MINISTRY OF EDUCATION AND TRAINING
MINISTRY OF NATIONAL DEFENSE
MILITARY MEDICAL UNIVERSITY
VI THUAT THANG
STUDY ON HISTOPATHOLOGICAL FEARURES,
IMMUNOHISTOCHEMISTRY AND METHYLATION
OF RASSF1A GENE IN PROSTATE CANCER
Speciality: Biomedical Sciences Code: 9720101
THE MEDICAL DOCTORAL THESIS
Ha Noi – 2018WORKS ARE COMPLETED AT MILITARY MEDICAL UNIVERSITY
Name of supervisors:
1. Prof. Dr. Nguyen Dinh Tao, MD. Ph.D
2. Dr. Nguyen Ngoc Hung, MD.
Reviewer 1: Reviewer 2: Reviewer 3: Assoc. Prof. Nguyen Van Hung, MD. Ph.D. Assoc. Prof. Quan Hoang Lam, MD. Ph.D. Assoc. Prof. Trinh Tuan Dung, MD. Ph.D.
The thesis will be protected before the Board of thesis
dissertation on the day: / /
A thesis can be found at: 1. National Library 2. Library of the Military Medical University
3
INTRODUCTION TO THESIS 1. Set the problem
Prostate cancer (Pca) is common in men over age of 65, the disease is occult. Most cases were detected accidentally by histopathological examination. In Vietnam, the standard age rate in 2012 was 3.4/100000 people and queued the 10th in cancer in men. Pca can be treated effectively if the disease is detected early. Therefore, the finding of early detection methods, accurate diagnosis, proper assessment of histopathological lesions by a standard and a uniform classification is essential to the development of therapeutic approaches and prognosis of this disease.
histopathological lately,
Several studies have shown that the RASSF1A gene methylation occurs at an early stage in the formation and progression of the Pca. Thus, RASSF1A gene methylation marker is being considered to Pca. In Vietnam, studies on DNA methylation in cancer have been carried out and immunohistochemistry studies of Pca according to the 2004 World Health Organization (WHO) classification of tumors of the prostate are lacking. Based on that, we do the following:
a. Study on some histopathological features of prostate carcinoma at Military Hospital 103 according to the 2004 WHO classification.
b. Determine the expression of some immunological markers and methylation status of the RASSF1A gene and compare with some histopathological features in prostate carcinoma. 2. New contributions of the thesis + Study on RASSF1A methylation status in some cases of adenocarcinoma of the prostate. + Compare RASSF1A methylation status with some histopathological features in prostate carcinoma. + Simultaneous staining using antibodies against PSA,
CK34betaE12, p63, CK7, CK5/6 and actin in prostate carcinoma. 3. Structure of the thesis
4
The thesis is 135 pages including: Introduction (2 pages); Chapter 1 Document overview: 40 pages; Chapter 2 Subjects and Methods: 20 pages; Chapter 3 Research results: 31 pages; Chapter 4 Discussion: 40 pages; Conclusion: 2 pages; Request: 1 page; List of articles: 1 page. The thesis has 27 tables of data, 1 diagram, 13 figues, 25 images, 148 references (25 in Vietnamese, 123 in English), annexes to research forms and patient lists.
Chapter 1. OVERVIEW OF DOCUMENTS 1.1. A brief description of the histology of the prostate and classification of prostate cancer 1.1.1. Histology
Histologically the structure of prostate consists of acinus, branched ducts and prostate urethral ducts. The acinus and ducts are lined by an secretary inner cell layer and an outer basal cell layer. The prostate gland does not have myoepithelium, the acinus and ducts are surrounded by smooth muscle, fibroblasts and collagen fibers. The acinus and ducts contain faint pink secretions and corpora amylacea. The main ducts are lined by urothelial epithelium. 2.1.2. The 2004 WHO histological classification of tumours of the prostate
Carcinoma of the prostate including adenocarcinoma, urothelial carcinoma, squamous cell carcinoma, basal cell carcinoma. 1.2. Carcinoma of the prostate, prostatic intraepithelial neoplasia and Gleason grading system 1.2.1. Adenocarcinoma of the prostate
* Adenocarcinoma: Glandforming Pca typically contain glands that are more crowded than in benign prostatic tissue, although there is overlap with certain benign mimickers of Pca. Glands of adenocarcinoma of the prostate typically grow in a haphazard fashion. Glands oriented perpendicular to each other and glands irregularly separated by bundle of smooth muscle are indicative of an infiltrative process. Another pattern characteristic of an infiltrative process is the presence of small atypical glands situated in between larger benign glands with the loss of glandular
5
differentiation and the formation of cribriform structures, fused glands, and poorly formed glands the distinction between benign glands based on the architectural pattern becomes more apparent. Tumors composed of soild sheets, cords of cells, or isolated individual cells characterized undifferentiated Pca.
Nuclei in Pca range from those indistinguishable from benign prostatic epithelium to those with overt malignancy. In most Pca, there are cytological difference in the malignant glands when compared to the surrounding benign glands. Nuclear enlargement with prominent nucleoli is a frequent the finding, althouth not every cancer cell will display these features. Some neoplastic nuclei lack prominent nucleoli, yet are enlarged and hyperchromatic. Pca nuclei, even in cancers which lack glandular differentiation, show little variability in nuclear shape or size from one nucleus to another. Mitotic figures may be relatively common in high grade cancer, yet are infrequent in lower grade tumors. The cytoplasm is brighter than the benign gland. The acinus may contain crystals, pink secretions or mucus. It can be seen that perineural invasion, mucinous fibroplasia, glomerulations.
* Primary urothelial carcinoma: the vast majority are high grade and are associated with an in situ components. A single cell pattern of pagetoids spreads or burrowing of tumor cells between the basal cell and secretary cell layers of the prostate is characteristic. With extensive tumor involvements, urothelial carcinoma fills and expands ducts and often develops central comedonecrosis. Stromal invasion is associated with a prominent desmoplastic stromal response with tumor cells arranged in small irregular nests, cords and single cells. * Squamous cell carcinoma: it is identical to squamous cell carcinoma of other origin. Adenosquamous carcinoma is defined by the presence of both glandular (acinar) and squamous cell carcinoma components.
* Basal cell carcinoma: the tumors comprising large basaloid nests with peripheral palisading and necrosis. Histologic criteria for malignancy that distinguish it from basal cell hyperplasia an
6
infiltrative pattern, extraprostatic extension, perineural invasion, necrosis and stromal desmoplasia. 1.2.2. Prostatic intraepithelial neoplasia of the prostate (PIN)
PIN is best characterized as a neoplastic transformation of the lining epithelium of prostatic ducts and acini. High grade PIN (HGPIN) is characterized by a more uniform morphologic alteration, The acini and ducts are lined malignant cells with a variety of architectural complexity and pattern. The individual cells are almost uniformly enlarged with increased nuclear/cytoplastic ratio. Therefore showing less variation in nuclear size than that seen in low grade PIN. 1.2.3. Gleason grading system Gleason grading system defines five histological patterns with decreasing differentiation.
Gleason pattern 1, 2, 3, 4, 5. Pca has a pronounced morphological heterogeneity and usually more than one histological pattern is present. The primary and secondary pattern, i.e. the most prevalent the second most prevalent pattern are added to obtained a Gleason scores. 1.4. Immunohistochemistry in prostate carcinoma
Immunohistochemistry is a special staining technique that uses specific antibodies to determine the presence of corresponding antigens on tissue sections or on cell types present in tissue.
Application of Immunohistochemistry to Pca is aimed at: helping to identify the origin of the tumor, identifying the type of histopathology, identifying the variants of adenocarcinoma, Gleason grading, distinguishing malignant lesions from benign lesions, identify the invasion and metastasis of cancer.
β
In this study, we performed immunohistochemical staining with monoclonal antibodies against PSA, CK34 E12, p63, CK5/6, CK7, actin. 1.5. DNA methylation in cancer 1.5.1. DNA methylation in prostate cancer
DNA methylation is an epigenetic mechanism that occurs by the addition of a methyl (CH3) group to DNA. In human genome,
7
RASSF1A
DNA methylation process is the covalent addition of the methyl group at the 5carbon of the cytosine ring resulting in 5 methylcytosine (5mC). In prostate cancer, tumor suppressor gene is often methylated. Methylationspecific PCR (MS PCR or MSP) is one of the most commonly used methods for gene/sequencespecific detection of DNA methylation. In this study, RASSF1A gene methylation was selected to analysis DNA methylation in prostate adenocarcinoma. 1.5.2. Methylation specific PCR
Principles: The DNA undergoes bisulfite conversion of cytosine to uracil and then the methylated sequences are selectively amplified with primers specific for methylation. . Chapter 2. OBJECTIVES AND RESEARCH METHODS
2.1. Objects 2.1.1. The group of patients for histopathological research
84 specimens of prostate carcinoma (84 patients) performed transurethral resection of the prostate (TURP) at Military Hospital 103 (from June 2008 to July 2017) that were diagnosed as primary carcinoma. Patients who have medical records; reports for histopathology; paraffin imbedded tissue samples are adequate to analysis.
Exclusion criteria: Secondary Pca and cases do not meet the need of criteria
to carry out an which mention above. 2.1.2. The group of patients for immunohistochemistry research + 31 tissue samples of primary prostate carcinomas. + 2 tissue samples of primary urothelial carcinomas Samples of cancer are adequate immunohistochemical staning and still have antigenicity. Immunohistochemical staining with monoclonal antibodies β
against PSA; CK34 E12; p63; CK5/6; CK7; actin. 2.1.3. The group of patients for RASSF1A methylation research
8
20 tissue samples of adenocarcinoma were identified by histopathology. 10 tissue samples of benign hyperplasia of the prostate (BHP) were identified by histopathology (in order to compare RASSF1A methylation rate in cancer with BHP) Tissue samples of adenocarcinoma and BHP are sufficient
for DNA methylation assay. 2.2. Research methods 2.2.1. Research design
Prospective study Sampling method: full and intentional sampling. Sample size: Calculated according to a ratio survey
/2 = 1.96; d = 0.1; p = 78%. Filled in the above formula, sample size is 66 samples. In fact, prostate cancer specimens obtained from TURP are usually small, and insufficience of quantification for many techniques. We have collected 84 samples. 2.2.2. Materials, chemicals, research equipment 2.2.2.1. Information gathering materials
α With = 5%; Z1(cid:0)
+ Medical records and reports of histopathology of the Histopathology Department – Military Hospital 103.
+ Collecting informations including: histopathological types; forms; variants of prostate carcinoma. Differentiation of the tumor is calculated by the Gleason grading system including Gleason pattern 1, 2, 3, 4, 5. The most prevalent pattern and secondary most prevalent pattern are added to obtain a Gleason score.
+ Malignant specific features, intraluminal features and adenocarcinomas associated with HGPIN were as follows: have or have no. Staining intensity of tumor cells were as follows: “faint” (1+); “moderate” (2+); “strong” (3+); “very strong” (4+).
9
Methylation of the RASSF1A gene was as follows: unmethylated (), methylated (+).
+ All the H.E staining specimens (84 patients) and immunohistochemical staining specimens (33 patients) were examined on optical microscope with Nguyen Manh Hung and Tran Ngoc Dung (Department of Histopathology of Military Hospital 103) with illustrated photos.
+ Methylation assay: 20 adenocarcinomas samples and 10 BHP samples were analyzed together with Vo Thi Thuong Lan (Hanoi University of Sciences). 2.2.2.2. Tissue samples of Pca which obtained from TURP Fragments were randomly submitted to the study.
2.2.2.3. Tissue sections were studied by using for H.E staining, immunohistochemical staining and methylation assay
β Fixation of the the tissue samples in a 10% neutral buffered formaldehyde solution, then embedded in paraffin blocks. Trimmed paraffin blocks are cut at 310 micrometers (5 micrometers is commomly used) to make the H.E stain and immunohistochemical stain. Tissue sections of 84 samples prostate carcinoma were stained for H.E to histopathological analysis; Tissue sections of 33 samples prostate carcinoma were stained for antibodies against PSA, CK34 E12, p63, CK5/6, CK7 and actin to immunohistochemical analysis. Number of samples for methylation analysis: 20 samples of
prostatic adenocarcinoma and 10 samples of BHP. 2.2.3. Techniques used in research 2.2.3.1. H.E staning technique H.E. stain was performed according to routine histological technique. The use of histopathological criteria and the 2004 WHO classification of tumors of the prostate. Use the Olympus CX21 optical microscope. 2.2.3.2. Immunohistochemical technique + Chemicals, antibodies, buffer solution, detection system (Leica, USA).
10
+ Evaluation of results according to McNeal et al. (1991). 2.2.3.3. Determine gene methylation status of RASSF1A + Some steps of the MSP technique
β DNA extraction from paraffinembedded specimens. The quality DNA of specimens were tested by Polymerase Chain Reaction (PCR) targeting house keeping gene, globin ; evaluation PCR product by electrophoresis on 1.5% agarose gel. Bisulfite conversion: The extracted DNA is treated with bisulfite and purified by Epitect Kit (Qiagen, Cat, No 59104).
β
Examination of genomic DNA before and after bisulfite treatment by PCR, amplifying globin gene . The PCR product was separated on a 1.5% agarose gel in order to test the ability of DNA treated with bisulfite.
and
MSP was performed with specific PCR primers to detect the methylation of the RASSF1A gene. RASSF1AM210F/RASSF1A M211R primer amplify methylated DNAspecific product (170 bp). Whereas, nested PCR was performed with RASSF1AUn F1/RASF1AUnR1 RASSF1AUnF2/RASSF1AUnR2 primers (Table 2.3) in order to amplify unmethylated DNAspecific product. Table 2.3. Primer sequences for PCR and MSP
Sequneces (5’3’)
Primers GLF GLR RASSF1AM210F RASSF1AM211R RASSF1AUnF1 RASF1AUnR1 RASSF1AUnF2 RASF1AUnR2 CAACTTCATCCACGTTCACC GAAGAGCCAAGGACAGGTAC GGGTTTTGCGAGAGCGCG GCTAACAAACGCGAACCG GGGGTTTTGTGAGAGTGTGTTTAG TAAACACTAACAAACACAAACC GAGAGTGTGTTTAGTTTTGTTTTTG CCACAAAACAAACCCCAACTTCAA
2.3. Data analysis: Data is processed by SPSS13 software.
11
2.4. Ethical issues in research: The ethical principles in research are guaranteed.
Chapter 3. RESEARCH RESULTS
3.1. Proportion of patients with prostate carcinoma by age group
The proportion of patients with prostate carcinoma was highest in the 7079 age group (42.86%); average age: 74.34 ± 9.27; no patients under the age of 40 years seen.
3.2. The results identify some histopathological features of
prostate carcinoma
3.2.1. Determine the types of prostate carcinoma according to the 2004 WHO histological classification of tumors of the prostate
Table 3.2. Types of histopathology
Types of histopathology 1. Adenocarcinoma Acinar adenocarcinoma Ductal adenocarcinoma 2. Urothelial carcinoma 3. Squamous cell carcinoma 4. Basal cell carcinoma Number (n = 84) 82/84 80/82 2/82 2/84 0 0 Ratio (%) 97,6 97,6 2,4 2,4 0 0
3.2.3. Ratio of adenocarcinoma associated with HGPIN
Table 3.4. Association between adenocarcinoma and HGPIN
Histopathology p Number (n=82) Ratio (%)
12
60 22 73,2 26,8 < 0,001 Adenocarcinoma associated with HGPIN Adenocarcinoma without associated with HGPIN
The difference between adenocarcinoma associated with HGPIN and without associated with HGPN was statistically significant (p <0.001).
3.2.4. Grouping Gleason score into differentiation categories
Table 3.5. Grouping Gleason score into differentiation categories Ratio (%) Differentiated grade No. of patients (n=82) Gleason score
2 – 4 5 7 8 – 10 14 58 10 17,1 70,7 12,2 Well differenciated Morderately differenciated Poorly differenciated
3.2.7. Distribution of malignant specific features of tumors
Table 3.8. Distributied ratio of adenocarcinoma according to malignant specific features (n=82)
Malignant specific features
Perineural invasion Mucinous fibroplasias Glomerulation Tumor has 2 or 3 malignant specific No malignant specific features seen Number 32 9 10 6 25 Ratio (%) 39% 11% 12,2% 7,3% 30,5%
Table 3.10. The distribution of tumors have malignant specific features and tumors have no malignant specific features by Gleason score group
13
Gleason score
Gleason 24 n=14 (%) Gleason 57 n=58 (%) Gleason 810 n=10 (%) Total (%)
Malignant specific features
Have 1/14 (7,2%) 47/58 (81%) 9/10 (90%) 57 (69,5%)
Have no 13/14 (92,8%) 11/58 (19%) 1/10 (10%) 25 (30,5%)
Total 14 (100%) 58 (100%) 10 (100%) 82 (100%)
P < 0.001
There is a correlation between Gleason score and malignant specific features. The higher the Gleason score, the higher the incidence of malignant specific features, with p <0.001.
3.2.9. Proportion of substances contained in malignant glands
Table 3.11. Rate of substances contained in malignant glands
Intraluminal features Number Ratio (%)
Crystalloids Pink acellular dense secretions Crystalloids/Pink acellular dense secretions No Crystalloids/Pink acellular dense seen 8 44 18 12 9,7 53,7 22 14,6
Table 3.12. Percentage distribution of tumors contain Crystalloids/Pink acellular dense secretions and tumors no contain Crystalloids/Pink acellular dense secretions according to the Gleason score group
Gleason score Total Gleason 24 n=14 (%) Gleason 57 n=58 (%) Gleason 810 n=10 (%)
14
Crystalloids & Pink dense secretions 14/14 (100%) 55/58 (94,83%)
Contain No contain Total 0/14 (0%) 14 (100%) 3/58 (5,17%) 58 (100%) 1/10 (10%) 9/10 (90%) 10 (100%) 70 (85,4%) 12 (14,6%) 82 (100%) p < 0,001
There is an association between Gleason score and Crystalloids
& Pink acellular dense secretions. The higher the Gleason score, the
lower the incidence of tumors contain Crystalloids & Pink acellular
dense secretions (p <0.001). 3.3. Results of immunohistochemistry study 3.3.1. Immunohistochemical staining of prostate carcinoma
β
94% of prostate carcinoma express PSA, the remaining 6% of prostate carcinoma does not express PSA but express CK34 E12 and p63 (2 cases of urothelial carcinoma). The tumor cells do not express CK7, CK5/6, actin.
3.3.3. Level of PSA expression of tumor cells
Table 3.16. Level of PSA expression of tumor cells
Histopathology
Level Level Ductal adenocarcinoma (n=2) Acinar adenocarcinoma (n=29)
() 0 (0%) 0/2
(1+) 7 (24,2%) 0/2
15
(2+) 13 (44,8%) 2/2
(3+) 9 (31%) 0/2
(4+) 0 (0%) 0/2
3.3.4. Distribution of PSA expression levels of tumor cells according to Gleason grade
Table 3.17. PSA expression levels according to Gleason grade (n = 31)
p Gleason score Leve l Grade 2 Grade 3 Grade 4 Grade 5 Total (%)
5 2 15 < 0,001 9 (1+) (2+) (3+) (4+) 7 (22,6%) 15 (48,4%) 9 (29%) 0 (0%)
There is an inverse relation between PSA and Gleason grade. The higher the Gleason grade, the lower the PSA expression (p <0,001). 3.3.5. The level of PSA expression in the tumor group of neural invasion Tumor cells expressed PSA at an average level of 53.85%. Tumor cells expressed PSA at an weak level accounts for 46.15%.
β 3.3.6. Expression of CK34 E12 and p63 of the basal cell β Table 3.19. Expression of 34 E12 and p63 of basal cell Markers (n=33)
16
34βE12 p63
Adenocarcinoma Benign part (+) (+) 31 31 Cancer part () ()
Urothelial cancer Benign part 2 (+) 2 (+)
Cancer part (+) (+)
β
β
β
Adenocarcinoma: benign part expressed CK34 E12 and p63, whereas the cancer part did not express CK34 E12 and p63. Urothelial carcinoma: benign part and cancer part expressed CK34 E12 and p63. 3.3.8. Status and level of CK7 and CK5/6 expression of benign and malignant urothelial cell
Table 3.21. Status and level of CK7 and CK5/6 expession of benign and malignant urothelial cells Markers (n=33)
CK7 CK5/6
Histopa thology Adenocarcinoma Benign part (4 +) () 31 31 Cancer part () ()
Urinary cancer Benign part (4+) () 2 2 Cancer part () ()
Adenocarcinoma: benign part expressed CK7, but did not express CK5/6. The cancer part did not express CK7 and CK5/6. Urothelial carcinoma: benign part expressed CK7, but did not express CK5/6. The cancer part did not express CK7 and CK5/6. 3.3.9. Status and level of actin expression of various stromal types
17
The mooth muscle of the prostate gland and vascula expressed actin. Fibrous cells, basal cells, endothelial cells did not express actin. 3.4. Results of RASSF1A methylation study 3.4.2. Results of evaluating the efficiency of bisulfite treatment
Results of genomic DNA before and after bisulfite treatment,
β amplifying globin gene by PCR was indicated in Figure 3.2
β globin gene from the Figure 3.2. PCR products amplified the
before (red band) and after DNA (yellow band) treated with bisulfite
of the PCa and BHP samples.
Prior to bisulfite treatment, samples were amplified PCR product
(250 bp). After treatment with bisulfite, PCR product was not
obserbed in gel eletrophoresis. Thus, genomic DNA were completely
treated with bisulfite.
3.4.3. Result of RASSF1A methylation in prostate cancer
Methylated and Unmethylated DNAspecific products was
detected in 11/20 (170 bp) and 20/20 (137 bp) Pca samples,
respectively. RASSF1A methylation ratio in prostate cancer is 11/20
specimens (55.0%)
18
Figure 3.. MSP product of PCa samples (P1P20) using methylated
primer (RASSF1AM210F/RASSF1AM211R) and unmethylated
primer (RASSF1AUnF2/ RASSF1AUnR2)
3.4.4. Result of RASSF1A methylation in begnin hyperplasia of
prostate
Figure 3.4. MSP result of BHP samples (B1B10)
Note: (m): methylated DNA, (u): unmethylated DNA. (): negative
control without DNA template
MSP analysis also revealed that the methylation of RASSF1A
was detected in 2/10 patients with BHP. Methylated and
unmethylated DNAspecific products was detected in 2/10 (170 bp)
and 10/10 (137 bp) BHP samples, respectively.
19
3.4.5. Relationship between RASSF1A methylation and pathological
characteristics in prostate cancer and begnin hyperplasia of prostate
3.4.5.1. RASSF1A methylation in Pca and BHP Table 3.3. RASSF1A methylation ratio in PCa and BHP
Sample Number (n=30) Methylation (%)
PCa 20 11/20 (55%)
BHP 10 2/10 (20%)
Methylation of RASSF1A was detected in 55% and 20% patients
with prostate cancer and begnin hyperplasia of prostate, respectively.
3.4.5.3. Methylation of RASSF1A methylation according to Gleason
score/tumor differentiation
ả B ng 3. 25. Ratio of RASSF1A methylation according to Gleason
Gleason score differentiation score/ differentiation of tumor Number (n=20) Methylation (%)
24 7 2 (28,6%) Well differentiated
57 8 4 (50%) Moderately differentiated
810 5 5 (100%) poorly differentiated
RASSF1 methylation increased according to Gleason score
differentiation of tumor.
20
3.4.5.4. Methylation of the RASSF1A gene and status of neural
invasion
21
Table 3.26. Methylation ratio according to status of neural
invasion
Neural invasion Methylation rate (n %) Total
(n=20)
7 Have 6/7 (85,7%)
13 No 5/13 (38,5%)
The rate of methylation in tumor groups of neural invasion is high
(85.7%).
3.4.5.5. Methylation of RASSF1A gene and PIN status in BHP
Table 3.27. RASSF1A gene methylation rate and PIN status in BHP
Total Methylation rate Histopathology (n=10) (n %)
BHP associated with low PIN 2/5 (40%) 5
BHP without associated with low PIN 0/5 (0%) 5
The methylation rate in BHP group associated with low PIN of 40%.
No methylation was detected in the BHP group with have no PIN.
Chapter 4. DISCUSSION
4.1. Percentage of patients with prostate cancer by age group
22
Proportion of patients with Pca increased with age group: group 70 79 accounts for the highest rate (42.86%), group 4049 accounts for very low rate (1.19%). Our result is consistent with Nguyen Viet Hai (2013), Ngo Van Trung (2004) and Gronberg H. (2003). 4.2. Some histopathological features of prostate carcinoma 4.2.1. Identify the histopathology types, forms, and variants + 2 out of 4 types of primary carcinomas were identified that were carcinoma. No squamous cell adenocarcinoma and urothelial carcinoma and basal cell carcinoma to be found. + Our result shows that primary adenocarcinoma accounts for 97.6%. This result is higher than Nguyen Van Hung (2004) but that is equivalent to the result of Nguyen Viet Hai (2013). Detection of primary urothelial carcinoma accounts for 2.4%. This result is consistent with Eble.J.N (2004) and Mostofi (1992). + Acinar adenocarcinoma accounts for 97.6%. This result is higher than Nguyen Van Hung (2004) and in accordance with previous announcements. Ductal adenocarcinoma accounts for 2.4%, which is a rare form. This result is consistent with published reports. + Ordinary acinar adenocarcinoma variant is the most frequent (87.5%), following pseudohyperplastic variant (7.5%) and atrophic variant (5%). Variant such as foamy gland, ring cells, colloid, oncocytic cell, lymphoepitheliomalike cell and carcinosarcoma were undetected. This result matches Eble J.N ‘s announcement (2004). 4.2.2. Adenocarcinoma associated with HGPIN + The incidence of adenocarcinoma associated with HGPIN is 73.2%. This result is in accordance with Nguyen Van Hung (2005) and Alexander E. E. (1996). + Adenocarcinoma associated with HGPIN accounted for the highest proportion in the Gleason score group 57 (66.7%), group 24 and group 810 were low proportion. This result was partially matched
23
with authors such as Eptain JL. Erbersdobler A. (1996), Hagman M. J. (1997), Montoroni R. (1999), Paradis V. (1999). 4.2.3. Gleason grades in adenocarcinoma + Gleason grade 3 usually meets in the primary patterns (46.3%), Gleason grade 4 is usually meets in the second patterns (43.9%), the remaining Gleason grades account for low proportion. This result is in accordance with Nguyen Van Hung (2005) and Samaratunga H. (2014). Eble. J. N (2004). + Gleason score of 57 was the highest (70.7%), this is consistent with Sgrignoli A. (1994), Samaratunga H. (2014). Gleason score 24 was 17.1%, that was higher than Nguyen Van Hung (2005) and Oesterling (1987). + Gleason score 810 was 12.2%, that was lower than Nguyen Viet Hai (2013) and Nguyen Van Hung (2005). In our opinion, this difference is due to the selection of subjects. 4.2.4. Malignant specific features + Malignant characteristics of tumors were 69.5%, tumors had no malignant characteristics were 30.5% (p <0.001). + The incidence of neural invasion tumor was 39%, this is lower than Eble's. J. N (2004). In our opinion, this result may be due to the selection of subjects. + The incidence of mucinous fibroplasia was 11%, and we found only in Gleason score 57 and 810. This result was consistent with Christian J. D. (2005). + Glomerulation likemalignant glands were 12.2%, and we found only in Gleason score 57 and Gleason score 810. 4.2.5. Luminal features + Tumors containing luminal features (85.4%) were higher than tumors noncontaining luminal features (p <0.001).
24
β
β
+ Tumors containing eosinophilic crystals (9.7%) were consistent with Young (2000) but higher than Nguyen Van Hung (2005). + Tumors containing dense pink secretions (53.7%), in accordance with Iczkowski K. A. (1999) and Christian J. D. (2005). + Tumors containing dense pink secretions/crystals (22%). Tumors do not contain dense pink secretions /crystals (14.6%). 4.3. The expression of some immunological markers and the methylation status of the RASSF1A gene in carcinoma of the prostate. 4.3.1. Expression of some immunological markers. * Prostate specific antigen (PSA) Primary carcinoma expresses PSA (94%), this result matched Nguyen Van Hung (2005) and Brimo F. (2012). Gleason grade 4 and 5 express PSA at week level, Gleason grade 3 expresses PSA at the moderate level, Gleason grade 2 expresses PSA at strong level. The higher the Gleason score (less differentiation), the less likely the expression (p <0.001). This result is consistent with the notice in the literature. Primary urothelial carcinoma does not express PSA, consistent with Eble.J.N. (2004) and previous studies. Tumors with neural invasion expresses PSA at the weak level (46.2%) and at the moderate level (53.8%). This result is consistent with Epstein J. I. (2014) and previous studies. * High molecular weight cytokeratins detected by CK34 E12β Malignant gland part does not exposed CK34 E12, in accordance with the published reports. + Benign gland part expresses CK34 E12 from m oderate level (69.7%) to strong level (30.3%), partly consistent with the results of researching of Nguyen Van Hung (2005). * p63 + Malignant gland part does not express p63.
25
β
+ Benign gland part and urothelial carcinoma express p63. Primary urothelial carcinoma: remnant basal cells of urothelial carcinoma in situ express CK34 E12 and p63. + Basal cells express p63 at the strong level (81.8%) and moderate level (18.2%). This matched previously reports. * Cytokeratin 7 The benign urothelial cells express CK7 at very high levels (100% cases). Meanwhile, primary urothelial carcinoma does not express CK7. This result is similar to Nguyen Van Hung ‘s result (2005). The use of CK7 will contribute to differentiating primary urothelial carcinoma of the prostate with secondary urothelial carcinoma and other types of carcinoma. * Cytokeratin 5/6 Benign prostatic epithelium, benign urothelial epithelium and malignant urothelial epithelium do not express CK5/6. This result consistent with published reports. Thus, the use of CK5/6 will help to identify the primary squamous cell carcinoma of the prostate with secondary squamous cell carcinoma. * Actin The smooth muscle of the prostate and the smooth muscle of the blood vessels express actin at a very high level. Basal cells, fibrocytes, endothelial cells do not express actin. This result is consistent with published reports. The use of actin will determine the invasion of tumor cells to wall of blood vessels. 4.4.2. Status of RASSF1A gene methylation in adenocarcinoma and begnin hyperplasia of prostate + Ratio of RASSF1A gene methylation in adenocarcinoma (55%) was higher in begnin hyperplasia of prostate (20%). + Ratio of RASSF1A methylation in adenocarcinoma of this study is consistent with some reports such as Bastian PJ (2004), Singal
26
̃ ươ ̣ ng Lan (2016).
(2004), whereas ratio of RASSF1A methylation is lower than one in reports of Woodson (2004), Kuzmin (2004), Liu L. (2002), Kawamoto (2007). + Ratio of RASSF1A methylation in adenocarcinoma of this study is higher than ratio of RASSF1A methylation in analysis of Syeed N. (2010), Vo Thi Th + Ratio of RASSF1A methylation in Gleason grade based group including grade 2, grade 3 and 4, grade 5 is 25%, 66.6% and 100%, respectively. Ratio of RASSF1A methylation increases according to Gleason score. Ratio of RASSF1 methylation in Well – differentiated group (Gleason score 24), Moderately differentiated group (Gleason score 57) and poorly differentiated group (Gleason score 810) is 28.6%, 50% and 100%, respectively. This result is partly consistent with report of Lui L. (2002). The methylation rate of neural invasive tumor group (85.7%) is higher than those without neural invasive tumor group (38.5%), this result is consistent with the result of Kang (2004). + RASSF1A methylation in begnin hyperplasia of prostate: The RASSF1A methylation rate in begnin hyperplasia of prostate is 20%. This result is consistent with reports of Singal (2004), Kawamoto K. (2007), Syeed N. (2010) and Vo Thi Thuong Lan (2016).
27
CONCLUSION
1. Some histopathological features of prostatic carcinoma + Two out of four types of primary prostatic carcinoma are identified, of which adenocarcinoma is very high (97.6%), urothelial carcinoma is low (2.4%).
In adenocarcinomas, most of them are acinar adenocarcinoma
(97.6%), ductal adenocarcinoma are less common (2.4%).
+ Ordinary acinar adenocarcinoma variant is very common
(85%), other variants are rare.
Adenocarcinoma associated with high grade prostatic
intraepithelial neoplasia was 73.2%.
Gleason grade 3 is the most common and occupying in the primary pattern of the tumor (46.3%), Gleason level 4 is the most common and occupying in the second pattern (43.9%).
+ Gleason score of 57 accounts for the highest percentage (70.7%), Gleason score 24 accounts for 17.1%. Gleason score 810 accounts for the lowest percentage (12.2%).
The tumors had malignant specific features (69.5%) higher than
the those without malignant specific features (30.5%).
2. Expression of some immunological markers and methylation status of RASSF1A gene and compared with some histopathological features
2.1. Expression of some immunological markers + Primary carcinoma of prostate expressed PSA with rate of 94%. 31 cases (100%) adenocarcinoma expressed PSA at different levels: 29% were strong expression (Gleason grade 2), 48.4% were expression (Gleason grade 3), 22.6% were weak morderate expression (Gleason grade 4 and Gleason grade 5). Tumors with neural invasion (13 cases) expressed PSA at a weak level (46.15%),
28
at a morderate level (53.85%). Primary urothelial carcinoma (2 cases) did not express PSA.
β β + Adenocarcinoma of the prostate: the malignant gland parts were negative to CK34 E12 and p63. In contrast, benign gland parts were positive to CK34 E12 and p63.
+ Urothelial carcinoma: the benign urothelial cells were positive to CK7, but were negative to CK5/6. Meanwhile, the malignant urothelial cells were negative to CK7 and CK5/6.
+ The smooth muscles of prostate (33 cases) were positive to
actin. Other connective tissues were negative to actin.
2.2. Methylation status of RASSF1A gene + The RASSF1A gene methylation rate in adenocarcinoma is
55% (11/20 cases).
The incidence of RASSF1A gene methylation in Gleason score 24, 57 and 810 were 28.6%, 50%, 100%, respectively. Tumors with neural invasion: 85.7% and tumors without neural invasion: 38.5%.
REQUEST
β
1. The 2004 World Health Organization of tumos of the prostate should be applied. Immunohistochemical staining with antibodies in against PSA, CK34 E12, p63, CK5/6, CK7, actin should be applied suspected cases.
2. When seeing high grade PIN on the H.E staining slides but has not cancer, need taking more tissue samples or parafin embeded tissue samples need to be cut more slides in order to avoid missing cancer.
3. Addition of DNA methylation markers on prostatic carcinoma such as methylation markers of GSTP1 gene… and RASSF1A gene marker should be used for larger specimens.
LIST OF FUBLISHED RESEARCH WORKS RELATED TO
THESIS
1. Vi Thuat Thang, Nguyen Dinh Tao, Vo Thi Thuong Lan et al. (2014), "Analysing the Hypermethylation of the RASSF1A in Prostate Cancer at Hospital 103", Journal of 108Clinical Medicine and Pharmacy.,(9): 7781.
2. Vi Thuat Thang, Nguyen Dinh Tao, Nguyen Ngoc Hung et al. (2017), "Study on some histopathological features and expression of some immunohistochemical markers in prostatic carcinoma", Journal of Military Pharmaco Medicine., (8):222227.
