cerebral low-grade glioma. J Clin Oncol 20:207620842002

  • 58

    Price SJJena RBurnet NGHutchinson PJDean AFPeña A: Improved delineation of glioma margins and regions of infiltration with the use of diffusion tensor imaging: an image-guided biopsy study. AJNR Am J Neuroradiol 27:196919742006

  • 59

    Sahm FCapper DJeibmann AHabel APaulus WTroost D: Addressing diffuse glioma as a systemic brain disease with single-cell analysis. Arch Neurol 69:5235262012

  • 60

    Sanai NBerger MS: Glioma extent of resection and its impact on patient outcome. Neurosurgery 62:7537642008

  • 61

    Sarubbo SDe Benedictis AMerler SMandonnet EBalbi SGranieri E: Towards a functional atlas of human white matter. Hum Brain Mapp 36:311731362015

  • 62

    Scherer J: The forms of growth in gliomas and their practical significance. Brain 63:1351940

  • 63

    Smith JSChang EFLamborn KRChang SMPrados MDCha S: Role of extent of resection in the long-term outcome of low-grade hemispheric gliomas. J Clin Oncol 26:133813452008

  • 64

    Smits ABaumert BG: The clinical value of PET with amino acid tracers for gliomas WHO Grade II. Int J Mol Imaging 2011:3725092011

  • 65

    Soffietti RBaumert BGBello Lvon Deimling ADuffau HFrénay M: Guidelines on management of low-grade gliomas: report of an EFNS-EANO Task Force. Eur J Neurol 17:112411332010

  • 66

    Watanabe MTanaka RTakeda N: Magnetic resonance imaging and histopathology of cerebral gliomas. Neuroradiology 34:4634691992

  • 67

    Yordanova YNMoritz-Gasser SDuffau H: Awake surgery for WHO Grade II gliomas within “noneloquent” areas in the left dominant hemisphere: toward a “supratotal” resection. Clinical article. J Neurosurg 115:2322392011

  • 68

    Zlatescu MCTehraniYazdi ASasaki HMegyesi JFBetensky RALouis DN: Tumor location and growth pattern correlate with genetic signature in oligodendroglial neoplasms. Cancer Res 61:671367152001

Disclosures

The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

Author Contributions

Conception and design: Zetterling, Smits. Acquisition of data: Zetterling, Berntsson, Latini, Larsson, Smits. Analysis and interpretation of data: Zetterling, Roodakker, Latini, Alafuzoff, Larsson, Smits. Drafting the article: Zetterling, Roodakker, Bernts son, Latini, Alafuzoff, Larsson, Smits. Critically revising the article: all authors. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Zetterling. Administrative/technical/material support: Roodakker

If the inline PDF is not rendering correctly, you can download the PDF file here.

Article Information

INCLUDE WHEN CITING Published online February 26, 2016; DOI: 10.3171/2015.10.JNS15583.

Correspondence Maria Zetterling, Department of Neuroscience, Neurosurgery, Uppsala University, Uppsala 75185, Sweden. email: maria. zetterling@akademiska.se.

© AANS, except where prohibited by US copyright law.

Headings

Figures

  • View in gallery

    Case 1. Preoperative coronal T2-FLAIR (A) and sagittal T2-weighted (B) MR images showing a nonenhancing tumor in the right frontal pole. Postoperative coronal T1-weighted (C) and sagittal T2-weighted (D) MR images showing complete resection of the tumor.

  • View in gallery

    Case 1. A: Preoperative delineated tumor on coronal T2-FLAIR MRI. B: Postoperative delineated cavity on coronal T1-weighted MRI. C: Overlay of the preoperative delineated tumor and postoperative delineated cavity showing the resected peritumoral margins on coronal T2-FLAIR MRI. D: Overlay of IDH1 image on coronal T2-FLAIR MR image. Figure is available in color online only.

  • View in gallery

    Case 1. An IDH1 image from the peripheral part of the tumor shows the brown-stained, infiltrating tumor cells following the white matter tracts. The peripheral parts of the white matter tracts show the highest tumor cell density. At the tumor margins, the white matter is occupied by a “single lane” of tumor cells. The tumor cells are visible outside the radiological border (inset). Figure is available in color online only.

  • View in gallery

    Case 1. An IDH1 image and corresponding coronal T2-FLAIR MR image from the core of the tumor. Left: In the tumor core, tumor cells are found at a high density in the gray matter. One area has a central part with relatively few tumor cells and vacuoles surrounded by a rim of densely packed tumor cells. Inset: The central part of the core consists of tumor cells and vacuoles. Right: The central part with tumor cells and vacuoles corresponds to low signal intensity on T2-FLAIR MRI, whereas the densely packed tumor cells correspond to a ring of high signal intensity on a T2-FLAIR sequence. Figure is available in color online only.

  • View in gallery

    Case 1. An area with a high density of tumor cells was detected outside the radiological border. Left: An IDH1 image with the radiological border delineated by a solid line and the tumor cells outside the radiological border indicated by a dashed line. Right: The corresponding preoperative T2-FLAIR image with the radiological tumor border delineated by a solid line and the area with tumor cells missed on MRI delineated by a dashed line. Figure is available in color online only.

  • View in gallery

    Case 2. IDH1 images from the peripheral part and core of the tumor. Black lines delineate the radiological border. A: An IDH1 image from a peripheral part of the tumor illustrating tumor cells that follow the white matter tracts. B: In the tumor core, tumor cells are found at high density in both the gray and white matter. Tumor cells are present outside the radiological border. There are fewer tumor cells at the distant side of the sulcus (right side of inset C) compared with the proximal side (left side of inset C). Tumor cells infiltrated along the white mater tracts, resulting in a higher density of tumor cells at the bottom of the sulcus (inset D), compared with the more rostral part of the gyrus (inset E). Figure is available in color online only.

  • View in gallery

    Case 2. Tumor cells are present outside the radiological border, which is indicated by a black line. The area marked with an asterisk is outside the radiological border. Fewer tumor cells occur farther away from the border. Tumor cells outside the radiological border occur more frequently closer to the border (lower inset). Farther from the border no tumor cells are found (upper inset). Figure is available in color online only.

  • View in gallery

    Case 3. Preoperative coronal T2-FLAIR (A) and sagittal T1-weighted (B) MR images showing a right-sided nonenhancing tumor in the premotor area. Postoperative coronal T1-weighted MR image (C) showing complete tumor resection.

  • View in gallery

    Case 3. IDH1 images from the core and the periphery of the tumor. A: In the tumor core, the tumor cells occur at a high density in both gray and white matter. B: There is a high tumor cell density in the proximal gray matter in connection to the pia overlaying the sulcus. In the adjacent gyrus, tumor cells are scarce, with a declining number of tumor cells farther from the bottom of the sulcus (that is, from left to right in the figure, magnified in insets C and D, respectively). C: More tumor cells are found close to the bottom of the sulcus. D: No tumor cells are found farther from the bottom of the sulcus. E: In the tumor periphery, tumor cells primarily follow the white matter tracts, at a slightly higher density at the border of the gray matter, and partly infiltrate the white matter tracts.

References

1

Amberger VRHensel TOgata NSchwab ME: Spreading and migration of human glioma and rat C6 cells on central nervous system myelin in vitro is correlated with tumor malignancy and involves a metalloproteolytic activity. Cancer Res 58:1491581998

2

Bauchet LEpidemiology of diffuse low-grade gliomas. Duffau H: Diffuse Low-Grade Gliomas in Adults LondonSpringer2013. 15

3

Berntsson SGFalk ASavitcheva IGodau AZetterling MHesselager G: Perfusion and diffusion MRI combined with 11C-methionine PET in the preoperative evaluation of suspected adult low-grade gliomas. J Neurooncol 114:2412492013

4

Capelle LFontaine DMandonnet ETaillandier LGolmard JLBauchet L: Spontaneous and therapeutic prognostic factors in adult hemispheric World Health Organization Grade II gliomas: a series of 1097 cases: clinical article. J Neurosurg 118:115711682013

5

Capper DZentgraf HBalss JHartmann Cvon Deimling A: Monoclonal antibody specific for IDH1 R132H mutation. Acta Neuropathol 118:5996012009

6

Choe ASGao YLi XCompton KBStepniewska IAnderson AW: Accuracy of image registration between MRI and light microscopy in the ex vivo brain. Magn Reson Imaging 29:6836922011

7

Claus EBBlack PM: Survival rates and patterns of care for patients diagnosed with supratentorial low-grade gliomas: data from the SEER program, 1973–2001. Cancer 106:135813632006

8

de Castro FBribián A: The molecular orchestra of the migration of oligodendrocyte precursors during development. Brain Res Brain Res Rev 49:2272412005

9

Dhermain FGHau PLanfermann HJacobs AHvan den Bent MJ: Advanced MRI and PET imaging for assessment of treatment response in patients with gliomas. Lancet Neurol 9:9069202010

10

Duffau H: The challenge to remove diffuse low-grade gliomas while preserving brain functions. Acta Neurochir (Wien) 154:5695742012

11

Duffau H: Long-term outcomes after supratotal resection of diffuse low-grade gliomas: a consecutive series with 11-year follow-up. Acta Neurochir (Wien) 158:51582016

12

Duffau H: Surgery of low-grade gliomas: towards a ‘functional neurooncology’. Curr Opin Oncol 21:5435492009

13

Duffau HCapelle L: Preferential brain locations of low-grade gliomas. Cancer 100:262226262004

14

Duffau HLopes MArthuis FBitar ASichez JPVan Effenterre R: Contribution of intraoperative electrical stimulations in surgery of low grade gliomas: a comparative study between two series without (1985–96) and with (1996–2003) functional mapping in the same institution. J Neurol Neurosurg Psychiatry 76:8458512005

15

Duffau HMandonnet E: The “oncofunctional balance” in surgery for diffuse low-grade glioma: integrating the extent of resection with quality of life. Acta Neurochir (Wien) 155:9519572013

16

Duffau HTaillandier L: New concepts in the management of diffuse low-grade glioma: Proposal of a multistage and individualized therapeutic approach. Neuro Oncol 17:3323422015

17

Ellingson BMMalkin MGRand SDConnelly JMQuinsey CLaViolette PS: Validation of functional diffusion maps (fDMs) as a biomarker for human glioma cellularity. J Magn Reson Imaging 31:5385482010

18

Eriksson SHFree SLThom MHarkness WSisodiya SMDuncan JS: Reliable registration of preoperative MRI with histopathology after temporal lobe resections. Epilepsia 46:164616532005

19

Fillmore HLVanMeter TEBroaddus WC: Membrane-type matrix metalloproteinases (MT-MMPs): expression and function during glioma invasion. J Neurooncol 53:1872022001

20

Fisher BJNaumova ELeighton CCNaumov GNKerklviet NFortin D: Ki-67: a prognostic factor for low-grade glioma?. Int J Radiat Oncol Biol Phys 52:99610012002

21

Gerin CPallud JDeroulers CVarlet POppenheim CRoux FX: Quantitative characterization of the imaging limits of diffuse low-grade oligodendrogliomas. Neuro Oncol 15:137913882013

22

Giese ABjerkvig RBerens MEWestphal M: Cost of migration: invasion of malignant gliomas and implications for treatment. J Clin Oncol 21:162416362003

23

Giese AWestphal M: Glioma invasion in the central nervous system. Neurosurgery 39:2352521996

24

Guichet POHugnot JPCellular origin of Grade II gliomas. Duffau H: Diffuse Low-Grade Gliomas in Adults LondonSpringer2013. 7590

25

Hassani ZHugnot JPDeciphering the molecular and cellular basis for dissemination of diffuse low-grade gliomas. Duffau H: Diffuse Low-Grade Gliomas in Adults LondonSpringer2013. 9198

26

Hayashida YHirai TMorishita SKitajima MMurakami RKorogi Y: Diffusion-weighted imaging of metastatic brain tumors: comparison with histologic type and tumor cellularity. AJNR Am J Neuroradiol 27:141914252006

27

Herholz KHölzer TBauer BSchröder RVoges JErnestus RI: 11C-methionine PET for differential diagnosis of low-grade gliomas. Neurology 50:131613221998

28

Ichimura KPearson DMKocialkowski SBäcklund LMChan RJones DT: IDH1 mutations are present in the majority of common adult gliomas but rare in primary glioblastomas. Neuro Oncol 11:3413472009

29

Ius TAngelini EThiebaut de Schotten MMandonnet EDuffau H: Evidence for potentials and limitations of brain plasticity using an atlas of functional resectability of WHO grade II gliomas: towards a “minimal common brain”. Neuroimage 56:99210002011

30

Iwama TYamada HSakai NAndoh TNakashima THirata T: Correlation between magnetic resonance imaging and histopathology of intracranial glioma. Neurol Res 13:48541991

31

Jakola ASMyrmel KSKloster RTorp SHLindal SUnsgård G: Comparison of a strategy favoring early surgical resection vs a strategy favoring watchful waiting in low-grade gliomas. JAMA 308:188118882012

32

Kaschten BStevenaert ASadzot BDeprez MDegueldre CDel Fiore G: Preoperative evaluation of 54 gliomas by PET with fluorine-18-fluorodeoxyglucose and/or carbon-11-methionine. J Nucl Med 39:7787851998

33

Kelly PJDaumas-Duport CKispert DBKall BAScheithauer BWIllig JJ: Imaging-based stereotaxic serial biopsies in untreated intracranial glial neoplasms. J Neurosurg 66:8658741987

34

Kim YHKim CHKim JSLee SKHan JHKim CY: Risk factor analysis of the development of new neurological deficits following supplementary motor area resection. J Neurosurg 119:7142013

35

Kinoshita MHashimoto NGoto TYanagisawa TOkita YKagawa N: Use of fractional anisotropy for determination of the cut-off value in 11C-methionine positron emission tomography for glioma. Neuroimage 45:3123182009

36

Kono KInoue YNakayama KShakudo MMorino MOhata K: The role of diffusion-weighted imaging in patients with brain tumors. AJNR Am J Neuroradiol 22:108110882001

37

Kracht LWMiletic HBusch SJacobs AHVoges JHoevels M: Delineation of brain tumor extent with [11C]l-methionine positron emission tomography: local comparison with stereotactic histopathology. Clin Cancer Res 10:716371702004

38

Laigle-Donadey FMartin-Duverneuil NLejeune JCrinière ECapelle LDuffau H: Correlations between molecular profile and radiologic pattern in oligodendroglial tumors. Neurology 63:236023622004

39

Leu Svon Felten SFrank SVassella EVajtai ITaylor E: IDH/MGMT-driven molecular classification of low-grade glioma is a strong predictor for long-term survival. Neuro Oncol 15:4694792013

40

Louis DNOhgaki HWiestler ODCavenee WKBurger PCJouvet A: The 2007 WHO classification of tumours of the central nervous system. Acta Neuropathol 114:971092007

41

Lunsford LDMartinez AJLatchaw RE: Magnetic resonance imaging does not define tumor boundaries. Acta Radiol Suppl 369:1541561986

42

Magnain CAugustinack JCReuter MWachinger CFrosch MPRagan T: Blockface histology with optical coherence tomography: a comparison with Nissl staining. Neuroimage 84:5245332014

43

Mandonnet EBiomathematical modeling of DLGG behavior. Duffau H: Diffuse Low-Grade Gliomas in Adults LondonSpringer2013. 447455

44

Mandonnet ECapelle LDuffau H: Extension of paralimbic low grade gliomas: toward an anatomical classification based on white matter invasion patterns. J Neurooncol 78:1791852006

45

Mandonnet EDelattre JYTanguy MLSwanson KRCarpentier AFDuffau H: Continuous growth of mean tumor diameter in a subset of grade II gliomas. Ann Neurol 53:5245282003

46

McGirt MJChaichana KLAttenello FJWeingart JDThan KBurger PC: Extent of surgical resection is independently associated with survival in patients with hemispheric infiltrating low-grade gliomas. Neurosurgery 63:7007082008

47

Mentlein RHattermann KHeld-Feindt J: Lost in disruption: role of proteases in glioma invasion and progression. Biochim Biophys Acta 1825:1781852012

48

Miyamoto YYamauchi JTanoue A: Cdk5 phosphorylation of WAVE2 regulates oligodendrocyte precursor cell migration through nonreceptor tyrosine kinase Fyn. J Neurosci 28:832683372008

49

Oishi KZilles KAmunts KFaria AJiang HLi X: Human brain white matter atlas: identification and assignment of common anatomical structures in superficial white matter. Neuroimage 43:4474572008

50

Olivier CCobos IPerez Villegas EMSpassky NZalc BMartinez S: Monofocal origin of telencephalic oligodendrocytes in the anterior entopeduncular area of the chick embryo. Development 128:175717692001

51

Pallud JDiffuse low-grade gliomas: What does “complete resection” mean?. Hayat MA: Tumors of the Central Nervous System DordrechtSpringer2011. 2:153161

52

Pallud JMandonnet EDuffau HKujas MGuillevin RGalanaud D: Prognostic value of initial magnetic resonance imaging growth rates for World Health Organization grade II gliomas. Ann Neurol 60:3803832006

53

Pallud JVarlet PDevaux BGeha SBadoual MDeroulers C: Diffuse low-grade oligodendrogliomas extend beyond MRI-defined abnormalities. Neurology 74:172417312010

54

Palm CAxer MGräßel DDammers JLindemeyer JZilles K: Towards ultra-high resolution fibre tract mapping of the human brain – registration of polarised light images and reorientation of fibre vectors. Front Hum Neurosci 4:92010

55

Pedeutour-Braccini ZBurel-Vandenbos FGozé CRoger CBazin ACostes-Martineau V: Microfoci of malignant progression in diffuse low-grade gliomas: towards the creation of an intermediate grade in glioma classification?. Virchows Arch 466:4334442015

56

Persson AIPetritsch CSwartling FJItsara MSim FJAuvergne R: Non-stem cell origin for oligodendroglioma. Cancer Cell 18:6696822010

57

Pignatti Fvan den Bent MCurran DDebruyne CSylvester RTherasse P: Prognostic factors for survival in adult patients with cerebral low-grade glioma. J Clin Oncol 20:207620842002

58

Price SJJena RBurnet NGHutchinson PJDean AFPeña A: Improved delineation of glioma margins and regions of infiltration with the use of diffusion tensor imaging: an image-guided biopsy study. AJNR Am J Neuroradiol 27:196919742006

59

Sahm FCapper DJeibmann AHabel APaulus WTroost D: Addressing diffuse glioma as a systemic brain disease with single-cell analysis. Arch Neurol 69:5235262012

60

Sanai NBerger MS: Glioma extent of resection and its impact on patient outcome. Neurosurgery 62:7537642008

61

Sarubbo SDe Benedictis AMerler SMandonnet EBalbi SGranieri E: Towards a functional atlas of human white matter. Hum Brain Mapp 36:311731362015

62

Scherer J: The forms of growth in gliomas and their practical significance. Brain 63:1351940

63

Smith JSChang EFLamborn KRChang SMPrados MDCha S: Role of extent of resection in the long-term outcome of low-grade hemispheric gliomas. J Clin Oncol 26:133813452008

64

Smits ABaumert BG: The clinical value of PET with amino acid tracers for gliomas WHO Grade II. Int J Mol Imaging 2011:3725092011

65

Soffietti RBaumert BGBello Lvon Deimling ADuffau HFrénay M: Guidelines on management of low-grade gliomas: report of an EFNS-EANO Task Force. Eur J Neurol 17:112411332010

66

Watanabe MTanaka RTakeda N: Magnetic resonance imaging and histopathology of cerebral gliomas. Neuroradiology 34:4634691992

67

Yordanova YNMoritz-Gasser SDuffau H: Awake surgery for WHO Grade II gliomas within “noneloquent” areas in the left dominant hemisphere: toward a “supratotal” resection. Clinical article. J Neurosurg 115:2322392011

68

Zlatescu MCTehraniYazdi ASasaki HMegyesi JFBetensky RALouis DN: Tumor location and growth pattern correlate with genetic signature in oligodendroglial neoplasms. Cancer Res 61:671367152001

TrendMD

Cited By

Metrics

Metrics

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 469 469 81
PDF Downloads 232 232 25
EPUB Downloads 0 0 0

PubMed

Google Scholar