Rectal proliferation and polyp occurrence in patients with familial adenomatous polyposis after sulindac treatment

GASTROENTEROLOGY 1994;106:362-366

Recta1 Proliferation and Polyp Occurrence in Patients With Familial Adenomatous Polyposis After Sulindac Treatment MARIA TERESA SPAGNESI,* FRANCESCO TONELLI,’ PIER0 DOLARA,* GIOVANNA ROSA VALANZANO,’ ALESSANDRO ANASTASI,’ and FRANCA BIANCHINI* *Department of Pharmacology and Toxicology and ‘Surgical Unit, Department of Clinical Physiopathology, Florence Italy

Sulindac, a nonsteroidal anti-inflammatory drug (NSAID), decreases the occurrence of polyps in patients with familial adenomatous polyposis (FAP). The effects of colectomy with ileorectal anastomosis (IRA) and sulindac treatment on recta1 mucosa proliferation and polyp occurrence were examined in patients with FAP. Methods: The number and size of recta1 polyps were measured with colonoscopy. The labeling index, the percentage of labeled cells per crypt compartment, was assessed in recta1 biopsy specimens with [3H]thymidine incorporation and autoradiography in 6 non-IRA and 14 IRA patients before and after treatment with 200 mg of sulindac/day for 60 days. Results: The IRA patients had a lower labeling index and a decrease in the percentage of labeled cells in the upper compatiment of the crypt (P < 0.01) relative to non-IRA sub jects. Sulindac did not influence the labeling index and the distribution of labeled cells along the crypt. On the contrary, a dramatic decrease in the size and number of polyps was observed after sulindac treatment (P < 0.001). Conclusions: The persistente of a abnormal mucosal proliferation after sulindac therapy, in spite of the reduction of polyp number, suggests caution in assuming a lower risk of recta1 cancer in patients with FAP. Bac/@ound/Aims:

F

amilial adenomatous polyposis (FAP), if left unattended, inevitably progresses to colorectal cancer. The current surgical treatments of this condition are total colectomy with ileoanal pul1 or ileorectal anastomosis (IRA). The latter

is a simpler

technique

and guarantees

satisfactory functional results in properly selected subjects. However, after IRA, patients with FAP must undergo lifelong surveillance of the recta1 stump for the early detection of new polyps and cancer.‘.’ It has been reported that after IRA, one third of patients with FAP have a complete spontaneous regression of recta1 polyps and lower frequenties of neoplastic transformation when compared with patients affected by the same disease who did not undergo 1RA.s However, surgical treatment with IRA does not annul the risk of colon cancer.” In fact, patients with FAP are characterized by a state of hyperproliferation in the colonic mucosa and by a shift of the proliferative zone to the upper portion

of the colonic

CADERNI,”

University of Florence,

crypts,5.” factors wel1 correlated

risk of cancer development.‘-”

We previously

with the showed

that after IRA, the proliferative pattern of the recta1 stump regresses to normality, but some indicators of an abnormal putrescine a 13%

control

of cel1 division,

and spermine, incidence

reported

for patients

of colon

subjects,2

cancer

after IRA

has been

with FAP in the 25 years following

surgery. ‘JI This frequency non-IRA

such as the polyamines

remain elevated.‘> Accordingly,

is lower than that recorded

but the individual

in

risk of developing

cancer is far from negligible. A few years ago it was reported that some nonsteroidal anti-inflammatory drugs (NSAIDs) were able to control the growth of desmoid tumors in patients affected by FAP or Gardner’s syndrome.‘“-‘” It was later shown that sulindac(ris-5-fluoro-2-methyl-l-~p-(methylsulfynil)benzylidenel

indene-3-acetic

acid), a widely

can also cause polyp regression.‘0-2i ies in patients 6-12 months sulindac

used NSAID,

In one of these stud-

with FAP, al1 polyps disappeared during of therapy with a dose of 300-400 mg

per day.2’ In other recent

studies,

a reduction

of the number and size of adenomas was reported a long-term therapy with sulindac.‘2B” This series of reports has induced considerable in colon cancer chemoprevention with NSAIDs,

after

interest also en-

couraged by the observation of a significant reduction colon cancer incidence in frequent aspirin users.”

in

We decided, therefore, to make additional measurements of mucosal proliferative activity in patients with FAP, before and after surgery, and to correlate the mucosal proliferation with the occurrence of polyps in the recta1 mucosa under the effect of sulindac.

Materials and Methods Patients The study was conducted on patients with FAP who either had no surgery or had undergone subtotal colectomy Abbreviations used in this paper: FAP, familial adenomatous polyp osis; IRA, ileorectal anastomosis. 0 1994 by the American Gastroenterological Association 0016-5065/94/$3.00

PROLIFERATION AND POLYPS AFTER WLINDAC

February 1994

and IRA. reactions

We excluded to NSAIDs

patients

or peptic

with

ulcers; patients

counts

<3500/mm’,

platelet

count

atinine

leve1 > 1.5 mg/dL; and pacients

had usjed NSAIDs

in the preceding

sent was obtained

from al1 subjects.

The eligible

patients

1 was composed

adenomatous

neal part of the rectum

subtotal

was performed

were treated

hemorrhoidal

with a Aexible Olympus saline solution. informed formed

and l-

sigmoidoscope after cleaning

about the therapeutic

was performed

(Olympus

Optica1 Co.,

the recta1 ampulla

of about

with a diameter surgery,

with a

of the patients,

the number

> 0.5 mm. In the group Houston

in the

valve at a distance

10 cm from the anal verge; in the IRA patients,

the polyps was calculated reference.

taking

were counted.

al1

The size of

the open biopsy forceps as a

At least three recta1 mucosal biopsy specimens

taken from each subject

at a mean distance

sections

previous

samples dently

following

for each patient

study,

by using

for human

Statistical

mucosal

group.

in vitro and

tween groups Statistical

On a set of 43 scores of labeling

subjects

side up. After processing the number

treatment

in a

both

in

were oriented each sample

for

of labeled cells in the crypt, the

of the labeled cells along the crypt, and the total of cells scored per crypt were measured. Data on cel1 were expressed

as labeling

index (number

of la-

beled cells in the crypt divided by number of cells scored times 100)~. Each crypt was also divided into three equal compartments (lower, middle, and upper), and the distribution of proliferating cells along the crypt was expressed as the percentage (of labeled cells in each compartment over the total labeled cells in the crypt section.

by two observers,

we had a

of 0.87 (P < 0.001).

with classica1 analysis of variante

Statistical

Package

(Statistical

MD) and for nonparametric

with the normal

Package

Graphic

comparisons

be-

score exact test using Statxact

(Cytel, Cambridge,

MA).

Results

with a pro-

ac least three specimens

The specimens

The

Analysis

Statgraphic

Figure we studied

every individual

were

specimens

methods,5-7

before and after sulindac

the IRA and non-IRA

proliferation

coefficient

to obtain

codes and were read indepen-

of the same specimens

correlation

crypt

and the counts

before and after therapy.

unknown

by two researchers.

Corp., Rockville,

in colon biopsy described

has been standardized

In the present

position number

was assessed

incorporation

autoradiography,

proliferation

were given

indices

work by US.‘>

vertically

were averaged

value of mucosal

of 7- 10 cm from

of Proliferation

proliferation

autoradiography, that

a mean of 20 -+ 1.4 (SE) longitudinal

was scored for each biopsy specimen,

Data were analyzed

Mucosal

cedure

IRA

of three specimens

using

[‘H)t:hymidine

NON-IRA

flgure 1. Individual values of proliferative actlvity (0) of recta1 mucosa in non-IRA (n = 6) and IRA (n = 14) patients with FAP before and after therapy.2, mean values 5 SE. **P = 0.01, non-IRA patients before therapy vs. IRA patients.

the anal verge.

Measurements

efter

per-

and the size

the polyps were counted

in the recta1 stump

before

therepy

For al1 subjects,

below the superior

polyps present

after therepy

(A.A.), who was not

regimen

The observer was asked to quantify

recta1 ampulla

1

(100

al1 sigmoidoscopies.

that did not undergo

2 bsfare

to sigmoidoscopy

The examination

One of the investigacors

of al1 recta1 polyps

4

vessels.

with sulindac

mg orally twice a day) and were subjected

Germany)

colectomy

on the extraperito-

for 2 months

before and after the treatment. Hamburg,

colorectal 2 was com-

10 cm from the anal verge, preserving

the blood flow from the superior Al1 patients

Group

after surgery was 6.2 years (range,

20 years). The anastomosis

14 i

Group

(9 male, 5 female; mean age, 27.7 years

years)) who underwent

IRA. The mean interval

16

years)); the diag-

of numerous

polyps and on family history.

{range, 12-49

or con-

16

with FAP (2 male, 4

12 -23

on the observation

posed of 14 patients

Informed

into cwo groups.

patients

female; mean age, 17.2 years [range, nosis was based

and cre-

who were pregnant

3 months.

were divided

of 6 non-IRA

l

with leukocyte

< 100,000/mm’,

363

XY

of allergie

a history

IN FAP

mucosa sulindac cantly patients

1 shows

the labeling

in non-IRA

and

IRA

treatment.

The

labeling

reduced

in IRA

(P = 0.01).

indices

patients

indices

patients

compared

A similar

result

of the recta1

before were with

and

after

signifinon-IRA

was obtained

by

our group previously in a study of 17 patients with ulcerative colitis before and after IRA.” In the present study, the non-IRA patients were slightly younger than the IRA patients (mean age, 17.2 vs. 27.7 years). However, because in the preceding study15 the age of the two groups was perfectly balanced, the observed decrease of proliferation after IRA of the present set of data is most likely caused by surgery and not by a slight variation in age. Besides, age would be expected to increase prolifera-

364

SPAGNESI ET AL.

GASTROENTEROLOGY Vol. 106, No. 2

We also measured the cellularity of the crypts, expressed as number of cells per crypt column, but we did not observe I

a

non-IRA

b’

a significant

patients

differente

between

or after sulindac

IRA

treatment

and

(data not

shown).

In

Contrary to the effects on cel1 proliferation, we observed a dramatic reduction in the number of polyps (Table

1). Nevertheless,

significantly

the size of the polyps

was not

varied.

Discussion

a

Most published

b

that changes ment

scientific

reports

in cel1 proliferation

have suggested

precede

of colon cancer both in humans

the develop-

and experimental

animals. Many studies have focused on cellular kinetics in colon mucosa with the aim of discovering early markers of neoplastic

X*

‘a

b’

nn

ments

transformation

and of devising

Accordingly,

patients

index and a displacement

with FAP have a higher labeling of the proliferation

surface of the colon crypts than normal macroscopically factors

normal rectosigmoid

have been associated

with

Figure 2. Distribution of percentage of labeled cells along the crypts of recta1 mucosa of non-IRA (a) and IRA (b) patients with FAP before ) therapy. Data are expressed as mean ? SE; **P -C 0.01, non-IRA vs. IRA patients.

and IRA patients.

even in

an increased activity

risk of is consid-

ered a protective factor against colon cancer development, as documented by the studies on Seventh-Day Adventist

CRYPTUIMPARTENTS

non-IRA

towards the

subjects,

mucosa. Both these

colon cancer.’ On the other hand, low proliferative

tion. Figure 1 also shows that sulindac erratic results on individual proliferation

treat-

that reduce the risk of colon cancer.5m”’

treatment had rates both in

In some patients,

we ob-

vegetarians,

a group with low mortality

from colon can-

25

cer. Our study shows that patients with FAP have a high proliferation rate and a displacement of the proliferative zone towards the surface of the crypts compared normal subjects. The present study also indicates IRA tends to revert to normality of patients with FAP. We previously

reported

the proliferation

that also patients

with that

pattern

with ulcera-

served a decreased proliferation and in others the reverse. The average effect of the treatment was, therefore, null. In the same subjects, we also studied the pattern of proliferation along the crypt, which is reported in Figure 2 as percentage of labeled cells in three crypt compartments.

tive colitis after IRA show a reduction in the labeling index of the recta1 stump and a change of the proliferation pattern, characterized by an increase in the percent of proliferating cells in the lower sections of the crypts.15

It was shown in fact that patients with FAP have a shift of proliferation towards the upper compartments of the crypt.5’6*‘5 In the present study, the percent distribution of labeled cells in the three compartments was increased in the lower compartment and decreased in the upper compartment in IRA patients compared with non-IRA patients (P < O.Ol), confirming previous results’> and showing that IRA normalizes the proliferative pattern of the mucosa and restores the physiological decrease of proliferation towards the surface of the crypt. Figure 2 also shows that sulindac did not modify the distribution of labeled cells along the crypt.

The mechanisms underlying these mucosal changes after IRA in patients with FAP and ulcerative colitis are stil1 not totally clear but are likely the results of variations in intestinal function and content (such as changes in short-chain fatty acids, pH, bile acids, and transit time), factors that modulate mucosal proliferative activ8,26-28 ity in experimental animals and humans. In the present study, we confirmed the previous observations that IRA tends to normalize the labeling index and proliferation patterns of patients with FAP. However, the treatment with sulindac had variable effects, with some patients having decreased proliferation and

PROLIFERATION AND POLYPS AFTER SLILINDAC IN FAP

February 1994

Table1. Effects of Sulindac Treatment

on Recta1 Polyps in Patients

Before and After Surgery

Before therapy

After therapy

No. of polyps Before surgery After surgety

Size (mm)

13.5 -t 7.8 (6-30) 10.08 % 8.05 (0-30)

365

No.

0.9 t 0.6 (0.5-2) 1.0 ? 0.57 (0.5-2)

of polyps

Size (mm)

2.5 2 2.4” (0-7) 2.53 t 3.4” (0.5-10)

0.5 i 0.4 (0-1) 0.57 2 0.54 (0.5-1.5)

NOTE. Results are expressed as mean 2 SD with range in parentheses. Sulindac treatment, 200 mg/day for 60 days. “P i 0.01, compared with the number of polyps in the same patients before therapy, using the normal score exact test.

others the reverse. As a consequente,

the average prolifer-

croscopical

ation rate did not vary significantly

after sulindac

sarily involve an intermediate

ment., neither

was the displacement

wards, higher

crypt

duration

the dosage

lower

of proliferation

levels influenced.

of the treatment than

treat-

In our study,

to-

early steps of the transformation

the

proliferation,

with sulindac

was shorter and

in some previous

reports’“.“.“;

in our study

was likely effective.

because it induced a very dramatic decrease in the expression of polyps, which totally disappeared under the effect of the drug in some of the patients before surgery. A regression and suppression of polyps and micropolyps in patients with FAP has been described after sulindac by other investigators There

as well.‘“-”

was an apparent

lack of association

between

cel1 proliferation and the occurrence of macroscopically evident polyps in our study. The reason for this dissociation is not clear. It has been suggested

that the disorder

in cel1 proliferation is under very strict in patients with FAP2’~‘o; it is possible

genetic control that the effects

of sulindac are thwarted by this factor. On the other hand, although NSAIDs black proliferation in mammalian

cel1 culture

with aspirin the rat.” Tlnerefore,

lines

increases it

in vitro,” proliferation

is possible

that

long-term

treatment

of colon

mucosa

sulindac

and

in

other

NSAIDs control some process leading to the formation of polyps that is independent of cel1 proliferation. For example, sulindac, being a good inhibitor of prostaglandin synthesis, tenance

may interfere

of the stroma

with the formation

and vessel structure

and main-

necessary

sulindac,

for

the polyps to grow. Whereas our data do not indicate an inhibitory effect of sulindac on colon cel1 proliferation, they show a clear inhibitory effect on macroscopically evident polyps. Wh’ether this effect is associated over long term with a lower risk of expression of cancer remains to be shown. A reduction in risk can be anticipated on the assumption of the existente of a multistep process leading from microa.denomas to polyps and eventually to adenomas and adenocarcinomas.1’.‘4 However, the transition from mi-

cancers may not necespolyp :stage. Because some process, such as altered

do not seem to respond

caution

that a reduction

however, other authors have documented a beneficial effect of sulindac in FAP also at lower dosage.‘” Moreover, the dosage of sulindac

lesions to full-grown

should

to the therapy with

be recommended

in the number

in assuming

of polyps after sulindac

is proof of a lower risk of cancer in patienrs

with FAP.

References 1. Heimann TM, Bolnick K, Aufses HA. Results of surgical treatment for familial polyposis coli. Am J Surg 1986; 152:276-278. 2. Bussey HJR, Eyers AA, Ritchie SM, Thornson JPS. The rectum in adenomatous polyposis: the St. Mark’s policy. Br J Surg 1985;72:29-35. 3. Heimann TM, Greenstein AJU,Bolnick K, Yoelson S, Aufses AH. Colorectal cancer in familial polyposis coli and ulcerative colitis. Dis Colon Rectum 1985; 28:658-661. 4. Nugent KP, Philips RK. Recta1 cancer risk in older patients with familial adenomatous polyposis and an Ileorectal anastomosis: a cause for concern. Br J Surg 1992;79:1204-1206. 5. Deschner EE, Lipkin M. Proliferative patterns in colonic mucosa in familial polyposis. Cancer 1975;35:413-418. 6. Deschner EE, Lipkin M. Study of human recta1 epithelial cells in vitro. RNA, protein, and DNA synthesis in polyps and adjacent mucosa. J Natl Cancer Inst 1970;40:1.75-185. 7. Lipkin M. Phase one and phase two proliferative lesions of colonic epithelial cells in diseases leading to colonic cancer. Cancer 1974; 34:878-888. 8. Cats A, Kleibeuker JH, Kuipers F. Hardonk MJ, Verschueren RCJ, Boersma W, Vonk RJ, Sluiter WJ, Mulder NH, Wolthers BG, de Vries GE. Changes in recta1 epithelial cell proliferation and intestinal bile acids after subtotal colectomy in familial adenomatous polyposis. Cancer Res 1992;52:3552-3557. 9. Terpstra TO, Blankenstein M, Dees J. ErIers GAM. Abnormal pattern of cell proliferation in the entrre colonic mucosa of patients with colon adenoma or cancer. Gastroenterology 1987;92:704708. 10. Lrpkin M, Blattner A, Gardner EJ, Bun RW, Lynch H, Deschner E. Winawer S, Fraumeni JF. Classlfication and risk assessment of individuals with familial polyposis, Gamer’s syndrome and familial non-polyposis colon cancer from “Hthymidine labeling patterns in colonic epithelial cells. Cancer Res 1984;44:42014207. Deschner EE, Lipkin M, Solomon C. Study of human recta1 epithell. lial cells in vitro. %thymidine Incorporatron into polyps and adjacent mucosa. J Natl Cancer Inst 1966;36:849-855. 12. Maskens AP, Deschner EE. Tritiated thymidine incorporation into epithelial cells of normal-appearing colorectal mucosa of cancer patients. J Nat1 Cancer Inst 1977;58:1221-1224. 13. Lipkin M, Blattner WE, Fraumeni JF, Lynch JF, Deschner EE, Winawer S. Tritiated thymidine labeling distributron as a marker for

366

14.

15.

16.

17. 18.

19.

20.

21. 22.

23.

24. 25.

GASTROENTEROLOGY Vol. 106, No. 2

SPAGNESI ET AL.

hereditary predisposition to colon cancer. Cancer Res 1983;43:1899-1904. Scalmati A. Roncucci L, Ghidini G. Biasco G, Ponz de Leon M. Epithelial cel1 kinetics in the remaining colorectal mucosa after surgery for cancer of the large bowel. Cancer Res 1990; 50:7937-7941. Tonelli F, Elianchini F, Lodovici M, Valanzano R, Caderni G, Dolara P. Mucosal cell proliferation of the recta1 stump in ulcerative colitis patients after ileorectal anastomosis. Dis Colon Rectum 1991; 34:385-390. Klein WA, Miller HH, Anderson M, Decosse JJ. The use of indomethacin, sulindac and tamoxifen for the treatment of desmoid tumors associated with familial polyposis. Cancer 1987; 6012863-2868. McAdam WAF, Goligher JK. The occurrence of desmoids in patients with familial polyposis coli. Br J Surg 1970; 149:215-218. Jones IT, Fazio VW, Weakley FL, Jagelman DG, Lavery IC, McGannon E. Desmoid tumors in familial polyposis coli. Ann Surg 1986;204:94-97. Belliveau P, Graham AM. Mesenteric desmoid tumor in Gardner’s syndrome treated by sulindac. Dis Colon Rectum 1984; 27:5354. Laybayle D, Fischer D, Vielh P, Drouhin F, Pariente A, Bories C, Duhamel 0, Trousset M, Attali P. Sulindac causes regression of recta1 polyps in familial adenomatous polyposis. Gastroenterology 1991; 101:635-639. Waddel RW, Ganser GF, Cerise EJ. Loughry RW. Sulindac for polyposis of the colon. Am J Surg 1989; 157:175-179. Winde G, Gumbinger HG, Osswald H, Kemper F, Bunte H. The NSAID sulindac reverses recta1 adenomas in colectomized patients with familial adenomatous polyposis: clinical results of a dose finding study on recta1 sulindac administration. Int J Coloreet Dis 1993;8:13-17. Giardello FM, Hamilton SR, Krush AJ, Piantodosr S, Hylind LM, Celano P, Booker SV, Rahj Robinson C, Offerhaus GJA. Treatment of colonic recta1 adenomas with sulindac in familial adenomatous polyposis. N Engl J Med 1993;328:1313-1316. Thun MJ, Namboodiri MM, Heath CW. Aspirin use and reduced risk of fatal colon cancer. N Engl J Med 1991;325:1593-1596. Lipkin M, Uehara K, Winawer S, Sauchez A, Bauer C, Philips R, Lynch H, Blattner WA, Fraumeni JF. Seventh-day adventist

26.

27.

28.

29.

30.

31.

32.

33. 34.

vegetarians have a quiescent proliferative activity in colonic mucosa. Cancer Lett 1985;26:139-144. Stadler J, Stern H. Yeung KS, Mcguire V, Furrer R, Marcon N. Brute WR. Effect of high fat consumption on cell proliferation activity of colorectal mucosa and on soluble faecal bile acids. Gut 1988; 29:1326-1331. Lupton JR. Coder DM, Jacobs LR. Long-term effects of fermentable fiber on rat colonic pH and epithelial cel1 cycle. J Nutr 1988; 118:840-845. Sakata T. Stimulatory effect of shortchain fatty acids on epithelial cell proliferation in the rat intestine: a possible explanation for trophic effects of fermentable fibre, gut microbes and luminal trophic factors. Br J Nutr 1987;58:95-103. Erisman MD, Scott JK, Astrin SM. Evidente that the familial adenomatous polyposis gene is involved in a subset of colon cancer with a complementable defect in c-myc regulation. Proc Natl Ac Sci 1989;86:4264-4268. Sasaki M, Okamoto M, Sato C, Sugio K, Soejima J, Iwama T, Ikeuchi T, Tonomura A, Miyachi M, Sasazuchi T. Loss of constitutional heterozygosity in colorectal tumors from patients with familial polyposis coli and those with non-polyposis colorectal carcinoma. Cancer Res 1989;49:4402-4406. Hial V, De Mello MCF, Horakova Z, Beaven MA. Antiproliferative activity of antiinflammatory drugs on two mammalian cell clones. J Pharmacol Exp Ther 177;202:446-454. Craven PA, Thornburg K, DeRubertis FR. Sustained increase of rat colonic mucosa during chronic treatment with aspirin. Gastroenterology 1988;94:567-575. McLellan EA, Bird RP. Aberrant crypts: preneoplastic lesions in the murine colon. Cancer Res 1988;48:6187-6192. Roncucci L, Medline A, Brute WR. Classification of aberrant crypt foei and microadenomas in human colon. Cancer Epidemiol Biomarkers Prevention 1991;1:57-60.

Received March 3, 1993. Accepted September 21, 1993. Address requests for reprints to: Piero Dolara, M.D., Department of Pharmacology and Toxicology, University of Florence, Viale Morgagni 65, 50134 Florence, Italy. Fax: (39) 554361613. Supported by a grant from Consiglio Nazionale delle Ricerche, fïnaliuato A.C.R.O., paper 267.