The Epidemiology of Cancer of the Small Bowel

Vol. 7, 243-251, March /998 Cancer Epidemiology, Biomarkers & Prevention 243

Review

The Epidemiology of Cancer of the Small Bowel

Alfred I. Neugut,’ Judith S. Jacobson, Sung Suh, Introduction Raja Mukherjee, and Nadir Arber Most biomedical research reports and review articles begin by Department of Medicine, College of Physicians and Surgeons, Columbia invoking the high incidence, mortality, or morbidity associated University, New York, New York 10032 [A. I. N., N. A.]; Division of with the disease under consideration. Cancer of the small bowel Epidemiology. School of Public Health, Columbia University, New York, New has a claim on our interest for the opposite reason. Although York [A. I. N., J. S. J., S. S., R. M.]; and Department of Gastroenterology, Tel located between the stomach and the large bowel, two of the Aviv Medical Center, Tel Aviv, Israel [N. A.] world’s most common cancer sites, the small bowel rarely develops a malignancy. However, in the United States and other countries in recent years, incidence rates of this cancer have Abstract been rising. Despite its anatomical location between two regions of Contrary to what its name implies, the small bowel con- stitutes 75% of the length of the alimentary tract and 90% of its high cancer risk, the small bowel rarely develops a mucosal surface area; yet in most industrialized countries, it is malignant tumor. However, in recent years, small bowel the site ofless than 5% of G12 malignancies (1-5). In the United cancer incidence rates have begun to rise. The purpose of States, 13 1 ,200 large bowel cancers and 22,400 stomach can- this review is to explore the descriptive and analytic cers but only 4,900 small bowel cancers are expected to have epidemiology of small bowel cancer for those factors that been diagnosed in 1997 (6). Clearly, the small bowel has some protect this organ and those factors associated with loss protection against cancer that adjacent organs lack. Studying of this protection. this protection and its failures may lead to preventive strategies Within the small intestine, the sites at the highest against cancer in other organs of the GI tract or elsewhere in the risk are the duodenum, for adenocarcinomas, and the human body. ileum, for carcinoids and lymphomas. In industrialized countries, small bowel cancers are predominantly Clinical Characteristics adenocarcinomas; in developing countries, lymphomas are much more common. The incidence of small bowel Small bowel cancer has been the subject of only six population- based descriptive epidemiological studies, each covering at cancer rises with age and has generally been higher least a decade’s accumulation of cases (2, 7-1 1). Hospital case among males than among females. series, also accumulated over decades, are therefore an impor- The risk factors for small bowel cancer include tant source of descriptive data about this rare cancer (3, 4, dietary factors similar to those implicated in large bowel 12-37). Although not population-based, these reports collec- cancer, cigarette smoking, alcohol intake, and other tively describe cases seen in hospitals around the world, and medical conditions, including Crohn’s disease, familial most of them provide more detail than is available from most adenomatous polyposis, cholecystectomy, peptic ulcer registries. disease, and cystic fibrosis. The protective factors may Approximately two-thirds of small bowel tumors are ma- include rapid cell turnover, a general absence of bacteria, lignant; more than 95% of these are adenocarcinomas, carci- an alkaline environment, and low levels of activating noids, lymphomas, or sarcomas. As Tables 1-3 indicate, the enzymes of precarcinogens. histology of these tumors is highly correlated with the anatomic Adenocarcinomas of the small and large bowel are subsite in which they arise. Adenocarcinomas, the most com- similar in risk factors and geographic distribution but not mon histological type in most Western populations, are pre- in recent time trends; colorectal cancer incidence rates in dominantly duodenal; carcinoids and lymphomas are predom- the United States have been falling since the mid-1980s. inantly ileal or jejunal. Sarcomas are more evenly distributed Small bowel lymphoma may be associated with infectious throughout the small bowel. agents, such as HIV. Given the differences in anatomic One reason why adenocarcinomas tend to arise in the duo- and geographic location among histological subtypes, denum may be that it is close to the ampulla of Vater. Although much may be learned from well-designed, histology- ampullary carcinomas are usually classified as tumors of the ex- specific epidemiological and genetic studies of cancer of trahepatic biliary tract rather than the small bowel, duodenal ad- the small bowel. enocarcinomas tend to cluster in the periampullary region. This clustering may implicate bile or its metabolites in the etiology of adenocarcinomas at this site (1 , 2, 38). However, among patients with Crohn’s disease, which generally affects the ileum rather than the more proximal small bowel, adenocarcinomas tend to occur in Received 7/25/97; revised I 2/8/97; accepted 12/18/97. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 2 The abbreviations used are: GI, gastrointestinal; SEER, Surveillance, Epidemi- I To whom requests for reprints should be addressed, at Division of Oncology, Columbia-Presbyterian Medical Center, 630 W. 168th Street, New York, NY ology and End Results; OR, odds ratio; CI, confidence interval; FAP, familial I0032. adenomatous polyposis; RR, risk ratio.

Table I Distribution of small bowel cancer b y histology. anatomic su bsite, and sex in po pulation-base d studies

Anatomic subsite b Sex Incidence Population/period (Ref.) Histology Total . y . Histology Duodenum Jejunum Ileum NOS” Male Female million

Los Angeles County” Adenocarcinoma 213 108 78 47 446 37.5 209 237 1972-1985 (2) Carcinoid 22 43 307 131 503 42.3 204 299 Sarcoma 22 45 35 50 152 12.8 54 98 Lymphoma 7 17 37 28 89 7.5 34 55 Total 264 213 457 256 1190 501 689

Nine SEER registries’ Adenocarcinoma 310 208 123 91 732 40.0 384 348 3.9 1973-1982 (7) Carcinoid 19 36 361 126 542 29.6 277 265 2.9 Sarcoma 26 70 51 85 232 12.7 119 113 1.2 Lymphoma II 72 126 103 312 17.0 193 119 1.6 Total 366 386 661 405 1832 973 845 9.6

Cancer registries of British Adenocarcinoma 215 115 63 128 521 41.9 Columbia. Alberta, Saskatchewan, Carcinoid 9 21 197 107 334 26.8 and Manitoba” Sarcoma 14 36 27 63 140 11.3 1975-1989 (8) Lymphoma 25 35 59 125 244 19.6 Total 263 207 346 423 1 244 682 562 1 1.0

Utah Cancer Registry’ Adenocarcinoma 29 16 13 22 80 24.4 53 27 3.0 1966-1990 (9) Carcinoid 5 17 63 51 136 41.5 84 52 6.5 Sarcoma 3 5 II 17 36 11.0 24 12 1.5 Lymphoma 5 17 20 30 72 22.0 48 24 2.5 Total 42 55 107 120 328 209 115 14.0

Nine SEER registries Adenocarcinoma 777 376 251 205 1609 842 767 6.5 1973-1991 (10) Carcinoid 114 125 951 493 1683 888 795 6.5 Total 891 501 1202 698 3292 1730 1562 13.0

a NOS, not otherwise specified.

I, Total includes 20 tumors with other histologies.

, Total includes 14 tumors with other histologies.

‘I Total includes 5 tumors with other histologies. Total includes 4 tumors with other histologies. I Figures exclude lymphomas and sarcomas.

the ileum (39, 40), suggesting that inflammation may increase risk populations. This correlation has become weaker in the past for these cancers. two decades.3 In most industrialized countries, small bowel lymphoma is Sex Differences. In most population-based registries, males relatively rare, arises mainly in the ileum, resembles the other have higher small bowel cancer incidence rates than females histological types of small bowel cancer in its association with (50). In three of four United States population-based studies, age, and has relatively good survival (41-43). However, in the male to female ratio of cases was elevated for all four main other parts of the world, lymphoma is the most common type of histological subgroups (Refs. 2, 7, 9, 10; Table I). In this small bowel cancer, may arise anywhere in the small bowel, is respect, small bowel cancer is similar to colorectal and stomach a disease of young adults, and has relatively poor survival (14, cancer (Ref. S 1 ; Table 4). 44-48). In industrialized countries, lymphoma of the small Age Distribution. Like the incidence of colorectal and stom- bowel is often grouped with other lymphomas and excluded ach cancer, the incidence of small bowel cancer rises with age from small bowel cancer studies. The small bowel lymphomas (1). The mean age at diagnosis is typically about 60 ± 10 years of developing countries and the Mediterranean region are often (Table 2) and is largely consistent across histological subtypes associated with immunoproliferative small intestinal disease in Western populations. However, the age at diagnosis of ad- (49). enocarcinoma tends to be somewhat older and that of lymphoma somewhat younger than those of other subtypes. Descriptive Epidemiology Race Distribution. Few data are available on the race distri- Geographical Distribution. The incidence rate of small bowel bution of small bowel malignancies within countries. Analyses cancer varies among populations. The geographic distribution of SEER data (7, 10, 1 1) and a review of the Los Angeles of this cancer differs from those of both stomach cancer (high County Cancer Surveillance Program (2) indicate a somewhat in Asian and Latin American countries and low among United higher incidence of adenocarcinoma and malignant carcinoid States whites) and large bowel cancer (high in the United States tumors in blacks (4. 1 and 4.6 per million, respectively, among and low in India and Latin America). Small bowel cancer rates males and 4.1 and 3.1 per million among females) than in are high among the Maori of New Zealand and among ethnic whites (4.0 and 3.9 per million, respectively, among males and Hawaiians, and low in India, Romania, and other parts of 3.4 and 3.7 among females; Ref. 1 1). This pattern is similar to Eastern Europe. Fig. 1 shows incidence rates among males and females reported by population-based tumor registries around the world (50). In the 1970s, Lowenfels (5) found a strong

linear correlation of small and large bowel cancer in different 3 Jacobson et al., unpublished observations.

Table 2 Distribution of small bowel cancer by histology, anatomic subsite, and sex in selected case series in the United States and Canada”

Anatomic subsite by Sex Average City/hospital/period (Ref.) Histology Total Duodenum Jejunum Ileum Histology Male Female age

New York, NY Adenocarcinoma 19 16 13 48 58.5 60s The New York Hospital- Carcinoid 0 0 1 1 1.2 SOs Cornell Medical Center Sarcoma 3 5 0 8 9.8 SOs 1932-1972 (3, 13) Lymphoma 4 12 9 25 30.5 54 Total 26 33 23 82

Memphis. TN” Adenocarcinoma 5 7 4 16 20.3 8 8 Baptist Memorial Memorial Carcinoid 4 I 26 31 39.2 16 15 1956-1976 (17) Sarcoma 3 5 2 10 12.7 7 3 Lymphoma 0 8 7 15 19.0 8 7 Total 12 21 39 79 39 33

Montreal, Quebec, Canada Adenocarcinoma 6 9 II 26 48.1 Royal Victoria Hospital Carcinoid 0 0 9 9 16.7 1935-1973 (18) Sarcoma I 1 3 5.6 Lymphoma 0 4 7 11 20.4 Total 7 14 28 54 >51

Boston, MA Adenocarcinoma 29 31 8 68 39.8 34 34 Massachusetts General Carcinoid 3 2 45 50 29.2 Hospital Sarcoma 2 3 7 12 7.0 1913-1976 (22) Lymphoma I 19 21 41 24.0 Total 35 55 81 171

Houston, TX” Adenocarcinoma 35 25 17 77 35.5 57 30 M. D. Anderson Hospital Carcinoid 7 6 51 64 29.5 48 41 1944-1982 (26) Sarcoma 3 9 15 27 12.4 18 15 Lymphoma 0 2 4 6 2.8 8 0 Total 45 42 87 217 131 86

Hartford, CT Adenocarcinoma 6 7 4 17 33.3 II 6 Hartford Hospital Carcinoid 0 0 20 20 39.2 8 12 1960-1983 (28) Sarcoma 0 6 2 8 15.7 6 2 Lymphoma 0 2 4 6 11.8 1 5 Total 6 15 30 51 26 25

New York, NY Adenocarcinoma 13 10 3 26 44.1 Mount Sinai Medical Carcinoid 0 1 6 7 11.9 Center Sarcoma 2 4 2 8 13.6 1970-1991 (36) Lymphoma 2 9 7 18 30.5 44 29 57 Total 17 24 18 59

Norfolk, VA Adenocarcinoma 13 8 7 28 53.8 Eastern Virginia Medical Carcinoid 3 13 17 32.7 School Sarcoma 0 2 1 3 5.8 1972-1991 (37) Lymphoma 0 2 2 4 7.7 27 25 62 Total 14 15 23 52

Los Angeles, CA Adenocarcinoma 26 II 5 42 68.9 Southern California Carcinoid 0 1 3 4 6.6 Permanente Medical Group Sarcoma I 5 2 8 13.1 1960-1989 (35) Lymphoma 1 3 3 7 I 1.5 Total 28 20 13 61 35 26 63 a Selection criteria were n > 50 and inclusion of all four major histological types. h Total includes 7 tumors of rare histology.

‘ Total includes 5 tumors of rare histology. d Total includes 10 adenocarcinomas, 25 carcinoids, 6 sarcomas, and 2 lymphomas with site unspecified.

those of large bowel and stomach cancer (Ref. 51; Table 4). lymphomas rose among both whites and blacks (10, 11). Black/white ratios for the rarer histological types vary, proba- Lymphoma incidence rates rose especially dramatically in bly because of small numbers. the 1980s, perhaps reflecting the spread of AIDS. Adeno- Time Trends. Three studies (7, 10, 1 1) have systematically carcinoma and carcinoid incidence rates among black males investigated changes over time in the incidence rates of also increased from 1987 to 1991 (9). Overall, from 1983 to small bowel malignancies in SEER. The age-adjusted mci- 1993, incidence rates of small bowel cancer increased from dence rate of malignant carcinoid tumors rose from 2.2 to 3.1 1.2 to 1.6 per 100,000 among white males, from 0.8 to 1.1 per million, but the rates of other tumor types changed very per 100,000 among white females. from I .8 to 2.4 per little from 1973 to 1982 (7). Subsequently, the incidence 100,000 among black males, and from 1.3 to 2.0 per 100,000 rates of adenocarcinomas, malignant carcinoid tumors, and among black females (Table 4).

Table 3 Disin bution of small bowel c ancer by histology, anatomic su bsite, and sex in selec ted case series i n Europe and Asia”

Anatomic subsite % b Sex Avera e Ciiy/hospiiaUperiod (Ref.) Histology Total . y g Histology Duodenum Jejunum Ileum Male Female age

Glasgow. Scotland” Adenocarcinoma 6 7 16 29 28.4 16 13 Royal Infirmary Carcinoid 0 1 15 16 15.7 8 8 SOs 1969-1983 (29) Sarcoma 0 2 3 5 4.9 1 4 Lymphoma 3 19 24 52 51.0 27 25 60s Total 9 29 58 106 52 50

Taipei. Taiwan Adenocarcinoma 16 5 9 30 29.7 19 II 60 National Taiwan University Carcinoid 0 1 2 3 3.0 3 0 59 Hospital Sarcoma S 14 7 26 25.7 14 12 51 1960-1989(3l Lymphoma 10 13 19 42 41.6 29 13 35 Total 31 33 37 101 65 36

Murcia. Spain Adenocarcinoma 13 10 3 26 37.7 Hospital Universitario Carcinoid 0 1 6 7 10.1 ‘Virgen de Ia Amxaca” Sarcoma 0 5 2 7 10.1 1977-1991 (34) Lymphoma 2 12 12 29 42.0 46 23 52.5 Total 15 28 23 69

‘, Selection criteria were a > SO and inclusion of all four major histological types.

I, Total includes 3 anaplastic tumors and 1 neurofibrosarcoma.

I- 4.5 4- 3.5 0 3- .-o 2.5 - 0 2- (c I .5 ______ Fig. 1. Age-standardized incidence rates of small bowel cancer among males and females in 0.5 - selected populations worldwide. 0

U Males 0 Females

Analytic Epidemiology Our case-control study used 19 adenocarcinoma and 17 Case-Control Studies malignant carcinoid tumors of the small bowel diagnosed at Although many studies have focused on specific risk factors for Columbia-Presbyterian Medical Center from 1980 to 1987 (53). small bowel malignancies, only three studies have compared small The control group for both case groups consisted of 52 patients undergoing benign non-GI surgical procedures, such as herni- bowel cancer cases to controls with respect to multiple risk factors orrhaphy and transurethral resection of the prostate. The cancer (52-54). Table S lists the main risk factors analyzed to date. One case-control study used decedents included in the diagnoses were reviewed by a single pathologist. Information 1986 National Mortality Follow-back Survey of the National on the cases and controls was collected by medical record review. Center for Health Statistics (52), which used a probability sample of 1% of United States deaths for 1986 along with all In the third study, 36 adenocarcinoma cases ages <65 deaths from certain rare cancers in 1985. The cases studied years were compared with 998 population controls for risk were 430 small bowel cancer deaths among white individuals factors assessed by interview (5 of the case interview respon- ages 25-74 years. Controls were 921 randomly selected white dents were surrogates; Ref. 54). All case diagnoses were patho- decedents in the 25-74 age range who died of causes other than logically confirmed. small bowel cancer. Questionnaires were sent to next-of-kin regarding demographic characteristics, dietary patterns, alcohol Behavioral Risk Factors and cigarette use, and other exposures. Because no information Diet. In an ecological study of small bowel cancer mortality was available on histological subtypes, the analysis of risk rates and food data supplied by the WHO, the correlation factors was conducted for cancer of the small intestine overall. coefficient was 0.61 (P < 0.005) with per capita daily animal

Table 4 Age-adjusted cancer incidence rates per 100,000 population by race adenocarcinomas (OR, 5.5; 95% CI, 1.4-21.4) and malignant and sex, SEER 1973-1993 carcinoid tumors (OR, 4.4; 95% CI, 1.1-18.5). Adjusting si- multaneously for cigarette smoking and alcohol lowered the Black Black White White females males females males observed ORs somewhat, but the basic findings remained the same. Wu et a!. (54) also found associations of adenocarcinoma Small bowel risk with ethanol intake (OR, 2.9; 95% CI, 1.2-7.1) and, al- 1973 1.2 1.8 0.7 0.9 though nonsignificant, with smoking (OR, 2.1; 95% CI, 0.8- 1983 1.3 1.8 0.8 1.2 5.1). 1993 2.0 2.4 1.1 1.6 Stomach Occupation/Socioeconomic Status. Small bowel cancer cases 1973 9.5 25.9 6.2 14.0 have been found to resemble controls in distribution of occu- 1983 8.5 22.0 5.0 10.8 pations, average family income, and educational attainment 1993 6.2 18.7 3.7 9.1 (52), and adenocarcinoma and malignant carcinoid cases have Large bowel been found to resemble controls in hospital payment type 1973 41.6 42.8 41.7 54.3 (private versus semiprivate versus ward; Ref. 53). 1983 49.0 61.0 43.8 60.4 1993 44.3 61.4 37.3 53.6 Other Medical Conditions Small Bowel Adenomas. Although the adenoma-carcinoma sequence in the large bowel is generally accepted (57), the Table 5 Comparison of risk factors for small and large bowel cancer rarity of both small bowel malignancies and small bowel Direction of effect on cancer screening make it difficult to demonstrate a similar sequence in risk the small bowel. However, a recent review points out many Risk factor In small In large similarities between the adenoma-carcinoma relationship in the bowel bowel large bowel and that in the small bowel (58). Of 255 sympto- matic small bowel adenomas reviewed, 76 (29.8%) showed Crohn’s disease + + + +“ + + invasive carcinoma. Like patients with colorectal neoplasia, Adenomas +++ those with small bowel carcinomas were, on the average, sev- FAP +++ ++++ Cholecystectomy + + eral years older than those with small bowel adenomas. Duo- Peptic ulcer disease + 0 denal site, large size, and villous histology were associated with Other intestinal cancers + + + + + + probability of containing invasive carcinoma. Almost all of the Other cancers ? + small bowel adenocarcinomas investigated showed associated Radiation + + adenomatous tissue. Small bowel adenomas and adenocarcino- BboodtypeA + ? mas had similar male/female ratios. Dietary intake of animal fat + + + Three case reports describe small bowel adenomas that Dietary intake of animal protein + + were left in place after pathological diagnosis (59-61). With Dietary intake of sugar + + follow-up at 4, 16, and 16 years, carcinoma was found in one Serum /3.-carotene - -

Tobacco use + + case after 16 years (61). Alcohol use + + FAP. Patients with familial adenomatous polyposis (FAP) Occupation ? + have multiple adenomas in the large bowel and, if untreated, Socioeconomic status ? ? close to 100% risk of colorectal cancer. These patients also tend a .. positive association (number of plus signs refers to strength or consistency to have multiple adenomas in the small bowel (62-68) and an of association); 0, no association; -, negative (inverse) association; ?, no credible increased risk of adenocarcinoma in the small bowel, especially evidence for or against an association. in the duodenum. A number of studies have explored the risk for small bowel adenocarcinoma in patients with FAP. Periam- pullary carcinoma was recently reported to be the most corn- fat consumption and 0.75 (P < 0.001) with per capita daily mon extracolonic malignancy in FAP patients (62); 2.9% of animal protein consumption (55). Chow et a!. (52) analyzed studied patients developed such a carcinoma. Of 1262 patients various dietary risk factors for small bowel cancer. Responses entered in the Polyposis Registry at St. Mark’s Hospital, 47 from both spouses and other surrogates for cases and controls developed duodenal cancer. Indeed, periampullary carcinoma yielded statistically significant ORs in the 2-3 range for intake was the most common cancer-related cause of death in this of red meat and salt cured/smoked foods. Wu et a!. (54) also population (63). Duodenal bile from 29 FAP patients was found found suggestions of an association with fried, smoked, or to be more mutagenic than that from 24 non-FAP patients (64). barbecued meat and fish, but none was statistically significant. Among 1,391 FAP patients registered in the Johns Hop- The strongest increase in risk was associated with sugar intake kins registry, with 18,679 person-years of follow-up, there were in nonalcoholic beverages (OR, 3.3; 95% CI, 1.2-9.4). Another 1 1 cases of duodenal and ampullary adenocarcinorna (65). The study found lower serum (3-carotene levels in patients with RR for duodenal adenocarcinoma given FAP was 330.8 (95% cancer of the small bowel than in their relatives (56). CI, 132.7-681.5), and the RR for ampullary adenocarcinoma Tobacco and Alcohol Use. Chow et al. (52) found that mdi- was 123.7 (95% CI, 33.7-316.7). Mean ages at diagnosis were viduals who had died of small bowel cancer did not differ in 53 for duodenal and 49 for ampullary carcinoma. Patients with alcohol or tobacco use from controls who had died from other FAP had no increased risk for gastric or non-duodenal small causes. Our study, based on medical record review (53), found bowel cancer. Among 247 patients with FAP who underwent an an OR of 6.2 (95% CI, 1.5-26.7) for the association of cigarette average of three upper 01 endoscopic screening examinations smoking with adenocarcinomas, and an OR of 5.8 (95% CI, in 1981-1991 at the Cleveland Clinic, 88% had duodenal 1.1-30.3) for the association of smoking with malignant carci- and/or gastric polyps; none had a carcinoma (66). noids. Likewise, alcohol consumption elevated risk for both In Peutz-Jeghers syndrome, a rarer familial disorder, pa-

tients have mucocutaneous pigmentation and multiple hamar- Radiation Therapy. Exposure to radiation therapy has been tomatous polyps of the GI tract. Their lifetime risk of devel- implicated as a possible source of small bowel malignancies oping small intestinal adenocarcinoma has been estimated as (80). A cohort study compared 49,828 women with cervical 2.4% ( 1 ): these patients also have an increased risk for other cancer treated with radiation therapy to 16,7 13 women treated cancers (69). without radiotherapy. The RR for small bowel cancer following Colorectal Cancer. Patients with hereditary nonpolyposis radiation therapy was 1.8 (95% CI, 1.1-2.0). Risk remained colorectal cancer also have an increased risk of small bowel elevated for up to 30 years following treatment. No data were provided regarding the histological types of the small bowel cancer (70, 71 ); lifetime risk is estimated at 1% among these cancers observed. patients, a significantly higher risk than that in the general population. However, in the general population, both malig- Crohn’s Disease. Over 100 cases of adenocarcinoma of the nancies also tend to occur in the same individuals. The 2581 small bowel following Crohn’s disease have been reported in small bowel malignancies registered in SEER between 1973 the medical literature since the initial report by Ginsburg et a!. and 1988 were analyzed to determine what second malignan- (81) in 1956. Although most adenocarcinomas occur in the cies were associated with small bowel cancer, overall and by duodenum, those associated with Crohn’s disease generally histological subtype during this period (72). RRs were calcu- occur in the ileum 20 years or more after the onset of the lated both including and excluding the first 6 months of Crohn’s disease (82). follow-up to assess medical surveillance bias. The RR for Several studies have been conducted to quantify this risk. colorectal cancer following small bowel adenocarcinoma was Darke et a!. (39) assigned a 0.3% probability to the chance occurrence of Crohn’s disease and small bowel cancer in the 5.0 (95% CI, 2.3-9.4) in males and 3.7 (95% CI, 1.3-8.0) in same individual. The RR of small bowel cancer in Crohn’s females. Risk of small bowel adenocarcinoma following cob- disease patients has been estimated as more than 100 times that rectal cancer was also elevated, with a RR of 7.1 (95% CI, of the general population (83). In Denmark, a retrospective 4.7-10.3) in males and 9.0 (95% CI, 6.0-12.9) in females. cohort analysis of 373 patients diagnosed with Crohn’s disease Other Cancers. In the same study (72), the risk of prostate between 1962 and 1987 and followed for 10-15 years found cancer was found to be elevated following a small bowel two cases of small bowel cancer, both in the ileum, compared malignant carcinoid tumor (RR. 2.3; 95% CI, 1.2-4.1) and vice to an expected 0.04 cases (40); thus, the RR of small bowel versa (RR, 2.6; 95% CI, 1.4-4.4). Of 128 small bowel adeno- cancer given Crohn’s disease was 50 (95% CI, 37.1-65.9; P carcinomas recorded in the British Columbia and Manitoba 0.0001). A similar study in Stockholm (83) found an RR of 15.6 cancer registries 1982-1991 (73), 54 were associated with other (95% CI, 4.3-40. 1). Crohn’s disease of the small bowel was not malignancies, including 38 diagnosed prior to, 8 synchronous found to increase the risk of other cancers, including colorectal with, and 6 subsequent to the diagnosis of the small bowel cancer, in these studies. cancer. The most common second malignancies were large In our case-control study, 4 of 19 small bowel adenocar- bowel ( 16 cases), prostate (9 cases), female genitalia (6 cases), cinoma cases arose in the ileum (53), and three of these were and lung (6 cases). During a mean of 6 years of follow-up associated with a history of Crohn’s disease (P < 0.004). Of 17 among 6 1 patients diagnosed with small bowel cancer, 7 pa- malignant carcinoid cases, 14 were in the ileum, and none were tients developed another primary cancer: 3 developed breast associated with Crohn’s disease. To explore which variables cancers, and 1 each developed rectal, renal, lymphoma, and predisposed Crohn’s disease cases to develop adenocarcinoma melanoma (35). of the small bowel, Lashner (83) compared 7 Crohn’s disease/ In a study of 49 patients with adenocarcinoma of the small adenocarcinoma cases seen at the University of Chicago to 28 intestine recorded in the Hawaii Tumor Registry from 1960 to Crohn’s disease controls matched to the cancer cases for age I 989, 1 1 had other cancers. Seventeen of the 49 patients were and sex. The OR for cancer in Crohn’s disease patients with Japanese ( I 2 males and S females; Ref. 74). Five Japanese jejunal lesions was 8. 1 (95% CI, I .6-39.3). Treatment with males had large bowel cancers; one of them also had prostate 6-mercaptopurine but not with prednisone or sulfasalazine also and stomach cancer; another had cancer of the ureter. The other increased the risk of small bowel cancer (OR, 10.8; 95% CI, patients with other cancers were 2 of the 12 non-Japanese 1. 1-108.7). In addition, Crohn’s disease patients with occupa- females (large bowel cancer and breast cancer), and 4 of the 20 tional exposure to aromatic compounds or to other potential non-Japanese males (myeloma, stomach cancer, ureteral can- carcinogens, such as asbestos or cutting oils, had an OR of 20.3 cer, and rectal cancer). (95% CI, 2.7-150.5) for small bowel cancer, compared to In a series of 9721 consecutive autopsies, 1 12 subjects patients without such exposures (84). Family history of GI were found with primary neoplasms of the small intestine, of cancer, urban residence, and cigarette smoking were not asso- which 40 were malignant (2 carcinoma, I sarcoma, 4 lym- ciated with risk. phoma, and 33 malignant carcinoids; Ref. 75). One of the Cholecystectomy and Bile. The concentration of small bowel noncarcinoid malignancies was found in association with an- adenocarcinomas in the duodenum and particularly in the peri- other malignancy; 13 (45%) of the malignant carcinoid tumors ampullary region implicates bile as a possible promoter of these were associated with another malignant neoplasm, although the cancers (1 , 2, 38). In our case-control study, 3 of 19 adenocar- sites of the additional tumors were not reported. cinoma cases and 3 of 17 carcinoid cases had a previous This observation suggests a prevalence of 4/1000 for all cholecystectomy, whereas none of 52 controls did (P < 0.004; undiagnosed small bowel cancer, which, given the overall Ref. 53). 1/100,000 incidence rate of this cancer, is surprisingly high but Peptic Ulcer Disease. Lightdale et a!. (85) suggested that the is probably accounted for by asymptomatic carcinoids. alkaline environment of the small bowel is part of the reason for Increased risks of small bowel cancer have also been its low cancer rate. Our study found that although none of 52 observed in association with Wilms’ tumor (76), Hodgkin’s controls had a history of PUD, 2 of 19 adenocarcinoma cases

disease (77), anal cancer (78), and squamous cell skin cancer (P = 0.02) and 3 of 17 malignant carcinoid cases (P = 0.002) (79). had such a history (53).

Other Risk Factors tion of nitrosarnines that may be carcinogenic in the acid Two recent studies have explored the association between environment of the stomach (85, 99). cystic fibrosis and small bowel cancer (86, 87). Four earlier 6. The small bowel has a well-developed protective local case reports had described adenocarcinoma of the ileum fol- IgA-mediated immune system. Patients with a deficiency lowing cystic fibrosis (88-91). One case was found in a cohort of this immunoglobulin have a higher than normal risk for of 412 cystic fibrosis patients, with 0.001 expected on the basis small bowel cancer (1 , 94). This mechanism may be more of population incidence rates (P = 0.003; Ref. 86). A retro- relevant to lymphomas than to other tumors. spective cohort study (87) of 28,5 1 1 patients with cystic fibro- 7. In the small bowel, carcinogens induce apoptosis and cells sis, followed from 1985 to 1992, found no elevation in the risk that contain genetic defects are quickly removed (94). of cancer overall but identified two small bowel cancers. The Expression of the bcl-2 gene, which prevents apoptosis, is authors noted that cystic fibrosis often involves the GI tract and abundant in the murine colonic crypt, but poor in the is associated with steatorrhea. They did not report the subsites murine small intestine crypt cells (100). or histologies of the cancers or estimate a specific risk ratio. 8. The small bowel has fewer stem cells (carcinogen target Taggart and Imrie (29) described 106 patients diagnosed cells) than the colon or the stomach (99). These stem cells with small bowel cancer, including 52 with lymphoma, diag- lie deep within the crypts and therefore are shielded from nosed between 1969 and 1983 in five Glasgow hospitals. Six of potential carcinogens by a mucous layer and other efficient these cases definitely, and seven possibly, had celiac disease. mechanisms that maintain the genetic integrity of the stem Adenocarcinoma has also been reported in association with cells (101). celiac sprue (1, 92). 9. The duodenum contains a water-soluble tumor-inhibiting Our study explored the association between blood group component (102). and small bowel malignancies; none of 36 cases, but 2 of 16 10. The small bowel contains low levels of activating enzymes (13%) controls with blood information, as well as 15% of a of precarcinogens. The microsomal enzyme benzopyrene systematic sample of 300 blood bank files, were type B (P < hydrolase has been identified and studied in experimental 0.09; Ref. 53). The cases and controls did not differ in distri- animals and humans. This enzyme converts the potent bution of Rh groups. carcinogen benzopyrene to a less active compound and is The associations of medical history and biological factors found in greatest concentration in the small bowel. Other with small bowel cancer suggest that this condition may have a enzymes of this type may protect the small bowel against familial basis, but no studies of familial small bowel cancer other carcinogens (103). (except in patients with Crohn’s disease) have been published. Among mice, a gene for susceptibility to small bowel cancer has been mapped to mouse chromosome 4 (93). Conclusion Although often studied as a single entity, small bowel cancer really represents four main histologic subtypes. Analysis of Why Is Small Bowel Cancer So Rare? small bowel cancer overall may be suggestive, but we must Like the spleen, the small intestine is remarkably resistant to ultimately undertake subtype-specific studies of both the de- carcinogenesis. When gastric carcinoma patients have a recur- scriptive and the analytic epidemiology of small bowel cancer. rence of cancer after gastrojejunostomy, the recurrence is 5 The physiological mechanisms described above should be stud- times more likely to be gastric than jejunal. However, when ied in relation to the main histological subtypes of small bowel peptic ulcer patients have a recurrent ulcer after gastrojejunos- cancer. Rising incidence rates of small bowel adenocarcinoma, tomy, the recurrence is nearly always jejunal (27). Likewise, carcinoid, and lymphoma also justify intensified scrutiny of the terminal ileum is highly resistant to tumor invasion from the these rare cancers, although these trends may merely reflect proximal colon. Only 0.7% of colon cancers have a local improvements in diagnosis. recurrence after an ileocolic anastomosis, compared to 14% A study of those genetic markers associated with carcino- after colocolic anastornosis (5). The association of small bowel genesis in the large bowel (104) and in small bowel adenomas cancers with other cancers and other diseases of the GI tract and adenocarcinomas (97) is long overdue. Like large bowel also suggests that developing a small bowel tumor reflects cancers, small bowel adenocarcinomas have been found to have failure of a significant protective system. K-ras mutations (105, 106). Several theories have been suggested to explain the low The geographical and temporal variations in the incidence incidence of small bowel tumors in humans: rates of small bowel cancer and the variations in distribution of histological subtypes suggest that environmental factors play an

1 . Small intestinal mucosal cells are lost and replaced very important role in the etiology of this disease. Small bowel rapidly; the rate is estimated at 1 g of intestinal mucosa adenocarcinoma may share certain risk factors with large bowel every 16 mm (94, 95). This rapid turnover may prevent adenocarcinoma. Indeed, this similarity has led clinicians to tumor growth. treat small bowel adenocarcinoma with the same therapeutic 2. Bacteria are relatively scarce in the normal small bowel. regimens that are used for large bowel cancer (107). Crohn’s Methylazoxymethanol produces colon cancer in normal disease and FAP are clearly associated with small bowel ade- animals but not in those that are germ-free (1, 96, 97). nocarcinomas of the ileum and duodenum respectively. How- 3. The rapid transit of the small bowel contents and the ever, these conditions account for only a small number of cases. sparseness of bacterial flora may minimize exposure to Factors affecting bile and its metabolites may account for a potential carcinogenic bacterial breakdown products (5, larger number of cases of both small and large bowel adeno- 98). This mechanism may be particularly relevant to ade- carcinoma, especially in industrialized countries. Infectious nocarcinoma. agents, perhaps including HIV, may be associated with small 4. Liquefied chyme may reduce mechanical trauma and pro- bowel lymphomas. Studies of populations such as those of New tect the small bowel from carcinogens (5, 99). Zealand and Hawaii, where different ethnic groups have very 5. The alkalinity of the small bowel contents prevents forma- different incidence rates of small bowel cancer, may shed some

light on genetic and environmental determinants of risk for this 27. Zollinger, R. M., Sternfeld, W., and Schreiber, H. Primary neoplasms of the disease. small intestine. Am. J. Surg., 15/: 654-658, 1986. Well-designed case-control studies with adequate power, 28. Ciccarelli, 0., Welch, J. P., and Kent, G. G. Primary malignant tumors of the small bowel. Am. J. Surg., 153: 350-354, 1987. obtaining data directly from living cases and controls, would be 29. Taggart, D. P., and Imrie, C. W. A new pattern of histological predominance most helpful, especially if they incorporated biological assays and distribution of malignant diseases of the small intestine. Surg. Gynecol. that could address some of the physiological hypotheses listed Obstet., 165: 515-518, 1987. above. 30. Brophy, C., and Cahow, C. E. Primary small bowel malignant tumors. Am. Surg., 55: 408-412, 1989. 31. Lee, W. J., Chang, K. J., Wang, S. M., Chen, K. M., and How, S. W. Primary References malignant tumor of the small intestine. J. Formosan Med. Assoc., 90: 776-781, 1. Schottenfeld, D., and Islam, S. S. Cancers of the small intestine. In: D. 1991. Schottenfeld and J. Fraumeni (eds.), Cancer Epidemiology and Prevention, pp. 32. Kusumoto, H., Takahashi, I., Yoshida, M., Maehara, Y., Watanabe, A., 806-812. New York: Oxford University Press, 1996. Oshiro, T., and Sugimachi, K. Primary malignant tumors of the small intestine: 2. Ross, R. K.. Hartnett, N. M., Bernstein, L., and Henderson, B. E. Epidemi- analysis of 40 Japanese patients. J. Surg. Oncol., 50: 139-143, 1993. ology of adenocarcinomas of the small intestine: is bile a small bowel carcinogen? 33. Baillie, C. T., and Williams, A. Small bowel tumours: a diagnostic challenge. Br. J. Cancer, 63: 143-145, 1991. J. R. Coil. Surg. Edinb., 39: 8-12, 1994. 3. Wilson, J. M., Melvin, D. B., Gray, G. F., and Thorbjarnarson, B. Primary 34. Garcia Marcilla, J. A., Sanchez Bueno, F., Aguilar, J., and Parrilla Paricio, P. malignancies of the small bowel: a report of 96 cases and review of the literature. Primary small bowel malignant tumors. Eur. J. Surg. Oncol., 20: 630-634, 1994. Ann. Surg.. 180: 175-179, 1974. 35. Frost, D. B., Mercado, P. D., and Tyrell, J. S. Small bowel cancer: a 30-year 4. Mittal, V. K., and Bodzin, J. H. Primary malignant tumors of the small bowel. review. Ann. Surg. Oncol., 1: 290-295, 1994. Am. J. Surg.. /40: 396-399, 1980. 36. Cunningham, J. D., Aleali, R., Aleali, M., Brower, S. T., and Aufses, A. H. 5. Lowenfels, A. B. Why are small-bowel tumors so rare? Lancet, 1: 24-26, Malignant small bowel neoplasms. Histopathologic determinants of recurrence 1973. and survival. Ann. Surg., 225: 300-306, 1997. 6. Parker, S. L., Tong, T., Bolden, S., and Wingo, P. A. Cancer statistics 1997. 37. Lambert, P., Minghini, A., Pincus, W., Kolm, P., and Perry, R. R. Treatment Ca. Cancer J. Clin. 47: 5-27, 1997. and prognosis of primary malignant small bowel tumors. Am. Surg., 62: 709- 7. Weiss, N. S., and Yang, C. Incidence of histologic types of cancer of the small 715, 1996. intestine. J. Nati. Cancer Inst., 78: 653-656, 1987. 38. Lowenfels, A. B. Does bile promote extra-colonic cancer? Lancet, 2: 239- 8. Gabos, S., Berkel, J., Band, P., Robson, D., and Whittaker, H. Small bowel 241, 1978. cancer in Western Canada. Int. J. Epidemiol., 22: 198-206, 1993. 39. Darke, S. G., Parks, A. G., Grogano, S. L., and Pollock, D. J. Adenocarci- 9. DiSario, J. A., Burt, R. W., Vargas, H., and McWhorter, W. P. Small bowel nomas and Crohn’s disease. A report of two cases and analysis of the literature. cancer: epidemiological and clinical characteristics from a population-based Br. J. Surg., 60: 169-175, 1973. registry. Am. J. Gastroenterol., 89: 699-701, 1994. 40. Munkholm, P., Langholz, E., Davidsen, M., and Binder, V. Intestinal cancer 10. Severson, R. K., Schenk, M., Gurney, J. G., Weiss, L. K., and Demers, R. Y. risk and mortality in patients with Crohn’s disease. Gastroenterology, 105: Increasing incidence of adenocarcinomas and carcinoid tumors of the small 1716-1723, 1993. intestine in adults. Cancer Epidemiol. Biomark. Prey., 5: 81-84, 1996. 41. Coit, D. G. Cancer of the small intestine. In: V. T. DeVita, Jr., S. HeIlman, 1 1. Chow, J. S., Chen, C. C., Ahsan, H., and Neugut, A. I. A population-based and S. A. Rosenberg (eds.), Cancer: Principles and Practice of Oncology, 4th Ed. study of the incidence of malignant small bowel tumors: SEER, 1973-90. Int. J. Philadelphia: J. B. Lippincott Co., 1993. Epidemiol., 25: 722-728, 1996. 42. Dawson, I. M. P., Comes, J. S., and Morson, B. C. Primary malignant 12. Adekunle, 0. 0., and Ajayi, 0. 0. Primary malignant small intestinal tumors lymphoid tumors of the intestinal tract: report of 37 cases with a study of factors in Nigeria. Int. Surg., 58: 708-71 1, 1973. influencing prognosis. Br. J. Surg., 40: 80-88, 1961. 13. Loehr, W. J., Mujahed. Z.. Zahn, F. D., Gray, G. F., and Thorbjarnarson, B. 43. Sweetenham, J. W., Mead, G. M., Wright, D. H., McKendrick, J. J., Jones, Primary lymphoma ofthe gastrointestinal tract: a review of 100 cases. Ann. Surg., D. H., Williams, C. J., and Whitehouse, J. M. A. Involvement of the ileocecal 170: 232-238, 1969. region by non-Hodgkins lymphoma in adults: clinical features and results of treatment. Br. J. Cancer, 60: 366-369, 1989. 14. Silberman, H., Crichlow, R. W., and Caplan, H. S. Neoplasms of the small bowel. Ann. Surg., 180: 157-161, 1974. 44. Al-Bahrani, Z. R., AI-Mondhiry, H., Bakir, F., and Al-Saleem, T. Clinical and pathologic subtypes of primary intestinal lymphoma. Cancer (Phila.), 52: 15. Croom, R. D., and Newsome, J. F. Tumors ofthe small intestine. Am. Surg., 1666-1672, 1983. 41: 160-167, 1975. 45. Khojasteh, A., and Haghighi, P. Immunoproliferative small intestinal disease: 16. Freund, H., Lavi, A., Pfeffermann, R., and Durst, A. L. Primary neoplasms portrait of a potentially preventable cancer from the Third World. Am. J. Med., of the small bowel. Am. J. Surg., 135: 757-759, 1978. 89: 483-490, 1990. 17. Miles, R. M.. Crawford, D., and Duras, S. The small bowel tumor problem. 46. Azar, H. A. Cancer in Lebanon and the Near East. Cancer (Phila.), 15: 66-78, Ann. Surg., /89: 732-740, 1979. 1962. 18. Coutsoftides, T., and Shibata, H. R. Primary malignant tumors of the small 47. Frand. V. I., and Ramot, B. Malignant lymphomata: an epidemiological intestine. Dis. Colon Rectum, 22: 24-26, 1979. study. Harefuah, 5: 83-86, 1963. 19. Awrich, A. E., Irish, C. E., Vetto, R. M., and Fletcher, W. S. A twenty-five 48. WHO. Alpha chain disease and related small intestinal lymphoma: a mem- year experience with primary malignant tumors of the small intestine. Surg. orandum. Bull. WHO, 54: 615-624, 1976. Gynecol. Obstet., 151: 9-14, 1980. 49. Ryan, J. C. Premalignant conditions of the small intestine. Semin. Gastroi- 20. Norberg, L., and Emas, S. Primary tumors of the small intestine. Am. J. Surg., ntest. Oncol., 7: 88-93, 1996. 142: 560-573, 1981. 50. Parkin, D., Muir, C., Whelan, S. L., Gao, Y. T., and Ferlay, J. Cancer 21. Gupta. S.. and Gupta, S. Primary tumors of the small bowel: a clinicopath- Incidence in Five Continents, 6th Ed., IARC Sci. PubI. No. 120. New York: ological study of 58 cases. J. Surg. Oncol., 20: 161-167, 1985. Oxford University Press, 1993. 22. Williamson, R. C. N., Welch, C. E., and Malt, R. A. Adenocarcinoma and 51. Ries, L. A. G., Kosary, C. L., Hankey, B. F., Harras, A., Miller, B. A., and lymphoma of the small intestine. Ann. 5mg., 197: 172-178, 1983. Edwards, B. K. (eds.). SEER Cancer Statistics Review 1973-93. Tables and 23. Barclay. T. H. C., and Schapira, D. V. Malignant tumors of the small Graphs. Bethesda, MD: National Cancer Institute, 1996. intestine. Cancer (Phila.), 51: 878-881, 1983. 52. Chow, W. H., Linet, M. S., McLaughlin, J. K., Hsing, A. W., CoChien, H. T., 24. Giuliani, A., Caporale, A., Teneriello, F., Alessi, G., Serpieri, S., and and Blot, W. J. Risk factors for small intestine cancer. Cancer Causes Control, 4: Sammartino, P. Primary tumors of the small intestine. mt. Surg.. 70: 33 1-334, 163-169, 1993. 1985. 53. Chen, C. C., Neugut, A. I., and Rotterdam, H. Risk factors for adenocarci- 25. Johnson, A. M., Harman, P. K., and Hanks, J. B. Primary small bowel nomas and malignant carcinoids of the small intestine: preliminary findings. malignancies. Am. Surg., 51: 31-36, 1985. Cancer Epidemiol. Biomark. Prey. 3: 205-207, 1994. 26. Martin, R. G. Malignant tumors of the small intestine. Surg. Clin. North Am., 54. Wu, A. H., Yu, M. C., and Mack, T. M. Smoking, alcohol use, dietary factors 66: 779-785, 1986. and risk of small intestinal adenocarcinoma. mt. J., Cancer, 70: 512-517, 1997.

55. Lowenfels, A. B., and Sonni, A. Distribution of small bowel tumors. Cancer 81. Ginsburg, L., Schneider. K. M., Dreizin, D. H., and Levinson, C. Carcinoma Left., 3: 83-86, 1977. of the jejunum occurring in a case of regional enteritis. Surgery (St. Louis). 39: 56. Smith, A. H., and WaIler, K. D. Serum a-carotene in persons with cancer and 347-351, 1956. their insmediate families. Am. J. Epidemiol., 133: 661-671, 1984. 82. Michelassi, F., Testa, G., Pomidor, W. I., Lashner, B. A., and Block, G. E. 57. Neugut, A. I., Jacobson, J. S., and DeVivo, I. Epidemiology of colorectal Adenocarcinoma complicating Crohn’s disease. Dis. Colon Rectum, 35: 654- adenomatous polyps. Cancer Epidemiol. Biomark. Prey., 2: 159-176, 1993. 661, 1993. 83. Lashner, B. A. Risk factors for small bowel cancer in Crohn’s disease. Dig. 58. Sellner, F. Investigations on the significance of the adenoma-carcinoma Dis. Sci., 37: 1179-1184, 1992. sequence in the small bowel. Cancer (Phila.), 66: 702-715, 1990. 84. Persson, P. G., Karlen, P., Bumell, 0., Leijonmarck. C. E.. Brostrom, 0., 59. Greenwald, E., Parker, J., Schultz, S., and Reed, E. Benign papillary adenoma Ahlbom, A., and Hellers, 0. Crohn’s disease and cancer: a population-based of the ampullary region of the duodenum with intussusception. Gastroenterology, cohort study. Gastroenterology, 107: 1675-1679, 1994. 43: 344-350, 1962. 85. Lightdale, C. I., Koepsell, T. D., and Sherlock, P. Small intestine. In: D. 60. Uppaputahangkule, V., Mass, L., and Gelzayd, E. Endoscopic diagnosis of Schottenfeld and I. F. Fraumeni, Jr. (eds). Cancer Epidemiology and Prevention, villous adenoma of the duodenum. Gastrointest. Endosc., 22: 97-98, 1976. pp. 692-702. Philadelphia: W. B. Saunders, 1982. 61. Hessler, P., and Braunstein, E. Adenocarcinoma of the duodenum arising in 86. Sheldon, C. D., Hodson, M. E., Carpenter, L. M., and Swerdlow, A. J. A a villous adenoma. Gastrointest. Radiol., 2: 355-357, 1978. cohort study of cystic fibrosis and malignancy. Br. I. Cancer, 68: 1025-1028, 62. Jagelman, D. G., DeCosse, J. J., Bussey, H. J. R., and the Leeds Castle 1993. Polyposis Group. Upper gastrointestinal cancer in familial adenomatous polyp- 87. Neglia. I. P., FitzSimmons, S. C., Maisonneuve, P., Schoni, M. H., Schoni- osis. Lancet, 1: 1149-1151, 1988. Affolter, F., Corey, M., and Lowenfels, A. B. The risk of cancer among patients 63. Spigelman, A. D., Talbot, I. C., Penna, C., Nugent, K. P., Phillips, R. K., with cystic fibrosis. N. EngI. I. Med., 332: 494-499, 1995. Costello, C., and DeCosse, J. J. Evidence for adenoma-carcinoma sequence in the 88. Roberts, J. A.. Tullett, W. M., Thomas, I. S. I., Galloway, D., and Stack, duodenum of patients with familial adenomatous polyposis. J. Clin. Pathol., 47: B. H. R. Bowel adenocarcinoma in a patient with cystic fibrosis. Scot. Med. J., 709-710, 1994. 31: 109, 1986. 64. Spigelman, A. D., Crofton-Sleigh, C.. Venitt, S., and Phillips, R. K. S. 89. Siraganian, P. A., Miller, R. W., and Swender, P. T. Cystic fibrosis and Mutagenicity of bile and duodenal adenomas in familial adenomatous polyposis. ileocarcinoma. Lancet, 2: I 158, 1987. Br. J. Surg., 77: 878-881, 1990. 90. Swender, P. T., Nathan, E. M., Sondheimer, I., and Kasowitz, M. H. Bowel 65. Offerhaus, G. J., Giardiello, F. M., Krush, A. J., Booker, S. V., Tersmette, carcinoma mimicking meconium ileus equivalent. Cystic Fibrosis Club Abstr., A. C., Kelley, N. C., and Hamilton, S. R. The risk of upper gastrointestinal cancer 23: 141, 1982. in familial adenomatous polyposis. Gastroenterology, 102: 1980-1982, 1993. 91. Redington, A. N., Spring, R., and Batten, I. C. Adenocarcinoma of the ileum 66. Church, J. M., McGannon, E., Hull-Boiner. S., Sivak, M. V., Van Stolk, R., presenting as non-traumatic clostridial myonecrosis in cystic fibrosis. Br. Med. J., Jagelman, D. G., Fazio, V. W., Oaldey, J. R., Lavery, I. C., and Milsom, J. W. 290: 1871-1872, 1985. Gastroduodenal polyps in patients with familial adenomatous polyposis. Dis. 92. Beagos, D. G., Kuan, S., Dobbins, J., and Ravikumar, T. S. Metachronous Colon Rectum, 35: 1 170-1 173, 1992. small-bowel adenocarcinoma in celiac sprue. J. Clin. Gastroenterol., 20: 233-236, 67. Nugent, K. P., Spigelman, A. D., Williams, C. B., Talbot, I. C., and Phillips, 1985. R. K. Surveillance of duodenal polyps and familial adenomatous polyposis: 93. Fijneman, R. J., and Demant, P. A gene for susceptibility to small intestinal progress report. J. R. Soc. Med., 87: 704-706, 1994. cancer, ssicl, maps to the distal part of mouse chromosome 4. Cancer Res., 55: 68. Spigelman, A. D., Williams, C. B., Talbot, I. C., Domizio, P., and Phillips, 3179-3182, 1995. R. K. S. Upper gastrointestinal cancer in patients with familial adenomatous 94. Kim, S. H., Roth, K. A., Moser, A. R., and Gordon, J. I. Transgenic mouse polyposis. Lancet, 2: 783-785, 1989. models that explore the multistep hypothesis of intestinal neoplasia. J. Cell Biol.. 69. DeCosse, J. J., and Jacobson, J. S. Polyps, polyposis, and benign tumors. in: 123: 877-893, 1993. R. E. Condon (ed), Surgery of the Alimentary Tract Vol. IV: Colon and Ano- 95. Boyle. W. J., and Brenner, D. A. Molecular and cellular biology of the small rectum, 3rd Ed. Philadelphia: W. B. Saunders, 1991. intestine. Curt. Opin. Gastroenterol.. 11: 121-127, 1995. 70. Watson, P., and Lynch, H. T. Extracolonic cancer in hereditary nonpolyposis 96. Laquer, G. K., McDaniel, E. 0., and Matsumoto, H. Neoplasms in female colorectal cancer. Cancer (Phila.), 71: 677-685, 1993. Fischer rats following intraperitoneal injection of methylazoxy-methanol. J. NatI. 71. Aarnio, M., Meckiin, i-P., Aaltonen, L. A., Nystrom-Lahti, M., and iarvinen, Cancer Inst., 39: 55, 1967. H. I. Life-time risk of different cancers in hereditary non-polyposis colorectal 97. Hill, M. J., Brasar, B. D., Hawksworth, G., Aries, V., Crother, i. S., and cancer (HNPCC) syndrome. mt. I. Cancer (Pred. Oncol.), 64: 430-433, 1995. Williams, R. E. Bacteria and etiology ofcanceroflarge bowel. Lancet, 1:95-100, 72. Neugut, A. I., and Santos, J. The association between cancers ofthe small and 1975. large bowel. Cancer Epidemiol. Biomark. Prey., 2: 551-553, 1993. 98. Arber, N., Neugut, A. I., Weinstein, I. B., and Holt, P. R. Molecular genetics 73. Ripley, D., and Weinerman, B. Increased incidence of second malignancies of small bowel cancer. Cancer Epidemiol. Biomark. Prey., 6: 745-748, 1997. associated with small bowel adenocarcinoma. Can. i. Gastroenterol., 11: 65-68, 99. Potten, C. S. Clonogenic, stem and carcinogen-target cells in small intestine. 1997. Scand. i. Gastroenterol., 104: 3-10. 1984. 74. Stemmermann, G. N., Goodman, M. T., and Nomura, A. M. Y. Adenocar- 100. Potten, C. S. The significance of spontaneous and induced apoptosis in the cinoma of the proximal small intestine. Cancer (Phila.), 70: 2766-2771, 1992. gastrointestinal tract of mice. Cancer Metastasis Rev., I 1: 179-195, 1992. 75. Alexander, i. W., and Altemeier, W. A. Association of primary neoplasms of 101. Potten, C. S. Stem Cells: Their Identification and Characterization. Edin- the small intestine with other neoplastic growths. Ann. Surg., 167: 958-964, burgh: Churchill Livingstone, 1983. 1968. 102. Arbuna, i., and de Cruicchi, E. T. A factor responsible for duodenal 76. Beales, I. L. P., and Scott. H. I. Adenocarcinoma of the duodenum with a resistance to cancer. Curt. Ther. Res., 40: 745-749, 1986. duodeno-colic fistula occurring after childhood Wilms’ cancer. Postgrad. Med. J., 103. Wattenberg. L. W. Carcinogen-detoxifying mechanisms in the gastrointes- 70: 933-936, 1994. tinal tract. Gastroenterology. 51: 932-935. 1966. 77. Henry-Amar, M. Second cancer after treatment for Hodgkin’s disease: a 104. Fearon, E., and Vogelstein, B. A genetic model for colorectal wmorigenesis. report from the international database on Hodgkin’s disease. Ann. Oncol., 4: Cell, 61: 759-767, 1990. 117-128, 1992. 105. Sutter, T., Arber, N., Moss, S. F., Findling, R. I., Neugut, A. I., Weinstein. 78. Frisch, M., Olsen, I. H., and Melbye, M. Malignancies that occur before and I. B.. and Holt, P. R. Frequent K-ras mutations in small bowel adenocarcinomas. after anal cancer clues to their etiology. Am. i. Epidemiol., 140: 12-19. 1994. Dig. Dis. Sci., 41: 115-118, 1996. 79. Frisch, M., and Melbye, M. New primary cancer after squamous cell skin 106. Younes, N., Fulton, N., Tanaka, R., Wayne, J., Straus, F. H., 2”, and cancer. Am. J. Epidemiol., 141: 916-922, 1995. Kaplan. E. L. The presence of K-I2 tax mutations in duodenal adenocarcinomas 80. Kleinerman, R. A., Boice, I. D., Jr., Storm, H. H., Sparen, P., Andersen, A., and the absence of ras mutations in other small bowel adenocarcinomas and Pukkala, E., Lynch, C. F., Hankey, B. F., and Flannery, J. T. Second primary carcinoid tumors. Cancer (Phila.), 79: 1804-1808, 1997. cancer after treatment for cervical cancer: an international cancer registries study. 107. Neugut, A. I., Marvin, M., Rella, V. A., and Chabot, J. A. An overview of Cancer (Phila.), 76: 442-452, 1995. adenocarcinoma of the small intestine. Oncology. 11: 529-536, 1997.

A I Neugut, J S Jacobson, S Suh, et al.

Cancer Epidemiol Biomarkers Prev 1998;7:243-251.

Updated version Access the most recent version of this article at: http://cebp.aacrjournals.org/content/7/3/243

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cebp.aacrjournals.org/content/7/3/243. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.