Annals of Oncology Advance Access published online on September 2, 2008
Annals of Oncology, doi:10.1093/annonc/mdn595
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Cancer risks in Crohn disease patients
1 Division of Molecular Genetic Epidemiology, German Cancer Research Center (DKFZ), Heidelberg, Germany
2 Center for Family and Community Medicine, Karolinska Institute, Huddinge, Sweden
* Correspondence to: K. Hemminki, MD, PhD., Division of Molecular Genetic Epidemiology, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 580, D-69120 Heidelberg, Germany. Tel: 49-6221-421800; Fax: 49-6221-421810. E-mail: k.hemminki{at}dkfz.de
 |
Abstract |
|---|
 Top Abstract introductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
Background: Patients diagnosed with Crohn disease (CD) are known to be at an increased risk of bowel cancers and lymphoma. CD is an autoimmune disease and we hypothesize that the patients are predisposed to a wider spectrum of cancers.
Patients and methods: A CD research database was constructed by identifying hospitalized CD patients from the Hospital Discharge Register and cancer patients from the Swedish Cancer Registry. Follow-up of 21 788 CD patients first hospitalized during the years 1964–2004 identified 1424 cancer cases. Standardized incidence ratios (SIRs) were calculated by comparing cancers in CD patients with subjects without CD.
Results: In addition to the known sites, many additional sites were in excess in CD patients. These included liver, pancreatic, lung, prostate, testicular, kidney and skin (squamous cell) cancers; nonthyroid endocrine tumors and leukemia. The previously established sites showed the highest SIRs; however, SIRs >2.0 were noted for the novel sites of the liver, testis and kidney. For testicular cancer, the SIR of seminoma was 2.74. Cancer risks were influences by age at first hospitalization for CD but whether the age effects were increasing or decreasing depending on the cancer type.
Conclusions: This large study identified many novel subsequent cancers in CD patients.
age at onset, autoimmunity, cancer risk, inflammatory bowel disease, subsequent cancer
|
introduction |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
Crohn disease (CD) is a common inflammatory bowel disease, in which nonpathogenic, commensal intestinal bacteria are thought to trigger a chronic dysregulated immune response against the mucosal barrier function [1, 2]. CD is considered an archetypical inflammatory barrier disease, which is characterized by disrupted barrier function maintenance by adaptive and innate immunity. Other diseases in this category include ulcerative colitis, atopic eczema, asthma, sarcoidosis and psoriasis [1]. Family history is a strong risk factors of CD, which has been observed both in family and twin studies [3–6]. Accordingly, several susceptibility genes for CD have been identified [1, 7], including nine associated loci detected in a single genome-wide scan [8].
CD patients are at an increased risk of many cancers, and colon cancer has been regarded as the most important cause of excess mortality in the patients [9]. A meta-analysis published in the year 2007 identified 34 studies and 60 122 CD patients who had been studied for subsequent cancers [10]. Colon and small bowel cancers and lymphomas were in excess; additionally, extraintestinal cancers were marginally in excess (relative risk 1.23) but no single site alone reached statistical significance. The medication for CD has changed over time and there is a need to follow the trends in cancer in the patients. While the traditional medication by 5-aminosalicylates is believed to provide some protection against colorectal cancer [11], immunosuppressive agents, including methotrexate and azathioprine, are carcinogenic [12, 13]. Around 1999, tumor necrosis factor (TNF) antagonists were taken into clinical use, particularly for patients with a severe disease, with emerging data on the associated risk for lymphoma [12–15].
We identified a total of 21 788 hospitalized CD patients throughout Sweden and followed their cancers from the National Cancer Registry. We carried out a detailed study on cancer risks in CD patients with a hypothesis that cancer risk could be shown at many sites, typical of an autoimmune disease. Compared with the similar studies recently reviewed [10], the present one is more than three times larger than any of the published ones outside Sweden; the one Swedish study with a matching size of CD patients considered only lymphohematopoietic neoplasms [16]. The size of the present study allowed us to consider risks at many cancer sites for which the earlier studies have been underpowered. We have no data on the medication but we followed separately CD patients who were first hospitalized after 1999 in order to search for indications of changing cancer trends after renewed medication.
|
patients and methods |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
The CD research database was constructed by linking several national Swedish registers. Statistics Sweden, the Swedish government-owned statistics bureau, provided the Multigeneration Register with data on the population, covering 11.5 individuals. Linkages were carried out to national census data, in order to obtain individual occupational status which was included because of its potential association to cancer. The final link was made by adding individual data from the Swedish Hospital Discharge Register that records data on all discharges with dates of hospitalization and diagnoses since the 1960s with a complete nationwide coverage since 1986. All linkages were carried out by the use of an individual national identification number that is assigned to each person in Sweden for their lifetime. This number was replaced by a serial number for each person in order to provide anonymity. The serial number was used to check that each individual was only entered once, for his or her first appearance with a CD diagnosis. The Cancer Registry records all new cases of cancer and close to 100% of the cases are histologically or cytologically confirmed at a full national coverage [17]. A four-digit diagnostic code according to the 7th revision of the International Classification of Diseases (ICD-7) was used.
outcome and adjustment variables
CD patients were retrieved from hospital discharges reported according to ICD-7 (1964–1968, codes 572.00 and 572.09), ICD-8 (1969–1986, code 563.00), ICD-9 (1987–1996, 555) and ICD-10 (1997–, code K50). A total of 21 788 CD patients were identified. Using codes for operations, we were able to identify 578 patients who had undergone colectomy; these patients were removed from the analysis because they were at a decreased risk or no risk of colon cancer. The follow-up of cancer was started in 1964 and extended to 2004. The individual variables controlled for in the analysis include gender, age at diagnosis (categorized <30, 30–39, 40–49, 50–59, 60–69 years), occupational status [six groups: (i) farmers, (ii) unskilled/skilled workers, (iii) white-collar workers (iv), professionals (v), self-employed and (vi) all others] and region [three groups: (i) large cities, Stockholm, Gothenburg and Malmo; (ii) Southern Sweden and (iii) Northern Sweden], allowing adjustment for possible regional differences in hospitalization rates.
statistical analysis
Person-years were calculated from the first hospitalization for CD until cancer diagnosis, death, emigration, or closing date, 31 December 2004. Standardized incidence ratios (SIRs) were calculated as the ratio of observed (O) to expected (E) number of cases. The expected number of cases was calculated for age (5-year groups), sex, period (5-year groups), region and socioeconomic status-specific SIRs for the population without CD.
|
results |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
Among 21 788 CD patients hospitalized for the first time during the years 1964–2004, a total of 1424 patients developed cancer after being hospitalized for CD, giving an overall SIR of 1.52 (‘All’) and an SIR of 1.29 for cancer diagnosed later than 1 year of the hospitalization (1+), as shown in Table 1. Only sites with at least 20 cases for the whole period were included; none of the rarer sites showed a significant difference, including Hodgkin's disease [‘All’: N = 8; SIR = 1.39; 95% confidence interval (CI) = 0.59–2.75]. We show separately data for ‘All’ and 1+ because for some cancers the risks were very high during the first year, probably because of a concomitant diagnosis of CD and cancer; however, significant increases in SIRs agreed for most sites for the ‘All’ and 1+ periods. We prefer to discuss the results for ‘All’ because even the cancers diagnosed within the first year were true diagnoses, verified as any other cancers in the Cancer Registry. The highest SIRs for ‘All’ was noted for small intestinal cancer (SIR = 13.82), followed by colon, liver, testis and kidney cancers and non-Hodgkin's lymphoma (SIRs between 2.0 and 3.0). Increased risks were observed even for common sites, such as the lung (1.50) and the prostate (1.19). For most significant sites, the SIR did not systematically change according to the length of the follow-up period, even when <1 year was excluded.
|
Some more detailed analyses were carried out for some sites (data not shown). Histology-specific analysis for small intestinal cancers showed that adenocarcinoma was increased more (‘All’: N = 26; SIR = 20.00; 95% CI = 13.05–29.34) than carcinoid tumors (N = 28; SIR = 13.46; 95% CI = 8.94–19.48). Adenocarcinomas were increased throughout the follow-up period, whereas carcinoid tumors were in excess only in follow-up periods <5 years. For testicular cancer, seminoma and nonseminoma histologies were analyzed separately. The excess was mainly due to seminoma (‘All’: N = 14; SIR = 2.74; 95% CI = 1.49–4.61); none of the SIRs for nonseminoma were significant. For nonthyroid (other) endocrine tumors, only parathyroid adenomas were in excess (‘All’: N = 31; SIR = 2.42; 95% CI = 1.64–3.43). Two specific leukemia subtypes showed an increase: acute myeloid leukemia (‘All’: N = 12; SIR = 2.02; 95% CI = 1.04–3.55) and chronic myeloid leukemia (‘All’: N = 10; SIR = 3.65; 95% CI = 1.74–6.74). For the former, however, the only significant period was <1 year after diagnosis, while for the latter no cases were noted in the early period (‘All 1+’: N = 10; SIR = 3.76; 95% CI = 1.79–6.94).
The effect of age at first hospitalization for CD is shown in Table 2. For all sites, a small increase in SIR was observed for those hospitalized at an old age. There were, however, large differences in the trends for hospitalization age by site. Colon, rectal and liver cancers were increased most in those hospitalized before age 25 years. Stomach, small intestinal, pancreatic and lung cancers; endocrine tumors and non-Hodgkin's lymphoma showed the highest SIRs in those hospitalized at ages >63 years.
|
In order to study periodic effects with respect to the changing therapeutic regimens, we analyzed separately cancer risk in CD patients who were first hospitalized before the 1990s, in 1990–1998 and 1999–2004 (Table 3). Separate analyses were done for those diagnosed <1 and 1–4 years after hospitalization. The overall SIRs were equal or higher for patients diagnosed after 1998 compared with those diagnosed earlier. The SIRs at individual sites were too few to allow conclusions about trends. Only considering the period 1–4 years after hospitalization, small intestinal and liver cancers, however, were increased more in the last compared with the earlier periods. Breast and endometrial cancers were increased only in the last period <1 year after hospitalization.
|
|
discussion |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
Many patients with autoimmune diseases, including those with CD, ulcerative colitis, rheumatoid arthritis, systemic lupus erythematosus and sarcoidosis, are at a risk of cancer [10, 18–26]. Data linkages between the Cancer Registry and the Hospital Discharge Register offer an excellent and unbiased opportunity to revisit cancer risk in CD patients. Using data on operations, we were able to remove the patients with colectomies from the study population. We hypothesized that the risks should be shown at many sites, typical of an autoimmune disease.
The diagnostic accuracy for hospitalized cases of CD has previously been estimated in Sweden by comparing the hospital discharge data with regional registers with individually reviewed patient data [16]. The concordance for the specific types of inflammatory bowel disease was 96%. According to that article, 69.2% of the patients in the regional registers with apparently complete catchment of CD patients were found in the Hospital Discharge Register. This figure is, however, an underestimate because the regional registers were started a decade before the Hospital Discharge Register and three decades before it reached a nationwide coverage. Nevertheless, for studies of the present kind, diagnostic accuracy rather than coverage of all cases is crucial, yet with the caveat that severe cases are most likely to be hospitalized.
The previously established sites, the small intestine, colon and lymphatic system (non-Hodgkin's lymphoma) were confirmed. Rectal cancer was also in excess for the whole follow-up period but not for the follow-up starting 1 year after diagnosis (1+). These sites showed the highest SIRs, notably the small intestine with an SIR of 13.82 and its adenocarcinoma histology with an SIR of 20.0. In the published meta-analysis, the relative risk of small intestinal cancer was 15.43 when large population-based studies were considered [10]. In this meta-analysis the risk for colon cancer was 2.59, compared with the present 2.93; no increase in rectal cancer was noted in the meta-analysis, compared with the present 1.55. The only other significant increase in the meta-analysis was lymphoma, with a relative risk of 1.42; in the present study only non-Hodgkin's lymphoma was significant with an SIR of 2.54. As there were 10 times more cases for non-Hodgkin's lymphoma than for Hodgkin's disease (with a nonsignificant SIR of 1.39) in the present study, the joint lymphoma risk would be higher in the present study than in the meta-analysis. The previous Swedish study, covering 19 024 hospitalized CD patients up to year 2000, reported an SIR of 1.55 for non-Hodgkin's lymphoma [16]. In that study, however, the cases diagnosed during the first year of follow-up were excluded, corresponding to the present SIR of 1.87 (‘All 1+’). Chronic myeloid leukemia was also in excess, SIR of 2.80, compared with the present SIR of 3.76 (‘All 1+’).
The spectrum of cancers that were increased after CD included many sites that have not been reported before [10]: liver (borderline increase has been reported before [27]), pancreas, lung, prostate, testicular, kidney and (squamous cell) skin cancers; endocrine tumors and leukemia. SIRs >2.0 were, however, noted for the novel sites of the liver, testis and kidney. For testicular cancer, seminoma rather than nonseminoma was in excess. Interestingly, a significant increase in lung cancer was observed only in the longest follow-up group. With increasing survival in CD, lung cancer may become another major cause of excess mortality, approaching colon cancer in significance. The high risks noted for non-Hodgkin's lymphoma and squamous cell skin cancer resemble the spectrum of cancers observed after immunosuppression, suggesting the underlying role for autoimmunity [28–31]. Recently, >30 susceptibility loci were described for CD, reinforcing the previous understanding of CD pathogenesis regarding dysregulated host–bacterial interactions through altered autophagy and innate immunity [6, 7, 32].
It is likely that CD is often diagnosed before the first hospitalization. In a previous Swedish study on inflammatory bowel disease patients diagnosed up to 1990, the median delay time for hospitalization was 0.5 years (mean 4.4 years) [16]. Thus, the data in Table 2 refer to the maximal age at diagnosis and those in Table 3 to the latest possible period for hospitalization. Cancer risks depended on the age at first hospitalization for CD but the trends varied extensively by cancer type. The SIRs for colon, rectal and liver cancers declined by age at hospitalization, and for the two latter ones no excess risk was seen in CD patients hospitalized after age 63 years. By contrast, stomach, small intestinal, pancreatic and lung cancers; endocrine tumors and non-Hodgkin's lymphoma showed the highest SIRs in those hospitalized at ages >63 years. Although we have no scientific explanation to these age-dependent differences, the sites most severely affected by fulminate inflammation showed the highest risk in those hospitalized at an early age. Risks at more distant sites, where the effects were mediated through systemic immunity, followed the overall age-dependent increase in cancer. These simplified conclusions, however, may be biased by the high cancer risks during the first year of CD diagnosis. We were also planning to follow the trends of cancers with respect to the changing spectrum of medication, particularly in regard to the introduction of TNF antagonists since 1999 [12–15]. The sparseness of individual cancers, however, did not allow conclusions about the trends in towards the end of the follow-up period.
In summary, the follow-up of CD patients showed a pattern of increased risk for many subsequent cancers that have not been reported before, probably because of the high statistical power of the present study.
|
funding |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
Deutsche Krebshilfe; the Bundesministerium für Bildung und Forschung (BMBF) for the National Genome Research Network plus (NGFN+); the Swedish Cancer Society; the Swedish Council for Working Life and Social Research; the European Union (LSHC-CT-2004-503465).
|
Acknowledgements |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
The used database was created by linking registers maintained at Statistics Sweden and the Swedish Cancer Registry.
Received for publication February 22, 2008. Revision received July 29, 2008. Accepted for publication July 30, 2008.
|
References |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionfundingAcknowledgementsReferences |
|---|
1. Schreiber S, Rosenstiel P, Albrecht M, et al. Genetics of Crohn disease, an archetypal inflammatory barrier disease. Nat Rev Genet (2005) 6:376–388.[Web of Science][Medline]
2. Sartor R. Mechanisms of disease: pathogenesis of Crohn's disease and ulcerative colitis. Nat Clin Pract Gastroenterol Hepatol (2006) 3:390–407.[CrossRef][Web of Science][Medline]
3. Orholm M, Fonager K, Sorensen HT. Risk of ulcerative colitis and Crohn's disease among offspring of patients with chronic inflammatory bowel disease. Am J Gastroenterol (1999) 94:3236–3238.[CrossRef][Web of Science][Medline]
4. Halfvarson J, Bodin L, Tysk C, et al. Inflammatory bowel disease in a Swedish twin cohort: a long-term follow-up of concordance and clinical characteristics. Gastroenterology (2003) 124:1767–1773.[CrossRef]
5. Halfvarson J, Jess T, Bodin L, et al. Longitudinal concordance for clinical characteristics in a Swedish-Danish twin population with inflammatory bowel disease. Inflamm Bowel Dis (2007) 13:1536–1544.[CrossRef][Web of Science][Medline]
6. Lewis CM, Whitwell SC, Forbes A, et al. Estimating risks of common complex diseases across genetic and environmental factors: the example of Crohn disease. J Med Genet (2007) 44:689–694.
7. Hampe J, Franke A, Rosenstiel P, et al. A genome-wide association scan of nonsynonymous SNPs identifies a susceptibility variant for Crohn disease in ATG16L1. Nat Genet (2007) 39:207–211.[CrossRef][Web of Science][Medline]
8. The Wellcome Trust Case Control Consortium. Genome-wide association study of 14,000 cases of seven common diseases and 3,000 shared controls. Nature (2007) 447:661–683.[CrossRef][Web of Science][Medline]
9. Askling J, Dickman PW, Karlen P, et al. Family history as a risk factor for colorectal cancer in inflammatory bowel disease. Gastroenterology (2001) 120:1356–1362.[CrossRef][Web of Science][Medline]
10. von Roon AC, Reese G, Teare J, et al. The risk of cancer in patients with Crohn's disease. Dis Colon Rectum (2007) 50:839–855.[CrossRef][Web of Science][Medline]
11. van Staa TP, Card T, Logan RF, Leufkens HG. 5-Aminosalicylate use and colorectal cancer risk in inflammatory bowel disease: a large epidemiological study. Gut (2005) 54:1573–1578.
12. Baumgart DC, Sandborn WJ. Inflammatory bowel disease: clinical aspects and established and evolving therapies. Lancet (2007) 369:1641–1657.[CrossRef][Web of Science][Medline]
13. Wolfe F, Michaud K. Biologic treatment of rheumatoid arthritis and the risk of malignancy: analyses from a large US observational study. Arthritis Rheum (2007) 56:2886–2895.[CrossRef][Web of Science][Medline]
14. Blomqvist P, Feltelius N, Lofberg R, Ekbom A. A 10-year survey of inflammatory bowel diseases-drug therapy, costs and adverse reactions. Aliment Pharmacol Ther (2001) 15:475–481.[CrossRef][Web of Science][Medline]
15. Askling J. Malignancy and rheumatoid arthritis. Curr Rheumatol Rep (2007) 9:421–426.[CrossRef][Medline]
16. Askling J, Brandt L, Lapidus A, et al. Risk of haematopoietic cancer in patients with inflammatory bowel disease. Gut (2005) 54:617–622.
17. Centre for Epidemiology. Cancer Incidence in Sweden 2005 (2007) Stockholm, Sweden: The National Board of Health and Welfare.
18. Hakulinen T, Isomaki H, Knekt P. Rheumatoid arthritis and cancer studies based on linking nationwide registries in Finland. Am J Med (1985) 78:29–32.[Web of Science][Medline]
19. Gridley G, McLaughlin JK, Ekbom A, et al. Incidence of cancer among patients with rheumatoid arthritis. J Natl Cancer Inst (1993) 85:307–311.
20. Mellemkjaer L, Linet MS, Gridley G, et al. Rheumatoid arthritis and cancer risk. Eur J Cancer (1996) 32A:1753–1757.[CrossRef]
21. Ekstrom K, Hjalgrim H, Brandt L, et al. Risk of malignant lymphomas in patients with rheumatoid arthritis and in their first-degree relatives. Arthritis Rheum (2003) 48:963–970.[CrossRef][Web of Science][Medline]
22. Askling J, Fored CM, Brandt L, et al. Risks of solid cancers in patients with rheumatoid arthritis and after treatment with tumour necrosis factor antagonists. Ann Rheum Dis (2005) 64:1421–1426.
23. Landgren O, Engels EA, Caporaso NE, et al. Patterns of autoimmunity and subsequent chronic lymphocytic leukemia in Nordic countries. Blood (2006) 108:292–296.
24. Landgren O, Linet MS, McMaster ML, et al. Familial characteristics of autoimmune and hematologic disorders in 8,406 multiple myeloma patients: a population-based case-control study. Int J Cancer (2006) 118:3095–3098.[CrossRef][Web of Science][Medline]
25. Landgren O, Engels EA, Pfeiffer RM, et al. Autoimmunity and susceptibility to Hodgkin lymphoma: a population-based case-control study in Scandinavia. J Natl Cancer Inst (2006) 98:1321–1330.
26. Baecklund E, Iliadou A, Askling J, et al. Association of chronic inflammation, not its treatment, with increased lymphoma risk in rheumatoid arthritis. Arthritis Rheum (2006) 54:692–701.[CrossRef][Web of Science][Medline]
27. Bernstein CN, Blanchard JF, Kliewer E, Wajda A. Cancer risk in patients with inflammatory bowel disease: a population-based study. Cancer (2001) 91:854–862.[CrossRef][Web of Science][Medline]
28. Birkeland S, Storm H, Lamm L, et al. Cancer risk after renal transplantation in the Nordic countries, 1964–1986. Int J Cancer (1995) 60:183–189.[Web of Science][Medline]
29. IARC. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans (1996) Lyon, France: IARC.
30. Adami J, Gabel H, Lindelof B, et al. Cancer risk following organ transplantation: a nationwide cohort study in Sweden. Br J Cancer (2003) 89:1221–1227.[CrossRef][Web of Science][Medline]
31. Vajdic CM, McDonald SP, McCredie MR, et al. Cancer incidence before and after kidney transplantation. JAMA (2006) 296:2823–2831.
32. Barrett JC, Hansoul S, Nicolae DL, et al. Genome-wide association defines more than 30 distinct susceptibility loci for Crohn's disease. Nat Genet (2008) 40:955–962.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||