CLINICOPATHOLOGIC PATTERN AND DNA METHYLATION

STATUS OF COLORECTAL CARCINOMA IN A GROUP OF

BANGLADESHI PATIENTS

Dr. Abu Khalid Muhammad Maruf Raza

MBBS

DEPARTMENT OF PATHOLOGY
Bangabandhu Sheikh Mujib Medical University
Dhaka, Bangladesh
2010
Contents Page
_________________________________________________ no

1 INTRODUCTION 01
2 REVIEW OF LITERATURE
a Anatomical consideration 04
b Physiology 08
c Tumors of the colon and rectum 09

3 MATERIALS AND METHODS 26

4 OBSERVATIONS AND RESULTS 33
5 DISCUSSION 54
6 SUMMARY AND CONCLUSION 62
7 BIBLIOGRAPHY 64
8 APPENDICES
I Certificate from The Chicago University i
II Clinical proforma ii
III Preparation of stains and chemicals iv
IV Tissue processing and staining procedures v
V Protocol for DNA extraction viii
VI Protocol for DNA methylation x
VII TNM staging system xii
VIII Master table xiv
IX Illustrations xviii
 LIST OF TABLE

Table Title Page no

no
2.3:1 Genes silenced by hypermethylation in colorectal cancer 18
2.3:2 TNM Classification of Colorectal Carcinoma by American Joint 22
Committee on Cancer (AJCC).
2.3:3 TNM staging for colorectal carcinoma. 23
4.1 Sex distribution in relation to age group. 35
4.2 Symptoms with tumour location of colorectal cancer cases. 36

4.3 Haemoglobin concentration in 50 cases of colorectal cancer 37

patients
4.4 CEA (Carcinoembryonic antigen) level in 50 colorectal 37
cancer cases
4.5 Site distribution of colorectal cancer from distal to proximal 38
4.6 Histological diagnosis of 50 cases of colorectal cancer 40

4.7 Histological grading of colorectal cancer cases with sex and 41

location
4.8 TNM (Tumour, Node, Metastasis) stage with sex and location 42
4.9 Signet ring cells in colorectal cancer patients 45

4.10 Circumferential margin involvement, Perineural invasion and 47

Tumour border configuration with TNM stage.
4.11 Selected genes that show methylation differences in 24 cases of 53
Colorectal carcinoma.
5.1 Age and sex distribution of CRC in different studies and the 54
present study.
5.2 Distribution of colorectal cancer in left colon and right colon. 57
 LIST OF FIGURES
Fig. Title Page
No no
2.1 Gross anatomy of large intestine 07
4.1 Bar diagram showing age distribution of colorectal cancer. 34
4.2 Growth pattern of colorectal cancer cases. 39
4.3 Various grade of tumour infiltrating lymphocytes in 50 cases. 43

4.4 Lymphovascular invasion in colorectal cancer patients 44

4.5 Extracellular mucin in colorectal cancer patient. 45
4.6 Scatter diagram showing total intensity in tumour tissue and its 48
corresponding normal tissue in all 24 cases
4.7 Scatter diagram showing average beta of all 48 samples 49
4.8 Histogram showing tumour differential score. 50
4.9 Bar diagram showing sources of variation of methylation data. 51
4.10 Variation of methylation data due to tumour differentiation. 52
1.0 INTRODUCTION

Colorectal cancer (CRC) is the third most common cancer in the world and the second

leading cause of cancer related deaths in the United States. Globally, the incidence of

CRC varies widely with higher incidence rates in North America, Australia and Northern

and Western Europe (Aljebreen, 2007). The lifetime risk of developing CRC is about 6%

or one in 18. Over 95% of these CRC is adenocarcinoma (Kim et al, 2010).

CRC is relatively uncommon in Indian sub continent. In India the incidence of colorectal

cancer was found to be 4.2 and 3.2 per hundred thousand for male and female population

respectively (Afroza et al, 2007).

The incidence of colorectal cancer in Bangladesh is not exactly known, it appears to be

common and occur in younger age group with slight male preponderence. Average age at

diagnosis is 10 years less than the developed countries. Rectal bleeding is the most

common symptoms and majority of the carcinoma were in the rectum (Hossain, 2007).

The peak incidence of colorectal carcinoma is in between the age of 60 and 69 years.

Fewer than 20% occur before the age of 50 years. Males are affected slightly more than

females (Turner, 2010). Colorectal carcinoma are rare before the age of 40 without

genetic predisposition or without predisposing factors (Hamilton, 2000).

Early detection of colonic cancers is a challenging task as because clinical symptoms

develop slowly. Per rectal bleeding is common. Many patients experience change in bowel

habit (Yawe et al, 2007). Screening tests like digital rectal examination, simple laboratory

1
investigations like estimation of CEA, estimation of haemoglobin, faecal occult blood test,

and visualization of the gut mucosa by sigmoidoscopy and colonoscopy examination may

be a help in the diagnosis (Aljebreen, 2007).

Colorectal cancer is a multifactorial disease process. Etiology contributing from

environmental factors including diatery factors, obesity, alcohol intake, smoking, life style

and genetic and epigenetic abnormalities. The molecular events that leads to CRC is

heterogenous and includes genetic and epigenetic abnormalities. Genetic events in

colorectal cancer is genetic alteration of the APC gene, mutations in the KRAS and P53

gene and abnormalities in the DNA mismatched repair genes (Turner, 2010).

Epigenetic changes, which is the heritable changes in gene function that are not due to

changes in the DNA sequence is an important pathway in the mechanism of tumerogenesis

in colorectal cancer. DNA methylation abnormalities is an important epigenetic changes in

CRC and become an area of great interest in the field of cancer research. The significance

of DNA methylation alteration in CRC has been reported widely. There is both

hypermethylation and hypomethylation abnomalities in various genetic loci in CRC (Kim

et al, 2010).

Available data indicates some differences in CRC in Bangladeshi population like lower

mean age of patients, predominently left sided tumour (Hossain, 2007). Correlation of

histopathological features with clinical data and detection of genetic abnormalities can give

better insight in this field.

2
1.1 AIM

To see the morphologic pattern, clinical features and DNA methylation changes in

colorectal carcinoma in a group of Bangladeshi patients.

1.2 OBJECTIVES

1. To see various histologic features of colorectal carcinoma.

2. To see different clinical presentation of colorectal carcinoma.

3. To see DNA methylation changes in colorectal cancer tissue and corresponding

surrounding normal colonic tissue.

3
2.0 REVIEW OF LITERATURE

2.1 Anatomy of large intestine

2.1.1 Gross anatomy:

The large intestine is about 1.5 meters long, extends from the distal end of ileum to anus.

It begins in the right iliac fossa as the caecum, from which the vermiform appendix arises.

The caecum becomes the ascending colon which passes upwards in the right lumbar region

and hypochondrium to the inferior aspect of the liver. It then bends to the left forming the

hepatic flexure (right colic flexure) and becomes the transverse colon. This loops across

the abdomen with an anterior-inferior convexity until reaches the left hypochondrium.

Where it curves inferiorly to form splenic flexure (left colic flexure) and becomes the

descending colon. It then forms sigmoid colon in the left illac fossa. The sigmoid colon,

descends deep into the pelvis and becomes the rectum, which ends in the anal canal at the

level of pelvic floor (Borley, 2008).

2.1.2 Blood supply of the colon:

Arterial supply: The arteries which supply the parts of the large intestine are from the

colic branch of the superior mesenteric artery. The hind gut derivatives are supplied by the

inferior mesenteric and rectal arteries (Borley, 2008).

4
The venous drainage: The vein of the large gut are the corresponding superior and

inferior mesenteric artery. The rectum and anal canal are drained by the rectal vein

(Borley, 2008).

Lymphatic drainage: The large gut is drained to the epicolic nodes, paracolic nodes, the

intermediate colic and terminal nodes (Borley, 2008).

2.1.3 The nerve supply:

The sympathetic supply of mid gut derivate are coeliac and mesenteric ganglia and the

parasympathetic supply is by the vagus nerve. The hind gut derivatives are supplied by the

superior hypogastric plexus. The rest of the portion is supplied by inferior hypogastrci

plexus and pudendal nerve (Borley, 2008).

2.1.4 Developmental anatomy of colon:

The caecum, the appendix, the ascending colon and the proximal two-third of the

transverse colon are derived from the midgut. The distal one-third of the transverse colon,

the descending colon, the sigmoid colon, the rectum and the upper portion of the anal canal

are derived from hindgut (Sadler, 1995).

2.1.5 Histology of the colon:

The large intestine is divided topographically into three segments ( Junquira et al, 1995):

5
 a. Colon,

b. Rectum and

c. Anal canal.

All the segments histologically have four coats;

a. Mucosa

b. Submucosa

c. Muscularis externa and

d. Serosa (Young and Heath, 2000).

The surface epithelium of the mucosa is made of tall columnar epithelium mixed with

goblet cells turn downward as a tubular gland. The lower portions of the tubular glands

have numerous goblet cells. The lamina propria is composed of fibrocollagenous tissue.

The submucosa is composed of loose connective tissue containing large blood vessels and

meissner plexus of nerve.

Muscularis externa in the colon shows some variation from the other parts of the gut. The

inner circular layer is complete and prominent. The outer longitudinal layer forms three

equidistant, longitudinal bands (Tinia coli).

Adventitia or serosa consists of thin connective tissue layer covered by the mesothelium

(Yound and Heath, 2000).

6
7
2.2 Physiology of the large intestine

The main function of the colon is absorption of water, Na+, and other minerals by removal

of about 90% of the fluid. It converts the 1000-2000ml of isotonic chyme that enters it

each day from the ileum to about 200-250ml of semisolid feces (Ganong, 2005).

Motility and secretion of the colon: When peristaltic wave reaches to the valve, it opens

and permits ilial chyme to squirt into the caecum. The movement of the colon include

segmentation contraction and peristalsis. Segmentation contraction mixes the content of

the colon and by exposing the more of the contents to the mucosa facilitates absorption.

Peristaltic waves propel the contents towards the rectum (Ganong, 2005).

Transit time in the colon: The first part of the test meal reaches the caecum in about 4

hours and all of the undigested food entered the colon in 8 to 9 hours. The first meal

reaches the pelvic colon in 12 hours (Ganong, 2005).

Absorption in the colon: Na+is actively transported out of the colon and water follows

along the osmotic gradient. K+and HCO3‫ ־‬are secreted into the colon. The absorptive

capacity of the colon makes it a suitable route of drug administration (Ganong, 2005).

Daefecation: Distention of the rectum with faces initiates reflex contraction of the

musculature and desire to defecate. The sympathetic nerve supply to internal anal sphincter

is excitatory whereas the parasympathetic nerve supply is inhibitory (Ganong, 2005).

8
2.3 Tumours of the colon and rectum

Many conditions, such as congenital disease, infection, inflammatory diseases, motility

disorders and tumours affect the large intestine (Turner, 2010). The colorectal cancer is

the second most common visceral cancer in the U.S.A. (Cooper, 1999). Virtually 98% of

colorectal carcinoma are adenocarcinoma (Turner, 2010). The World Health Organization

classified colorectal cancer histologically as follows (Hamilton. 2000).

2.3.1 WHO classification of tumors of colon and rectum: (Hamilton, 2000)

1. Non-neoplastic polyp

i. Hyperplastic polyp

ii. Peutz-Jeghers polyp

iii. Juvenile polyp

2. Epithelial tumors

1. Adenoma

i. Tubular adenoma

ii. Villous adenoma

iii. Tubulovillous adenoma

iv. Serrated adenoma

9
2. Intraepithelial neoplasia (Dysplasia) associated with Chronic inflammatory

diseases

Low-grade glandular intraepithelial neoplasia

High grade glandular intraepithelial neoplasia

3. Carcinoma

i. Adenocarcinoma

ii. Mucinous carcinoma

iii. Signet ring carcinoma

iv. Small cell carcinoma

v. Adenosquamous carcinoma

vi. Medullary carcinoma

vii. Squamous carcinoma

viii. Undifferentiated carcinoma

4. Carcinoid tumour

5. Mixed carcinoid-adenocarcinoma

6. Non-epithelial tumor

i. Lipoma

ii. Leiomyoma

iii. Gastrointestinal stromal tumor

10
 iv. Leiomyosarcoma

v. Angiosarcoma

vi. Malignant lymphoma

vii. Malignant melanoma

7. Secondary tumors

2.3.2 Colorectal carcinoma

Colorectal cancer (CRC) is the third most common malignant neoplasm worldwide and the

third most common malignancy and one of the leading causes of cancer death in women

and men in the United States. The lifetime risk of CRC among women and men at average

reaches 6% or one in 18 (Kim, 2010). The peak incidence of colorectal carcinoma is

between 60 and 79 years. Fewer than 20% cases occur before the age of 50 years (Turner,

2010).

2.3.2.1 Incidence of colorectal carcinoma

Colorectal cancer (CRC) is the third most common cancer and the second leading cause of

cancer related death in the United States (Aljebreen, 2007). In the USA and Western

Europe, colorectal cancer constitutes approximately 10% of all malignancies (Leon et al,

2004). It represents 9.4% of all incident of malignancy in men and 10.1% in women

globally. CRC is not equally common through out the world. The incidence varies in

various geographical part of the world. High incidence seen in Europe, America,

Australia, New Zealand with incidence ranges from 25.3/100000 to 45.8/100000

11
population (Boyle and Langman, 2000). Compare to developed countries, lower incidence

of colorectal cancer is seen in developing countries like Africa and Asia, including central

and south Asia and India. Incidence ranges from 2/100000 to 8/100000 population

(Notani, 2001). The epidemiological study of colorectal cancer done by Keating et al

(2003) among 2272 pathological report in New Zealand showed the incidence of colorectal

carcinoma was 46.8/1000000 for men and 43.5/100000 for women. Ayyub et al (2000), in

a large hospital of Saudi Arabia has shown the incidence were 4.6/100000 in male and

4.4/100000 in female population.

2.3.2.2 Age and Sex distribution of Colorectal carcinoma

Colorectal cancer is a disease of advanced aged population. Greater number of CRC are

found in the six decades with a median age of about 62 years (Riddell, 2003). Keating et

al (2003) done a study on the epidemiology of colon cancer in New Zealand. They found

in their study that the mean age of women was 69.4 years and mean age of men was 68.5

years. Turner (2010) have shown nearly the same result in their study as the peak

incidence of colorectal carcinoma was in between ages 60-70 and only 20% cases occur

before the age of 50 years. Ayyub et al (2002) studied on clinicopathological trends in

colorectal cancer on 160 cases in Saudi Arabia. They showed that the mean age was

56.3±14.98 years.

Colorectal cancer is extremely rare in paediatric age group. Afroza et al (2007) reported a

11 year old Bangladeshi boy with primary mucinous adenocarcinoma in the rectum.

12
Colorectal cancer affects men and women almost equally (Boyle and Langman 2000).

Some study have shown male predominance (riddell et al, 2003). Keating et al (2003)

showed in their study that the male and female ratio were almost equal. Fireman et al

(2005) showed female predominance in their study as among 624 cases, 271 were male

and 353 were female.

2.3.2.3 Clinical features of Colorectal carcinoma

Patients with colorectal cancer have usually presented with abdominal pain, alteration of

bowel habit, loss of weight, vomiting, frequently with colic, anorexia, bleeding per rectum,

lump, indigestion and acute on chronic obstruction (Hamilton, 2000). Ayyub et al (2002)

mentioned on their study that most of the patient presented with the symptom of abdominal

pain, altered bowel habit, per rectal bleeding weight loss, intestinal obstruction and

constipation. Duration of symptom varied from one month to 2 years.

2.3.2.4 Investigations in CRC

The laboratory investigation include the blood and biochemical parameters like tumour

markers (CEA, CA-19-9), and also the visualization of the lesion through endoscope.

Ayyub et al (2002) studied clinicopathological trends in colorectal carcinoma showed

anaemia (Hb% lOgm/dl or less) in 55% of the patients. The distribution of anaemia varied

significantly in various sub site of primary colorectal carcinoma. About 30% patients had

anaemia with left sided tumour and 70% of patients in right side colon cancer. They also

measure CEA level in serum of colorectal cancer patients. CEA level varied from 1 to

13
850.2 with a mean of 48.62. Leon et al (2004) studied trend of incidence, subtype

distribution and staging of colorectal carcinoma in the 15 years experience of a specialized

cancer registry, showed more then 50% colorectal carcinoma can be detected by

sigmoidoscopy and rest needed pancolonscopy for the choice of screening for individual at

the risk of colorectal carcinoma.

2.3.2.5 Sub site distribution

The right colon is considered from caecum to spleenic flexure; left colon includes

descending colon, sigmoid colon and rectum (Gomez et al, 2004). The sub site-specific

variation is seen in the development of colorectal carcinoma. Two-third of colorectal

carcinoma occur in the left colon and the rest in the right colon (Leon et al, 2004). Ayyub

et al (2002) showed majority of the colorectal carcinoma 68.2% in the rectum and sigmoid

colon; and 22.5% colorectal carcinoma originating from ascending and transverse colon.

Gomez et al (2004) showed total 31% of the cancer were in right side. The study also

cited the differences in the anatomical distribution between sex, in female 48% were in

right and 41% were left sided colon cancer in comparison to 59% cases with left colon

cancer and 52% cases with right colon cancer in male.

2.3.2.6 Histological subtypes of colorectal carcinoma

Most of the colon cancer is adenocarcinoma. In the study of Keating et al, (2003)

showed 94.7% cases of adenocarcinoma, 4.2% were mucinous adenocarcinoma, 0.4%

were signet ring cell carcinoma and less then 1% cases were adenosquamous, squamous

and miscellaneous carcinoma. Hossain (2007) found 74% of cases were usual

14
adenocarcinoma, 23% cases mucinous adenocarcinoma, 1.5% cases each signet ring cell

carcinoma and small cell carcinoma.

2.3.3 Pathogenesis

Environmental and genetic factors both contribute to colorectal cancer development.

Genetic susceptibility may be the soil on which subsequent environmental factors act

(Riddell, 2003).

2.3.3.1 Environmental factors:

Epidemiologic studies have indicated that meat consumption, smoking, sedentary lifestyle

and alcohol consumption are risk factors for colorectal carcinoma. Inverse associations

include vegetable consumption, prolonged use of non-steroidal anti inflammatory drugs,

oestrogen replacement therapy and physical activity (Hamilton, 2000).

2.3.3.2 Molecular carcinogenesis:

The combination of molecular events that lead to colonic adenocarcinoma is heterogenous

and includes genetic and epigenetic abnormalities (Turner, 2010). The genetic and

epigenetic pathways involved are:

1. APC/β-catenin pathway which is associated with adenoma-carcinoma sequence.

2. Loss of p53.

3. K-ras mutation.

4. Delated in colon cancer gene (DCC gene).

15
 5. Telomerase activity.

6. Microsatellite instability pathway which is associated with DNA mismatch repair.

7. DNA methylation abnormalities (Hypomethylation and Hypermethylation).

2.3.3.3 DNA methylation in cancer

Cancer results from the accumulation of mutation in the genes. In addition to genetic

mutation, this epigenetic change has been included as an alternative mechanism to cancer

development. Epigenetics refers to change in the pattern of gene expression by

mechanisms other than alterations in the primary nucleotide sequence of a gene(Herman,

2003). DNA methylation is the enzymatic attachment of methyl group to the 5th carbon of

the cytosine base (Samarakoon, 2010). Methylation usually occurs in the CpG islands, a

cytosine guanosine rich region in the DNA. In humans, DNA methylation is carried out by

a group of enzymes called DNA methyltransferases. The letter “p” here signifies that the C

and G are connected by a phosphodiester bond. Methylation patterns in tumor cells are

significantly different from those in normal cells (Herman, 2003).

Silencing or inactivation of tumour suppressor gene is seen in cancers mostly due to

hypermethylation. Hypermethylation profiling over more than 15 tumor types (colon,

stomach, pancreas, liver, kidney, lung, breast, ovary, endometrium, kidney, bladder, brain,

and leukemia and lymphomas) has shown that all the metabolic pathways are affected by

promoter hypermethylation-associated silencing. Hypomethylation leads to the

inappropriate and increased levels of gene expression in tumors such as oncogene

activation (Samarakoon, 2010).

16
2.3.3.4 DNA methylation in colorectal cancer:

Colorectal cancer (CRC) arises as a consequence of the accumulation of genetic and

epigenetic alterations in colonic epithelial cells during neoplastic transformation (Kim,

2010).

Hypermethylation in the CpG islands of the tumor suppressor gene promoters can lead to a

complete block of transcription and inactivates the tumor suppressor genes. DNA

hypomethylation could also drive neoplastic progression and transformation. It may make

chromosomes more susceptible to breakage and oncogene activation (Kim, 2010).

2.3.3.4.1 Genes inactivated by promotor hypermethylation in colorectal cancer

Hypermethylated genes are associated with colorectal neoplasia includes the tumour

suppressor, mismatch-repair and cell-cycle regulatory genes (Wong, 2007). Table 2.3.1

shows functions and frequency of different genes involved by hypermethylation in

colorectal cancer.

17
Table: 2.3.1 Genes silenced by hypermethylation in colorectal cancer (Wong, 2007)

Genes Function Frequency (%)

APC Signal transduction, beta-catenin regulation 10–50

CDH13 Cell signalling (cell recognition and adhesion) 30–40

CDKN2A Cell-cycle regulation 15–30

CHFR Mitotic stress checkpoint 30–40

HIC1 Regulation of DNA damage responses 80

HPP1 Transmembrane transforming growth factor 80

(TGF)- β antagonist

LKB1 Cell signalling, cell polarity 5–10

MGMT Repair of DNA guanosine methyl adduct 30–40

MLH1 Mismatch repair 10–20

Genes Function

p14ARF Cell-cycle regulation 20–30

RASSF1A DNA repair, cell-cycle regulation >50

SOCS1 Cell signaling 5–10

THBS1 Angiogenesis 10–20

TIMP3 Matrix remodelling, tissue invasion 10-30

18
2.3.3.4.2 DNA hypomethylation in colorectal cancer

Hypomethylation may make chromosomes more susceptible to breakage and therefore lead

directly to genomic instability. DNA hypomethylation can also lead to the activation of

oncogenes such as S100A4 metastasis-associated gene in colorectal carcinoma and can

lead to loss of imprinting (LOI) which can drive cellular proliferation in cancer. The

clearest example of this phenomenon is hypomethylation of IGF2/H19 seen in about 40%

of colorectal cancer tissue (Wong, 2007). Hypomethylation of CDH3 promotor, CD133,

LINE-1 are also seen in colorectal cancer (Kim, 2010).

2.3.4 Gross morphology of colorectal carcinoma

The tumor of the proximal colon tends to grow as polypoid and exophytic mass and in the

distal colon the mass grow as annular and encircling manner. The gross morphology of the

lesions are fungating. annular, tubular and ulcerated. The fungating growth is cauliflower

like growth, some time ulceration is seen in the tip of the fungating growth. The annular

growth encircling whole of the circumference and produce obstructive feature clinically.

The tubular growth is flat (linitis plastica). The ulcerated growth form an ulcer in the gut

wall. The superficial spreading lesion is difficult to identify grossly (Turner, 2010). The

cut surface of the gut wall shows grayish white tissue replacing the bowel wall. Highly

mucinous tumour have a gelatinous glaring appearance, and layers of mucus may separate

the layers of the bowel wall (Turner, 2010).

19
2.3.5 Microscopic features of colorectal carcinoma

2.3.5.1 Adenocarcinoma: About eighty percent of colorectal carcinomas are histologically

characterized by good gland formation and varying degrees of differentiation, from well to

moderately differentiated. Well differentiated carcinoma have >95% glandular structure,

moderately differentiated carcinoma have 50-95% glands, poorly differentiated carcinoma

have 5-50% glandular structure and undifferentiated carcinoma have <5% glandular

structures (Hamilton 2000)

2.3.5.2 Mucinous adenocarcinoma: Mucinous and Signet-ring adenocarcinoma account

for 10% of colorectal cancer. This designation is used if >50% of the lesion is composed of

mucin. This variant is characterized by pools of extracellular mucin that can contain

malignant epithelium as acinar structure, strips of cells or single cells (Hamilton,2000)

2.3.5.3 Signet ring cell adenocarcinoma: Signet-ring and mucinous cell carcinoma

account for approximately 10% of colorectal cancer. This variant of adenocarcinoma is

definied by the presence of >50% of the tumour cells with prominant intracytoplasmic

mucin (Hamilton, 2000)

2.3.5.4 Small cell carcinoma: A rare variant is small cell carcinoma which composes <1%

of colorectal cancer (Cooper, 1999)

2.3.5.5 Adenosquamous carcinoma: These unusual tumours show features of both

squamous carcinoma and adenocarcinoma either as separate areas within the tumour or

admixed pattern (Hamilton, 2000).

20
2.3.5.6 Squmous cell carcinoma: Squamous cell carcinoma of the colon is very rare. To

make diagnosis of the squamous cell carcinoma in the colon, there must be no other site of

the squamous cell carcinoma in the body and no involvement of cloacogenic squamous

lined mucosa (Rosai, 2004).

2.3.5.7 Medullary carcinoma: This rare variant is characterized by the sheets of

malignant cells with vesicular nuclei, prominent nucleoli and abundant pink cytoplasm

exhibiting prominent infiltration by intraepithelial lymphocytes (Hamilton, 2000).

2.3.5.8 Undifferentiated cancer: Undifferentiated cancer is uncommon, accounting for

approximately 1 % of the colorectal carcinoma. They are malignant epithelial neoplasm

that have no glandular structure or other features to indicate definite differentiation

(Hamilton, 2000).

2.3.6 Staging of colorectal cancer:

The major role of proper staging of colorectal carcinoma is to provide information to

physician regarding patient's prognosis and the need for adjuvant therapy. For many years,

pathologists used the classic Dukes' classification(1932), Astler-Coller classification(1954)

and the TNM classification(Turner, 2010). Table 2.3.2 shows TNM classification and

Table 2.3.3 shows TNM staging of colorectal carcinoma (Turner, 2010).

21
Table:2.3.2 TNM Classification of Colorectal Carcinoma by American Joint

Committee on Cancer (AJCC) (Turner, 2010).

TUMOUR
Tis In situ dysplasia or intramucosal carcinoma
T1 Tumor invades submucosa
T2 Tumor invades into, but not through, muscularis propria
T3 Tumor invades through muscularis propria
T3a Invasion <0.1 cm beyond muscularis propria
T3b Invasion 0.1 to 0.5 cm beyond muscularis propria
T3c Invasion >0.5 to 1.5 cm beyond muscularis propria
T3d Invasion >1.5 cm beyond muscularis propria
T4 Tumor invades adjacent organs or visceral peritoneum
T4a Invasion into other organs or structures
T4b Invasion into visceral peritoneum
REGIONAL LYMPH NODES
NX Lymph nodes cannot be assessed
N0 No regional lymph node metastasis
N1 Metastasis in one to three regional lymph nodes
N2 Metastasis in four or more regional lymph nodes
DISTANT METASTASIS
MX Distant metastasis cannot be assessed
M0 No distant metastasis
M1 Distant metastasis or seeding of abdominal organs

22
Table :2.3.3 TNM staging for colorectal carcinoma (Turner, 2010)

Stage-I Tl N0 M0 or T2, N0, M0

Stage-II T3 N0 M0 or T4, N0, M0

Stage-III T any Nl M0 or T any,N2,N3, M0

Stage-IV T any N any Ml

2.3.7 Grading of colorectal cancer:

Grading is done on the basis of differentiation of tumour cells. Differentiation refers to the

extent to which neoplastic cells resemble comparable normal cells (Kumar et al, 2010).

The formation of glands (acini) is the basis for grading system in CRC. Well differentiated

carcinoma have >95% glandular structure, moderately differentiated carcinoma have 50-

95% glands, poorly differentiated carcinoma have 5-50% glandular structure and

undifferentiated carcinoma have <5% glandular structures. Mucinous carcinoma and signet

ring cell carcinoma by definition are poorly differentiated cancer (Hamilton, 2000).

2.3.8 Prognosis of colorectal cancer: (Rosai, 2004)

There are several clinical and pathological parameters which determines prognosis. Most

important is the tumour stage. 5 years survival rate in TNM stage I disease is 90 to 98

percent. About 80 percent 5 year survival in TNM stage II and less than 10 percent in

stage III with many lymph nodes involvolvement and in stage IV (Riddell, 2003). Some

factors related to prognosis are given below:

23
Age: Tumour occuring in very young and old patient have poor prognosis.

Sex: The prognosis is better in female than in male.

CEA serum level: >5.0 ng/ml have been shown to have adverse prognosis.

Tumour location: Tumour located in the sigmoid colon and rectum have the worse

prognosis.

Local extent: Lymph node metastasis have worse prognosis than local invasive.

Tumour size: There is a little relationship between the size of the tumour and prognosis.

Tumour edge: The non-polypoid edge of the tumour have the worse prognosis than

polypoid cancer.

Obstruction: Obstruction is an indicator of worse prognosis.

Perforation: It is due to invasion to the wall shows worse prognosis.

Tumour margin and inflammatory reaction: Pushing margin with inflammatory

response have better prognosis.

Vascular invasion: When venous invasion is present, the 5-years survival decreases

markedly.

Perineural invasion: Reflects the advanced stage of the disease.

Surgical margins: Presence of tumour in surgical margin is the single most critical factor

for recurrence of carcinoma.

24
Tumour thickness: Central depressed area of the tumour is correlated with the presence of

lymph node and liver metastasis and also with bad prognosis.

Microscopic tumour type: Mucinous carcinoma, Signet ring cell carcinoma and

anaplastic carcinoma have worse prognosis.

Acinar morphology: Microacinar pattern of growth is associated with poor prognosis.

Tumour angiogenesis: Tumour angiogenesis predict recurrence and is associated with

decreased survival in colorectal carcinoma.

Mucin-related antigen: Colorectal carcinoma that express the mucin-associated antigens

sailyl-Tn and sailyl-Lewis antigen have more aggressive clinical course.

HLA-DR expression: Patient with tumour having strong HLA-DR expression show better

prognosis.

HCG expression: Effect as adverse indicator of prognosis.

Bcl-2 protein expression: Associated with better prognosis.

DNA ploidy: The DNA aneuploidy and risk of recurrence of survival.

Allelic loss of chromosome 18q: Have strongly negative prognostic significance.

25
3. MATERIAL AND METHOD

3.1 Place and period of study

a. The clinical and histopathological part of this study was carried out at the

Department of Pathology, Bangabandhu Sheikh Mujib Medical University (BSMMU),

Dhaka during the period of November 2009 to July 2010. A total of 50 cases of

colorectal adenocarcinoma from all ages and both sexes were included in the study.

b. The DNA methylation part of this study performed on 48 samples (24 tumour tissue

and 24 corresponding healthy tissue) was carried out in collaboration at the Department

of Health Studies, Biological Sciences Division (BSD) of The University of Chicago

Medical Center, Chicago , IL , USA (Appendix-I).

3.2 Collection of the case

A total of 50 cases of colorectal carcinoma were collected for study. For selection,

inclusion criteria described below was followed. Cases were collected from

Bangabandhu Sheikh Mujib Medical University (BSMMU) and private hospitals and

clinics in Dhaka city.

3.3 Inclusion criteria

a. Histologically confirmed cases of adenocarcinoma of the colon.

b. Cases with complete clinical information.

c. Availability of fresh unfixed colorectal cancer specimen.

d. Availability of normal uninvolved colonic tissue away from the tumour.

26
3.4 Exclusion criteria

a. Clinically suspected colorectal carcinoma subsequently proved to be non-malignant

lesions after histological examination.

b. Non Hodgkin lymphoma and other non epithelial tumors of the colon.

c. Cases without clinical data.

d. Formalin fixation or delay in receiving specimen.

3.5 Collection of clinical information

Detailed clinical information was obtained by taking history and recorded in clinical

proforma (Appendix-II). Filling up of the clinical proforma was performed in all cases

either from patient's attendants statement and/or from patients file.

3.6 Collection of the tissue block for histopathological examination

Fresh unfixed specimen were obtained after surgical resection. Large plastic container

with lid were used for collection. Containers were properly labeled with identification

number, name of the patient, address, type of specimen and date and place of operation.

3.6.1 Fixation of sample for Histopathological examination:

After collecting blocks for methylation study (describeb next), the unfixed specimen

were transferred to 10% formalin for overnight fixation. The next morning the

specimen were examined during gross cut up and tissue blocks were taken according to

the following procedure (Rosai, 2004)

27
a. Recording of gross features of the specimen:

i. Specimen type, length of the specimen and amount of mesentery.

ii. Tumour characteristics: Tumor size including thickness, extent around bowel,

tumour configaration(exophytic, infiltrative, ulcerating), presence of necrosis or

hemorrhage, tumour number, extent through bowel wall, serosal involvement,

invasion of adjacent organs and distance of the tumour from the proximal and

distal resection margins.

iii. Number of lymph nodes found and size of the largest node.

b. Sections of tissue blocks:

i. Three blocks from tumour including the serous layer.

ii. One block from tumour with surrounding normal mucosa.

iii. Proximal line of resection.

iv. Distal line of resection

v. Appendix, if included in the specimen.

vi. All the lymph nodes.

vii. Polyp/ Suspicious areas.

3.6.2 Tissue processing for routine histopathological examination:

All the tissue blocks were submitted for routine processing and paraffin

Embedding (appendix-III). The tissue processing and staining was performed following

the standard protocol (appendix-IV). Microscopic examination of routine paraffin

sections stained with haematoxyline and eosin staining method were carried out during

which relevent points were recorded.

28
3.6.3 Routine microscopic examination

Routinely stained sections examined first under low power and then under high power

magnification. The following points were noted during examination:

i. Histologic type of growth.

- Adenocarcinoma(NOS).

- Mucinous.

- Signet ring cell.

- Small cell.

ii. Histologic grade (applicable in adenocarcinoma (NOS) type).

- Well differentiated.

- Moderately differentiated.

- Poorly differentiated.

iii. Extent of invasion.

-Submucosa.

-Muscle coat.

-Serosa.

-Pericolic fat.

iv. Proximal and distal resection margins.

v. Circumferential ( radial) margin.

vi. Lymphatic (small vessel) invasion.

vii. Venous (large vessel)invasion.

viii. Perineural invasion.

ix. Tumor border configaration

- Pushing.

29
 - Infiltrative.

x. Lymphocytic response

Grade 0 - <10 lymphocytes per high-power field (HPF).

Grade 1- Tumour infiltrating lymphocytes (TIL) level of 10–50 HPF.

Grade 2 -Tumour infiltrating lymphocytes (TIL) level 50-100 per HPF.

Grade 3- Tumour infiltrating lymphocytes (TIL) level >100 per HPF.

xi. Extracellular mucin content

-No extracellular mucin.

-1-49% extracellular (Tumour cell) mucin.

->50% extracellular (Tumour cell) mucin.

xii. Signet ring cells (Ogino et al, 2008)

- No signet ring cell.

- 1-49% of tumour cell.

- >50% of tumour cells.

xiii. Lymph node status- number of lymph nodes involved.

The tumour was classified according to the World Health Organization classification

(Hamilton, 2000). The tumour staging was done using TNM classification (Turner,

2010).

30
3.7 DNA methylation study

3.7.1 Tissue sample collection for DNA methylation study

Paired unfixed tissue samples, one each from tumour and healthy mucosa were taken in

DNAse free 1.5 ml eppendorf tube (Ambion Catalogue # AM12450, Ambion

company, USA). Dimention of the tissue blocks were 4 to 5 mm. These eppendorf

were kept in -20°c and sent to the department of Health Studies, Biological Sciences

Division of The University of Chicago Medical Center, Chicago , IL , USA in dry ice.

DNA extraction and methylation work done at the Department of health studies (BSD),

The University of Chicago, Chicago, IL, USA under aggrement between the

department of pathology, BSMMU and Department of health studies (BSD), The

University of Chicago, Chicago, IL, USA.

3.7.2 DNA extraction from colonic tissue:

DNA extraction were done using Gentra Purgene Tissue kit (Qiagen Catalogue #

158622, Qiagen company, USA) (Appendix-V).

3.7.3 DNA Methylation by Infinium Assay:

Principle:

The DNA sample is first treated with sodium bisulphite which converts the

unmethylated cytosine bases to uracil. The methylated cytosines on the other hand,

remains unaffected. DNA methylation in this study was based on Infinium assay

31
system using The Human Methylation 27 BeadChip (Inf HumanMeth27, RevB BC

Kit, Catalogue # WG-311-1201 introduced by Illumina Inc, USA) ( Appendix-VI).

3.8 Statistical analysis:

Histopathological portion: All the necessary and relevant data were recorded

methodically and meticulously as far as possible in the clinical proforma. Relevant data

were analyzed by standard statistical method.

DNA methylation portion: The raw image data was processed by Bead Studio
software provided by Illumina inc. USA, to generate the average beta values. Average
beta is the methylation status of a gene locus. It is calculated by intensity of the
methylated sequence(x)/ [intensity of the methylated sequence (x)+ intensity of the
unethylated sequence (y) ]. The ratio is between 0 to 1, where 0= extremely
hypomethylated and 1= extremely hypermethylated. Unpaired t-test was done to
compare the average beta between the tumour & normal tissue. DiffScore was
calculated from the p-value of the un-paired t-test. Delta Beta was calculated as beta of
tumor tissue minus beta of normal tissue.

32
4.0 OBSERVATIONS AND RESULTS

This cross sectional study was undertaken to see the morphologic pattern and DNA

methylation changes in colorectal carcinoma in a group of Bangladeshi patients. Clinical

data were recorded. DNA methylation of non tumorous areas of colon of the same patient

was also determined.

A total of 50 cases of colorectal cancer with their corresponding normal mucosa were

included in the study. As molecular genetic laboratory facility is not available at present in

BSMMU, the DNA methylation part was done in the molecular biology laboratory of the

Department of Health Studies (BSD) of the University of Chicago,Chicago, IL, USA with

aggrement between two department (Appendix-I).

4.1 Age and sex distribution

4.1.1: Age distribution of the colorectal carcinoma cases

The age range was from 19 years to 84 years with a mean age of 46.6 ± 14.8 years. The

patients were divided into 8 groups on the basis of decades. Out of 50 cases maximum

number 12 (24%) of patients belonged to the age group 50-59 years, followed by 11(22%)

cases each in 30-39 years and 40-49 years groups, 6(12%) cases in 20-29 years group, 5

(10%) cases in 60-69 years group, 3 (6%) in 70-79 years group and 1 (2%) case each in

10-19 years and 80-89 year groups. Figure 4.1 shows age distribution of colonic cancer

patient. From this figure, peak age of CRC appears to be 30 to 60 years in our population.

33
Figure 4.1 : Bar diagram showing age distribution of colorectal cancer.
Number of cases

Age in decades

4.1.2 Male to female ratio

In this study out of total 50 cases, 29 (58%) cases were male and 21 (42%) cases were

female with male to female ratio of 1.4:1, male patients were higher than female patients.

Table 4.1 shows male and female patients with CRC in different age groups.

34
Table 4.1 : Sex distribution in relation to age group.

Age in years Total number of No. of Male No. of Female

cases
10-19 1(2%) 1(2%) -

20-29 6(12%) 4(8%) 2(4%)

30-39 11(22%) 5(10%) 6(12%)

40-49 11(22%) 6(12%) 5(10%)

50-59 12(24%) 7(14%) 5(10%)

60-69 5(12%) 3(6%) 2(4%)

70-79 3(6%) 3(6%) -

80-89 1(2%) - 1(2%)

Total=8 50(100%) 29(58%) 21(42%)

4.2 Clinical presentation at the time of attending to doctor

At the time of first consultation, majority of the patients 20(40%) had per-rectal bleeding,

8(16%) had abdominal pain, 7(14%) had altered bowel habit, 6(12%) had generalised

weakness, anorexia and pallor, 6(12%) had combined per rectal bleeding and abdominal

pain and 3(8%) had palpable abdominal mass. The duration of symptoms ranged from one

month to 24 months. Table 4.2 shows clinical presentation of colorectal cancer cases.

35
Table 4.2: Symptoms with tumour location of colorectal cancer cases.

Symptoms Cases Cases in different

location
N=50
Right Left colon
colon
Per rectal bleeding 20 (40%) 1 19

Abdominal pain 8 (16%) 6 2

Altered bowel habit 7 (14%) 4 3

Generalized weakness, anorexia and pallor 6(12%) 4 2

Per rectal bleeding and abdominal pain 6(12%) 4 2

Palpable abdominal mass 3(6%) 2 1

4.3 Haemoglobin concentration in blood

Haemoglobin levels were recorded in 50 cases. It ranged from 8.3 to 14.6 gm/dl.

Haemoglobin level <10gm/dl considered anaemia in both sex (Ayyub,2002).

Haemoglobin was below 10 gm/dl in 13 (26%) cases and in 37(74%) cases it was more

than 10 gm/dl. In 13 cases of below 10gm/dl, 8(62%) were female and 5(38%) were male.

In 37 cases of hemoglobin level more than 10gm/dl, 24(65%) cases were male patient and

13(35%) were female. Table 4.3 shows haemoglobin concentration in blood of 50

colorectal cancer patient.

36
Table 4.3: Haemoglobin concentration in 50 cases of colorectal cancer patients

Hb (gm/dl) N=50 Male Female

<10 13(26%) 5(38%) 8(62%)

>10 37(74%) 24(65%) 13(35%)

4.4 CEA (Carcinoembryonic antigen) level in 50 colorectal cancer cases

CEA levels were recorded in 50 cases. It ranged from 0.31 to 672 ng/ml. CEA <5ng/ml

considered normal (Aljebreen, 2007). In 25 (50%) cases CEA was below 5 ng/ml and in

25 (50%) it was more than 5 ng/ml (Table 4.3).

Table 4.4: CEA (Carcinoembryonic antigen) level in 50 colorectal cancer cases

CEA (ng/ml) N=50 Male Female

<5 25 15(30%) 10(20%)

>5 25 19(38%) 6(12%)

37
4.5 : Sub site distribution of colonic cancer

Distribution of cancer among 50 cases according to the site of the colon affected were as

follows: 33 (66%) case were in rectum, 6 (12%) were in the ascending colon , 3 (6%) in

hepatic flexure of colon, 2 (4%) each in caecum, descending colon, sigmoid colon and in

the transverse colon (Table 4.5). Of the total 50 cases 37 (74%) were in the left colon

(splenic flexure to rectum) and 13 (26%) were in the right colon (caecum to upto splenic

flexure). Distribution of colon cancer in 29 male cases were : 17 (34%) in rectum, 4 (8%)

in ascending colon, 2 (4%) each in transverse colon, hepatic flexure of colon and caecum

and one(2%) each in sigmoid colon and descending colon. In 21 female cases, 16

(32%)cases were in rectum, 2 (4%) cases in ascending colon, one(2%) case each in

sigmoid colon, descending colon and hepatic flexure of colon.

Table 4.5: Site distribution of colorectal cancer from distal to proximal

Variable N=50 Male Female

Rectum 33 17(34%) 16(32%)

Sigmoid colon 2 1(2%) 1(2%)

Descending colon 2 1(2%) 1(2%)

Transverse colon 2 2(4%) -

Hepatic flexure of colon 3 2(4%) 1(2%)

Ascending colon 6 4(8%) 2(4%)

Caecum 2 2(4%) -

38
4.6 Size and shape of the lesion

Morphology of the growth were recorded by examining the surgically resected specimen.

Maximum diameter of the growth ranged form 3 cm to 10 cm. Grossly 25 (50%) tumours

were ulcerated growth, 22 (44%) were polypoid or exophytic growth and 3(6%) were

infiltrative or annular growth. Figure 4.2 shows growth pattern of colorectal cancer

patients.

Figure 4.2 Growth pattern of colorectal cancer cases.

3%

Ulcerating
Exophytic
44% 50%
Infiltrative

4.7 Histological features of colorectal cancer cases

In 50 cases of colorectal adenocarcinoma, 44(88%) were adenocarcinoma (NOS) and

6(12%) were mucinous adenocarcinoma. Of 44 cases of adenocarcinoma (NOS), 23 (46%)

were male and 21 (42%) cases were female. All 6 (12%) cases of mucinous

adenocarcinoma were male (Table 4.6).

39
Table 4.6: Histological diagnosis of 50 cases of colorectal cancer

Histological type N=50 Male Female

Adenocarcinoma 44 (88%) 23 21

Mucinous adenocarcinoma 6(12%) 6 -

4.7.1: Histological grading of colorectal cancer

Of 50 colorectal cancer cases, 39(78%) cases were moderately differentiated and 11(22%)

cases were poorly differentiated. In this present study no well differentiated adeno

carcinoma was observed. All the mucinous carcinoma were included in the poorly

differentiated group (Hamilton, 2000). Male predominance was observed in both

moderately differentiated and poorly differentiated cases. Among the 39 moderately

differentiated cases, 22 cases were seen in male and 17 cases in female and of 11 poorly

differentiated cases, 7 cases were seen in male and 4 in female.

Of thirty nine moderately differentiated cases, 10(20%) cases originated in right colon and

29 (58%) cases in the left colon. Of 11 poorly differentiated carcinomas, 4(8%) cases were

in the right colon and 7(14%) cases were in the left colon. Table 4.7 shows histological

grade of colorectal cancer in 50 cases with location and sex distribution.

40
Table 4.7: Histological grading of colorectal cancer cases with sex and location:

Tumour differentiation Number Male Female Right Left

colon colon

Moderately differentiated 39(78%) 22(44%) 17(34%) 10(20%) 29(58%)

Poorly differentiated 11(22%) 7(14%) 4(8%) 4(8%) 7(14%)

4.7.2 Clinical stage of cases of colorectal cancer

Staging of colonic cancer were done in all the cases. It was done on the basis of TNM

system (Turner, 2010). The maximum number of cases 22 (44%) were in stage III, 14

(28%) cases were in stage II and 13 (26%) cases in stage I and 1 (2%) case in stage IV.

Out of 22 stage III cases, 11(22%) cases were male and 11(22%) cases were female. In

14(28%) cases of stage II, 9(18%) cases were male and 5(10%) cases were female. In stage

I there was a total 13(26%) cases, 8(16%) cases were male and 5(10%) cases were female.

One (2%) stage IV case were male patient. Table 4.8 shows TNM stage in 50 colorectal

cancer cases with sex distribution and location.

41
Table 4.8: TNM (Tumour, Node, Metastasis) stage with sex and location

TNM stage Number of Male Female Right Left

cases colon colon

Stage I 13(26%) 8(16%) 5(10%) 2(4%) 11(22%)

Stage II 14(28% 9(18%) 5(10%) 6(12%) 8(16%)

Stage III 22(44%) 11(22%) 11(22%) 8(16%) 14(28%)

Stage IV 1(2%) 1(2%) - 1(2%) -

TNM=Tumour Node Metastasis

4.7.3 Lymph node metastasis in 48 cases of colorectal cancer

Out of 50 resected specimen, 49 cases contained lymph nodes. Lymph nodes number

ranged from four to twenty five. Histologically lymph node metastasis were present in

23(47%) cases. Among the 23 nodal metastasis cases, 17(74%) cases metastasized from

moderately differentiated primary tumour and 6(26%) cases from poorly differentiated

primary carcinoma.

42
4.7.4 : Tumour infiltrating lymphocytes (TIL) in colorectal cancer patients

Tumour infiltrating lymphocytes (TIL) were observed in all the 50 colorectal cancer cases

and catogorized into four groups, nil or absent, grade- 1, grade-2 and grade- 3. 20(40%)

cases were in grade -1, 18(36%) cases were in grade- 2, 4(8%) cases were in grade- 3 and

the rest 8(16%) cases has no tumour infiltrating lymphocytes. Figure 4.3 show percentage

of tumour infiltrating lymphocytes in 50 colorectal cancer cases.

Figure 4.3: Various grade of tumour infiltrating lymphocytes in 50 cases.

8% 16%

Nil
Grade-1
36%
Grade-2
40% Grade-3

43
4.7.5: Lymphovascular invasion in colorectal cancer patients

Lymphovascular invasion was present in 16(32%) cases and absent in 34 (68%) cases

Figure 4.4 lymphovascular invasion in CRC cases.

Figure 4.4: Lymphovascular invasion in 50 colorectal cancer cases

32%
Present
68% Absent

4.7.6: Extracellular mucin in colorectal cancer patients

Out of 50 colorectal cancer patients percentage of extracellular mucin in the tumour tissue

were recorded. 1-49% areas of extracellular mucin was present in 18(36%) cases, >50%

extracellular mucin was present in 6(12%) cases and no extracellular mucin was found in

26 (52%) cases. Figure 4.5 shows distribution of extracellular mucin in colorectal cancer

cases.

44
Figure 4.5 Extracellular mucin in colorectal cancer patient.

>50% Extracellular mucin

12%

1-49%
52% No Extracellular
Extracellular
36% mucin
mucin

4.7.8: Signet ring cells in 50 colorectal cancer cases

Percentage of signet ring cells in the tumour tissue were recorded. 1-49% of the tumour

cells were signet ring cells in 17(34%) cases and no signet ring cells in 33 (66%) cases.

No cases having more than 50% signet ring cells was present in this series. Table-4.9

shows percentage of signet ring cells in 50 colorectal cancer cases

Table 4.9: Signet ring cells in colorectal cancer patients

Percentage of signet ring cells Number of cases (%)

Nill (%) 33(66%)

1-49% 17(34%)

>50% -

45
4.7.9: Circumferential margin, perinural invasion and tumour border

configuration in colorectal cancer cases

Histologically out of 50 colorectal cancer cases circumferential margin were involved in 12

(24%) cases. Of twelve involved cases 8 (67%) were stage in III and 4 (33%) were in

stage II disease. In uninvolved 38 cases, 14 (37%) cases were in stage III, 13 (34%) cases

were in stage I, 10 (26%) were in stage II and one (3%) case in stage IV.

Out of 50 colorectal cancer cases, perineural invasion was present in 15(30%) cases. Out

of these 15 cases 10 (66%) cases had stage III disease, 3 (20%) cases had stage II disease,

one (7%) case each in stage I and stage IV. Perinural invasion was absent in 35(70%)

cases. In which 12(24%) cases each were in stage I and stage III and 11(22%) cases in

stage I disease.

In 50 cases, tumour border configurations were examined microscopically. Infiltrative

pattern were seen in 26 (52%) cases and pushing pattern in 24(48%) cases. In 26(52%)

infiltrative pattern cases, 15 (58%) cases were in stage III, 6 (23%) cases in stage II, 4

(15%) cases in stage I and one (4%) case in stage IV. In 24 cases having pushing border, 9

(37%) cases were in stage I, 8 (33%) cases in stage II, 7 (33%) cases in stage IV. Table

4.10 shows Circumferential margin involvement, Perineural invasion and Tumour border

configuration with TNM stage

46
Table 4.10: Circumferential margin involvement, Perineural invasion and

Tumour border configuration with TNM stage.

TNM Circumferential margin Perinural invasion Tumour border

stage configuration
Involved Not Present Absent Infiltrative Pushing
involved
N=12 N=15 N=35 N=26 N=24
N=38

Stage-1 - 13(34%) 1(7%) 12(34%) 4(15%) 9(37%)

Stage-2 - 10(26%) 3(20%) 11(32%) 6(23%) 8(33%)

Stage-3 4(33%) 14(37%) 10(66%) 12(34%) 15(58%) 7(30%)

Stage-4 8(67%) 1(3%) 1(7%) - 1(4%) -

4.8 DNA Methylation in 24 colorectal cancer tissue and

corresponding normal colonic tissue

Genome-wide methylation status was interrogated at nucleotide resulotion in 27500 CpG

loci. In figure 4.6 scatter diagram shows a linear correlation between the total intensity

(methylated + unmethylated) of tumour and normal tissue. Strong positive correlation

was seen which indicates there is no bias in the microarray intensity data. Figure 4.6

shows total intensity of ethylation data in all 24 cases.

47
Figure 4.6: Scatter diagram showing total intensity in tumour tissue and its
corresponding normal tissue in all 24 cases

vs Index

50000

40000
Tumour Intensity

30000

20000

10000

0 10000 20000 30000 40000 50000 60000

Normal Intensity

4.8.1 Methylation changes in tumour and surrounding normal tissue

There was significant difference in the methylation status between tumour and sorrounding

normal tissue.

4.8.1.1 The average beta

Average beta is the methylation status of a gene locus. It is calculated by intensity of the

methylated sequence(x)/ [intensity of the methylated sequence (x)+ intensity of the

unethylated sequence (y) ]. The ratio is between 0 to 1, where 0= extremely

hypomethylated and 1= extremely hypermethylated. Figure 4.2 shows scatter plot of

average beta for tumour tissue against that of corresponding normal tissue in 24 cases. The

figure shows that in the tumor tissue, there are a number of loci that are hypomethylated

compared to the corresponding normal tissue and a number of loci that are

48
hypermethylated. The central line represents regression line and other two lines on either

side represents the boundary for two fold change

Fig 4.7: Scatter diagram showing average beta of all 48 samples

vs Index

Tumour
1

0.80
Tumour AVG_Beta

0.60

0.40

0.20

0 Normal
0 0.20 0.40 0.60 0.80 1
Normal AVG_Beta

4.8.1.2 Tumour differential score

Tumour differential score indicates how significant the differenceis between the

methylation status of the normal and tumour tissue. The more the positive value, the more

that is hypermethylation. The more negative the more hypomethylation. Usually more

than 30 is significant and indicates hypermethylation. Figure 4.4 shows some cases

beyond +30 which are hypermethylated loci and many data beyond -30 which indicates

hypomethylated loci. It also shows majority of the data clustering around 0 which

indicates no differential methylation pattern in many of the cases.

49
Figure 4.8: Histogram showing tumour differential score.

Histogram for Tumour.DiffScore

12000

10000
# of Occurrences

8000

6000

4000

2000

-300 -200 -100 0 100 200 300

Tumour.DiffScore

4.8.2 Factors responsible for the change in methylation data

Bar diagram ( Fig 4.6) shows the significance of the different sources of variation in the

total methylation data done by ANOVA test. It indicates the factors that influence the

variation of the methylation data in 48 samples. It shows significance of factors

responsible for the methylation changes in tumour tissue and normal tissue . Difference of

the tissue (whether tumour or normal), sex (male or female, case id i.e case to case

variation and location of the lesion (proximal or distal colon).

50
Fig 4.9: Bar diagram showing sources of variation of methylation data.

Another important factor that is responsible for the changes in the methylation is tumour

differentiation. In this study in 24 cases 6 cases were poorly differentiated tumour and rest

were moderately differentiated tumour. There were changes in methylation pattern due to

differentiation (Figure 4. 7).

51
Figure 4.10 Variation of methylation data due to tumour differentiation.

4.8.3 Genes involved in the methylation difference in colonic tumour

tissue and normal colonic tissue

Some of the candidate genes are identified in this study which show methylation difference

in case of tumour tissue compared to normal colonic tissue. But which of these hypo or

hypermethylated genes play role in CRC and what are their exact role are yet to be

determined and needs further study. Table 4.11 show some genes that are methylated in

colorectal cancer in this study.

52
Table 4.11: Selected genes that show methylation differences in 24 cases of
Colorectal carcinoma.

Genes Tumour AVG. Beta Normal AVG Beta DiffScore Delta Beta

EYA4 0.516264 0.149085 362.88 0.367179

TFPI2 0.422315 0.100107 362.88 0.322208

GATA4 0.569414 0.243687 362.88 0.325726

CDKN2A 0.422516 0.159246 362.88 0.263270

DCC 0.566206 0.289243 362.88 0.276963

MDF1 0.550374 0.083746 362.88 0.461493

ITGA4 0.630950 0.122244 362.88 0.508707

GAD2 0.430238 0.155448 362.88 0.274790

ADCY4 0.474352 0.151537 362.88 0.322815

CCNA1 0.663816 0.300486 362.88 0.363330

DAB21P 0.455048 0.036997 362.88 0.41805

53
5.0 DISCUSSION
In this study, 50 cases of colorectal cancer were analyzed to find out histomorphological

features, selected clinical data and DNA methylation changes in 24 cases.

The mean age of the 50 cases was 47±14.8 years. The age range was from 19-84 years

with male and female ratio of 1.4:1. 58% of the cases were below the age of 50 years.

Peak incidence of colorectal cancer in this study were 50-59 years which is lower than that

of western and other countries (Table 5.1). Turner (2010) found only 20% of the cases

below 50 years. Keating et al (2003) found only 6.3% cases below 50 years. (Table-5.1).

Table 5.1 Age and sex distribution of CRC in different studies and the present

study.

Peak Mean age Range of Cases <50 Male :

Investigator/ Country incidence (Years) age (Years) years female
(Years)
Turner (2010),USA 60-70 <20% 1.2:1
Gomez et al (2004), UK 60-70 70±11 01-90 1.5‫׃‬1
Riddel et al (2003), USA 60-70 60 3‫׃‬2
Keating et al (2003), 70-80 69 6.3 Equal
New Zealand
Hossain(2007) 40-49 44.1±16.2 11-75 53% 1.6:1
Bangladesh
Present study 50-59 46.6±14.8 19-84 58% 1.4:1

54
The mean age of colorectal cancer in this present study indicate that colorectal carcinoma

is relatively common in lower age group in our country. Though incidence of colorectal

cancer in Bangladesh is not exactly known, it appears to be common in younger age group.

This may be due to both environmental factors and genetic factors. Leon et al (2004)

commented on recent trends in colorectal cancer. They observed rapid increase of CRC in

the developing countries. Gomez et al (2004) commented a tendency of right sided shift

from left sided colorectal cancer distribution. Keating et al (2003) suggested right sided

cancer to be less well differentiated than the left sided CRC. Presently we do not know

whether the CRC changing trends in Bangladeshi population i.e early age of onset, Right

colon shift etc.

It was also observed in present study that there is slight male preponderance regarding

colorectal cancer cases. Keating et al (2003) found equal gender distribution in their study.

Rectal bleeding was the commonest presentation (40% cases) at the time of first

consultation. The other presenting complaints were abdominal pain (16% cases), altered

bowel habit (12% cases), anorexia, pallor and generalized weakness (12% cases). Palpable

abdominal mass, partial intestinal obstruction and weight loss were present in addition to

the major complaints. Though abdominal pain has been described by some to be the

commonest symptom in colorectal cancer (Aljebreen, 2007), it was not the case in this

present study. Only 8(16%) cases had this feature. One possible explanation is negligence

on the part of the patient and frequent use of analgesics.

In this study, 20(40%) patients presented with per rectal bleeding, is similar to other study

(Hossain, 2007). Per rectal bleeding was observed in 17(85%) cases of left colon cancer.
55
Most of the cases with per rectal bleeding, tumour was present in rectum which explains

partly that per rectal bleeding may be most common symptoms in rectal cancer.

Anaemia was present in 13(26%) cases. Among these anaemic patients with colorectal

cancer 8 cases were female and 5 cases were male. About 69% of cases with cancers in

the right sided colon were anaemic and 31% of the patients with left sided cancer had

anaemia. Ayyub et al(2002) observed anaemia in 55% of their cases and also mentioned

70% cases were in right sided colon and 30% cases in left sided colon. This present study

shows lower haemoglobin levels for right sided colon cancer.

CEA (Carcinoembryonic antigen) level in serum were ranged from 0.39 ng/ml to 672

ng/ml with mean level 26.6. CEA was elevated (>5 ng/ml) in 25(50%) cases. Rosai

(2004) commented CEA more than 5 ng/ml is the adverse prognosis factor in CRC. In this

present study, in 25 cases having CEA more than 5 ng/ml, 17(68%) cases were in stage III

and one case was in stage I. This shows majority of the patients having elevated CEA are

in high stage disease. Aljebreen (2007) found 32% of cases of colorectal cancer with

elevated CEA level. Ayyub et all (2002) found CEA level ranged from 1 to 850.2 ng/ml

with a mean of 48.62.

The sub site distribution of colorectal cancer in this study shows 37(74%) cases were in the

left colon and 13(26%) cases in the right colon. Recent trend of shifting of CRC towards

right colon observed by Gomez et al (2004) is not supported by this present study.

However it is similar to the studies done by other researchers. This is shown in table 5.2.

56
Table-5.2: Distribution of colorectal cancer in left colon and right colon.

Investigator Left colon Right colon

Riddell et al(2003), USA 75% 25%

Leon et al (2004), Italy 70% 30%

Ayyub et al (2002), Saudi Arabia 70% 30%

Hossain(2007), Bangladesh 63% 37%

Present study 74% 26%

In Among 37 cases of left colon cancer cases, 33 cases were in rectum. This shows higher

percentage of rectal cancer in this study.

Size and shape of the tumour were recorded in all 50 cases. The size of the tumour ranged

from 3 cm to 10 cm. Pattern of growth were, 25(50%) cases ulcerating, 22(44%) cases

exophytic/polypoid and the rest 3(6%) infiltrative. 18(72%) of the ulcerated lesion was

present in rectum. This may be one of the probable cause of per rectal bleeding in most of

the rectal cancer cases.

On histological examination of 50 colorectal cancer cases, 44(88%) of cases were usual

adenocarcinoma and 6(12%) cases were mucin secreting adenocarcinoma. Among these

39(78%) cases were moderately differentiated and 11(22%) cases were poorly

differentiated cancer. No well differentiated cancer was reported in this series. Ekem et al

(2008) found 37% moderately differentiated, 17% well differentiated and 6% poorly

differentiated cancer. Keating et al (2003) found right sided tumour were less well

differentiated than left sided tumour. In this study. of thirty nine moderately differentiated

cases, 10(26%) originated in right colon and 29 (74%) in the left colon. Of the 11 poorly
57
differentiated carcinomas, 4 were in the right colon and 7 were in the left colon. In this

study left colon cancer were less well differentiated than right side.

Tumour infiltrating lymphocytes were observed in all 50 cases. 20(40%) cases were in

grade 1, 18(36%) cases were in grade 2, 4(8%) cases were in grade 3 and the rest 8(16%)

cases has no tumour infiltrating lymphocytes. In grade III TIL cases, only one were in

stage III compared to 2 in stage I. In no TIL cases, 4 were in stage III disease compared to

2 in Stage I disease. It shows higher TIL is associated with lower stage though conclusion

can not be made from this small sample size.

Lymphovascular invasion was present in 16(32%) cases and absent in 34 (68%) cases.

Ekem et al (2008) found lymphovascular invasion in 33% cases. Out of 16 positive cases

12(75%%) cases had stage III disease and 3(19%) cases had stage II and one (6%) case in

stage IV disease, which indicates higher stage disease in lymphovascular invasion positive

tumour. As stage is the major factor in predicting prognosis (Rosai, 2004). In this study,

majority of the cases in stage III disease show lymphovascular invasion.

This partly explains the importance of reporting lymphovascular invasion and Tumour

infiltrating lymphocytes in surgical pathology report.

Percentage of signet ring cells in the tumour tissue were recorded. 17(34%) cases show 1-

49% signet ring cells and 33 (66%) cases with no signet ring cells. No case having more

than 50% signet ring cells was present in this series.

58
Out of 50 colorectal cancer cases circumferential margin were involved in 12 (24%) of

the cases. Of twelve involved cases 8 cases were in stage III and 4 cases were in stage II

disease. Perineural invasion was present in 15(30%) cases. Out of 15 cases 10 cases had

stage III disease, 3 cases had stage II disease and one case each in stage I and stage IV.

Tumour border configuration were examined microscopically. Infiltrative pattern was seen

in 26 (52%) cases and pushing pattern were seen in 24(48%) cases. In 26 infiltrative

pattern cases, 15 cases in stage III, 6 cases in stage II, 4 cases in stage I and one case in

stage IV. This study shows higher stage disease in circumferential margin involved,

perinural invasive and tumour with infiltrative border cases. Rosai (2004) commented

circumferential margin, perineural invasion and infiltrative tumour border are adverse

prognostic factors in colorectal cancer (CRC).

TNM Staging (Turner, 2010) of colorectal cancer was done in all the cases. The

maximum number of cases 22 (44%) were in stage III, 14 (28%) were in stage II and 13

(26%) in stage I and one (2%) in stage IV. This study shows only 26% stage 1 cases

which indicates delay in diagnosis of the colorectal cancer cases. Derwinger et al (2010)

found 41% cases in stage III, 36% cases in stage II, 12% cases in stage IV and 10% cases

in stage I.

In this present study 49 colon cancer specimens were accompanied by lymph nodes. Of

these 23(47%) had nodal metastasis. It indicates patients are in advanced stage at the time

of attending physician. In 23 nodal metastasis cases 17(74%) were moderately

differentiated and 6(26%) cases were poorly differentiated tumour.

59
In addition to environmental factors, genetic and epigenetic mechanisms (DNA

methylation) are involved in the pathogenesis of Colorectal carcinoma (Turner, 2010).

DNA methylation abnormalities is such important epigenetic changes in CRC.

Methylation abnormalities are also observed in other tumours like stomach , pancrease,

breast, lung cancers (Samarakoon, 2010).

In this study DNA methylation changes was anlyzed in 24 cases of CRC tumour tissue and

corresponding normal colonic tissue. This study shows methylation difference between

the tumour and normal colonic tissue. A number gene loci are hypermethylated compared

to normal tissue and a number of gene loci are hypomethylated compared to normal tissue

(Figure 4.8).

This study also shows the factors responsible for methylation changes in tumour tissue.

Methylation difference is observed between individual cases, location of tumour (right

colon tumour tissue is compared with left colon tumour tissue) and between male and

female tumour tissue. Methylation difference was also observed between poorly

differentiated and moderately differentiated adenocarcinoma (Figure 4.10).

Some of the candidate genes showing hypo or hypermethylation is also identified and

shown in table 4.11 in result observation section. Kim et al (2010) also found in their

study that GATA4 gene is 70% methylated in CRC compared to only 6% methylated in

normal colonic tissue and ITGA4 gene is 92% methylated in CRC compared to 13%

methylated in normal tissue.

60
The exact role of DNA methylation in molecular pathogenesis of cancer is not fully

understood. It needs further study to conclue the role of DNA methylation in the

pathogenesis of colorectal carcinoma.

61
6.0 Summary and conclusion

6.1 Summary

This study was carried out to find histomorphological features, selected clinical data in 50

colonic cancer cases and DNA methylation changes in 24 of these cases. In this study the

mean age was 47±14.8 years and 58% of the cases were below the age 50 years which was

10 years less than that described by western observers. Male cases appears to be slightly

more than females.

Clinical features and related investigations are more or less similar that of other countries.

Recent trend show right shift (more cancers in the right side than left sided colon) of CRC

in western countries but is is not the findings in this study.

Adenocarcinoma of usual pattern was most common type, other common type was mucin

secreting adenocarcinoma. 78% cases were moderately differentiated and 22% cases were

poorly differentiated. No well differentiated cases were observed in this series indicates

higher grade tumours in this study subjects. Most common stage was stage III (44%) cases

and 24% cases were in stage I. Lymphnode metastasis was present in 47% cases which

indicates delay in diagnosis and advanced stage of disease.

DNA methylation changes was seen in tumour tissue compared to normal tissue. Other

factors for methylation changes were location of the tumour, differentiation of the tumour

and case to case variation. Some genes were identified which are methylated in colorectal

cancer tissue but needs further study with large sample size to specify the association.

63
6.2 Conclusion

Histomorphological and selected clinical data presented in this study are more or less

similar to that in other published reports, with the exception of mean age of the cases

which is about one decade less than western population. DNA methylation changes are

observed in the tumour tissue. Some genes are also identified but needs further study to

confirm the association and their role in pathogenesis.

63
7.0 BIBLIOGRAPHY

Afroza, A, Hasan, S, Rukunuzzaman, M, Hussain, SA, and Amin, R 2007, ‘Carcinoma-

Rectum in an 11 Years old Boy’, Mymensingh Medical Journal, 16(2 Suppl), 70-

72.

Aljebreen, AM 2007, ‘Clinico-Pathological Patterns of Colorectal Cancer in Saudi Arabia:

Younger with an Advanced Stage Presentation’, The Saudi Journal of

Gastroenterology, 33, 84-87.

Ayyub, MI, Alradi, AO, Khazeinder, AM, Nagi, AH and Maniyar, IA 2002,

‘Clinicopathological trends in colorectal cancer in a tertiary care hospital’, Saudi

Medical Journal, 23, 160-163.

Borley, NR 2008, ‘The large intestine’ in Gray's Anatomy, 40th edition, Elsevier company,

Pheladelphia, USA, 1137-1160.

Boyle, P and Langman, JS 2000, ‘ ABC of colorectal cancer: Epidemiology’, British

Medical Journal, 321, 805-808.

Cooper, H. S (ed) 1999, ‘Intestinal Neoplasm’ in Diagnostic Surgical Pathology, 3rd

edition, Sternberg, S. S. Lippincott Williams and Wilkins, Philadelphia, 2, 1413-

1461.

Derwinger, K, Kodeda, K, Bexe-Lindskog, E And Taflin, H 2010, ‘Tumour differentiation

grade is associated with TNM staging and the risk of node metastasis in colorectal
cancer’, Acta Oncologica, 49, 57–62.
Ekem, TE, Bahadir, B, Gun, BD, Bektas, S, Kertis, G, Yurdakan, G and Ozdamar, SO
2008, ‘Colorectal carcinomas: Clinicopathologic investigation, correlation with
expression of estrogen and progesterone receptors’, Turkish Journal of Cancer, 38,
118-122.
64
Fireman, Z, Neiman, E, Mouch, SA And Kopelman, Y 2005, ‘Trends in incidence of

colorectal cancer in Jewish and Arab population in central Israel’, Digestion, 72,

223-227.

Ganong, W.F.(ed) 2005, ‘ Regulation of Gastrointestinal function’ in Review of Medical

Physiology, 22nd edition, The Mcgraw-hill companies, New York. pp 479-513.

Gomez, D, Dalai, Z, Raw, E, Roberts, C and Lyndon, PJ 2004, ‘Anatomical distribution

of colorectal cancer over 10 year period in a district general hospital: is there a true

rightward shift?’, Postgraduate Medical Journal, 80, 667-669.

Hamilton, SR, Vogelstein, B, Kudo, S, Riboli, E, Nakamura, S, Hainaut, P, Rubio, CA,

Sobin, L H, Fogt, F, Winawer, SJ, Goldgar, DE and Jass, JR 2000, ‘Carcinoma of

the colon and rectum’ in Pathology and Genetics of Tumour of the Digestive

System, Hamilton S.R. and Aaltonen, L. A, IARC press, Lyon, France, pp103-142.

Herman, JG and Baylin, SB 2003, ‘Gene Silencing in Cancer in Association with Promoter
Hypermethylation’, The New England Journal of Medicine, 349, 2042-54.

Hossain, T 2007, ‘Clinicopathological study of colorectal carcinoma’. MD Thesis,

Bangabandhu Sheikh Mujib Medical University.

Junquira, LC, Carneiro, J and Kelly, RO 1995, Basic Histology, 9th edition, Lange Medical

Publishers, California, 397-402.

Keating, J, Pater. P, Lolohea, S and Wickremesekera, K 2003, ‘The epidemiology of

colorectal cancer: what can we learn from the New Zealand Cancer Registry’,

The New Zealand Medical Journal, 116, 1-8.

65
Kim, SK, Lee, J and Sidransky, D 2010, ‘DNA methylation markers in colorectal cancer’,

Cancer Metastasis Review,29, 181-206.

Leon, MPD, Marino, M, Benatti. P, Rossi, G, Menegatti, M, Pedron, M, Gregorio, C D,

Losi, L, Borghi, F, Scarsclli, A, Ponti, G, Roncari, B, Zangardi, G, Abbati, G,

Ascari, E, and Roncucci, L 2004, ‘ Trend of incidence, sub site distribution and

staging of colorectal neoplasms in the 15-year experience of a specialized cancer

registry’, Annals of Oncology, 15, 940-946.

Notani, PN 2001, ‘Global variation in cancer incidence and mortality’, Current Science,

81, 467-468.

Ogino, S, Kawasaki, T, Nosho, K, Ohnishi, M, Suemoto, Y, Kirkner, GJ and Fuchs, CS

2008, ‘LINE-1 hypomethylation is inversely associated with microsatellite
instability and CpG island methylator phenotype in colorectal cancer’, International
Journal of Cancer, 122, 2767–2773

Riddell, RH, Petras, RE, Williams, GT and Sobin, LH 2003, ‘Epithelial neoplasia of the

intestines’ in Tumors of the Intestines , Third series,Fascicle 32, Armed forces

institute of pathology,Washington,DC, pp85-240.

Ropponen, KM, Eskelenen, MJ, Lipponen, PK, Alhava, E and Kosma, VM 1997,

‘Prognostic Value of Tumour Infiltrating Lymphocytes (TIL) in Colorectal Cancer’,

Journal of Pathology,182, 318-324.

Rosai, J.(ed) 2004, ‘Gastrointestinal tract’ in Ackerman's Surgical Pathology, 9th edition,

Mosby company, St Luis, 1, 776-825.

Sadler, T.W.(ed), 1995. Langman's Medical Embryology, 7th edition, Willams and

Wilkins, Baltimore, 329-331.

66
Samarakoon, PS 2010, ‘ Epigenomics and Genome wide Methylation Profiling’, Sri Lanka

Journal of Bio-Medical Informatics , 1, 53-62.

Turner, JR 2010, ‘The Gastrointestinal Tract’ in Robbins and Cotran Pathologic Basis of

Disease, 8th edition, Kumar, V. Abbas, A.K. Fausto, N and Aster, J. C . W B

Saunders company, Philadelphia,Pennsylvania, 822-825.

Wong, JJL, Hawkins, NJ and Ward, RL 2007, ‘Colorectal cancer: a model for epigenetic

tumorigenesis’, Gut, 56, 140-148.

Yawe, KT, Bakari, AA, Pindiga, UH and Mayun, AA 2007, ‘Clinicopathological pattern

and challenges in the management of colorectal cancer in sub-Sahar Africa’,

Journal of Chinese Clinical Medicine, 2, 688-694.

Young, B and Heath, JW 2000, ‘Gastrointestinal tract’ in Fundamental Histopathology, 4th

edition, Harcout publishers Ltd, Churchill livingstone, pp 249-273.

67
 Acknowledgements

I convey my respect and gratitude to my teacher honorable professor Dr.

Mohammed Kamal, Chairman, Department of Pathology, Bangabandhu Sheikh
Mujib Medical University (BSMMU), Dhaka and supervisor of this study for
his guidance, valuable suggestions, constant supervision, affectionate advice
and whole hearted co-operation in making this study a success. Without his
vigilance and care this research would have been difficult.

I would like to express my sincere gratitude to my teacher Professor (Dr.)

Ashim Ranjan Barua for his valuable suggestions and co-operation.

I would pay my deep gratitude to my teacher Professor (Dr.) Kamrul Hasan

Khan for his valuable advice and co-operation.

I am thankful to my teacher Dr. Tamanna Choudhury, Associate professor,

Department of Pathology, BSMMU, Dhaka for her valuable suggestions and
co-operation.
I am greatly thankful to Professor Pran Gopal Datta, Vice-Chancellor, BSMMU, Dhaka for
providing me an opportunity to perform this study in this University.

I am grateful to my teacher and co-supervisor, Dr. Ferdousy Begum, Assistant

Professor, Department of Pathology, BSMMU, for her advice, meticulour
correction and active participation in this study.

I am highly grateful to my teachers Dr. AKM Nurul Kabir, Dr. Sultana

Gulshana Banu, Dr. Suraiya Enam, Dr. Saequa Habib and Dr. Shabnom
Akhter, Assistant Professors, Department of Pathology, BSMMU for their help
and co-operation.
I must thank and pay my gratitudes to DR. Muhammad G. Kibriya and Dr. Farzana
Jasmine, Research associate- Assistant Professors, Department of Health Studies (BSD),
The University of Chicago Medical Center, Chicago , IL , USA, for their constant
encouragement, support, suggestions and immediate response to my e-mails regarding this
study.

I am very grateful to Prof. Zahidul Haq, Professor, Department of Surgery, BSMMU for
his help, co-operation and suggestions for collecting specimen for this study.

I would like to extend my sincere thanks to all my colleagues specially Dr. Himel, Dr.
Tahera, Dr. Nizam, Dr. Babul, Dr. Taufiq, Dr. Rupash, Dr. Alamgir, Dr. Shahadat and all
other residents for their advice, sympathy and co-operation during my study.

I would like to appreciate Mr. Osman Goni Chowdhury, Mr. Ruhi Das Roy and other staff
of the Department of Pathology, BSMMU for their help and co-operation.

I am thankful to Mr. Shelabrata Barua (Jewel) and Md. Faruk hossain who took the trouble
to computer compose and typing this thesis.

I acknowledge all patients involved in this study and their relatives for co-operation and
providing me valuable informations that were helpful in the research work.

I wish to express my highest regards and heartiest gratitude to my beloved parents Md.
Muzibur Rahman Sarkar and Mrs. Khaleda Begum who always wanted me to be a good
and honest teacher. My two younger brothers who always supported me and help me to
take any desision. And also my students who inspired me to be a good teacher and
researcher.

All credit belongs to almighty Allah.

Septeber, 2010 Dr. Abu Khalid Muhammad Maruf Raza
 APPENDIX-II
Clinical proforma

Title: Study on pathological aspects of colorectal carcinoma.

Identification:

1.Case number-
2.Lab number-
3.Date-
4.Name-
5.Age-
6.Sex-
7.Adress-
8.Occupation-
Presenting complaints:

Tumor Location:

Family history:

Socio-economic history:

Relevent investigation: Haemoglobin-

ESR-
CEA-
Others-
Gross examination: Specimen type-
Tumor type-
Tumor configuration-
Tumor size-
Mesorectum-

Microscopic examination:
ii
 Histologic type-

Histologic grade-
Extent of invasion-
Margins- Proximal-
Distal-
Circumferential-
Lymphovascular invasion-
Perinural invasion-
Tumour infiltrating lymphocytes-
Tumor border configuration-
Signet ring cells-
Extracellular mucin-
Additional pathologic findings-

Diagnosis:

Consultant pathologist:

iii
 APPENDIX -III

PREPARATION OF STAINS AND CHEMICALS:

HAEMATOXYLIN AND EOSIN STAINS:

a) Harris’ Haematoxylin:

Haematoxylin crystal 5.00 gm

Absolute alcohol 50.ml
Ammonium alum 100.00 gm
Distilled water 950. ml
Mercuric oxide 2.50 gm
Glacial acetic acid 40.00 ml

Procedure: Haematoxylin crystals were dissolved in alcohol at 56° C in an oven. Alum

was also dissolved by heating. These two solutions were mixed thoroughly and brought to

boil rapidly. Then it was removed from the flame and mercuric oxide was added slowly.

The solution was then allowed to cool rapidly in cold water. The solution was ready to use

after cooling. It was kept in room temperature and filtered before use. 2-4 ml of glacial

acetic acid was added per 100 ml of solution to enhance the precision of nuclear stain.

b) Acid alcohol:

Hydrochloric acid (pure) 1 ml

70% ethyl alcohol (mixed thoroughly) 99 ml

c) Eosin solution:

Eosin (water soluble) 1gm

Distilled water 100 ml

Procedure : Eosin dissolved in distilled water and mixed thoroughly. It was stored at room
temperature and filtered before use. Solution was made more intense by adding glacial
acetic acid in the proportion of 0.2 ml for 100 ml solution before use.
iv
 APPENDIX-IV

TISSUE PROCESSING AND STAINING PROCEDURE

Tissue processing was done in Department of Pathology, BSMMU by standard protocol in

automatic tissue processor (BAVIMED 2050, BAVIMED Laborgeneratebau GmBH,

Birkeau, GERMANY).

Processed tissue than properly embedded on melted paraffin for making blocks and

sections. The sections were stained with haematoxylin and eosin for microscopic

examination.

Paraffin embedding: Metallic moulds were used for this purpose. These moulds were first

lubricated with liquid paraffin. Then melted paraffin was poured into it. The tissue was

carefully embedded in proper plane at the bottom of the mould. The respective number of

the tissue was inserted into the paraffin wax by the side of the mould. The melted paraffin

was then allowed to harden at room temperature. After hardening, the mould was removed

and blocks were trimmed properly to mount on a block holder. Then the blocks were kept

in ice chamber of a refrigerator for sometime before cutting the sections.

Section cutting: Each block of tissue mounted on the holder was fitted in the microtome

machine. The microtome knife was properly sharpened before. A water bath with regulated

temperature of 45° to 50°C was used for floatation. Sections were cut at 4-5 micron

thickness. Ribbons of good sections were selected and floated on luke warm water in the

water bath. The sections were then taken on albuminized glass slides. The slides were kept

in inclined position to drain off excess water and allowed to dry at room temperature.

Staining: All the slides for histopathological examination were stained by routine

Haematoxylin and Eosin method

v
SPTEPS OF HAEMATOXYLIN AND EOSIN STAINING:

a) Deparaffinization :

i) Slides were kept in hot air oven 15 minutes

ii) Xylol- I 5 minutes

iii) Xylol –II 5 minutes

b) Hydration :

i) Absolute alcohol – I 3 minutes

ii) Absolute alcohol –II 2 minutes

iii) 95% alcohol 2 minutes

iv) 70% alcohol 2 minutes

v) 50% alcohol 2 minutes

vi) Running tap water 2 minutes

c) Staining with Haematoxylin and Eosin:

Harris’s Haematoxylin 10 minutes

1% acid alcohol 1 minutes

Section were placed in running tap water till the sections become blue. Counter staining

was done by immersing the section in 1% watery solutions of eosin for 1 minute.

d) Dehydration:

The sections after Haematoxylin and Eosin staining were dehydrated in

following manner.

i) 50% alcohol 02 minutes

ii) 70% alcohol 02 minutes

iii) 80% alcohole 02 minutes

iv) 95% alcohol 02 minutes
v) absolute alcohol I 02 minute
vi) absolute alcohol II 02 minutes

vi
e) Clearing:
a) Xylol – I - 5 minutes
b) Xylol – II - 5 minutes
c) Xylol – III - 2 minutes.
6) Mounting:

Mounting was done with DPX using No. 1 cover slip.

Results:

Cell nuclei - Blue

Cytoplasm - Pink
Collagen fibre - Pink
RBC - Bright red

vii
 APPENDIX-V

Protocol for DNA extraction and purification from colonic tissue (Gentra Puregene
Handbook 04/2010, QIAGEN company, USA (www.qiagen.com). :

1. Dissect tissue sample quickly and freeze in liquid nitrogen. Grind frozen tissue in liquid
nitrogen with a mortar and pestle. Work quickly and keep tissue on ice at all times,
including when tissue is being weighed.

2. Dispense 300 μl Cell Lysis Solution into a 1.5 ml grinder tube on ice, and add the
ground tissue from the previous step. Complete cell lysis by following step 2a or 2b below:

2 a. Heat at 65°C for 15 min to 1 h.

2 b. If maximum yield is required, add 1.5 μl or Puregene Proteinase K, mix by inverting

25 times, and incubate at 55°C for 3 h or until tissue has completely lysed. Invert tube
periodically during the incubation. The sample can be incubated at 55°C overnight for
maximum yields.

3. Add 1.5 μl RNase A Solution, and mix the sample by inverting 25 times. Incubate at
37°C for 15–60 min.

4. Incubate for 1 min on ice to quickly cool the sample.

5. Add 100 μl Protein Precipitation Solution, and vortex vigorously for 20 s at high speed.

6. Centrifuge for 3 min at 13,000–16,000 x. The precipitated proteins should form a tight
pellet. If the protein pellet is not tight, incubate on ice for 5 min and repeat the
centrifugation.

7. Pipet 300 μl isopropanol into a clean 1.5 ml microcentrifuge tube and add the
supernatant from the previous step by pouring carefully. Be sure the protein pellet is not
dislodged during pouring. If the DNA yield is expected to be low (<1 μg) add 0.5 μl
Glycogen Solution

viii
8. Mix by inverting gently 50 times.

9. Centrifuge for 1 min at 13,000–16,000 x g.

10. Carefully discard the supernatant, and drain the tube by inverting on a clean piece of

absorbent paper, taking care that the pellet remains in the tube.

11. Add 300 μl of 70% ethanol and invert several times to wash the DNA pellet.

12. Centrifuge for 1 min at 13,000–16,000 x.

13. Carefully discard the supernatant. Drain the tube on a clean piece of absorbent paper,

taking care that the pellet remains in the tube. Allow to air dry for 5 min.

The pellet might be loose and easily dislodged. Avoid over-drying the DNA pellet, as the

DNA will be difficult to dissolve.

14. Add 200 μl DNA Hydration Solution and vortex for 5 s at medium speed to mix.

15. Incubate at 65°C for 1 h to dissolve the DNA.

16. Incubate at room temperature overnight with gentle shaking. Ensure tube cap is tightly

closed to avoid leakage. Samples can then be centrifuged briefly and transferred to a

storage tube.

ix
 APPENDIX-VI

Protocol DNA methylation (Infinium Assay Methylation Protocol Guide, Catalog # WG-

901-2701 Part # 11322371 Rev. A,Illumina Inc, USA.)

1. Bisulphite conversion of DNA material:

After DNA extraction bisulphite conversion of DNA was done by using the EZ DNA

MethylationTM Kit (Zymo Research, Catalogue # 500, USA). The kit is based on the reaction

that takes place between unconverted cytosine and sodium bisulfite where cytosine is converted

into uracil. With subsequent amplification this converted Uracil becomes T (thimine). So instesd

of unmethylated C we get T in the sequence. 1 µg DNA from each tissue sample were used for

bisulfite conversion. These bisulphite converted DNA were used immediately for methylation

assay.

2. Amplification of DNA:

The Bisulfite converted DNA samples were denatured by alkali to open up the strands and then

was isothermally amplified using Multi-sample Amplification. Master Mix provided by Illumina

by overnight incubation.

3. Fragmentation, precipitation and hybridization of DNA:

The amplified product was fragmented by a controlled enzymatic process (FMS reagent provided

by Illumina. The process uses end-point fragmentation to avoid overfragmenting the samples.

These fragmented DNA was precipitated by adding and spinning with isopropanol. Then the

x
precipitated DNA was resuspended in hybridization buffer (RA1 reagent provided by Illumina).

After resuspension, the fragmented DNA was heat denatured. Then samples were put on

methylation bead chip. Twelve samples are applied on to each BeadChip, which keeps them

separate with an IntelliHyb seal. The prepared BeadChip is incubated overnight in the Illumina

Hybridization Oven at 48° c.

4. Wahing, extension and staining of Beadchips:

On the following day unhybridized and non-specifically hybridized DNA were removed by

washing, and the chips were prepared for staining. A single base extension reaction of the

hybridized methylated or unmethylated sequence was done by incorporating fluor tagged

necleotide.

5. Imaging Bead Chip:

This was done by using Illumina Bead Array Reader software. Intensity of methylation was

calculated by intensity of the methylated sequence(x)/ intensity of the methylated sequence (x) +

intensity of the umethylated sequence (y). If both the alleles are methylated, intensity = x/x+0=1

and if both the alleles are unmethylated then intensity = 0/0+y=0 and if one allele is methylated

and the other is unmethylated then intensity is=x/x+y.

xi
 Appendix - VII

American Joint Committee on Cancer (AJCC) TNM Classification of

Colorectal Carcinoma (Turner, 2010).

TUMOUR
Tis In situ dysplasia or intramucosal carcinoma
T1 Tumor invades submucosa
T2 Tumor invades into, but not through, muscularis propria
T3 Tumor invades through muscularis propria
T3a Invasion <0.1 cm beyond muscularis propria
T3b Invasion 0.1 to 0.5 cm beyond muscularis propria
T3c Invasion >0.5 to 1.5 cm beyond muscularis propria
T3d Invasion >1.5 cm beyond muscularis propria
T4 Tumor invades adjacent organs or visceral peritoneum
T4a Invasion into other organs or structures
T4b Invasion into visceral peritoneum

REGIONAL LYMPH NODES

NX Lymph nodes cannot be assessed
N0 No regional lymph node metastasis
N1 Metastasis in one to three regional lymph nodes
N2 Metastasis in four or more regional lymph nodes

DISTANT METASTASIS
MX Distant metastasis cannot be assessed
M0 No distant metastasis
M1 Distant metastasis or seeding of abdominal organs

xii
TNM staging for colorectal carcinoma (Turner, 2010)

Stage-I Tl N0 M0 or T2, N0, M0

Stage-II T3 N0 M0 or T4, N0, M0

Stage-III T any Nl M0 or T any,N2,N3, M0

Stage-IV T any N any Ml

xiii
 Appendix – VIII
Master Table

Hb (mg/dl)

Differentia

Signet ring

Circumfer
Size (Cm)

TIL grade
Diagnosis

Ext. Muc

Tumour

LN/ Inv
Age/ Sex

margin
border
Location
Lab. No.

invasion

invasion

ential
Case ID

Tumour
Growth
tion
C/F

(ng/ml)
Sl. No.

pattern

stage
CEA

PN
cell

LV
1 C1 B-7750/09 35/F p/r/b 11.1 8.01 Rec 6 Ade M 2 No No Pre P Inf Inv 11/11 Ul III

2 C2 B-114-16/10 19/M p/r/b 10.8 3.14 Rec 4 Ade M N No No A P Inf Inv 7/0 Exo II

3 C3 KA-119-23 75/M p/r/b 11 672 Rec 3 Ade M N No No A A Push Inv 7/3 Ul III

4 C4 B-174/10 68/M pain 11 10.9 Cae 10 Muc P 1 >50% 1-49% A A Push Free 10/0 Exo II

5 C5 B-107-8/10 41/M p/r/b 7 3.9 Tran 6 Muc P N >50% 1-49% A A Push Free 4/0 Exo I
and
pain
6 C6 B-268/10 52/F/ p/r/b 9.3 16.8 Rec 4 Ade P 1 No No A A Push Free 8/0 Exo I

7 C7 B-273/10 55/M p/r/b 11.2 79.64 Rec 6 Muc P N >50% 1-49% A A Push Inv 0/0 Inf II

8 C8 B-327/10 25/M p/r/b 10.8 3.9 Rec 4 Ade M 1 No No Pre P Inf Free 7/4 Exo III

9 C9 B-331/10 52/M pain 10 4.1 Asc 6 Ade M 1 1-49% 1-49% Pre A Inf Free 11/1 Ulc III

10 C10 KA-266-69 65/F a/b 11.1 9.3 Rec 8 Ade M 1 1-49% No A A Inf Free 11/5 Exo III

11 C12 B-444/10 45/M p/r/b 13 18.34 Rec 5 Ade M 1 1-49% No Pre A Inf Free 4/2 Ulc III

12 C13 KA-399-402 45/M p/r/b 9.3 9.1 Rec 7 Ade M 1 No No A A Inf Free 4/0 Ulc II

13 C14 KA-406-07 43/M p/r/b 10.1 11.1 Rec 5 Ade M N No No A A Inf Free 7/0 Exo I

14 C15 KA423-26- 35/F Pain 11 68.9 Rec 5 Ade P 1 1-49% 1-49% A A Inf Free 12/1 Ulc III

15 C16 B-566/10 35/M p/r/b 11 13.7 Rec 6 Ade M 1 No No Pre P Infiltr Free 7/1 Exo III
ative

xiv
 Appendix – VIII
Master Table

Hb (mg/dl)

Differentia

Signet ring

Circumfer
Size (Cm)

TIL grade
Diagnosis

Ext. Muc

Tumour

LN/ Inv
Age/ Sex

margin
border
Location
Lab. No.

invasion

invasion

ential
Case ID

Tumour
Growth
tion
C/F

(ng/ml)
Sl. No.

pattern

stage
CEA

PN
cell

LV
16 C17 B-589/10 24/M Pain 12.7 0.31 Cae 10 Muc P 1 >50% No A A Push Inv 20/15 Ulc III
0
17 C19 B-960/10 59/M p/r/b 10.6 1.9 Hepa 6 Ade P 2 No No A A Push Free 20/0 Exo II

18 C20 B-1126/10 28/M a/b 10.6 3.1 Rec 9 Ade M 1 No No A A Push Free 14/0 Exo I

19 C21 B-1370/10 53/M p/r/b 10.2 24.6 Rec 5 Ade M 2 1-49% No A A Inf Free 7/0 Exo I

20 C22 B-1433/10 84/F mass 11.6 4.44 Rec 4 Ade M 1 1-49% No A A Push Free 11/0 Ulc I

21 C23 B-1467/10 30/M a/b 12 3.9 Rec 7 Ade M 1 No 1-49% Pre P Inf Free 19/1 Ulc III

22 C25 B-1706/10 27/F p/r/b 11.1 3.1 Hepa 6 Ade M 1 1-49% No pre P Inf Free 23/1 Ulc III
and
pain
23 C26 KB-705-08 48/F p/r/b 12 3.19 Rec 4 Ade M 3 No No A A Push Free 10/0 Exo I

24 C27 B-1851/10 36/M p/r/b 9 3.9 Rec 10 Ade M 1 No No A A Inf Free 12/0 Exo I

25 C28 B-1995/10 28/F p/r/b 13.1 3.9 Rec 4 Ade M 3 1-49% 1-49% pre A Inf Free 4/0 Ulc II

26 C30 KC-61- 59/F pain 11.8 41.2 Rec 7 Ade M 3 1-49% No A A Inf Free 8/0 Exo I
63/10
27 C31 BH-779- 50/M p/r/b 11.6 3.9 Asc 6 Ade M 2 1-49% No A A Inf Free 4/0 Exo II
80/10 and
pain
28 C32 KC-320-23 42/M a/b 13 1.81 Rec 3 Ade M 1 1-49% 1-49% A A Push Free 2/0 Exo I

29 C33 B-2397- 58/F a/b 12.8 5.5 Rec 6 Ade M 1 No No Pre P Inf Free 15/0 Exo II
99/10
30 C34 B-2415- 37/M p/r/b 14.5 314 Des 5 Muc P N >50% 1-49% Pre P Inf Involve 17/17 Inf III
16/10 d

xv
 Appendix – VIII
Master Table

Hb (mg/dl)

Differentia

Signet ring

Circumfer
Size (Cm)

TIL grade
Diagnosis

Ext. Muc

Tumour

LN/ Inv
Age/ Sex

margin
border
Location
Lab. No.

invasion

invasion

ential
Case ID

Tumour
Growth
tion
C/F

(ng/ml)
Sl. No.

pattern

stage
CEA

PN
cell

LV
31 C35 B-2626/10 45/F a/b 11.2 7.49 Rec 8 Ade P N 1-49% 1-49% A A Inf Free 15/1 Exo III

32 C36 B-2613/10 45/F a/b 12 234 Des 4 Ade M N No No Pre P Inf Inv 25/20 Ulc III

33 C37 KD-236-41 40/F w/p 13.1 3.38 Rec 5 Ade M 2 No 1-49% Pre A Push Free 9/3 Ulc III

34 C38 KD236-41- 73/M p/r/b 13.5 24.8 Hepa 3 Ade M 2 A A Push Free 15/0 Ulc I
and 1-49% 1-49%
pain
35 C39 B-2888/10 50/F pain 5.4 2.17 Sig 5 Ade P 2 No 1-49% A P Push Free 16/0 Ulc II

36 C40 KD-715-19 66/M p/r/b 14.6 2.1 Rec 4 Ade M 1 No No A P Push Free 10/0 Ulc I

37 C41 KD-809-11 30/F p/r/b 12.7 4.1 Rec 7 Ade M 2 No No A A Push Free 8/0 Exo I
and
pain
38 C42 B-2976/10 50/F w/p 10.9 3.75 Rec 5 Ade M 2 1-49% No A A Pushi Free 7/2 Ulc III
ng
39 C43 B-3376- 28/M p/r/b 13.8 13.57 Rec 5.5 Muc P 1 >50% 1-49% A A Inf Inv 5/2 Ulc III
79/10 and
pain
40 C44 B-3374/10 52/M mass 8.8 4.34 Rec 5 Ade M 1 No No A A Push Free 15/9 Ulc III

41 C45 KD-868-71 50/M a/b 6.9 1.63 Tran 4.5 Ade M 3 No No A A Push Free 4/3 Exo III

42 C46 B-3395-96 63/F p/r/b 12.7 38.9 Rec 5 Ade M 2 No No A A Push Free 5/0 Ulc II

43 C47 B-3469/10 38/F a/b 10 29.3 Asc 8 Ade M 2 No No A P Inf Free 7/0 Ulc III

44 C48 B-3484- 50/M w/p 11.6 256 Asc 3 Ade M 2 1-49% No Pre P Inf Free 7/4 Ulc IV
85/10
45 C49 B-3604- 48/F w/p 10.7 5.1 Rec 4 Ade M 2 1-49% No A A Push Free 6/0 Exo II
05/10
xvi
 Appendix – VIII
Master Table

Hb (mg/dl)

Differentia

Signet ring

Circumfer
Size (Cm)

TIL grade
Diagnosis

Ext. Muc

Tumour

LN/ Inv
Age/ Sex

margin
border
Location
Lab. No.

invasion

invasion

ential
Case ID

Tumour
Growth
tion
C/F

(ng/ml)
Sl. No.

pattern

stage
CEA

PN
cell

LV
46 C50 B-3738- 65/M w/p 8 15.9 Asc 10 Ade M 2 1-49% 1-49% A A push Inv 11/0 Exo II
39/10
47 C51 B-3906/10 39/F w/p 8.6 1.9 Rec 6 Ade P 2 1-49% 1-49% Pre P Inf Inv 9/9 Inf III

48 C52 B-4307/10 39/F mass 8.1 1.09 Sig 5 Ade M 2 No No Pre A Inf Free 7/0 Ulc II

49 C53 B-4328- 72/M a/b 12.7 4.5 Rec 4 Ade M 2 No No A A Push Inv 5/0 Ulc II
29/10
50 C54 B-4407/10 38/F w/p 6 32.19 Asc 6 Ade M 2 No No pre A Push Inv 9/1 Ulc III

M=male, f= female, p/r/b= per ectal bleeding, w/p= weakness, pallor, anorexia, mass= abdominal mass, a/b= altered bowel habit, Rec= Rectum, Asc= Asending colon, Tran= Transeverse
colon, Des= descending colon, sig= Sigmoid colon, cae= caecum, Ade= Adenocarcinoma usual type, Muc= Mucinous carcinoma. P= Present, A= absent, push= pushing margin, Inf=
Infiltrating margin, Ulc= Ulcerated, Exo= Exophytic, Inf= infiltrative. Inv= Involved, Til =tumour infiltrating lymphocytes, Ext. Muc= Extra cellular mucin, LV invasion= Lymphovascular
invasion, PN invasion= Perineural invasion. LN= lymphnode.

xvii