ArticlePDF Available

Abstract

Background Lower-grade gliomas may be indolent for many years before developing malignant behaviour. The reasons mechanisms underlying malignant progression remain unclear. Methods We collected blocks of live human brain tissue donated by people undergoing glioma resection. The tissue blocks extended through the peritumoral cortex and into the glioma. The living human brain tissue was cut into ex vivo brain slices and bathed in 5-aminolevulinic acid (5-ALA). High-grade glioma cells avidly take up 5-aminolevulinic acid (5-ALA) and accumulate high levels of the fluorescent metabolite, Protoporphyrin IX (PpIX). We exploited the PpIX fluorescence emitted by higher-grade glioma cells to investigate the earliest stages of malignant progression in lower-grade gliomas. Results We found sparsely-distributed ‘hot-spots’ of PpIX-positive cells in living lower-grade glioma tissue. Glioma cells and endothelial cells formed part of the PpIX hotspots. Glioma cells in PpIX hotspots were IDH1 mutant and expressed nestin suggesting they had acquired stem-like properties. Spatial analysis with 5-ALA conjugated quantum dots indicated that these glioma cells replicated adjacent to blood vessels. PpIX hotspots formed in the absence of angiogenesis. Conclusion Our data show that PpIX hotspots represent microdomains of cells with high-grade potential within lower-grade gliomas and identify locations where malignant progression could start.
Accepted Manuscript
© The Author(s) 2021. Published by Oxford University Press, the Society for Neuro-Oncology and the
European Association of Neuro-Oncology.
This is an Open Access article distributed under the terms of the Creative Commons Attribution
License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse,
distribution, and reproduction in any medium, provided the original work is properly cited.
Multicellular ‘hotspots’ harbour high-grade potential in lower-
grade gliomas
Alastair J. Kirby1, José P. Lavrador2, Istvan Bodi1,3, Francesco Vergani2, Ranjeev Bhangoo2,
Keyoumars Ashkan1,2, Gerald T. Finnerty1,4,*
1Department of Basic and Clinical Neuroscience, King‟s College London, De Crespigny
Park, London SE5 8AF, UK; 2Department of Neurosurgery, King‟s College Hospital NHS
Foundation Trust, Denmark Hill, London SE5 9RS, UK; 3Department of Clinical
Neuropathology, King‟s College Hospital NHS Foundation Trust, Denmark Hill, London
SE5 9RS, UK; 4Department of Neurology, King‟s College Hospital NHS Foundation Trust,
Denmark Hill, London SE5 9RS, UK.
ORCID IDs: Kirby 0000-0003-1525-0902; Finnerty 0000-0003-1082-0128
*Corresponding Author and lead contact: G.T.F. (gerald.finnerty@kcl.ac.uk), Tel: +44 20
3299 8352
Funding: Medical Research Council (UK) (MR/N013700/1), Psychiatry Research Trust,
Inman Charity.
Conflict of Interest: AJK is CEO of Vivisco Scientific. Vivisco Scientific made no financial
contribution to this study. GTF is a scientific consultant for Vivisco Scientific.
Authorship: AJK & GTF designed the study. JPL, FV, RB, KA collected samples or data.
AJK did the experimental work. AJK and GTF analysed the data. IB made a
neuropathological diagnosis from fixed samples. AJK & GTF wrote the manuscript with
input from JPL, IB, FV, RB and KA.
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
Abstract
Background: Lower-grade gliomas may be indolent for many years before developing
malignant behaviour. The reasons mechanisms underlying malignant progression remain
unclear.
Methods: We collected blocks of live human brain tissue donated by people undergoing
glioma resection. The tissue blocks extended through the peritumoral cortex and into the
glioma. The living human brain tissue was cut into ex vivo brain slices and bathed in 5-
aminolevulinic acid (5-ALA). High-grade glioma cells avidly take up 5-aminolevulinic acid
(5-ALA) and accumulate high levels of the fluorescent metabolite, Protoporphyrin IX (PpIX).
We exploited the PpIX fluorescence emitted by higher-grade glioma cells to investigate the
earliest stages of malignant progression in lower-grade gliomas.
Results: We found sparsely-distributed „hot-spots‟ of PpIX-positive cells in living lower-
grade glioma tissue. Glioma cells and endothelial cells formed part of the PpIX hotspots.
Glioma cells in PpIX hotspots were IDH1 mutant and expressed nestin suggesting they had
acquired stem-like properties. Spatial analysis with 5-ALA conjugated quantum dots
indicated that these glioma cells replicated adjacent to blood vessels. PpIX hotspots formed in
the absence of angiogenesis.
Conclusion: Our data show that PpIX hotspots represent microdomains of cells with high-
grade potential within lower-grade gliomas and identify locations where malignant
progression could start.
Keywords: brain tumor, glia, malignant progression, vessel co-option, nestin
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
Key points
Malignant progression of lower-grade gliomas studied in slices of living human brain
tissue
Lower-grade gliomas contain hotspots of nestin-positive glioma cells and endothelial
cells
Glioma hotspots may signify the earliest stages of malignant progression
Importance of the study
Lower-grade gliomas may be indolent for many years before developing malignant
behaviour. The mechanisms underlying malignant progression remain unclear. We studied
where malignant progression starts in gliomas by using living human brain tissue donated by
people undergoing glioma resection. High-grade glioma cells were labelled fluorescently and
imaged in living ex vivo human brain slices. We found sparsely-distributed groups of glioma
cells with high-grade features clustered around small blood vessels, which we termed
hotspots. The hotspots occur in the absence of angiogenesis. We propose that the hotspots are
the seedbeds for malignant progression.
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
Introduction
Gliomas are the commonest primary brain tumor.1 A subset of diffuse gliomas are indolent
for many years before they adopt more aggressive behaviour.2 Management of these lower-
grade diffuse gliomas is a significant clinical problem because they usually affect young
adults. The median survival varies between 5.2 - 7.2 years with only 20% of patients
surviving for two decades.3 Due to the young age at presentation, lower-grade gliomas cause
a large loss of “potential years of life”.4
Greater understanding of the mechanisms underlying malignant progression of lower-grade
diffuse gliomas is key to developing new treatments. Previous work has focused on molecular
mechanisms, such as new mutations,5,6 changes in gene expression7,8 and altered signalling
pathways.9 However, these studies do not identify the earliest stages of malignant
progression. The issue is that they compare a time before malignant progression with a time
when malignant progression is established. The studies do not tell us how the transition
between the two tumor states occurs. Accordingly, we don‟t know which molecular changes
initiate malignant progression, which act to sustain it, and which are the result of malignant
progression.
The site where malignant progression begins could give important information about the
underlying mechanisms. Identifying the spatial origins of malignant progression would open
opportunities to study how it is initiated and the factors that sustain it. However, currently, it
is not known whether malignant progression can start in any part of a lower-grade diffuse
glioma or whether it occurs at specific locations where the glioma microenvironment fosters
malignant progression.
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
We investigated the spatial origins of malignant progression at a cellular level by using ex
vivo human brain tissue. In common with other tumors, high-grade glioma cells exhibit
metabolic reprogramming.10 Glioma cells avidly take up 5-aminolevulinic acid (5-ALA),
which is upstream in the heme biosynthesis pathway, and metabolize it to the fluorescent
molecule, Protoporphyrin IX (PpIX).11 High-grade glioma cells accumulate high levels of
PpIX, which makes them fluorescent.11 Neurosurgeons have taken advantage of 5-ALA-
induced PpIX fluorescence to aid resection of higher-grade gliomas intraoperatively.12
We exploited the PpIX fluorescence emitted by higher-grade glioma cells to investigate the
earliest stages of malignant progression in lower-grade gliomas. We collected blocks of live
human brain tissue donated by people undergoing glioma resection and prepared ex vivo
brain slices, which preserved the architecture of the glioma and peritumoral cortex. The
human brain slices were bathed in 5-ALA to enable us to image cells in the tissue with PpIX
fluorescence. We identified sparsely distributed clusters of PpIX fluorescent cells, which we
refer to as PpIX hotspots. The PpIX hotspots in lower-grade glioma tissue contained nestin
positive (nestin+) glioma cells and endothelial cells. Our data suggest that the PpIX hotspots
represent microdomains of cells with high-grade potential within lower-grade gliomas.
Materials and Methods
Ethical approval and consent
The UK Human Research Authority (https://www.hra.nhs.uk/) approved the collection of ex
vivo brain samples following a favorable opinion from the South West Research Ethics
Committee (REC approval code: 18/SW/0022).
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
The ex vivo brain samples were donated by participants who underwent a craniotomy for
either tumor resection or intractable epilepsy at King‟s College Hospital between March 2018
and December 2019 (Table 1). All participants gave informed consent prior to their surgery.
A person was eligible for the study if (s)he required brain surgery that necessarily involved
removal of brain tissue to treat the participant‟s disease. Individuals who had multiple brain
diseases or previous chemotherapy/radiotherapy were excluded. Malignant progression of
diffuse lower-grade gliomas was identified preoperatively by neuroimaging, e.g. contrast
enhancement, increased perfusion, and then confirmed by finding focal, high-grade features
neuropathologically. No participants were prescribed anticonvulsant medications, which
reduce 5-ALA synthesis in glioma cells.13
Live human brain tissue collection
Participants with suspected WHO grade III or grade IV gliomas were given 5-ALA
(Gliolan®, 20 mg/kg) 2 - 4 hours prior to their surgery.
The location of the brain sample was recorded with the intraoperative neuronavigation system
(StealthViz, Medtronic, Minneapolis, USA) and a brightfield image of the surface of the
brain (Zeiss OPMI Pentero 900 or Zeiss KINEVO 900 operating microscope) before the
sample was taken (Fig. 1A, B). PpIX fluorescence was detected by switching the operating
microscope to fluorescence mode (excitation 400 410 nm, emission 620 - 710 nm).
Before removing the block of tissue, the surface of the brain was cooled with chilled Ringer‟s
solution. The ex vivo tissue blocks were excised with a scalpel and immediately placed in
cooled dissection artificial cerebral spinal fluid (aCSF) comprising in mM: 108 choline-Cl, 3
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
KCl, 26 NaHCO3, 1.25 NaHPO4, 25 D-glucose, 3 Na pyruvate, 2 CaCl2, 1 MgCl2 and 0.1
Heparin, and was bubbled with 95 % O2 / 5 % CO2.
The ex vivo brain tissue was transferred to the laboratory in a transportation system that kept
the samples cool while bathing them in dissection aCSF constantly gassed with 95 % Oxygen
/ 5 % Carbon Dioxide.14 After arriving in the laboratory, the ex vivo tissue blocks were bathed
in cooled dissection aCSF and sliced into 300 500 µm thick sections using a vibratome
(Campden Instruments) (Fig. 1C, D). After every fifth section, a 1 mm thick section was cut
and fixed with 4 % paraformaldehyde in 1 mM phosphate buffered saline for 2 - 4 hours at 4
oC for neuropathology.
Ex vivo imaging
The ex vivo brain slices were transferred to an incubation chamber (Scientific Systems
Design, Ontario, Canada). The artificial CSF in the incubation chamber comprised (mM):
120 NaCl, 3 KCl, 23 NaHCO3, 1.25 NaHPO4, 10 D-glucose, 2 CaCl2, and 1 MgCl2 bubbled
with 95 % Oxygen / 5 % CO2. The artificial CSF was gradually warmed to 37 C and 1 mM
5-aminolevulinic acid (5-ALA, Sigma: 5451-09-2) was added prior to imaging. The ex vivo
samples were imaged on an interface recording chamber (Scientific Systems Design, Ontario
Canada) mounted on an Olympus BX51WI epifluorescence microscope. The ex vivo
fluorescence and brightfield images were imaged with a Spot RT sCMOS cooled 5MP
camera (RT39M5, Spot Imaging, USA) controlled by Spot Advanced imaging software (Spot
Imaging, USA).
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
5-ALA conjugated quantum dots
We conjugated 5-ALA to fluorescence quantum dots nanocrystals following the
manufacturer‟s instructions (ThermoFisher Qdot Probes; Supplementary Methods). The 5-
ALA-conjugated Qdots were incubated with ex vivo brain tissue in artificial CSF with 1 mM
5-ALA for four hours. The slices were imaged on an Olympus BX51WI epifluorescence
microscope using a Spot RT sCMOS cooled 5MP camera before fixation in 4 % PFA for 20
minutes.
Ex vivo PpIX image analysis
PpIX fluorescence were quantified in FIJI (https://imagej.net/Fiji) using a custom-written
script (Supplementary Methods, Supplementary Fig. 1).
Neuropathological processing
Brain tissue slices, 1 mm thick, were fixed in 10 % formalin and embedded in paraffin
blocks. Immunohistochemistry and neuropathological assessment for brain tumors were
performed by the Neuropathology Department, King‟s College Hospital. Sections were
embedded in paraffin blocks and 4 - 5 µm sections were used for immunohistochemistry.
Neuropathology slides were imaged on an Olympus VS120 Slide scanner and quantified
using a custom-written script in FIJI (Supplementary Methods, Supplementary Fig. 2).
WHO grade of ex vivo human brain samples
The glioma diagnosis was based on the histopathological features of the diagnostic biopsy
from the tumor core. The ex vivo research samples were graded by a Consultant
Neuropathologist (IB) who was blind to the PpIX fluorescence data (Table 1).
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
We identified two ex vivo samples from WHO grade IV gliomas that had no or infrequent
tumor cells (< 10 cells.mm-2). These samples were combined with tissue from one case of
focal cortical dysplasia and one case of metastatic adenocarcinoma and used as controls.
Immunofluorescence
After live imaging was completed, brain slices were fixed, resectioned (50 µm thick) and
immunofluorescence staining was performed using standard techniques (Supplementary
Methods). Resectioned slices were imaged on an inverted confocal microscope (Nikon AR1)
using a 20x air objective (NA 0.75) and pinhole size of 1.2 Airy units. Z-stacks with 5 µm
steps were acquired through the entire slice. The same gain and laser power were used to
image each of the fluorophores. Images were acquired using NIS-elements (Nikon,
Amsterdam, Holland) and analysed with FIJI software (https://imagej.net/Fiji).
Statistical analysis
We could not perform all tests on all samples. Therefore, the data used for each figure panel
is listed in Supplementary Figure 2. Statistical analysis and graphing were undertaken in
Graph Pad Prism 7. Data were described by their mean ± standard error or median and
interquartile range. Statistical tests were two-tailed and had a threshold for type 1 statistical
error of α < 0.05. Means were compared with t-tests or a repeated measures ANOVA if the
data fulfilled the assumptions for parametric tests. A Mann-Whitney U-test was used when
parametric tests were not appropriate. The linear regression function was used to fit lines to
data that fulfilled the normality and equal variance conditions. A generalized linear model
running under R (R project for Statistical Programming, https://www.r-project.org/) was used
to fit a linear relationship (quasipoisson family) between PpIX+ cells and glioma grade.
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
Results
Neuropathological examination of lower-grade gliomas reveals that malignant progression
starts focally within the tumor.15 Therefore, we investigated whether our ex vivo brain tissue
samples, which included the invasive edge of gliomas, had the same neuropathological
features as the diagnostic biopsies (Fig. 1E). We found that our ex vivo brain samples were
frequently classified as having the neuropathological features of a WHO grade II glioma
when the diagnostic biopsy was WHO grade III (ex vivo sample/diagnostic biopsy; 4/9 WHO
grade III, Table 1). We refer to our ex vivo brain samples as WHO II {III} when the ex vivo
sample was graded as WHO II and the diagnostic biopsy graded as WHO III. In participants
with glioblastoma (WHO grade IV), no glioma cells (NGC) were present in two patients‟
research samples collected from the six participants with glioblastoma.
We reasoned that we could use brain tissue with WHO grade II neuropathological features
that was collected from lower-grade gliomas to study the origins of malignant progression.
PpIX ‘hot-spots’ at the edge of lower-grade gliomas
Tumor cells exhibit increasing metabolic changes as they become more malignant.10 The
altered cellular metabolism is a source of biomarkers for tumor cells. Neurosurgeons use the
increase in PpIX fluorescence emitted by glioma cells intraoperatively to improve resection
of higher-grade gliomas.12 More recently, it has been proposed that macroscopic PpIX
fluorescence visible during surgery to resect lower-grade gliomas indicates an area of
malignant progression.16 Therefore, we explored which cells exhibit PpIX fluorescence in
adult lower-grade gliomas. We cut blocks of ex vivo human brain tissue donated by patients
undergoing glioma resection into brain slices to enable us to image cellular PpIX
fluorescence in living tissue.
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
PpIX is a naturally occurring molecule. Therefore, we first studied PpIX fluorescence in
brain tissue that was not infiltrated by glioma cells to establish baseline levels of PpIX
fluorescence. We found that PpIX fluorescence in the ex vivo brain tissue that had no glioma
cells (NGC) frequently formed tubular structures resembling blood vessels (Fig. 2A,
Supplementary Fig. 3A, B). We quantified the cells exhibiting PpIX fluorescence using
Particle Analysis in FIJI (Supplementary Methods, Supplementary Fig. 1). Ex vivo brain
samples that had not been infiltrated by glioma contained cells that exhibited PpIX
fluorescence (PpIX fluorescent cells = 7.7 ± 0.8 cells.mm-2, n = 4 patients comprising 2
glioblastomas with no glioma cells in research sample, 1 focal cortical dysplasia, 1 metastatic
adenocarcinoma) (Fig. 2B - C). Henceforth, any cells exhibiting PpIX fluorescence are
termed PpIX+.
We then studied ex vivo human brain tissue infiltrated by glioma cells. In lower-grade
gliomas, PpIX+ cells formed bright cylindrical clusters, which we refer to as „PpIX hotspots‟
(Fig. 2A - B). In contrast, the PpIX+ cells at the edge of WHO grade IV gliomas appeared
more uniformly spread (Fig. 2A, Supplementary Fig. 1). We found that the density of PpIX+
cells increased with WHO grade (PpIX+ cell density, median [IQR]: NGC, 7.7 [6.0 - 9.0]
cells.mm-2; WHO II, 10.9 [8.7 14.4] cells.mm-2; WHO II {III}, 13.6 [5.3 21.9] cells.mm-2;
WHO III, 18.1 [13.8 26.0] cells.mm-2; WHO IV, 19.5 [14.7 24.9] cells.mm-2; generalised
linear model (quasipoisson), slope = 3.1 ± 0.8, t = 3.76, P = 0.002, n = 18 brain samples
comprising 14 gliomas and 4 NGC controls) (Fig. 2C). We concluded that, firstly, PpIX+
cells in high-grade gliomas were present at a higher density than in low-grade gliomas and,
secondly, PpIX+ cells in low-grade gliomas tended to clump together to form cylindrical
structures.
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
Molecular alterations in glioma cells are a better guide to clinical prognosis than the
neuropathological WHO grade.17 Therefore, we investigated the relationship between PpIX
fluorescence and the 2016 WHO molecular classification of gliomas.18 We found that
differences in the density of PpIX+ cells across molecular subtypes, when not accounting for
grade, were not statistically significant (median [IQR]: IDH1mut/1p19qco-del 14.4 [8.7 21.1]
cells.mm-2; IDH1mut/1p19qintact 18.0 [13.8 22.2] cells.mm-2; IDH1wt 19.0 [14.7 24.9]
cells.mm-2 ; One-way ANOVA, F(2, 10) = 0.86, P = 0.45, n = 13 gliomas) (Fig. 2D).
Collectively, our findings indicate that, firstly, PpIX+ cell density more closely reflects the
WHO grade of the ex vivo samples rather than its molecular subtype and, secondly, PpIX+
cell density increases with malignant progression. This is consistent with the idea that
metabolic changes in tumor cells become more pronounced as tumors become malignant.10
PpIX fluorescence increases sublinearly with glioma infiltration
We next asked whether the density of PpIX+ cells was related to glioma cell infiltration or to
the density of proliferating cells in low-grade gliomas and high-grade gliomas. Glioma cells
were labelled for mutant IDH1 protein (IDH1mut) or for nestin19. The ex vivo brain tissue
samples were allocated to low-grade and high-grade glioma groups based on the WHO grade
of the diagnostic biopsies. We investigated whether the diagnostic grade of the glioma was
related to cellular changes at its the invasive edge (Fig. 3A). We found a power relationship
between PpIX+ cells and tumor infiltration for high-grade gliomas (Fig. 3B; HGG slope =
0.13 ± 0.04, R2 = 0.65, P = 0.02, n = 8 WHO grade III or IV gliomas). However, there was no
power relationship for low-grade gliomas (LGG slope = 0.27 ± 0.14, R2 = 0.48, P = 0.21, n =
6 WHO grade II or grade II {III} gliomas).
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
We next investigated whether the density of PpIX+ cells was related to the density of
proliferating cells expressing Ki-67 in our ex vivo samples from the tumor edge, again using
the diagnostic WHO groups. The density of PpIX+ cells in low-grade and high-grade gliomas
did not increase with the density of proliferating cells (Fig. 3C; LGG slope = -0.18 ± 0.29, R2
= 0.09, P = 0.57, n = 6 WHO grade II or II {III} gliomas; HGG slope = 0.07 ± 0.05, R2 =
0.35, P = 0.22, n = 6 WHO grade III or IV gliomas).
We revisited the relationship between PpIX fluorescence and glioma infiltration. WHO
grading of a biopsy gives a snapshot of the cellular features of the glioma, but is less accurate
guide to prognosis than molecular profiling.17 The IDH1mut molecular marker demarcates a
subset of WHO grade II and grade III gliomas with a better prognosis.17 These WHO grade II
and grade III gliomas are termed lower-grade.17 We reasoned that we could use the IDH1mut
molecular marker to reduce variability between gliomas and that this would give us a better
way to assess whether the density of lower-grade glioma cells is related to PpIX+ cell density.
We found a power relationship between the density of IDH1mut cells and of PpIX+ cells in our
ex vivo samples from lower-grade gliomas (Fig. 3D; slope = 0.24 ± 0.06, R2 = 0.77, P =
0.009, n = 7 WHO grade II, II {III} or III gliomas). The power coefficient of 0.24 indicated
that the density of PpIX+ cells increases as approximately the fourth root of the density of
lower-grade glioma cells. We concluded that PpIX+ cells increased with glioma cell number
if analysis was restricted to IDH1mut lower-grade gliomas. However, the increase was
markedly sublinear suggesting the glioma cell number is a weak determinant of PpIX
fluorescence.
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
Nestin+ glioma cells and CD34+ endothelial cells contribute to PpIX ‘hotspots’ in IDH1mut
diffuse gliomas
We had observed that PpIX fluorescence outlined vessel-like structures in control brain
samples. Therefore, we investigated the contributions of glioma cells and endothelial cells to
PpIX fluorescence. Our analysis had shown that PpIX fluorescence increases sublinearly as
the number of IDH1mut glioma cells increases. Therefore, we asked whether nestin+ cells in
lower-grade IDH1mut gliomas were linked to PpIX fluorescence. We found that the density of
PpIX+ cells grew exponentially with the density of nestin+ cells (Fig. 4A, B; slope = 0.002 ±
0.0004, R = 0.78, P = 0.008, n = 7 IDH1mut gliomas). This suggests that nestin+ glioma cells
contribute to PpIX fluorescence. However, endothelial cells may express nestin when
replicating.20 Glioma cells tend to be rounded whereas endothelial cells tend to be elongated.
Therefore, we applied a shape (circulatory) filter to separate the effects of elongated nestin+
cells from rounded nestin+ cells on PpIX fluorescence (Fig. 4A, C). We found a strong
relationship between rounded nestin+ cells and PpIX+ fluorescence at the edge of IDH1mut
lower-grade gliomas (Fig. 4C; slope = 0.003 ± 0.001, R2 = 0.85, P = 0.003, n = 7 IDH1mut
gliomas), but no correlation with elongated nestin+ cells (Fig. 4D; slope = 0.23 ± 0.03, R2 =
0.29, P = 0.22, n = 7 IDH1mut gliomas). Taken together, our findings suggest that the density
of PpIX+ cells in lower-grade gliomas grows exponentially as the density of nestin+ glioma
cells increases.
We next studied the contribution of endothelial cells to PpIX hotspots by staining the tissue
for endothelial cell marker, CD34 (Supplementary Fig. 3C). We found no relationship when
we compared the density of CD34+ cells with the density of PpIX+ cells in IDH1mut gliomas
(Fig. 4E; R2 = 0.04, P = 0.51, n = 13 comprising 12 gliomas and 1 focal cortical dysplasia).
However, CD34 is also expressed on reactive microglia and on a subset of glioma cells.21,22
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
Therefore, we applied the shape (circulatory) filter again to distinguish elongated endothelial
cells from rounded microglia and glioma cells (Fig. 4A). After applying the shape filter, we
found a linear relationship between the density of putative CD34+ endothelial cells and the
density of PpIX+ cells (Fig. 4F; slope = 0.23 ± 0.05, R2 = 0.74, P = 0.006, n = 8 IDH1mut
gliomas). Endothelial cells that are entering into a proliferative state express nestin.20
However, application of the shape (circulatory) filter did not uncover a relationship between
elongated nestin+ cells and PpIX+ cells in lower-grade gliomas (Fig. 4D; slope = 0.02 ± 0.02,
R2 = 0.07, P = 0.48, n = 9 IDH1mut gliomas). Collectively, our data suggested that endothelial
cells contributed to PpIX hotspots in lower-grade gliomas. However, the endothelial cells in
PpIX hotspots were not proliferating in large numbers. Moreover, there was no correlation
between the densities of rounded CD34+ cells and PpIX+ cells (Fig. 4G; slope = slope = 0.01
± 0.02, R2 = 0.06, P = 0.61, n = 8 IDH1mut gliomas) suggesting the activated microglia were
not major contributors to the PpIX signal.
Our data indicated that the PpIX+ cells in PpIX hotspots were a combination of nestin+ glioma
cells and endothelial cells. Nestin is commonly used as a marker for glioma stem/progenitor
cells.23 However, nestin is also expressed by other types of activated glial cells, such as
reactive astrocytes and activated microglia.24,25 We estimated the PpIX+ cell density
attributable to putative glioma cells by subtracting the average contribution of PpIX+ putative
endothelial cells from the PpIX+ cell total (Methods). In lower-grade IDH1mut gliomas, we
found that the number of PpIX+ putative glioma increased with the number of nestin+ cells
(Fig. 4H; slope = 0.04 ± 0.01, R2 = 0.65, P = 0.03, n = 7 IDH1mut gliomas). We concluded
that PpIX fluorescence is more common in nestin+ glioma cells.
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
5-ALA-conjugated quantum dots accumulate around blood vessels and in glioma cells
Our data indicated that nestin+ glioma cells and endothelial cells contributed to PpIX
fluorescence. We next investigated the spatial relationship of cells in PpIX hotspots. PpIX
fluorescence photobleaches rapidly. Therefore, we conjugated 5-ALA to quantum dots (5-
ALA-QD), which remain fluorescent after fixation, and incubated ex vivo human brain slices
with 5-ALA-QD (Fig. 5A, B).
Low levels of 5-ALA-QD were taken up by IDH1mut glioma cells (Fig. 5C). We observed
IDH1mut glioma cells that had accumulated 5-ALA-QD and were also proliferating next to
blood vessels (Fig. 5D). 5-ALA-QD were taken up by GFAP+ cell processes surrounding
blood vessels (Fig. 5E - F) and within the blood vessel wall (Fig. 5E, G). These findings
suggest that 5-ALA-QD are taken up by lower-grade glioma cells, astrocytic endfeet on
vessel walls and endothelial cells. We concluded that PpIX+ lower-grade glioma cells
replicate in close proximity to blood vessels.
Discussion
We used live human brain tissue to investigate the spatial origins of malignant progression in
lower-grade gliomas. At the edge of lower-grade diffuse gliomas, we observed sparsely-
distributed clusters of 5-ALA-induced PpIX+ cells that we termed PpIX „hotspots‟. These
PpIX hotspots contained nestin+ glioma cells and endothelial cells. The cells in PpIX hotspots
have characteristics, which suggest that PpIX hotspots may signify the earliest stages of
malignant transformation.
High intracellular levels of PpIX indicate reprogramming of the haem biosynthesis pathway.
This metabolic reprogramming is a feature of high-grade glioma cells and enables them to be
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
visualised macroscopically during surgery.12 In contrast, only a minority of low-grade
gliomas exhibit macroscopic PpIX fluorescence and, when present, is thought to indicate that
malignant progression has occurred.16,26,27 This is consistent with the idea that metabolic
reprogramming becomes more pronounced with malignant progression.10 Our data suggest
that microscopic PpIX hotspots are present in lower-grade glioma tissue in the absence of
macroscopic PpIX fluorescence and without marked increases in proliferation.
We found that the densities of PpIX+ cells increased exponentially with the density of nestin+
cells in lower-grade gliomas. In the healthy central nervous system, nestin is expressed by
neural progenitor/stem cells, but not by differentiated neurons and glial cells.28,29 Nestin has
been used as a marker for glioma stem cells in glioblastomas.23 More work is needed to
quantify the proportion of IDH1 cells that are PpIX+ and nestin+. However, our data suggest
that a subset of PpIX+ lower-grade glioma cells not only exhibit metabolic reprogramming,
but are also in a less differentiated cellular state.
Transcriptomic studies indicate that diffuse lower-grade gliomas originate from neural
progenitor-like cells.30,31 These neural progenitor-like cells replicate, unlike the vast majority
of diffuse lower-grade glioma cells.30,31 Our data indicate that the glioma cells in PpIX
hotspots can divide. Therefore, our findings combined with the transcriptomic studies30,31
suggest that PpIX hotspots include neural progenitor-like cells.
We found that endothelial cells contributed to PpIX hotspots in ex vivo lower-grade glioma
tissue. Notably, our neuropathological and molecular data indicate that the endothelial cells in
PpIX hotspots were not proliferating. Our results are consistent with experiments on
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
endothelial cell cultures that have shown that quiescent endothelial cells accumulate PpIX,
although to a lesser extent than proliferating endothelial cells.32
The PpIX hotspots frequently had a tubular or branched structure resembling blood vessels.
We found no evidence of microvessel proliferation in WHO grade II glioma tissue. This
suggests that the glioma cells in PpIX hotspots were growing along established blood vessels,
which is termed vessel co-option.33,34 Molecules released by endothelial cells of small blood
vessels create perivascular niches, which attract and sustain glioma cells.35-37 Niches have
been subdivided on the basis of the signalling mechanisms operating within the niche.38 Our
data suggest that the PpIX hotspots in lower-grade gliomas represent an invasive niche
involving nestin+ glioma cells, which may proliferate around small blood vessels.
The glioma cells in the PpIX hotspots exhibit several hallmarks of cancer, such as less
differentiated cellular state, ability to divide and metabolic reprogramming, which indicate
high-grade potential and suggest that PpIX hotspots may be seedbeds for malignant
progression. PpIX hotspots were sparse, but widely distributed, suggesting that malignant
progression can start at many locations within lower-grade gliomas. Each PpIX hotspot
provides a microdomain, which could fuel malignant transformation. Understanding the
spatial origins of malignant progression at a cellular level will provide a basis for new
treatments to prevent it and to reduce the risk of glioma recurrence after resective surgery.
Acknowledgements
We thank Natalie Long for assistance with recruitment.
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
References
1. Ostrom QT, Cioffi G, Gittleman H, et al. CBTRUS Statistical Report: Primary Brain
and Other Central Nervous System Tumors Diagnosed in the United States in 2012-
2016. Neuro Oncol. 2019; 21(Suppl 5):v1-v100.
2. Kirby AJ, Finnerty GT. New strategies for managing adult gliomas. J Neurol. 2020.
3. Claus EB, Walsh KM, Wiencke JK, et al. Survival and low-grade glioma: the
emergence of genetic information. Neurosurg Focus. 2015; 38(1):E6.
4. Rouse C, Gittleman H, Ostrom QT, Kruchko C, Barnholtz-Sloan JS. Years of
potential life lost for brain and CNS tumors relative to other cancers in adults in the
United States, 2010. Neuro Oncol. 2016; 18(1):70-77.
5. Suzuki H, Aoki K, Chiba K, et al. Mutational landscape and clonal architecture in
grade II and III gliomas. Nat Genet. 2015; 47(5):458-468.
6. Bai H, Harmanci AS, Erson-Omay EZ, et al. Integrated genomic characterization of
IDH1-mutant glioma malignant progression. Nat Genet. 2016; 48(1):59-66.
7. Mazor T, Pankov A, Johnson BE, et al. DNA Methylation and Somatic Mutations
Converge on the Cell Cycle and Define Similar Evolutionary Histories in Brain
Tumors. Cancer Cell. 2015; 28(3):307-317.
8. Ceccarelli M, Barthel FP, Malta TM, et al. Molecular Profiling Reveals Biologically
Discrete Subsets and Pathways of Progression in Diffuse Glioma. Cell. 2016;
164(3):550-563.
9. Leventoux N, Augustus M, Azar S, et al. Transformation Foci in IDH1-mutated
Gliomas Show STAT3 Phosphorylation and Downregulate the Metabolic Enzyme
ETNPPL, a Negative Regulator of Glioma Growth. Sci Rep. 2020; 10(1):5504.
10. Faubert B, Solmonson A, DeBerardinis RJ. Metabolic reprogramming and cancer
progression. Science. 2020; 368(6487).
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
11. Stummer W, Stocker S, Novotny A, et al. In vitro and in vivo porphyrin accumulation
by C6 glioma cells after exposure to 5-aminolevulinic acid. J Photochem Photobiol B.
1998; 45(2-3):160-169.
12. Stummer W, Pichlmeier U, Meinel T, et al. Fluorescence-guided surgery with 5-
aminolevulinic acid for resection of malignant glioma: a randomised controlled
multicentre phase III trial. Lancet Oncol. 2006; 7(5):392-401.
13. Hefti M, Albert I, Luginbuehl V. Phenytoin reduces 5-aminolevulinic acid-induced
protoporphyrin IX accumulation in malignant glioma cells. J Neurooncol. 2012;
108(3):443-450.
14. Kirby AJ, Lavrador JP, Bodi I, et al. Ex vivo ultrasonic samples of human brain
tumors in the molecular era. Neurooncol Adv. 2020; 2(1):vdaa014.
15. Paulus W, Peiffer J. Intratumoral histologic heterogeneity of gliomas. A quantitative
study. Cancer. 1989; 64(2):442-447.
16. Widhalm G, Kiesel B, Woehrer A, et al. 5-Aminolevulinic acid induced fluorescence
is a powerful intraoperative marker for precise histopathological grading of gliomas
with non-significant contrast-enhancement. PloS one. 2013; 8(10):e76988.
17. Schiff D, Van den Bent M, Vogelbaum MA, et al. Recent developments and future
directions in adult lower-grade gliomas: Society for Neuro-Oncology (SNO) and
European Association of Neuro-Oncology (EANO) consensus. Neuro Oncol. 2019;
21(7):837-853.
18. Louis DN, Perry A, Reifenberger G, et al. The 2016 World Health Organization
Classification of Tumors of the Central Nervous System: a summary. Acta
Neuropathol. 2016; 131(6):803-820.
19. Kitai R, Horita R, Sato K, et al. Nestin expression in astrocytic tumors delineates
tumor infiltration. Brain Tumor Pathol. 2010; 27(1):17-21.
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
20. Suzuki S, Namiki J, Shibata S, Mastuzaki Y, Okano H. The neural stem/progenitor
cell marker nestin is expressed in proliferative endothelial cells, but not in mature
vasculature. J Histochem Cytochem. 2010; 58(8):721-730.
21. Ladeby R, Wirenfeldt M, Dalmau I, et al. Proliferating resident microglia express the
stem cell antigen CD34 in response to acute neural injury. Glia. 2005; 50(2):121-131.
22. Galloway M. CD34 expression in glioblastoma and giant cell glioblastoma. Clin
Neuropathol. 2010; 29(2):89-93.
23. Singh SK, Hawkins C, Clarke ID, et al. Identification of human brain tumour
initiating cells. Nature. 2004; 432(7015):396-401.
24. Tamagno I, Schiffer D. Nestin expression in reactive astrocytes of human pathology.
J Neurooncol. 2006; 80(3):227-233.
25. Elmore MR, Najafi AR, Koike MA, et al. Colony-stimulating factor 1 receptor
signaling is necessary for microglia viability, unmasking a microglia progenitor cell in
the adult brain. Neuron. 2014; 82(2):380-397.
26. Goryaynov SA, Widhalm G, Goldberg MF, et al. The Role of 5-ALA in Low-Grade
Gliomas and the Influence of Antiepileptic Drugs on Intraoperative Fluorescence.
Front Oncol. 2019; 9:423.
27. Jaber M, Ewelt C, Wolfer J, et al. Is Visible Aminolevulinic Acid-Induced
Fluorescence an Independent Biomarker for Prognosis in Histologically Confirmed
(World Health Organization 2016) Low-Grade Gliomas? Neurosurgery. 2019;
84(6):1214-1224.
28. Lendahl U, Zimmerman LB, McKay RD. CNS stem cells express a new class of
intermediate filament protein. Cell. 1990; 60(4):585-595.
29. Wiese C, Rolletschek A, Kania G, et al. Nestin expression--a property of multi-
lineage progenitor cells? Cell Mol Life Sci. 2004; 61(19-20):2510-2522.
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
30. Tirosh I, Venteicher AS, Hebert C, et al. Single-cell RNA-seq supports a
developmental hierarchy in human oligodendroglioma. Nature. 2016; 539(7628):309-
313.
31. Venteicher AS, Tirosh I, Hebert C, et al. Decoupling genetics, lineages, and
microenvironment in IDH-mutant gliomas by single-cell RNA-seq. Science. 2017;
355(6332).
32. Wyld L, Burn JL, Reed MW, Brown NJ. Factors affecting aminolaevulinic acid-
induced generation of protoporphyrin IX. Br J Cancer. 1997; 76(6):705-712.
33. Holash J, Maisonpierre PC, Compton D, et al. Vessel cooption, regression, and
growth in tumors mediated by angiopoietins and VEGF. Science. 1999;
284(5422):1994-1998.
34. Kuczynski EA, Vermeulen PB, Pezzella F, Kerbel RS, Reynolds AR. Vessel co-
option in cancer. Nat Rev Clin Oncol. 2019; 16(8):469-493.
35. Calabrese C, Poppleton H, Kocak M, et al. A perivascular niche for brain tumor stem
cells. Cancer Cell. 2007; 11(1):69-82.
36. Watkins S, Robel S, Kimbrough IF, Robert SM, Ellis-Davies G, Sontheimer H.
Disruption of astrocyte-vascular coupling and the blood-brain barrier by invading
glioma cells. Nat Commun. 2014; 5:4196.
37. Zagzag D, Esencay M, Mendez O, et al. Hypoxia- and vascular endothelial growth
factor-induced stromal cell-derived factor-1alpha/CXCR4 expression in
glioblastomas: one plausible explanation of Scherer's structures. Am J Pathol. 2008;
173(2):545-560.
38. Hambardzumyan D, Bergers G. Glioblastoma: Defining Tumor Niches. Trends
Cancer. 2015; 1(4):252-265.
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
FIGURE LEGENDS
Figure 1 Ex vivo human brain tissue
A) Intraoperative image of exposed brain surface of participant with a glioblastoma. A
neuronavigation probe lies on the left superior frontal gyrus where the sample was taken
from. B) Intraoperative coronal MRI snapshot taken with the neuronavigation system. Red
dot denotes the location of the probe shown in panel A superior to the tumor. C) Block of
brain tissue resected from participant in A & B. D) Montage of brightfield images (5x) of a
neocortical brain slice cut from block shown in C. Slice spans white matter to pia. E) H&E
and Ki-67 staining of the diagnostic biopsy (top row) and ex vivo brain sample (bottom row).
Figure 2 PpIX ‘hot-spots’ in lower-grade gliomas
A) PpIX+ cells in ex vivo human brain tissue from control group with no glioma cells (NGC)
and from invasive edge of WHO II - IV gliomas. NGC image, patient had a glioblastoma. B)
Images (x40) of tubular PpIX structures in WHO grade III glioma (upper) and no glioma
cells (NGC) from a participant with focal cortical dysplasia (lower). C) PpXI+ cell density at
glioma edge for different tumor grades. D) PpIX+ cell density for each glioma molecular
subgroup.
Figure 3 PpIX expression in low-grade and high-grade gliomas
A) IDH1mut, Nestin and Ki-67 stains of the edge of WHO II (upper panels) and WHO III
gliomas (lower panels). B) PpIX+ cell density increases with tumor infiltration in low-grade
gliomas (LGG, red circles, 6 IDH1mut gliomas) and in higher grade gliomas (HGG, blue
squares, 6 IDH1wt and 2 IDH1mut gliomas). C) No power relationship between PpIX+ cell
density and Ki-67+ cell density in lower-grade gliomas (red line). LGG, 6 IDH1mut gliomas;
HGG, 4 IDH1wt and 2 IDH1mut gliomas. D) Power relationship between numbers of PpIX+
cells and IDHmut cells for all WHO glioma grades.
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
Figure 4 Cellular expression of PpIX at the edge of gliomas
A) Shape (circulatory) filter divides image data for Nestin and CD34 stains (left panels) into
elongated cells (middle panels) and rounded cells (right panels). B) Number of PpIX+ cells
grow exponentially as number of nestin+ cells increases in WHO grade II tissue from the edge
of WHO grade II and II{III} gliomas. C) Exponential relationship between numbers of PpIX+
cells and rounded nestin+ cells at the edge of WHO grade II and II{III} gliomas. D) No
correlation between numbers of elongated nestin+ cells and PpIX+ cells in lower-grade
gliomas. E) No correlation between CD34+ cells and the density of PpIX+ cells in WHO grade
II - IV gliomas when no shape filter is applied. F) Linear relationship between numbers of
elongated CD34+ cells and PpIX+ cells in gliomas. G) No correlation between numbers of
PpIX+ cells and rounded CD34+ in gliomas. H) Number of putative glioma cells that exhibit
PpIX fluorescence increases with the number of nestin+ cells.
Figure 5 5-ALA-Quantum dot (5-ALA-QD) uptake in glioma cells and small blood
vessels
A) 5-aminolevulinic acid (5-ALA) was conjugated to quantum dot (QD) nanoparticles via a
polyethylene glycol (PEG) linker. B) 5-ALA-QD incubated with ex vivo brain slices. T0,
before incubation. T4, after 4 hours. Tubular structures were imaged with QD fluorescence or
PpIX fluorescence ex vivo. C) 5-ALA-QD, IDH1mut and Ki-67 imaged in ex vivo slice.
Arrows indicate fluorescence colocalization. White box in merged image enlarged in D. D)
Proliferating, lower-grade glioma cells (Ki-67+ and IDH1mut) that had taken up 5-ALA-QD
adjacent to blood-vessel-like structure. E) 5-ALA-QD colocalize with GFAP+ cells. F)
Blood-vessel-like structure outlined by GFAP staining shows accumulation of 5-ALA-QD
along vessel wall. G) Blood-vessel-like structure outlined by GFAP staining that did not
accumulate 5-ALA-QD.
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
Table 1: Tumour information and intraoperative PpIX fluorescence for each
participant
I
D
A
ge
S
e
x
Tumor Information
Site
Diagnostic grade
Ex
viv
o
Gr
ade
Molecular Markers
1
65
M
Right
Frontal
Oligo
III
II
IDH1mut,
1p/19q co-del,
ATRX retained
Negative
2
29
F
Right
Frontal
Oligo
II
II
IDH1mut,
1p/19q co-del,
ATRX retained,
TERTpmut,
Negative
3
39
F
Left
Frontal
Oligo
III
III
IDH1mut,
1p/19q co-del, ATRX
retained
Negative
4
31
F
Left
Frontal
Oligo
III
III
IDH1mut,
1p/19q co-del, ATRX
retained
Negative
5
34
M
Left
Frontal
Astro
III
II
IDH2mut (R172C),
ATRX lost, TERTpwt
Negative
6
47
M
Right
Frontal
Oligo
III
II
IDH1mut,
1p/19q co-del, ATRX
retained, TERTpmut
Positive
7
27
F
Left
Frontal
Oligo
II
II
IDH1mut,
1p/19q co-del, ATRX
retained, TERTpmut
Negative
8
43
F
Right
Temporal
Oligo
II
II
IDH1mut,
1p/19q co-del, ATRX
retained
Negative
9
55
F
Right
Temporal
Astro
III
III
IDH1mut,
ATRX lost, TERTpwt
Negative
1
0
55
M
Right
Frontal
Oligo
III
II
IDH1mut,
1p/19q co-del, ATRX
retained
Negative
1
1
52
F
Left
Parietal
GBM
IV
IV
IDH1wt,
ATRX retained
Positive
1
2
69
M
Right
Occipital
GBM
IV
IV
IDH1wt,
ATRX retained,
TERTpmut
Positive
1
3
57
M
Right
Frontal
GBM
IV
NG
C
IDH1wt,
ATRX retained,
TERTpmut,
Positive
1
4
54
F
Right
Frontal
GBM
IV
NG
C
IDH1wt, TERTpmut
Positive
1
5
62
M
Right
Temporal
GBM
IV
IV
IDH1wt,
ATRX retained,
Positive
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
TERTpmut,
1
6
27
F
Left
Frontal
GBM
IV
IV
IDH1wt,
ATRX retained,
TERTpmut,
Positive
1
7
21
M
Right
Frontal
Focal Cortical
Dysplasia
NG
C
CD34 Negative
Negative
1
8
28
M
Right
Frontal
Astro
III
III
IDH1mut,
ATRX lost
Positive
1
9
58
M
Left
Parietal
Metastatic adeno-
carcinoma
NG
C
N/A
Negative
2
0
69
F
Left
Temporal
Astro
III
III
IDH1wt,
ATRX retained,
TERTpwt,,
Negative
Abbreviations: Oligo, Oligodendroglioma; Astro, Astrocytoma; GBM, Glioblastoma; IDH1wt,
no R132H mutation in Isocitrate Dehydrogenase 1; IDH1mut, R132H mutation in Isocitrate
Dehydrogenase 1 unless stated; 1p/19q co-del, co-deletion affecting chromosomes 1p and
19q; ATRX, ATP-dependent helicase ATRX; TERTp, Telomerase Reverse Transcriptase
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
Figure 1
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
Figure 2
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
Figure 3
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
Figure 4
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
Accepted Manuscript
Figure 5
Downloaded from https://academic.oup.com/noa/advance-article/doi/10.1093/noajnl/vdab026/6131481 by guest on 10 February 2021
... To this end, early maximal safe surgical excision represents initial treatment, since reduction of the tumor volume decreases the risk of MT [8] and thus prolongs overall survival (OS) [9][10][11][12][13]. Moreover, because intratumoral heterogeneity is frequent in LGG, large resection increases the chances of detecting possible microfoci of high-grade glioma within the neoplasm, enabling better adaptation of the next strategy according to extensive histomolecular data [14][15][16]. In other words, from an oncological perspective, variability in LGG is considerable not only at any given moment, then over months or years, but also from one patient to another one depending on the intrinsic glioma behavior (partly related to the genetic subtype), and depends also on the time of diagnosis in the natural history of the disease [17]. ...
... Moreover, the WHO classification does not take into account the existence of a major intratumoral heterogeneity which is very frequent in LGG. As demonstrated by multiple samples or when extensive surgical resections have been performed "en bloc", distinct histomolecular components have often been observed within the same tumor [14][15][16]31,64,65]. Moreover, recent research using intraoperative image-guided biopsies, genetic analyses with RNA sequencing, and whole-exome sequencing reported a gene expression pattern and mutational landscape of the PTZ that were distinct from those seen in the tumor core and peripheral brain tissue [66]. ...
Article
Full-text available
Diffuse low-grade glioma (LGG) is a rare cerebral cancer, mostly involving young adults with an active life at diagnosis. If left untreated, LGG widely invades the brain and becomes malignant, generating neurological worsening and ultimately death. Early and repeat treatments for this incurable tumor, including maximal connectome-based surgical resection(s) in awake patients, enable postponement of malignant transformation while preserving quality of life owing to constant neural network reconfiguration. Due to considerable interindividual variability in terms of LGG course and consecutive cerebral reorganization, a multistage longitudinal strategy should be tailored accordingly in each patient. It is crucial to predict how the glioma will progress (changes in growth rate and pattern of migration, genetic mutation, etc.) and how the brain will adapt (changes in patterns of spatiotemporal redistribution, possible functional consequences such as epilepsy or cognitive decline, etc.). The goal is to anticipate therapeutic management, remaining one step ahead in order to select the optimal (re-)treatment(s) (some of them possibly kept in reserve), at the appropriate time(s) in the evolution of this chronic disease, before malignization and clinical worsening. Here, predictive tumoral and non-tumoral factors, and their ever-changing interactions, are reviewed to guide individual decisions in advance based on patient-specific markers, for the treatment of LGG.
... As the name states, diffuse low-grade gliomas (DLGG)-World Health Organization (WHO) grade 2 diffuse astrocytic and oligodendroglial tumors [1]-are in essence poorly circumscribed and intrinsically heterogeneous [2][3][4][5] tumors that progressively infiltrate the brain. Regardless of the common initial slow growth rate, these tumors will ineluctably become more aggressive with malignant transformation if left untreated [6]. ...
... Additionally, concerning oligodendrogliomas, but using en bloc resected tumors, different tumor cell and vessel densities were identified throughout all the cases [4]. A more recent study revealed the existence of sparse, but widely distributed, protoporphyrin IX "hotspots" within low-grade gliomas, which exhibited some malignancy features, such as less differentiated cellular state, ability to divide, and metabolic reprogramming [5]. ...
Article
Full-text available
Diffuse low-grade gliomas (DLGGs) are heterogeneous and poorly circumscribed neoplasms with isolated tumor cells that extend beyond the margins of the lesion depicted on MRI. Efforts to demarcate the glioma core from the surrounding healthy brain led us to define an intermediate region, the so-called peritumoral zone (PTZ). Although most studies about PTZ have been conducted on high-grade gliomas, the purpose here is to review the cellular, metabolic, and radiological characteristics of PTZ in the specific context of DLGG. A better delineation of PTZ, in which glioma cells and neural tissue strongly interact, may open new therapeutic avenues to optimize both functional and oncological results. First, a connectome-based “supratotal” surgical resection (i.e., with the removal of PTZ in addition to the tumor core) resulted in prolonged survival by limiting the risk of malignant transformation, while improving the quality of life, thanks to a better control of seizures. Second, the timing and order of (neo)adjuvant medical treatments can be modulated according to the pattern of peritumoral infiltration. Third, the development of new drugs specifically targeting the PTZ could be considered from an oncological (such as immunotherapy) and epileptological perspective. Further multimodal investigations of PTZ are needed to maximize long-term outcomes in DLGG patients.
... 4,5 Although a mean diameter of less than 1 cm beyond tumor contrast enhancement has been suggested with 5-ALA-guided resection, 4 there is evidence for both residual tumor even after no fluorescence was left at the time of surgery 5 and for tumor cells in nonfluorescent tissue samples. 6,7 Our group strongly supports a patient-centered approach to tumor resection, in an individualized oncofunctional paradigm, where the decision to push the extent of the tumor resection to the functional boundaries and the acceptability of any potential neurological deficits is made jointly by the clinical team and the patient. However, we recognize that by definition the functional limits vary from patient to patient and from institution to institution, 8 thus limiting the comparison of patient-based and oncological-based outcomes. ...
... However, we believe that such consensus should be based on intraoperative 5-ALA fluorescence (such as 1-2 cm beyond fluorescence) as resections that extend purely beyond the contrast enhancement risk leaving tumor tissue behind. [5][6][7] Funding This study did not receive any funding or financial support. ...
Article
Full-text available
To the Editor: We read with great enthusiasm the article published by Khalafallah et al.1 The authors address an utmost point of importance in contemporary glioma surgery: an operational definition for supratotal resection (SpTR) in glioblastoma (GBM). We believe this is a potential landmark article on this subject because it paves the way toward uniformization of SpTR concept across different institutions allowing for comparison of outcomes. This is crucial for validation and affirmation of SpTR as the goal for surgical treatment of GBM.
Article
Full-text available
Gliomas are hard to treat. Their prognosis has improved little over the past few decades. Fundamental therapeutic challenges such as treatment resistance, malignant progression, and tumour recurrence persist. New strategies are needed to advance the management and treatment of gliomas. Here, we focus on where those new strategies could emerge. We consider how recent advances in our understanding of the biology of adult gliomas are informing new approaches to their treatment.
Article
Full-text available
IDH1-mutated gliomas are slow-growing brain tumours which progress into high-grade gliomas. The early molecular events causing this progression are ill-defined. Previous studies revealed that 20% of these tumours already have transformation foci. These foci offer opportunities to better understand malignant progression. We used immunohistochemistry and high throughput RNA profiling to characterize foci cells. These have higher pSTAT3 staining revealing activation of JAK/STAT signaling. They downregulate RNAs involved in Wnt signaling (DAAM2, SFRP2), EGFR signaling (MLC1), cytoskeleton and cell-cell communication (EZR, GJA1). In addition, foci cells show reduced levels of RNA coding for Ethanolamine-Phosphate Phospho-Lyase (ETNPPL/AGXT2L1), a lipid metabolism enzyme. ETNPPL is involved in the catabolism of phosphoethanolamine implicated in membrane synthesis. We detected ETNPPL protein in glioma cells as well as in astrocytes in the human brain. Its nuclear localization suggests additional roles for this enzyme. ETNPPL expression is inversely correlated to glioma grade and we found no ETNPPL protein in glioblastomas. Overexpression of ETNPPL reduces the growth of glioma stem cells indicating that this enzyme opposes gliomagenesis. Collectively, these results suggest that a combined alteration in membrane lipid metabolism and STAT3 pathway promotes IDH1-mutated glioma malignant progression.
Article
Full-text available
Background Gliomas are composed of multiple clones of tumor cells. This intratumor heterogeneity contributes to the ability of gliomas to resist treatment. It is vital that gliomas are fully characterized at a molecular level when a diagnosis is made to maximize treatment effectiveness. Methods We collected ultrasonic tissue fragments during glioma surgery. Large tissue fragments were separated in the operating theatre and bathed continuously in oxygenated artificial cerebrospinal fluid to keep them alive. The ex vivo tissue fragments were transferred to a laboratory and incubated in 5-aminolevulinic acid (5-ALA). 5-ALA is metabolised to Protoporphyrin IX (PpIX), which accumulates in glioma cells and makes them fluorescent. The molecular and neuropathological features of the PpIX fluorescent ultrasonic tissue fragments were studied. Results We show that PpIX fluorescence can rapidly identify tissue fragments infiltrated by glioma in the laboratory. Ultrasonic tissue fragments from the tumor core provided molecular and neuropathological information about the glioma that was comparable to the surgical biopsy. We characterized the heterogeneity within individual gliomas by studying ultrasonic tissue fragments from different parts of the tumor. We found that gliomas exhibit a power relationship between cellular proliferation and tumor infiltration. Tissue fragments that deviate from this relationship may contain foci of more malignant glioma. The methylation status of the O6-methlguanine DNA methyltransferase (MGMT) gene promoter varied within each glioma. Conclusion Ex vivo ultrasonic tissue fragments can be rapidly screened for glioma infiltration. They offer a viable platform to characterize heterogeneity within individual gliomas, thereby enhancing their diagnosis and treatment.
Article
Full-text available
Objectives: Intraoperative tumor visualization with 5-aminolevulinic acid (5-ALA) induced protoporphyrin IX (PpIX) fluorescence is widely applied for improved resection of high-grade gliomas. However, visible fluorescence is present only in a minority of low-grade gliomas (LGGs) according to current literature. Nowadays, antiepileptic drugs (AEDs) are frequently administered to LGG patients prior to surgery. A recent in-vitro study demonstrated that AEDs result in significant reduction of PpIX synthesis in glioma cells. The aim of this study was thus to investigate the role of 5-ALA fluorescence in LGG surgery and the influence of AEDs on visible fluorescence. Patients and Methods: Patients with resection of a newly diagnosed suspected LGG after 5-ALA (25 mg/kg) administration were initially included. During surgery, the presence of visible fluorescence (none, mild, moderate, or bright) within the tumor and intratumoral fluorescence homogeneity (diffuse or focal) were analyzed. Tissue samples from fluorescing and/or non-fluorescing areas within the tumor and/or the assumed tumor border were collected for histopathological analysis (WHO tumor diagnosis, cell density, and proliferation rate). Only patients with diagnosis of LGG after surgery remained in the final study cohort. In each patient, the potential preoperative intake of AEDs was investigated. Results: Altogether, 27 patients with a histopathologically confirmed LGG (14 diffuse astrocytomas, 6 oligodendrogliomas, 4 pilocytic astrocytomas, 2 gemistocytic astrocytomas, and one desmoplastic infantile ganglioglioma) were finally included. Visible fluorescence was detected in 14 (52%) of 27. In terms of fluorescence homogeneity (n = 14), 7 tumors showed diffuse fluorescence, while in 7 gliomas focal fluorescence was noted. Cell density (p = 0.03) and proliferation rate (p = 0.04) was significantly higher in fluorescence-positive than in fluorescence-negative samples. Furthermore, 15 (56%) of 27 patients were taking AEDs before surgery. Of these, 11 patients (73%) showed no visible fluorescence. In contrast, 10 (83%) of 12 patients without prior AEDs intake showed visible fluorescence. Thus, visible fluorescence was significantly more common in patients without AEDs compared to patients with preoperative AED intake (OR = 0,15 (CI 95% 0.012–1.07), p = 0.046). Conclusions: Our study shows a markedly higher rate of visible fluorescence in a series of LGGs compared to current literature. According to our preliminary data, preoperative intake of AEDs seems to reduce the presence of visible fluorescence in such tumors and should thus be taken into account in the clinical setting.
Article
Full-text available
Background: Approximately 20% of low-grade gliomas (LGG) display visible protoporphyrin fluorescence during surgery after 5-aminolevulinic acid (5-ALA) administration. Objective: To determine if fluorescence represents a prognostic marker in LGG. Methods: Seventy-four consecutive patients with LGG (World Health Organization 2016) were operated on with 5-ALA. Fluorescent tissue was specifically biopsied. Tumor size, age, Karnofsky index, contrast-enhancement, fluorescence, and molecular factors (IDH1/IDH2-mutations, Ki67/MIB1 Index, 1p19q codeletions, ATRX, EGFR, p53 expression, and O6-methylguanine DNA methyltransferase promotor methylation), were related to progression-free survival (PFS), malignant transformation-free survival (MTFS) and overall survival (OS). Results: Sixteen of seventy-four LGGs (21.6%) fluoresced. Fluorescence was partially related to weak enhancement on magnetic resonance imaging and increased (positron emission tomography)PET-FET uptake, but not to Karnofsky Performance Score, tumor size, or age. Regarding molecular markers, only EGFR expression differed marginally (fluorescing vs nonfluorescing: 19% vs 5%; P = .057). Median follow-up was 46.4 mo (95% confidence interval [CI]: 41.8-51.1). PFS, MTFS, and OS were shorter in fluorescing tumors (PFS: median 9.8 mo, 95% CI: 1.00-27.7 vs 45.8, 31.9-59.7, MTFS: 43.0 [27.5-58.5] vs 64.6 [57.7-71.5], median not reached, P = .015; OS: 51.6, [34.8-68.3] vs [68.2, 62.7-73.8], P = .002). IDH mutations significantly predicted PFS, MTFS, and OS. In multivariate analysis IDH status and fluorescence both independently predicted MTFS and OS. PFS was not independently predicted by fluorescence. Conclusion: This is the first report investigating the role of ALA-induced fluorescence in histologically confirmed LGG. Fluorescence appeared to be a marker for inherent malignant transformation and OS, independently of known prognostic markers. Fluorescence in LGG might be taken into account when deciding on adjuvant therapies.
Article
Metabolism as cancer progresses Numerous cancer-specific alterations in metabolism have been identified but have not yet resulted in an effective anti cancer therapeutic. In a Review, Faubert et al. discuss how metabolism changes as cancer develops from a small, premalignant lesion to an aggressive primary tumor and then metastasizes. Metabolic vulnerabilities likely change with cancer progression, making the identification of general cancer-associated metabolic features difficult. The authors propose that a more targeted approach to tissues and vulnerabilities identified in patients may be more effective. Science , this issue p. eaaw5473
Article
The Central Brain Tumor Registry of the United States (CBTRUS), in collaboration with the Centers for Disease Control and Prevention and National Cancer Institute, is the largest population-based registry focused exclusively on primary brain and other central nervous system (CNS) tumors in the United States (US) and represents the entire US population. This report contains the most up-to-date population-based data on primary brain tumors available and supersedes all previous reports in terms of completeness and accuracy. All rates are age-adjusted using the 2000 US standard population and presented per 100,000 population. The average annual age-adjusted incidence rate (AAAIR) of all malignant and non-malignant brain and other CNS tumors was 23.41 (Malignant AAAIR = 7.08, non-Malignant AAAIR = 16.33). This rate was higher in females compared to males (25.84 versus 20.82), Whites compared to Blacks (23.50 versus 23.34), and non-Hispanics compared to Hispanics (23.84 versus 21.28). The most commonly occurring malignant brain and other CNS tumor was glioblastoma (14.6% of all tumors), and the most common non-malignant tumor was meningioma (37.6% of all tumors). Glioblastoma was more common in males, and meningioma was more common in females. In children and adolescents (age 0-19 years), the incidence rate of all primary brain and other CNS tumors was 6.06. An estimated 86,010 new cases of malignant and non-malignant brain and other CNS tumors are expected to be diagnosed in the US in 2019 (25,510 malignant and 60,490 non-malignant). There were 79,718 deaths attributed to malignant brain and other CNS tumors between 2012 and 2016. This represents an average annual mortality rate of 4.42. The five-year relative survival rate following diagnosis of a malignant brain and other CNS tumor was 35.8%, and the five-year relative survival rate following diagnosis of a non-malignant brain and other CNS tumors was 91.5%.
Article
The finding that most grade II and III gliomas harbor isocitrate dehydrogenase (IDH) mutations conveying a relatively favorable and fairly similar prognosis in both tumor grades highlights that these tumors represent a fundamentally different entity from IDH wild-type gliomas exemplified in most glioblastoma. Herein we review the most recent developments in molecular neuropathology leading to reclassification of these tumors based upon IDH and 1p/19q status, as well as the potential roles of methylation profiling and CDKN2A/B deletional analysis. We discuss the epidemiology, clinical manifestations, benefit of surgical resection, and neuroimaging features of lower-grade gliomas as they relate to molecular subtype, including advanced imaging techniques such as 2-hydroxyglutarate magnetic resonance spectroscopy and amino acid PET scanning. Recent, ongoing and planned studies of radiation therapy and both cytotoxic and targeted chemotherapies are summarized, including both small molecule and immunotherapy approaches specifically targeting the mutant IDH protein.
Article
All solid tumours require a vascular supply in order to progress. Although the ability to induce angiogenesis (new blood vessel growth) has long been regarded as essential to this purpose, thus far, anti-angiogenic therapies have shown only modest efficacy in patients. Importantly, overshadowed by the literature on tumour angiogenesis is a long-standing, but continually emerging, body of research indicating that tumours can grow instead by hijacking pre-existing blood vessels of the surrounding nonmalignant tissue. This process, termed vessel co-option, is a frequently overlooked mechanism of tumour vascularization that can influence disease progression, metastasis and response to treatment. In this Review, we describe the evidence that tumours located at numerous anatomical sites can exploit vessel co-option. We also discuss the proposed molecular mechanisms involved and the multifaceted implications of vessel co-option for patient outcomes.
Article
Tumor subclasses differ according to the genotypes and phenotypes of malignant cells as well as the composition of the tumor microenvironment (TME). We dissected these influences in isocitrate dehydrogenase (IDH)-mutant gliomas by combining 14,226 single-cell RNA sequencing (RNA-seq) profiles from 16 patient samples with bulk RNA-seq profiles from 165 patient samples. Differences in bulk profiles between IDH-mutant astrocytoma and oligodendroglioma can be primarily explained by distinct TME and signature genetic events, whereas both tumor types share similar developmental hierarchies and lineages of glial differentiation. As tumor grade increases, we find enhanced proliferation of malignant cells, larger pools of undifferentiated glioma cells, and an increase in macrophage over microglia expression programs in TME. Our work provides a unifying model for IDH-mutant gliomas and a general framework for dissecting the differences among human tumor subclasses.