UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

Three decades of gastroenterology in Soweto South Africa: from descriptive toscientific observations

Segal, I.

Publication date2002

Link to publication

Citation for published version (APA):Segal, I. (2002). Three decades of gastroenterology in Soweto South Africa: from descriptiveto scientific observations.

General rightsIt is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s)and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an opencontent license (like Creative Commons).

Disclaimer/Complaints regulationsIf you believe that digital publication of certain material infringes any of your rights or (privacy) interests, pleaselet the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the materialinaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letterto: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. Youwill be contacted as soon as possible.

Download date:09 Jun 2021

https://dare.uva.nl/personal/pure/en/publications/three-decades-of-gastroenterology-in-soweto-south-africa-from-descriptive-to-scientific-observations(7f66f3c1-a5b2-4d1d-b17e-5134161855ab).html

Chapterr 8

Colonicc Cell Proliferation in Two Different Ethnic Groups

withh Contrasting Incidence of Colon Cancer:

iss there a Difference in Carcinogenesis ?

Vann 't Hof A, Gilissen K, Cohen RJ, Taylor L,

Haffajeee Z, Thornley AL, Segal I

Gutt 1995;36:691-695

69 9

Abstract t

Mostt studies on colorectal carcinogenesis suggest a field defect, preceding overt development

off cancer. The low incidence of adenomatous polyps in the African population, however,

suggestss that there may be an alternative route for cancer development. The aim of the study

wass to discover if the difference in incidence of colorectal cancer in Africans compared with

thee white population is reflected in a different pattern of cell proliferation. Histological

normall mucosa from 30 patients (15 white South African (W), 15 South African Africans

(A))) with confirmed colon cancer were examined. Proliferating cells were detected using the

K.i-677 antigen. In addition, cell proliferation data were obtained, from 30 age matched

controlss (15 Africans, 15 white South Africans), without colorectal disease. The African

controlss were significantly younger (mean (SD) {A: 42 (20), W: 66 (13), p

patients.. It was also found that most cancers were unrelated to polyps.12 These findings

suggestt that in Africans, colorectal cancer may have a different genesis to that of the

Westernisedd population.

Dataa from cell kinetic parameters can give insight into the mechanism of colorectal

carcinogenesis.. It was suggested that hyperplastic polyps are not related to the histogenctic

processs of colorectal cancer, because there was no significant difference in cell proliferation

inn comparison with normal mucosa.1'5 Deschner et al found that the proliferative pattern in

cryptss of mice and rats were different. It was suggested that the direction of the shift in the

majorr zone of proliferation is related to the type of lesion present in the colonic mucosa.

Downwardd migration seemed to be related with intramucosal formed cancers, and upward

migrationn with adenomatous polyps.

Thiss study compares the proliferative pattern of histologically normal mucosa from African

andd white patients, with and without colonic cancer. The aim of the study was to assess if the

loww incidence of colorectal adenomatous polyps and cancer in Africans is reflected in a

differentt pattern of cell proliferation.

Methods s

Specimens Specimens

Specimenss were obtained from 30 patients with confirmed colonic carcinoma, selected from

thee index files of the department of histopathologgy at the University of Witwatersrand. Only

normall appearing colonic mucosa, obtained from the tumour free resection margins was

examined,, and mucosa within a distance of 2 cm from the tumour was excluded. The surgical

specimenss were routinely fixed in formalin and processed.

Sectionss of 4 microns were cut and dried overnight to ensure adequate adhesion before

staining. .

Inn addition, biopsy specimens were obtained from 30 age matched control patients (15

Africans,, 15 white patients), without colorectal disease. All patients underwent endoscopy

betweenn 9:00 am and 2:00 pm, after whole gut irrigation. Patients signed an informed consent

formm before the procedure. Biopsy specimens were taken at a mean of 20 cm (Africans) and

off 26 cm (white patients) from the anal ring. Immediately after the biopsy, the specimens

weree coded and orientated mucosal side up on filter paper and formalin fixed. Four micro

71 1

thickk sections were cut from paraffin wax blocks. Patients with histologically abnormal

mucosaa were excluded.

ImmunohistochemImmunohistochem is tty

Proliferatingg cells were detected, using the MIB1 antibody (DIANOVA GmbH). The MIB1

antibodyy reacts with the Ki-67 nuclear antigen and has the advantage that it can be used in

routinelyy processed, formalin fixed, paraffin wax sections.1

Sectionss were dewaxed, hydrated, and trypsiniscd (0-1% trypsin) for 15 minutes at 37°C.

Whilee in a citrate buffer (pH 6-0), slides were exposed to microwave irradiation for a period

off nine minutes (three times three minutes). Endogenous peroxidase was blocked, using 0.3%

H2O22 in methanol, for 20 minutes. After washing in phosphate buffered saline the sections

weree covered with normal horse serum for 20 minutes. The slides were then incubated in the

primaryy antibody (MIB1, dilution 1: 10) for 60 minutes at 37°C. Antibody detection was

madee with an avidinbiotin horseradish peroxidase complex (VECTASTAIN, ABC) with

diaminobenzidinee as the chromogen. The sections were lightly counterstained with

haematoxylin. .

Quantification Quantification

Onlyy whole length, longitudinal crypts were selected. These were defined as crypts with the

musculariss mucosa at the bottom and the surface at the top. Scoring of nuclei was done

manually,, by only one observer, with the use of a light microscope (X400). Nuclei were

countedd as positive if any nuclear staining was present.

Analysis Analysis

Thee parameters used were: the mean number of labelled nuclei per crypt, the mean number of

totall nuclei per crypt, and the labelling index, which is the ratio of labelled (MIB1 positive) to

thee total (MIB1 positive + MIB1 negative) number of nuclei. In addition, the crypt was

dividedd into five equally sized compartments (from base to surface) and the number of

(labelled)) nuclei in each compartment was calculated. This permitted the calculation of a

labellingg index for each of the five compartments.

StatisticalStatistical analysis

Onee way analysis of variance was used for comparison of the mean values of labelling

indices.. When data were not normally distributed, the Mann-Whitney test was used.

72 2

Results s

PatientsPatients with colorectal cancer

Thirtyy patients (15 South African Africans, 15 South African whites), were selected. The

Africann group was significantly younger. In the Africans four patients had well differentiated

tumourss compared with one in the white group. For seven patients the exact distance of the

mucosaa to the tumour was not known (Table 1).

TableTable I: Demographic and colonic cell proliferation data of Africans and White patients with colorectalcolorectal cancer

Number r Agee (y) (mean (SD)) (range) Sex x Distancee from rumour (mean Duke'ss classification

A A B B C C D D

Differentiation n Well l Moderate e Poor r

Location n Caecum m Ascending g Transverse e Descending g Sigmoid d Rectum m Unknown n

Noo of crypts counted/patient

(SD)) )

meann (SD)) (range) Positivee no of cells/crypt (mean (SD)) (range) Labelingg index (x 100) (mean Labelingg index (x 100) (mean

compartimentt 1 compartimentt 2 compartimentt 3 compartimentt 4 + 5

(SD))) (range) (SD)) )

Africans s

15 5 48(18)(27-79) )

9 M , 6F F 10(8)) (n=12)

0 0 5 5 8 8 2 2

4 4 9 9 2 2

3 3 ----1 1 5 5 4 4 2 2 9(3)) (6-15)

233 (9) (6-42) 155 (5)(6-25)

28(8) ) 29(10) ) 14(8) ) 4(6) )

Whitee patients

15 5 700 (12)(44-86)* 8 M , 6F F

12(9) (n= l l )t t

1 1 4 4 9 9 1 1

1 1 12 2 2 2

2 2 1 1 2 2 --3 3 2 2 5 5 9(2)(6-12) )

244 (9) (13-39)f 16(6)(8-27)t t

344 (14)f 311 (12)t 133 (6)t 2(2)+ +

** o

PatientsPatients without colorectal disease

Theree was no statistical significant difference in age between patients with colorectal cancer

andd controls from the same ethnic group. The indications for endoscopy included microcystic

anaemia,, abdominal pain, diarrhoea, and perirectal bleeding. In one white patient, we found

evidencee of angiodysplasia and four other patients (two Africans, two white patients) had

diverticula.. On histological examination one African patient had melanosis coli and another

showedd an ovum of schistosoma in the section. These abnormalities were thought not to

influencee the cell proliferation data (Table II) .

TableTable 2: Demographic and colonic cell proliferation data of Africans and White patients withoutwithout colorectal cancer

Number r Agee (y) (mean (SD)) (range) Sex x Indicationn for endoscopy Anaemia a

abdominall pain constipation n diarrhoea a losss of weight perirectall bleeding unknown n

Macroscopicc pathology none e angiodysplasia a diverticula a

Microscopicc pathology none e melanosiss coli schistosomiasis s

Timee of biopsy Distancee from anal ring (cm) ( Noo of crypts counted/patient (

meann (SD)) neann (SD)) (range)

Positivee no of cells/crypt (mean (SD)) (range) Labelingg index (x 100) (mean Labelingg index (x 100) (mean

compartimentt 1 compartimentt 2 compartimentt 3 compartimentt 4 + 5

(SD))) (range) (SD)) )

Africans s

15 5 422 (20) (16-75)

7M,, 8 F

8 8 1 1 0 0 1 1 1 1 4 4 0 0

13 3 0 0 2 2

13 3 1 1 1 1

100 15-13 00 20(6) )

9(3)(5-15) ) 7(6)) (3-25) 6(4)(2-18) )

15(7) ) 9(8) ) 4(6) ) 0(1) )

Whitee patients

15 5 666 (13) (30-82)* 4M,, 11 F

2 2 7 7 1 1 3 3 0 0 1 1 1 1

12 2 1 1 2 2

15 5 0 0 0 0

088 55-14 15 26(4)t t 9(3)(5-15) )

13(4)(8-18)* * 111 (3)(7-17)*

23(6) ) 21(8) ) H (6) )

l ( l ) t t pp

Quantitativee results

PatientsPatients with colon cancer

Thee total number of labelled nuclei and the total labelling index do not differ significantly

betweenn the two ethnic groups. Although there is a trend towards a higher mean labelling

indexx of the upper crypt compartments in the mucosa from Africans, it is not statistically

significantt (Table I, Fig 1).

Black pattern s wit h cance r DD Whit e patient s wit h canca r

bas e e

1 1

Cryp tt compartmen t surfac e

22 3 4+5

FigureFigure I: Comparison of labelling index (mean (SD)) in eacheach crypt compartment between the two ethnic groups withwith colon cancer (compartiment I is located at the base ofof the crypt, compartment 5 at the surface).

Inn a subgroup of 23 patients (12 Africans, 11 white patients) in which the distance of the

mucosaa to the turnout was known, the total labelling index of mucosa within a distance of 5

cmm from the tumour, was 19 (+5) for the African patients and 27 (+2) for the white patients.

Labellingg indices from mucosa located more than 5 cm away from the carcinoma was 12 (+4)

andd 15 (+4) respectively, which is significantly lower (p

Patientss without colorectal disease

Tablee II summarises epithelial cell proliferation variables for patients without colorectal

cancer.. The white patients had a significantly higher mean total labelling index than Africans

(phasee II proliferative lesion). In addition, the proliferative pattern of white patients without

evidencee of colorectal cancer, showed a comparatively large amount of dividing cells in

compartmentt 2, compared with the African controls (phase 1 proliferative lesion) (Fig 2).

Thiss is in agreement with the results from other studies. They also showed that patients at

increasedd risk of colon cancer may show a proliferative pattern of the mucosa, which is

characterisedd by phase I and phase II proliferative lesions, in the absence of any colorectal

abnormality. .

SO O

40 0

X X

-SS 30 c c II 20 a a Ji Ji

TableTable IV: Influence of age on epithelial cell proliferation in patients without colorectal disease disease

5.3(3.3) ) 13.7(6.5) ) 7.5(5.6) ) 2.8(3.9) ) 0.33 (0.5) 6 6

11.3(2.3) ) 24.88 (6.2) 20.33 (5.0)

9.22 (4.5) 1.0(0.9) )

8.33 (6.7)* 17.8(8.2)* * 14.8(12.7)* * 7.33 (9.2)* 0.5(1.0)* * 9 9

11.0(4.0) ) 21.6(5.5) ) 21.11 (9.2) 11.7(6.6) ) 0.99 (0.8)

Youngg (>67) Old (>= 67) Africanss (n) 11 4

Totall labeling index (LI) (xx 100)

LII compartment 1 LII compartment 2 LII compartment 3 LII compartment 4 +5

Whitee patients (n) Totall labeling index (LI)

(xx 1000) LII compartment 1 LII compartment 2 LII compartment 3 LII compartment 4 +5

** p=NS. Data shown as mean (SD).

Muchh information about colorectal carcinogenesis comes from studies concerning cell

proliferation.. In 1973, Lipkin described phase I and phase II proliferative lesions in the

mucosaa from patients at increased risk of colorectal cancer. He suggested that neoplastic

lesionss develop in stages: the adenoma-carcinoma sequence. In support of this hypothesis,

clinicall and histopathological data indeed have shown that most, if not all, malignant

colorectall carcinomas arise from pre-existing adenomas." These studies mainly concern the

Westernisedd populations. In this group, colorectal cancer is the second most common cancer.

Inn this study, cell proliferation data were obtained from a group of South African Africans, in

whichh colorectal cancer is one of the lowest in the world. Moreover, macroscopic

adenomatouss polyps are seldom seen. This led to the hypothesis that the adenoma carcinoma

sequencee does not occur in this ethnic groups.12 It was suggested that cancer could have arisen

dee novo or from flat mucosal lesions in apparently normal mucosa. Little is known about the

changess in cell proliferation that occur in de novo carcinogenesis or in carcinomas that arise

fromm flat adenomas. Animal studies suggest that the direction of the shift in the major zone of

proliferationn is related to the type of lesion: downward migration seemed to be related with

intramucosall formed cancers, and upward migration with adenomatous polyps.8 In humans,

downwardd shift would not be evident as such, as the major zone of cell proliferation is already

locatedd at the base of the crypt. Thus, a pattern of cell proliferation would be expected to be

foundd in mucosa from Africans with colon cancer, in which no significant changes in mucosal

celll proliferation is present. In this study, however, the mucosa from Africans with colorectal

77 7

cancerr showed an even more pronounced upward shift of the proliferating cells (phase I

proliferativee lesion) compared with the changes in proliferative pattern, which occur in

mucosaa from South African white patients with cancer (Fig 3). This suggests that colorectal

cancerr in Africans is also preceded by adenomas. We reviewed all slides in an attempt to find

moree evidence of the adenoma carcinoma sequence in Africans. In two tumours from African

patients,, it was obvious that the tumour arose from a tubulovillous adenoma. Another African

patientt had a tubulovillous polyp with severe dysplasia located proximal from the tumour.

Thee white group showed the same percentage of tumours (two of 15, 13%), in which the

originatingg polyp could be found.

Slack patients with cancer CC Black patients without cancer

SO O

«0 0

8 8 00 30 "a "a c c

Labe

lling

g

33

S

n n

--

'i i

White patients with cancer aa White patients without cancer

"" ^ N

11 1 l ^ fU 1 bis » »

1 1

Cryptt compartment surfaca

22 3 4*5

base e

1 1 Cryptt compartment surfac e

22 3 «4-5

FigureFigure 3: Proliferation pattern of (A) Africans and (B) white patients with colon cancer, comparedcompared with controls without colorectal disease. Note the more pronounced phase 1 and phasephase 11 proliferation lesions in the mucosa from Africans with cancer.

Iff the adenoma-carcinoma sequence were a common event, however, it might be expected that

adenomatouss polyps would be more common in African patients. This does not seem to

occur.122 A possible mechanism to consider is that colonic cancer may behave more

aggressivelyy in this ethnic group. In this way, a single precursor lesion, which may only be

1-22 mm. in size, may be rapidly replaced by overt cancer, before being detected. In this study

itt was found that mucosa from Africans with colon cancer showed a more pronounced phase 1

andd phase II proliferative lesions despite a significantly younger age. We suggest that the

mucosall changes in Africans progress from a very low risk pattern (as seen in controls) to a

highh risk pattern in a comparatively short period of time, consistent with a more aggressive

78 8

tumourr behaviour than occurs in white patients. Also other authors suggest a different,

possiblyy more aggressive tumour behaviour in Africans. Michelassi et al found different

karyorypicc characteristics of the neoplastic cells in tumours from Africans.19 Dayal found that

thee African race was an independent risk factor, associated with a poor survival from

colorectall cancer . This study suggests that there are at least two varieties of the

adenoma-carcinomaa sequence: one in which clearly visible adenoma s are gradually replaced

byy malignant tumour (the most likely mechanism in white patients), and one in which a less

easyy detectable precursor lesion (?flat adenoma) is rapidly replaced by overt cancer (proposed

mechanismm in Africans). It is suggested that de novo carcinogenesis is a variant of the

adenoma-carcinomaa sequence. This model is consistent with the hypothesis of Hill et al21 and

Fearonn and Vogelstein , who describe a stepwise model for colorectal tumorigenesis.

Interventionn at various stages of tumour development, either by (in) activation of certain

oncogeness or suppressorgenes,2 or by the absence or presence of environmental factor A, B,

orr C," could explain which variety of adenoma-carcinoma sequence occurs. Studies

evaluatingg the genetic basis of colorectal cancer in low risk communities may uncover

importantt data and assist in elucidating clues to colorectal cancer pathogenesis.

References s

1.. Risio M, Coverlizza S, Ferrari A, Candelaresi GI, Rossini FP. Immunohistochemical

studyy of epithelial cell proliferation in hyperplastic polyps, adenomas and

adenocarcinomaa of the large bowel. Gastroenterology 1988; 94: 899-906.

2.. Fearon ER, Vogelstein B. A genetic model for colorectal rumorigenesis Cell 1990; 61:

759-67. .

3.. Lipkin M. Phase I and phase 2 proliferative lesions of colonic epithelial cells in diseases

leadingg to colonic cancer. Cancer 1974; 34 (suppl): 878-88.

4.. Lane N. The precursor tissue of ordinary large bowel cancer. Cancer Res 1976; 86:

2669-72. .

5.. O'Brien MJ, O'Keane iC, Zauber A, Gottlieb LS, Winawer SJ. Precursors of colorectal

carcinoma,, biopsy and biologic markers. Cancer 1992; 70 (suppl): 1317-27.

6.. Hamilton SR. The adenoma-adenocarcinoma sequence in the large bowel: variations on

aa theme. J Cell Biochem 1992; 16G (suppl): 41-6.

79 9

7.. Maskens AP, Dujardin-Loits RM. Experimental adenomas and carcinomas of the large

intestinee behave as distinct entities: most carcinomas arise de novo in flat mucosa.

Cancerr 1981;47:81-9.

8.. Deschner EE, Maskens AP. Significance of the labelling index and labelling distribution

ass kinetic parameters in colorectal mucosa of cancer patients and DMH treated animals.

Cancerr 1982; 50: 1136-41.

9.. Shamsuddin AKM, Weiss I, Phelps PC, Trump BF. Colon epithelium IV: human colon

carcinogenesis.. Change in human colon mucosa adjacent to and remote from

carcinomass of the colon. J Natl Cancer Inst 1981; 66. 413-9.

10.. Burkitt DP. Epidemiology of cancer of the colon and rectum. Cancer 1971; 28: 3-13.

11.. Bremner CG, Ackcrman LV. Polyps and Carcinoma of the large bowel in the South

Africann Bantu. Cancer 1970; 26: 991-9.

12.. Segal I, Cooke SAR, Hamilton DG, Ou Tim L. Polyps and colorectal cancer in South

Africann Africans. Gut 1981; 22: 653-7.

13.. Key G, Becker MHG, Duchrow M, Schluter C, Gerdes J. New Ki-67 equivalent murine

monoclonall antibodies (MIB 1-3) generated against recombinant parts of the Ki-67

antigen.. Anal Cell Pathol 1992; 4: 18 1.

14.. Kyzer S, Mitmaker B, Gordon PH, Schipper H, Wang E, Proliferative activity of

colonicc mucosa at different distances from primary adenocarcinoma as determined by

thee presence of statin: a non-proliferation-specific nuclear protein. Dis Colon Rectum

1992;35:879-83. .

15.. Roncucci I, Ponz de Leon M, Scalmad A, Pratissoli S, Perini M, Chahin NJ. The

influencee of age on colonic epithelial cell proliferation. Cancer 1988; 62: 2373-7.

16.. Deschner EE, Godbold J, Lynch HT. Rectal epithelial cell proliferation in a group of

youngg adults. Influence of age and genetic risk for colon cancer. Cancer 1988; 61:

2286-90. .

17.. Rozen P, Fireman Z, Fine N, Cheaft A, Lubin F. Rectal epithelial proliferation of first

degreee relatives of sporadic colon cancer patients. Cancer Lea 1990; 51: 127-32.

18.. Risio M. Cell proliferation in colorectal tumor progression: an immunohistochemical

approachh to intermediate biomarkers. J Cell Biochem 1992; 16G (suppl): 79-87.

19.. Michelassi F, Vannucci LE, Montag AG, Dytch HE, Bibbo M. Nuclear morphometric

measurementss in rectal adenocarcinoma cells of patients of different races Anal Quant

CytolHistoll 1989; 11: 173-6.

80 0

20.. Dayal H, Polissar I, Yang CY, Dahlberg S. Race, socioeconomic status, and other

prognosticc factors for survival from colo-rectal cancer. J Chron Dis 1987; 40: 857-64.

21.. Hill MJ, Morson BC, Bussey HJR. Aetiology of adenoma carcinoma sequence in large

bowel.. Lancet 1978; i: 245-7.

81 1