Luận án nghiên cứu phân lợi mô bệnh học u thần kinh đệm lan tỏa của não theo who 2007 ta

MINISTRY OF EDUCATION AND TRAINING

MINISTRY OF HEALTH

HANOI MEDICAL UNIVERSITY

NGUYEN SY LANH

RESEARCH ON THE HISTOPATHOLOGICAL
CLASSIFICATION OF DIFFUSE GLIOMAS OF THE BRAIN
USING THE CLASSIFICATION OF THE WORLD HEALTH
ORGANIZATION (WHO) IN 2007
Specialized: Anatomical Pathology and Forensic Medecine
Major: Biomedical Science

Code: 9720101

SUMMARY OF DOCTORAL THESIS

HA NOI - 2022


THE SCIENTIFIC WORK WAS COMPLETED
AT HANOI MEDICAL UNIVERSITY

Academic supervisors:
Assoc. Prof. Dr. Nguyen Phuc Cuong
Prof. Dr. Nguyen Thuy Huong

Reviewer 1:

Reviewer 2:

Reviewer 3:
The thesis will be defended in front of the school-level doctoral
thesis examination committee held at Hanoi Medical University.
Time of organization: ……hours….date……month….. 2022

The thesis can be found at:
Vietnam National Library
Hanoi Medical University Library


PUBLISHED RESEARCH WORKS RELATED
TO THE THESIS
1.

2.

3.

Nguyen Sy Lanh, Nguyen Phuc Cuong, Nguyen Thuy Huong.
Applying the classification of the World Health Organization
(WHO) in 2007 in the diagnosis of diffuse gliomas of the
cerebral hemisphere at Viet Duc Hospital from 6/2014 to
10/2017. Journal of Vietnamese Medicine. Episode 461 December - special issue - 2017: 335-344.
Nguyen Sy Lanh, Nguyen Phuc Cuong, Hoang Xuan Su,
Nguyen Thuy Huong. Application of immunohistochemistry in
evaluating the expression characteristics of IDH1 and IDH2
genes in diffuse gliomas of brain. Vietnamese Journal of Internal
Medicine. Issue 18/2020: 15-23.

Nguyen Sy Lanh, Nguyen Phuc Cuong, Nguyen Thuy Huong.
Study on histopathological classification and some
immunohistochemical markers of diffuse gliomas of the brain
using the World Health Organization classification in 2007.
Vietnam Medical Journal. Episode 513 - April - Issue 1 - 2022:
249-253.


1
INTRODUCTION
Brain tumor is the conventional term for tumors in the skull, in the central
nervous system. According to the definition of the International Agency for
Recording of Cancer (IARC), the annual incidence of brain tumors ranges from 3
to 5 per 100,000 population and this number is increasing. The frequency of the
disease is mainly seen in 2 age groups from 3 to 12 and 40 to 70 years old.
Gliomas occur in any location in the brain such as the frontal lobes, temporal
lobes, parietal lobes, occipital lobes, pons, brainstem, and cerebellum. Gliomas
includes low-grade gliomas and high-grade gliomas. Diffuse gliomas or highgrade gliomas that typically grows, diffuse, and increase in malignancy over time
or from the time of presentation presents as a highly malignant glioma such as
glioblastomas.
In order to effectively apply the treatment methods, we must make the
correct histopathological diagnosis and histopathological types using the new
classification of the World Health Organization. Then, we have to evaluate each
specific case whether there are favorable factors with the treatment methods such
as: IDH1 mutant; 1p19q co-deleted; p53 mutant or p53 gene overexpression on
immunohistochemical staining. From that fact, we conducted a study on the topic
" Research on histopathological classification of diffuse gliomas of the brain
using the classification of the World Health Organization (WHO) in 2007" with
the following purposes:
1. Description of histopathological and immunohistochemical characteristics of

diffuse gliomas of the brain using the classification of the World Health
Organization in 2007.
2. Analysis of the relationship between the expression of immunohistochemical
markers with histopathological types and histopthological grade in the group of
patients studied.
NEW CONTRIBUTIONS OF THE THESIS
- The thesis has new contributions that are given the rate of histopathological
types and histopathological grade of diffuse gliomas of the brain by using WHO
classification in 2007.
- The first study in Vietnam using many immunohistochemical markers (7
markers) for diffuse gliomas, in which there are new and highly significant
markers for clinical practice such as GFAP, OLIG2, IDH1, INA, P53, Ki67 and
ATRX. Giving the positive rate of these markers in the research results helps
pathologists have a look and apply in diagnostic practice.
- The study also showed the relationship between pairs of histochemical markers
contributing to the grouping of gliomas for treatment and prognosis of patients.
PRACTICAL VALUE OF THESIS
- This study helps to make an accurate diagnosis of histopathological type,
grade and grouping prognosis of patients based on histopathological-molecular
subtypes of each patient. This helps the treatment and prognosis of patients to be
reasonable for each specific patient.
- The study also opens up a new direction for individualized analysis of each
patient, helping to apply appropriate, safe and economical treatment methods for
patients, avoiding under- or over-treatment.


2
STRUCTURE OF THE THESIS
The thesis consists of 124 pages: Introduction 2 pages; Overview
document 37 pages; Research subjects and methods 15 pages; Research results 32

pages; Discussion 35 pages; Conclusion 2 pages and Recommendations 1 page.
The thesis has 48 results tables, 4 charts and 42 images, 157 references in
Vietnamese, English and French.
Chapter 1
OVERVIEW DOCUMENT
1.1. Epidemiology
With an annual incidence of 6-7 cases per 100,000 population, malignant
gliomas are the most common CNS primary tumour, accounting for more than half
of adults and the one of the top 10 causes of cancer death. Every year in France,
3,000 patients with malignant glioma are seen. In the US, the incidence of brain
tumors is 4.5/100,000 population, the mortality rate ranks fifth after other liver
cancer, lung cancer, stomach cancer and esophageal cancer. In Vietnam, according
to statistics in 2000, the rate of brain tumors accounted for 1.3/100,000 people and
recorded a number of cases. Data from Globocan Vietnam in 2020, brain tumor
occupies the 15th position.
1.2. Histopathological classification
1.2.1. Classification of World Health Organization
The WHO classification is based on two theories that define the
histopathological types of tumors depending on the predominant tumor cell type
and the histopathological grade of the tumor based on the dedifferentiation
properties. Each tumor has an official WHO name, ICD-O code.
Table 1.1: World Health Organization Classification 2007
Type
Astrocytic tumors
Piloytic astrocytoma
+Pilomyxoid astrocytoma
Subependymal giant cell astrocytoma
Pleomorphic xanthoastrocytoma
Diffuse astrocytoma
+ Fibrillary astrocytoma

+ Protoplasmic astrocytoma
+ Gemistocytic astrocytoma
Anaplastic astrocytoma
Glioblastoma
+ Giant cell glioblastoma
+ Gliosarcoma
Gliomatosis cerebri
Oligodendroglial tumors
Oliogodendroglioma
Anaplastic oligoastrocytoma
Oligoastrocytic tumors
Oligoastrocytoma
Anaplastic oligoastrocytoma

ICD code

Histological
Grade

9421/1
9425/3*
9384/1
9424/3
9400/3
9420/3
9410/3
9411/3
9401/3
9440/3
9441/3

9442/3
9381/3

I
II
I
II
II
II
II
II
III
IV
IV
IV
IV

9450/3
9451/3

II
III

9382/3
9382/3

II
III

3
1.2.3. Histopathological classification of gliomas
- The first edition of the International Classification of Diseases (ICD) in
1993. We are currently using the ICD-10, which entered use in the United States in
2015, and is revised annually.
Histopathological grading according to World Health Organization
- The WHO histological grading system is also based on the same
histopathological criteria as the St. Anne-Mayo as follows: nuclear abnormalities
(atypial nuclear); Multiplication; Proliferation of endothelial cells, not
angiogenesis; Tumor necrosis. Staging system for malignancy according to St.
Anne-Mayo has four grades of tumor: Grade 1 are tumors that do not meet any of
the criteria; Grade 2 are tumors that have one criterion, usually “Nuclear atypia”
criterion; Grade 3 are tumors with two criteria, usually: “Atypical nuclear” and
“Mitose”; Grade 4 are tumors with three or four criteria. The WHO classification
also defines the histopathological types of the tumor according to the predominant
tumor cell type and the histopathological grade based on the signs of tumoral cell
dedifferentiation. This classification was modified in 1993, 2000, 2007 and 2016
to incorporate biological and molecular genetic factors.
1.4. Immunohistochemical markers in diffuse gliomas of the brain
1.4.1. Glial Fibrillary Acidic Protein (GFAP)
- GFAP is the main glial fiber, specific for astrocyte differentiation, is an
immunohistochemical marker that has a useful role in identifying tumor cells of
glial origin. GFAP is a reliable marker in diagnostic practice for gliomas, which is
significant in determining astrocyte differentiation. Tumors were defined as
astrocytomas if the tumor cells had abnormal nuclear characteristics and the tumor
cell cytoplasm reacted to the GFAP marker to varying degrees. Reactive astrocytes
also express the GFAP marker with multiple cytoplasmic and elongated
cytoplasmic branches.
1.4.2. Oligodendrocyte transcription factor (OLIG2)
- OLIG2 is a recently identified transcription factor that plays a role in the

differentiation of oligodendrocytes. Studies in human brain tumors indicate
significantly higher OLIG2 mRNA expression in oligodendrogliomas compared
with other types of gliomas. OLIG2 expression on immunohistochemical staining
in oligodendrogliomas confirmed this. Studies have also shown that using the
OLIG2 antibody on immunohistochemical staining can help differentiate the
histopathologic types of brain gliomas, although the distinction between
oligodendrogliomas and astrocytomas of the brain may not be really obvious.
1.4.3. Isocitrate Dehydrogenase (IDH)
- The genes for IDH1 and IDH2 are located on chromosomes 2q33.3 and
15q26.1. Determination of IDH mutation status is a valuable factor in the
diagnosis as well as in the prognosis for patients with glioma. Mutations in the
IDH1 gene are much more common than the IDH2 gene in 85% and 3% of grade
2 gliomas, respectively. When IDH1 mutation, more than 93% at the R132H
position. The product of the mutation has been used to produce selective antiR132H antibodies and is used in routine diagnostic practice with formalin-fixed
and paraffin-embedded tissue samples. IDH2 mutation at codon position 172
(R172K). Other rare IDH mutations can only be identified by gene sequencing.


4
1.4.4. Alpha internexin (INA)
- The alpha-internexin (INA) is a type of intermediate fiber present in the
cells of the central nervous system whose dominant gene is located on the long
arm of chromosome 10 (10q24.33). INAs are expressed to varying degrees in
oligodendrogliomas that have 1p19q co-deleted but retain the long arm of
chromosome 10. All gliomas with 1p19q co-deleted have mutations in the IDH
gene. Expression of the INA marker is a good prognostic factor, increasing
susceptibility to chemotherapy.
1.4.5. P53
- The gene p53, located on the short arm of chromosome 17, encodes for a
nuclear protein that is responsible for cell cycle regulation but also for

proliferation control. Using immunohistochemical staining with monoclonal or
polyclonal antibodies on parafin-embedded specimens, not only known mutant
Protein53 is detected as well as wild-type Protein53. Protein53 overexpression,
including mutant and wild type, plays an important role and was very common
in grade 2 and 3 astrocytomas or secondary glioblastomas and unstable to the
oligoastrocytomas.
1.4.6. Ki67
- Ki67 is a nuclear antigen, appearing between the G1, S, G2 phases and
the entire M phase of the cell cycle, which is characteristically detected on
immunohistochemical staining. The gene encoding the Ki-67 protein was
discovered by Scholzen and Gerdes in 2000. The expression of the Ki-67 protein
is associated with the proliferative activity of tumor cell populations in
malignancies. Ki 67 is also a marker for tumor invasion. The role of the Ki-67
indexes in determining prognosis has been studied quite a bit before, so it is
considered as a potential marker for tumors of the central nervous system.
1.4.7. Alpha thalassemia X-linked mental retardation (ATRX)
- The ATRX regulatory gene is located on chromosome Xq21.1 and
encodes a nuclear protein weighing 280 kDa, which is involved in many cellular
functions, including DNA recombination, repair, and transcriptional regulation.
The human ATRX mutation leads to the development of α-thalassemia, an Xlinked mental retardation syndrome, and other genetic conditions. ATRX
mutations are present in at least 15 types of human tumors, including
neuroblastoma, osteosarcoma, and neuroendocrine tumors of the pancreas. The
ATRX mutation is strongly associated with the IDH mutation and strongly
associated with KD of the CpG island methylation phenotype.
1.5. Histopathology-immunohistochemistry subtypes and pathogenesis of
diffuse gliomas
- In clinical practice, we will encounter combinations of three markers
IDH1, INA and P53 according to D Figarella-Branger et al as follows:
+ Groups of IDH1(-)/INA(-)/P53(+) include mainly: glioblastomas and a
few anaplastic astrocytomas.

+ Groups of IDH1(+)/INA(-)/P53(+) include mainly: anaplastic
astrocytomas, a number of glioblastomas, and a number of anplastic
oligodendrogliomas.


5
+
Groups
of
IDH1(+)/INA(+)/P53(+)
include:
anaplastic
oligodendrogliomas, a few cases of glioblastomas and a few cases of
astrocytomas.
+ Groups of IDH1(+)/INA(+)/P53(-) include: olgiodendrogliomas and a
small number of anaplastic oligodendrogliomas.
+ Groups IDH1(+)/INA(-)/P53(-) include mainly: astrocytomas and a few
oligodendrogliomas.
+ Groups IDH1(-)/INA(-)/P53(-) include: a small number of astrocytomas
and oligodendrogliomas.
+ Groups of IDH1(-)/INA(+)/P53(+) include: a very small number of
anaplastic oligodendrogliomas and anaplastic astrocytomas.
+ Groups of IDH1(-)/INA(+)/P53(-) include a very few cases of
olgidendrogliomas.
Chapter 2
OBJECT AND METHOD OF RESEARCH
2.1. OBJECT
- Including 216 patients with diffuse gliomas who underwent surgical
resection at Viet Duc Friendship Hospital, with the pathological diagnosis of
diffuse gliomas from grade 2 to grade 4 according to WHO classification in

2007, in the period from June 2014 to January 2020.
2.1.1. Standard patient
The patient must meet all of the following criteria:
- Had brain tumor surgery at Viet Duc Friendship Hospital.
- Histopathological result was diffuse gliomas.
- The medical record has full information on clinical symptoms, with MRI
and/or CT scan.
- There are full slides and parafin blocks.
- The tissue sample is still large enough in quantity, ensuring enough for
immunohistochemical staining.
- Tissue samples are still antigenic when immunohistochemical staining is
based on positive and negative control staining.
2.1.2. Patient elimination standard
- The patient underwent surgery to remove the brain tumor but there were
not enough specimens for immunohistochemical staining.
- There is not enough information on medical records, MRI or CT scans,
specimens and archival parafin blocks.
- Tissue samples are not in sufficient quantity for immunohistochemical
staining and no longer show antigenicity based on positive and negative control
staining.
2.1.3. Study size and method of the study
- Study size: 216 patients who met the selection criteria for inclusion in the
study.
- Method of the study: non-probability, purposeful, ensuring full selection
criteria and not getting caught in exclusion criteria. The patients had a complete
medical history, underwent brain tumor resection, were confirmed as diffuse
glioma of the brain, and had enough samples in terms of quantity and quality to
perform immunohistochemistry staining.

6
2.1.4. Time and place of study
- Period: From June 2014 to January 2020.
- Location: Department of Pathology, Viet Duc Friendship Hospital.
2.2. Research Methods
2.2.1. Research design
Cross-sectional, retrospective and prospective descriptive studies
2.2.2. Indicators and variables used in the study
- Age group:
- Gender: number and percentage of men and women.
- Some clinical symptoms in patients with diffuse glioma of the brain:
- Location of diffuse glioma of the brain:
- Tumor size: based on the size on preoperative MRI or CT scan, measured
in millimeters.
- Histopathological type and histological grade: determining the
histopathological type and histological grade of diffuse gliomas of the brain
according to the WHO classification in 2007.
- Mitotic ratio: multiplier divided by /10 high power field (HPF), counting
at the most active areas of the tumor.
- Tumor necrosis: to evaluate the presence or not of necrotic tissue on the
specimen, including: tumoral necrosis and hypotrophic necrosis.
- Expression rate of immunohistochemical markers of diffuse gliomas
(n=130)
- Expression rate of immunohistochemical phenotypes applied to diffuse
gliomas (n=130)
- The relationship between immunohistochemical markers with
histopathological type and histopathological grade
- The relationship between immunohistochemical phenotypes of IDH1,
INA and P53 markers with histopathological type and histopathological grade
- General characteristics of the research subjects: Collect data on age and

gender; Some common clinical and paraclinical signs of patients with diffuse
gliomas of the brain through treatment records, CT or CT scans and detailed
design of the research record.
- Histopathological study:
* Macroscopic research:
+ Measure the size of the patient in 3 dimensions with the smallest unit
being millimeters.
+ Comment on the characteristics: color, hardness, cystization or necrosis
and border with the normal area.
* Histopathological research
+ The slides were read on an optical microscope with a magnification of 50
- 400 times.
+ Evaluation of mitotic index: Select the area with the most mitotic activity
area on each specimen.
+ Formula for calculating the mitotic index of 10 HPF according to WHO:
(
)


7
+ Evaluation of tumor necrosis index: Find the presence of tumor necrosis,
including 2 types of neoplastic necrosis, map and hypotrophic necrosis.
+ Formula to calculate tumor necrosis index:
(
)
(
)
+ Determine the histopathological type of diffuse gliomas according to the
classification criteria of the World Health Organization classification in 2007.
+ Determination of histologic grade of diffuse glioma according to the

2007 World Health Organization classification.
- Immunohistochemical research: All immunohistochemical staining were
performed at the Department of Pathology, Viet Duc Friendship Hospital.
+ Antibodies and antibody concentrations: Antibodies used in the study
by BioSP, Sigma and Dako. Antibody dilutions according to the manufacturer's
instructions.
+ Evaluation of results:
- Conditions for evaluate results: There are negative control samples,
external positive controls and internal positive controls. Compare with
Hematoxylin eosin stain to know clearly where antigens need to be determined
in the nucleus, cytoplasm or cell membrane such as: The markers GFAP, INA
and IDH1 were positive stain in the cytoplasm; The markers Oligo2, P53, Ki-67
and ATRX were positive stain at the cell nucleus.
- Evaluate the result: Negative: only green and Positive: brown yellow.
+ How to evaluate the results of immunohistochemical staining
- The markers GFAP, IDH1, ATRX, Oligo2, INA, Ki67, P53 were
evaluated for intensity, positive charecteristics and quantified.
- GFAP and IDH1 markers were evaluated with the following criteria:
Coloring in tumor cells cytoplasm. Positive expression is when the tumor cell
cytoplasm has a yellow-brown color with sufficient intensity to be seen under
the optical microscope. The result is negative when the cytoplasm of tumor cells
is not brown yellow. Coloration of less than 1% of tumor cells was considered
negative. Color staining 1-10% of tumor cells are considered weakly positive
(+). The staining of 11 - 50% of tumor cells is considered to be moderately
positive (++). Color staining > 51% of tumor cells was considered strongly
positive (+++).
- Oligo2 and P53 markers were evaluated with the following criteria:
Coloring in tumor cell nucleus. Expression is positive when the tumor cell
nucleus was brown yellow with sufficient intensity to be seen under the optical
microscope. The result was negative when the tumor cells nucleus was not

brown yellow. Coloration of less than 1% of tumor cells was considered
negative. Color staining 1-10% of tumor cells are considered weakly positive
(+). The staining of 11 - 50% of tumor cells is considered to be moderately
positive (++). Color staining > 51% of tumor cells was considered strongly
positive (+++).
- INA markes was evaluated with the following criteria: Coloring in
tumor cells cytoplasm. Positive expression was when the tumor cell cytoplasm
has a yellow-brown color with sufficient intensity to be seen under the optical
microscope. The result was negative when the cytoplasm does not stain brown


8
yellow. Coloration of less than 1% of tumor cells was considered negative.
Color staining 1-10% of tumor cells were considered weakly positive (+). Color
staining of > 11% of tumor cells was considered strongly positive (+++).
- The ATRX marker is evaluated with the following criteria: Coloring in
the nucleus of tumor cells. Expression is positive when the tumor cell nucleus is
not brown yellow with sufficient intensity to be discerned under the optical
microscope. The result is negative when the tumor cell nucleus becomes brown
yellow. Loss of staining in less than 10% of tumor cells was considered
negative. Loss of staining expression 11-90% of tumor cells were considered
weakly positive (+). Loss of staining expression > 90% of tumor cells was
considered strongly positive (+++).
- For Ki-67: Ki67 was positive when any nuclei of tumor cells are brown
yellow with sufficient intensity to be seen under the optical microscope. The
Ki67 proliferation index (Ki67-LI) was recorded as the percentage of tumorpositive cells per tumor cell count after counting at least 1000 cells in a high
power- field (x 400). In the region of increased mitotic activity, positive standard
color of the staining method, have internal control, then calculate the percentage
(%).
+ The formula to calculate the proliferation index Ki67:

(

)

(
)
- Take microscopic photos of typical illustrative specimens.
Positive test and negative test
Positive control: GFAP: Reactive astrocytes in the normal brain
parenchyma around the tumor; OLIG2: oligodendrocytes; IDH1: Macrophages,
compared with PCR results and sent to a control test in the French Republic;
Ki67: The epithelium of the tonsils is benign; INA: Neuron; P53: Serous
papillary carcinoma of the ovary; ATRX: astrocytoma
Negative control: Do not coat the first antibody on the slide for all cases of
negative control slide staining.
2.4. Data processing using the statistical software SPSS 16.0
Data were collected and processed using the statistical software SPSS 16.0.
- Descriptive statistics:
- Statistical analysis:
2.5. Limit error
- Take the whole or maximum specimen and represent all areas of the
lesion based on detailed macroscopic evaluation in case of large specimens.
Invite the instructor to consult with a group of Pathologists who have experience
in difficult cases. Histochemical staining results are positive and negative. Data
is carefully checked before entering and processing.
2.6. Ethics in research
- Research is carried out on tissue samples or parafin blocks of research
patients, only for the purpose of improving the quality of diagnosis, treatment,
assessment of disease prognosis and quality of life for patients. The research
objective was approved by the Proposal Evaluation Committee and approved by

the Ethics Committee in Biomedical Research, Hanoi Medical University by


9
'Acceptance from the Ethics Council in Biomedical Research' NUMBER:
187/HDĐHYHN and consensus of Viet Duc Friendship Hospital. All
information extracted from patients and medical records are kept confidential,
only for research purposes. Research results are not used for commercial
purposes, absolutely do not disclose information about patients in the study.
There was no distinction between study patients and non-study patients....
Chapter 3
STUDY RESULTS
3.1. Characteristics of age, gender and some common clinical and
laboratory signs of patients with diffuse gliomas of the brain
3.1.1. Characteristics of distribution according to age group
- The mean age in the study was 47.42±13.53 years old. In which, the most
common age group is from 31 to 40 years old, accounting for 25.46%. Second is
the age group from 41 to 50 years old (23.15%). The age group ≤ 20 accounts
for the lowest percentage, accounting for 0.5%, the rest is 21.30% in the age
group > 60, 18,98% in the 51 - 60 group and 10.65% in the 21 - 30 age group.
3.1.2. Characteristics of distribution according to sex
- In total 216 patients. Men accounted for a higher proportion of 53.70%
and women accounted for 46.30%. Male/Female ratio = 1.16/1.
3.2. Histopathological images of diffuse glioma of the brain
3.2.1. Distribution of histopathological types using the World Health
Organization Classification 2007
Table 3.6: Distribution of histopathological types using the WHO
classification 2007 (n=216)
Type
Number of patients

Ratio %
Oligoastrocytoma
19
8.80
Oligodendroglioma
20
9.26
Diffuse astrocytoma
18
8.33
Anaplastic oligoastrocytoma
17
7.87
Anaplastic oligodendroglioma
29
13.43
Anaplastic astrocytoma
31
14.35
Glioblastoma
82
37.96
Total
216
100.00
Comment: Glioblastoma was the most common, accounting for 37.96%.
Anaplastic astrocytoma accounted for the second with 14.35%. In addition,
some other common types such as: anaplastic oligodendroglioma accounted for
13.43%, oligodendroglioma accounted for 9.26%, oligoastrocytoma accounted
for 8.80%, diffuse astrocytoma is 8.33%. The least common histopathological

type is anaplastic oligoastrocytoma accounting for 7.87%.
3.2.2. Histological distribution characteristics of diffuse gliomas
- The most common grade 4 glioma was 37,96%, the second most common
grade 3 accounted for 35.65% and grade 2 accounted for at least 26.39%.


10
3.2.3. Characteristics of the mitotic index of histopathological types and
histopathological grade of diffuse glioma of the brain
Table 3.8: Mitotic index of histopathological types of diffuse gliomas of the
brain using the WHO Classification 2007 (n=216)
Mitotic ratio /10
Type
Number
Min Max
HPF
Oligoastrocytoma
19
1,42±0.61
1
3
Oligodendroglioma
20
1,5±0,76
1
3
Diffuse astrocytoma
18
1,11±0,32
1

2
Anaplastic oligoastrocytoma
17
7,06±3,29
1
18
Anaplastic oligodendroglioma
29
8,31±3,75
5
22
Anaplastic astrocytoma
31
8,13±3,93
4
23
Glioblastoma
82
19,65±14,44
6
82
Total
216
10,70±11,83
1
82
Comment: Glioblastomas have the highest mitotic index of 9.65±14,44.
Anaplastic gliomas have the second highest mitotic index, including: anaplastic
astrocytomas are 8.13±3,93; anaplastic oligodendrogliomas are 8.31±3.75 and
anaplastic olioastrocytomas are 7.06±3.29. Groups with lower mitotic index

include: oligodendrogliomas are 1.5±0.76; oligoastrocytomas are 1.42±0.61. The
lowest mitotic index in the diffuse astrocytomas was 1.11±0.32.
Histopathologically mitotic index of diffuse gliomas (n=216)
- Grade 4 gliomas have the highest mitotic rate of 19.7±14.44, the rest are grade
3 gliomas of 8.0±3.72 and grade 2 gliomas are 1.4±0.61.

Figure 3.1: The ROC curve finds the cut off point of the mitotic index to
distinguish between grade 2 and grade 3 diffuse gliomas.
Comment: Area under the curve is 0.992, with P<0.001. The cut off value for
the mitotic ratio between the grade 2 gliomas and grade 3 gliomas is 3.5
mitotic/10 HPF with J=0.987 (the range of the mitotic ratio with sensitivity and
specificity >0.8 is from 2, 5 mitotic/10HPF to 5.5 mitotic/10 HPF).

Figure 3.2: The ROC curve finds the cut off point of the mitotic ratio to
distinguish grade 3 gliomas and grade 4 gliomas.


11
Comment: The area under the curve is 0.895, with P<0.001. The cut off value
for the mitotic ratio of the difference between grade 3 and grade 4 is 8.5 mitotic/
10 HPF with J=0.631 (the range of the mitotic ratio with sensitivity and
specificity > 0.6 is from 7, 5 mitotic/10 HPF to 12.5 mitotic/10 HPF).
3.3. Immunohistochemical characterization of diffuse gliomas of the brain
3.3.1. Characterization of the expression of GFAP marker in diffuse gliomas
- The average value of the expression level of the GFAP marker is
77.76±20.11% (5-98). In which, GFAP has a strong expression level, accounting
for 87.50% and 0.93% of GFAP cases are positive with mild or negative level.
Moderate disclosure accounted for 11.57%.
3.3.2. Characterization of the expression of OLIG2 marker in diffuse gliomas
- The average value of the expression level of the OLIG2 marker is

83.53±16,41% (5-98). The strong positive rate of the OLIG2 marker accounted for
93.06% and the mild and moderate expression level accounted for 0.46% and 6.48%.
3.3.3. Characterization of the expression of IDH1 marker in diffuse gliomas
- In 216 cases, there were 126 cases with IDH1-marker showing for
58.33% and 90 cases not revealing with IDH1-marking, accounting for 41.67%.
3.3.4. Characterization of the expression of INA markers in diffuse gliomas
- In total 216 cases of the study, 130 cases were immunohistochemically
stained with INA marker, the results showed that 45 cases were positive,
accounting for 34.62% and 85 cases were negative, accounting for 65.38%.
3.3.5. Characterization of the expression of ATRX markers in diffuse gliomas
- In 216 studied patients, 41 cases were stained with ATRX marker, in
which the rate of loss of ATRX marker accounted for 36.59% and 63.41% of
cases did not lose expression with ATRX.
3.3.6. Characterization of the expression of P53 markers in diffuse gliomas
- In 216 case study patients, the number of revealed cases with P53 was
156, accounting for 72.22%, 2.6 times higher than the number of negative cases,
accounting for 27.78%.
3.3.7. Characterization of the expression of Ki67 markers in diffuse gliomas
- In 216 studied patients, the number of cases with Ki67 expression was
216, accounting for 100.0% at different levels, proving that the samples for
immunohistochemistry all showed guaranteed antigenicity.
3.3.8. Characterization of the expression of a combination of
immunohistochemical markers IDH1, INA and P53 of diffuse gliomas
- In 216 cases, there were 130 cases where the full set of 3 markers IDH1,
INA and P53 were completed. The group with IDH1 (-), INA (-), p53 (+)
accounted for the highest rate of 28.46%; ranked second is the group with IDH1
(+), INA (-), p53(+) accounting for 20.00%. The group with IDH1 (-), INA (+),
p53 (-) accounted for the lowest rate of 0.77%.

12
3.4. The relationship between the expression of immunohistochemical markers of
diffuse glioma with histopathological type and histological grade
3.4.1. The relationship between IDH1 marker expression with
histopathological type and histopathological grade
The relationship between IDH1 marker expression and histopathological type
(n=216)
- The group of oligoastrocytoma, oliodendroglioma, astrocytoma and
anplastic oligodendroglioma have a higher positive rate of IDH1 than other
groups with statistical significance with p < 0.05. The IDH1-positive
glioblastomas have a lower positive rate for IDH1 than other groups with
statistical significance with p<0.001. The group of anaplastic oligoastrocytoma
and anaplastic astrocytoma had a lower IDH1-positive rate than other groups,
but there was no statistical significance with p>0.05.
The relationship between IDH1 marker expression and histopathological
grade (n=216)
- Grade 2 gliomas, the probability of being positive for IDH1 marker was
higher than other groups with OR of 9.52 in the confidence interval from 3.87 to
23.45, statistically significant with p<0.001.
- Grade 3 gliomas, the possibility of being positive for IDH1 marker was
higher than other groups with OR of 1,409 in the confidence interval from 0.80
to 2.50, not statistically significant with p=0.239.
- Grade 4 gliomas, the possibility of being positive for IDH1 marker was
lower with other groups with OR 0.16 in the confidence interval from 0.09 to
0.29, statistically significant with p<0.001.
3.4.2. The relationship between the expression of immunohistochemical
marker INA with histopathological type and histopathological grade
The relationship between INA marker expression and histopathological type
(n=130)
- The group of oligoastrocytoma, oliodendroglioma, astrocytoma, and

anplastic have a higher positive rate with the INA marker than other groups,
with statistical significance with p<0.05. Glioblastomas have a lower positive
rate with the INA marker than other groups, with statistical significance with
p<0.001. Anaplastic oligodendrogliomas and anaplastic astrocytomas have a
lower IDH1-positive rate than other groups, but there is no statistical
significance with p>0.05.
The relationship between INA marker expression and histopathological grade
(n=130)
- Grade 2 gliomas have a higher probability of being positive for the INA
marker than other groups with an OR of 3,63 in the confidence interval from
1.67 to 7,89 statistically significant with p<0.001.
- Grade 4 gliomas have a lower probability of being positive for the INA
marker than other groups with an OR of 0.03 in the confidence interval from
0.004 to 0.21, statistically significant with p<0.001.
- Grade 3 gliomas have no statistically significant difference with other
groups in terms of ability to express INA marker with P = 0.124.


13
3.4.3. The relationship between expression of ATRX marker with
histopathological type and histopathological grade
The relationship between expression of ATRX marker with histopathological
type (n=41)
- In histopathological types, only the diffuse astrocytomas lost ATRX expression
higher than other groups with statistical significance with p<0.05.
The relationship between ATRX marker expression and histopathological
grade (n=41)
- Grade 2 gliomas have a lower positive rate for ATRX than other groups with
p<0.05 with statistical significance. That is, the loss of ATRX expression was more
common in the group of grade 2 gliomas than in the rest.

- Groups of grades 3, 4 the difference in ATRX expression with other
groups was not statistically significant with p>0.05.
3.4.4. The relationship between the expression of P53 marker with
histopathological type and histopathological grade

The relationship between expression of P53 marker and histopathological type
(n=216)
- Oligodendrogliomas have a lower positive rate for P53 than the rest with
an OR of 0.07 in the CI confidence interval from 0.02 to 0.23, this difference is
statistically significant with p<0.001. The P53 group revealed a higher rate of
P53 than other groups with an OR of 2.48, in the confidence interval from 1,23
to 4,99 with p=0.01, statistically significant. Groups of oligoastrocytoma;
astrocytoma, anaplastic oligoastrocytoma, anaplastic oligodendroglioma,
anaplastic astrocytoma have a high expression rate of p53 but the difference was
not statistically significant with p > 0.05.
The relationship between P53 marker expression and histopathological grade (n=216)
- Grade 2 gliomas have a lower probability of revealing the P53 marker
than other groups with an OR of 0.32 in the CI confidence interval from 0.17 to
0.52 with P<0.001, statistically significant.
- Grade 4 gliomas a higher ability to reveal the P53 marker than other
groups with an OR of 2,48 in the confidence interval of 1,23 to 4,99 with
p=0.01, statistically significant.
- Grade 3 gliomas have a higher ability to express P53 than other groups with
an OR of 1,28 in the CI confidence interval from 0.68 to 2.41, but this difference is
not statistically significant because p = 0.449 > 0.05.
3.4.5. The relationship between the expression of Ki67 marker with
histopathological type and histopathological grade
The relationship between the expression of Ki67 marker and histopathological
type (n=216)
- The rate of expression of Ki67 imprint of glioblastoma is highest

compared with other groups at 32.41±12,90. The group of anaplastic gliomas
account for the second highest percentage, anaplastic oligoastrocytoma are
16.65±7,87; anaplastic oligodendrogliomas are 13.07±5,48; anaplastic


14
astrocytomas are 14.94±5,07. The group of low-grade diffuse glioma has the
lowest Ki67 expression rate: oligoastrocytomas
are 3.79±1.62;
oligodendrogliomas are 4.25±1,65 and astrocytomas are 3.28±1,32.
The relationship between the expression of Ki67 marker and histopathological
grade (n=216)
- The rate of Ki67 expression with grade 2 diffuse astrocytomas are:
3.8±1.57, grade 3 gliomas are: 14.6±6.01 and grade 4 gliomas are: 32.4±12.90.

Figure 3.3: The ROC curve finds the Ki67 cut off point to distinguish grade 2
and grade 3 diffuse gliomas.
Comment: The area under the curve is 0.994 > 0.9 which is very good with
P<0.001. The cut off value of the % positive index of the Ki67 marker that
distinguishes grade 2 gliomas and grade 3 gliomas is 6.5% with J=0.917 (range
of the Ki67 ratio with sensitivity and specificity >0.8 (from 5.5% to 7.5%).

Figure 3.4: ROC curve to find the Ki67 cut off point to distinguish grade 3 and
grade 4 diffuse gliomas.
Coment: The area under the curve is 0.921 > 0.9 which is very good with
P<0.001. The Cut off Ki67 value that distinguishes grade 3 from grade 4 is
23.5% with J = 0.689 (range Ki 67 with sensitivity and specificity > 0.6 is from
16 to 23.5).
3.5. The relationship between the expression of immunohistochemical
phenotype IDH1, INA and P53 of diffuse gliomas with histopathological

type and histopathological grade
3.5.1. The relationship between the expression of immunohistochemical
phenotype IDH1(+), INA (+) and P53(+) with histopathological grade
(n=130)
- The ability to express of immunohistochemical phenotype IDH1(+),
P53(+) and INA (+) in the group of grade 2 gliomas are higher than in other
groups with OR of 2,53, statistically significant with p< 0.05. The ability to
express of immunohistochemical phenotype IDH1(+), P53(+) and INA (+) in the
group of grade 4 gliomas are lower than that of other groups with OR 0.08,
statistically significant with p<0 .05.


15
3.5.2. The relationship between the expression of immunohistochemical
phenotype IDH1(+), INA (-) and P53(+) with histopathological grade (n=130)
- The ability to express of immunohistochemical phenotype IDH1(+), INA
(-) and P53(+) in different groups of grades 2, 3 and 4 diffuse gliomas are not
statistically significant with all p > 0.05.
3.5.3. The relationship between the expression of immunohistochemical
phenotype IDH1(-), INA (-) and P53(+) with histopathological grade (n=130)
- The ability to express of immunohistochemical phenotype IDH1(-), INA
(-) and P53(+) in the group of grade 2 diffuse gliomas are lower than that of
other groups with statistical significance with OR 0.12 and p< 0.05. The ability
to express of immunohistochemical phenotype IDH1(-), INA (-) and P53(+) in
the group of grade 4 diffuse gliomas are statistically significant with OR of 5.47
and p<0.05.
Chapter 4
DISCUSSION
4.1. Characteristics of age, gender and some common clinical and
laboratory signs of patients with diffuse gliomas of the brain

4.1.1. Characteristics of distribution according to age group
- The most common age group in the study is from 30 to 40 years old. The
mean age is 47.42±13.53 years, which indicates that the age concentration of
diffuse glioma in the cerebral hemisphere is from 34 to 60 years old. Regarding
the distribution by age group, it is more common in the age group from 30 to 60
years old, similar to other authors such as Nguyen Phuc Cuong, Tran Chien and
Tran Minh Thong, there is no change in the distribution of age groups from the
years 2000 to present. Research by Le Van Phuoc at Cho Ray Hospital (n=109)
found similar results, the majority of astrocytomas see in the 31-60 age group
and the mean age was 37,57 ± 16.96 years old.
4.1.2. Characteristics of distribution according to sex
- In our study, the number of men infect is higher than that of women with
the rate of men accounting for 53.70%, women accounting for 46.30%, the ratio
of men to women was 1.16. This shows that the prevalence of diffuse glioma of
the brain is higher in men than in women, in the range of 10% to 30%.
Regarding the gender ratio, our results are quite similar to other studies in the
country in recent years. Author Pham Tuan Dung sees 54% males and 46%
females. Nguyen Duy Hung sees 57.7% males and 42.3% females. Nguyen
Ngoc Bao Hoang sees 54.8% males and 45 females .2%. Author Heather E.
Leeper et al. in 2015 (n= 159) found that male account for 57% and female
accounted for 43% .33 The results of our study are similar to the results of other
authors above. The world over the past 30 years has shown that the incidence
rate is higher in men than in women, ranging from 1.1/1 to 1.35/1.
4.1.3. Some common clinical signs of diffuse gliomas of the brain
- In this study, we monitored 168 patients with clinical and subclinical
signs. The results showed that the most common symptom of headache
accounted for 75,00%, then the symptom of limb weakness was 35.12%, and
epilepsy was 15.48%. In a study by Pham Tuan Dung in 2017 (n = 52), headache
was encountered in 92.3%, followed by hemiplegia 38.5%, epilepsy 28.8% and

16
vomiting 25%.30 Research by Nguyen Ngoc Bao Hoang in 2017 (n = 126), the
most common signs of headache were 46.8%, focal neurological signs 18.3%,
signs of nausea 7.9%.
4.1.4. Some common laboratory features of diffuse gliomas of the brain
- Location of the tumor: The most common location of diffuse glioma in
our study was the frontal lobe 61.90%, then the temporal lobe 37.50% and the
parietal lobe 16.67 %. Research by Nguyen Duy Hung in 2018 (n= 85) found the
position of frontal lobe 68.2%, temporal lobe 48.2% and parietal lobe 8.2%.31
Author Heather E. Leeper et al. 2015 A study of 159 cases found that the frontal
lobes accounted for 56%, the multiple lobes accounted for 13%, the temporal
lobes 11%, the parietal lobes 13% and the insula accounted for 7%.
- Tumor size: In our study, the average tumor size was: 54.19±16.25 mm,
the largest tumor was 105 mm, the smallest tumor was 10 mm. Research by
Nguyen Thi Thom in 2019 (n= 70) found that tumors > 5cm accounted for 51%,
tumors <= 5 cm accounted for 49%. Research by Le Van Phuoc (n=109) found
that low histopathological grade tumors have a small size 5.03±1.55 cm, tumor
with high histopathological grade tumors is 5.48±1.88 cm.
4.2. Histopathological features of diffuse gliomas of the brain
4.2.1. Distribution of histopathological subtypes using the
WHO
Classification in 2007
- The results of our study found that all microscopic histopathological types
of diffuse glioma of the brain were classified using the WHO classification in
2007. The type of glioblastoma is the highest rate of 37,96%, which explains
why is a group of tumors with 2 sources of pathogenesis from advanced lowgrade glioma and primary glioblastoma. The second group anaplastic glioma
tumors is 35.65%, including: anaplastic oligoastrocytoma, anaplastic
oligodendroglioma, anaplastic astrocytoma. Diffuse gliomas, grade 2 group
includes: olioastrocytoma, oligodendroglioma and astrocytoma. This group

accounted for the lowest rate of 26.39%. Compared with the authors of Japan,
the US, Europe and China, our rate is quite similar in the rate of
oligoastrocytoma and much higher than that of astrocytoma or pure
oliodendroglioma and this is also one of the main factors. This is also one of the
factors that makes the WHO 2016 classification of diffuse glioma no longer
divided into groups named according to each tumor cell type. The results are
slightly different from the study by Nguyen Phuc Cuong in the years 2000.
Eckel-Passow and et al in a study of 1056 cases of diffuse glioma of the brain
gave the rate of glioblastoma 44.5%, oliodendroglioma 19.4% and astrocytoma
accounted for the lowest rate of only 18, 8%. Rodriguez Pereira of Spain in a
study in 2000 with 134 cases including: astrocytoma 42 accounted for 31%,
anaplastic astrocytoma 25 accounted for 19%, oliodendroglioma 42 accounted
for 31%, anaplastic oliodendroglioma 3 accounted for 2.2% and glioblastoma is
25 accounts for 19%.
4.2.2. Distribution characteristics of histopathological grade of diffuse gliomas
- The research results show that the most common glioblastoma or grade 4
glioma is 37,96%. Diffuse glioma grade 2 accounts for at least 26.39%. The
distribution of histological grade, we encountered from grade 2 to grade 4, at most
grade 4 and then grade 3. Compared with previous studies in 2001 also at Viet Duc


17
Friendship Hospital, they I found it quite similar, grade 4 accounted for 22 out of 60
cases of diffuse glioma. Compared to a 2009 study by Dinh Khanh Quynh, the
author also encountered all histological grades from grade 2 to grade 4, at most
grade 4 accounted for 34.87% and grade 3 accounted for 28.4% of the cases. We
have quite similar results with the research results of Khalid and et al, Wakimoto and
et al.This also shows that diffuse glioma of the brain is often detected at a rather late
stage, when it has turned to grade 3 or grade 4.
4.2.3. Characteristics of the mitotic index by histopathological types and

histopathological grade of diffuse gliomas of the brain
- The results of our study showed a very clear spectrum of the number of
mitotic/0HPF according to the microtypes. The group of oligoastrocytomas
mitotic are 1.42±0.61/10HPF and the smallest number of encounters is 1 and the
largest is 3. The mitotic number of oligodendrogliomas are 1.5±0.76/10HPF, the
smallest number of encounters is 1 and the largest is 3. The diffuse astrocytomas
mitotic are 1.11±0.32/10 HPF, the smallest is 1 and the largest is 2. The above
results show that the mitotic/10 HPF for grade 2 gliomas group are consistent in
2 mitotic/10 HPF, and the number of mitotic for oligodendrogliomas are about 1
higher than the average astrocytoomas. The group anaplastic oligoastrocytooma
are mitotic/10 HPF of 7.06±3,29, the smallest number of cases is 4 and the
largest is 18. The group anaplastic oligodendroglioma are 8.31±3.752 mitotic/10
HPF, the smallest is 5 and the largest is 22. Anaplastic astrocytoma of mitotic/10
HPF are 8.13±3,93, the smallest is 4 and the maximum is 23. Anaplastic gliomas
of mitotic/10 HPF range from 4 to 23 and is most common in the 7 to 8
mitotic/10 HPF. Anaplastic oligodendrogliomas are higher than that of the
anaplastic astrocytoma with an average of about 0.5 dedifferentiation. The
number of glioblastoma mitotic/10 HPF are 19.65±14,438, the smallest ratio is 6
and the largest is 82. The mitotic number of glioblastomas are from 10 to 30 and
the most common is 20 mitotic/10 HPF. Stephen W. Coons studied 517 cases of
diffuse glioma of the brain in terms of the mitotic index and found: grade 2
astrocytoma (n = 51), with a mitotic index of 5.1 mitotic/10 HPF; grade 4
astrocytoma (n=292), is a mitotic index of 26.1 mitotic/ 10 HPF; grade 1
gliomas (n=9), no mitotic and grade 2 oligodendroglioma (n=62), is a mitotic
index of 3.1/10 HPF; grade 3 oligodendroglioma (n=22), is a mitotic index of
2.2/10 HPF; grade 4 diffuse gliomas (n=14), has a mitotic index of 1.4/10 HPF.
4.2.4. Tumor necrosis characteristics of diffuse gliomas
- The results of our study show that 56.94% of cases of diffuse gliomas has
necrosis, in which palisading necrosis is the most common, accounting for 38.43%,
followed by hypotrophic necrosis is 13.89%. Author F G Barker et al., 1996 studied

275 cases of glioblastoma and found that 88% had tumor necrosis.
4.3. Positive characteristics of immunohistochemical markers and the
relationship between the characteristics of immunohistochemical markers
with histopathological type and histopathological grade of diffuse gliomas
4.3.1. Characterization of GFAP marker expression and relationship with
histopathological type and histopathological grade of diffuse gliomas
- In our study, the positive rate for GFAP marker with strong level was
87.50%, moderate positivity accounted for 11.57% and only 0.93% was mildly
positive and negative. From there, it was shown that most of the glioma had