Scandinavian Journal of Gastroenterology. 2014; Early Online, 1–7
ORIGINAL ARTICLE
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Windsor on 07/10/14 For personal use only.
Increased risk of colorectal neoplasia in adult patients with cystic fibrosis: a matched case–control study ILANA GORY1, GREGOR BROWN1, JOHN WILSON2, WILLIAM KEMP1, ELDHO PAUL3 & STUART K ROBERTS1 1
Department of Gastroenterology, The Alfred Hospital, Melbourne, Australia, 2Department of Respiratory Medicine, The Alfred Hospital, Melbourne, Australia, and 3Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Australia
Abstract Objective. As the lifespan of patients with cystic fibrosis (CF) increases, concerns regarding the occurrence of bowel cancer have arisen. However, previous cohort studies have yielded contradictory results as to the risk of colorectal cancer (CRC) in the CF population. The aim of this study was to determine whether CF is associated with an increased risk of colorectal neoplasia. Methods. We performed a matched case–control study at a statewide CF center. Cases were adult CF patients undergoing colonoscopy in a 5-year period from 2007 to 2012. Controls were matched in a 2:1 ratio for age, gender, and colonoscopy indication. Results. A total of 50 patients with CF who underwent colonoscopy were identified. Among CF patients, there were 5 (10%) cases with CRC, 1 (2%) with ileal adenocarcinoma, 13 (26%) with adenomas, and 16(32%) with advanced adenomas. In contrast, there was 1 (1%) case with CRC, 11 (11%) with adenomas, and 6 (6%) with advanced adenomas identified among controls. Compared to controls, CF was associated with a 10-fold increased risk of CRC (odds ratio [OR] = 10.0, 95% confidence interval [CI]: 1.2–85.6; p = 0.03), threefold increased risk of adenomas (OR = 3.34, 95%CI: 1.23–9.08; p = 0.018], and sevenfold increased risk of advanced adenomas (OR = 6.95, 95% CI: 2.30–21.01; p = 0.001). Moreover, the quality of bowel preparation was significantly worse in CF patients with a fourfold (p = 0.01) increased likelihood of a poor preparation. Conclusions. Our data suggest that CF patients are at significantly increased risk of CRC compared to age, sex, and colonoscopy-indication matched controls. Consideration should be given to the introduction of a CRC surveillance program in the CF population.
Key Words: adenomas, case–control, colorectal cancer, cystic fibrosis
Introduction Cystic fibrosis (CF) is the most common life-limiting genetic condition in Caucasians with a gene carrier frequency of 1 in 25 and affecting 1 in every 2500 live births [1]. It is a multisystem disorder characterized by a defect in the CF transmembrane conductance regulator (CFTR), thereby affecting several organs including the airways, pancreatic ducts, bile ducts and intestine [2]. Over the past three decades, the median survival of CF patients in developed countries has more than doubled due to advances in the early diagnosis and
treatment of CF including improvements in pulmonary, nutritional, and lung transplantation care [3-7]. Consequently, concerns surrounding age-related conditions such as colorectal cancer (CRC) in the CF population have heightened in recent times [8,9]. Increasing data suggest that CF predisposes to cancer, and in particular gastrointestinal malignancy [10,11]. Since the first report in 1982 of a cholangiocarcinoma in a CF patient [12], multiple case reports have been published of neoplasms in CF involving the digestive tract, mainly involving the pancreas, small bowel, and colon [8,13-19]. In 1993, Sheldon et al.
Correspondence: Stuart Roberts, Department of Gastroenterology, The Alfred Hospital, Commercial Rd, Melbourne 3004, Australia. Tel: +61 3 9076 3375. Fax: +61 3 9076 2194. E-mail: [email protected]
(Received 1 June 2014; revised 13 June 2014; accepted 13 June 2014) ISSN 0036-5521 print/ISSN 1502-7708 online 2014 Informa Healthcare DOI: 10.3109/00365521.2014.936035
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Windsor on 07/10/14 For personal use only.
2
I. Gory et al.
published a cohort study that described a possible association between adenocarcinoma and CF, particularly involving the ileum and pancreas [20]. Several larger studies have further examined the risk of cancer in CF patients reporting an increased risk of between three- to six-fold in the CF population [10,11,21-23]. However these studies have yielded inconsistent results and have been mostly focused on the composite target of digestive tract malignancies. Moreover, in some of these studies limited histological data were available and information on genotype and family history of malignancy was incomplete. Nevertheless, a growing body of evidence mostly derived from case reports [8,15-18], limited case series [19], and lung transplant cohorts [9,11] strongly suggest that there is an increased risk of CRC in CF. In this context, we observed anecdotally that CRC was occurring in CF at a relatively young age at our statewide CF referral center. Hence, we conducted an exploratory case–control study to evaluate the relative risk of colorectal neoplasia among adult patients with CF.
Patients and methods Study subjects and design This is a retrospective exploratory case–control study of patients with the diagnosis of CF from The Alfred Hospital, which is a statewide adult CF referral center (catchment 3 million) and the largest CF center in Australia with over 300 patients undergoing assessment at least three times annually by a respiratory physician. Eligible case patients included adults ‡18 years with a confirmed diagnosis of CF based on clinical, sweat, and genetic testing who underwent colonoscopy for any indication between June 2007 and June 2012. The control group consisted of patients who underwent colonoscopy during the same period without the diagnosis of CF.
Identification of cases All patients who undergo colonoscopy at The Alfred Hospital are recorded on the endoscopy database. Using a combination of this system, medical records, and the hospital coding data, we identified all CF patients who had undergone a colonoscopy during the 5-year period from June 1 2007 to June 1 2012. Demographical and epidemiological data was obtained by examination of hospital medical records.
Identification of controls For each case, we randomly selected two controls from the endoscopy database blinded to the results of the colonoscopy findings. Using this system we matched for age, gender, and indication of colonoscopy. For each case subject, two controls who met all the above matching criteria could be found. Controls were not matched for lung function or nutritional status. Colonoscopy indications and findings Indications for colonoscopy were recorded as follows: family history of malignancy, rectal bleeding, irondeficiency anemia, abdominal pain, previous polyps, diarrhea, constipation, and abnormal imaging. Possible confounders were also recorded. These include personal or family history of cancer and a history of lung transplantation. Endoscopy findings of colorectal and intestinal malignancy, and size, number, and location of polyps were recorded. Hospital electronic medical records were then examined to confirm histological diagnosis of adenocarcinoma and adenomas and their anatomical location were identified. In addition, the quality of the bowel preparation as subjectively reported by the endoscopist on the endoscopy report (i.e., poor, average, good) was recorded. Poor preparation was defined as solid or adherent stool that was unable to be washed away and therefore the preparation was considered inadequate to exclude colonic lesions. Average preparation was defined as opaque fluid and some adherent stool but was able to be washed away and provide reasonable views to exclude lesions. Good preparation was defined as clear or opaque fluid only with good views of the entire colon. Definitions Colorectal adenocarcinoma and adenomas were defined histopathologically using tissues retrieved at surgery or colonoscopic polypectomy. Advanced adenomas were defined as adenomas that were either >10 mm, demonstrated a villous component, and had high-grade dysplasia, multiplicity, or early invasive cancer. Statistical analysis Statistical analysis was conducted using SAS software version 9.2 (SAS Institute, Cary, NC, USA). Data were summarized using mean ± standard deviation (SD) or median (range) for continuous variables and number (%) for categorical variables. Risk factors for
Colorectal cancer in patients with cystic fibrosis
20% of the cohort. Other common indications included abdominal pain (16%), loss of weight (10%), and family history of CRC (10%) (Table I).
Table I. Baseline demographics and indication for colonoscopy.
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Windsor on 07/10/14 For personal use only.
Characteristic
CF (n = 50)
Age (years), mean ± SD 37.6 ± 10.9 Male sex % 54 Indication for colonoscopy Bleeding 22 (45%) Rectal bleeding 10 (20%) Fecal occult blood test 7 (14%) Iron-deficiency anemia 5 (10%) Abdominal pain 8 (16%) Family history 5 (10%) Loss of weight 5 (10%) Diarrhea 4 (8%) Constipation 3 (6%) Polyp follow up 1 (2%) Abnormal imaging 1 (2%)
3
Control (n = 100) 37.6 ± 10.8 54
Colorectal adenocarcinoma 44(45%) 20 (20%) 14 (14%) 10 (10%) 16 (16%) 10 (10%) 10 (10%) 8 (8%) 6 (6%) 2 (2%) 2 (2%)
Among the 50 CF patients, there were 5 (10%) cases of colorectal malignancy and 1 (2%) case of ileal adenocarcinoma. In contrast, there was only 1 (1%) case of CRC among the 100 controls. The location and size of the tumors are detailed in Table II. When compared to controls, a diagnosis of CF was associated with a 12-fold increased risk of malignancy identified at colonoscopy (OR = 12.0, 95% CI: 1.45–99.67; p = 0.02]. CF was still associated with a 10-fold increased risk of CRC after exclusion of the single case of ileal adenocarcinoma (OR = 10.0, 95% CI: 1.2–85.6; p = 0.03).
Abbreviations: CF = Cystic fibrosis; SD = Standard deviation.
the development of CRC in CF patients were assessed using conditional logistic regression, taking into account the matched design, with results reported as odds ratios (OR) and 95% confidence intervals (CI). Subgroup analysis was performed in patients with colorectal adenomas to compare histology, number, and size of polyps identified. Statistical significance was set at a two-sided p-Value of 0.05.
Colorectal adenomas There were 13 (26%) cases with adenomatous polyps identified among the 50 CF patients compared with 11 (11%) among the 100 controls. In comparison with controls, CF was associated with a threefold increased risk of adenoma development (OR = 3.34, 95% CI: 1.23–9.08; p = 0.018). On histopathology, CF patients were more likely to have a villous component compared with controls (69% vs. 27%); however this was not statistically significant (p = 0.99). Other characteristics including size and number of adenomas were similar between the groups (Table III). In total, 50 polyps were identified among the 24 patients (Figure 1). The histology of adenomas was predominantly tubulovillous (low-grade dysplasia) among CF patients and tubular adenoma among controls (Table III).
Study conduct This study was conducted in accordance with the Helsinki Declaration of 1975 and approved by the Ethics Committee of the Alfred Hospital, Melbourne, Australia, prior to commencement of the study. Results Between 2007 and 2012, 50 patients with CF who had colonoscopies were identified of which 54% were male with a mean age of 37.6 (range 20–70) years. The clinical characteristics of the CF and control groups are shown in Table I. The major indication for colonoscopy was rectal bleeding, being present in
Advanced adenomas There were 22 patients with advanced adenomas (i.e. >10 mm, villous component, high-grade dysplasia,
Table II. Summary of findings in patients with adenocarcinoma. Age CF 46 48 46 46 50 33 Control 70
Gender
Indication
Site
Size (mm)
CF gene
Male Female Female Male Male Male
Family history Rectal bleeding Positive FOBT Positive FOBT Iron-deficiency anemia Abdominal pain
Sigmoid Rectal Sigmoid Cecum Ascending Ileum
15 45 30 40 60 50
DF508/DF508 DF508/DF508 DF508/DF508 DF508/DF508 DF508/DF508 DF508/g542x
No No No No No No
Female
Iron-deficiency anemia
Ascending
50
-
No
Abbreviations: CF = Cystic fibrosis; FOBT = Fecal occult blood test.
Transplant
4
I. Gory et al.
Table III. Summary of findings in patients with colorectal adenomas. Characteristic
Overall (n = 24)
CF (n = 13)
Control (n = 11)
p-Value*
Age (years), mean ± SD Male sex % Size (mm) Median (range) >10 mm (%) Number of polyps, median (range) Histology Villous component Tubular adenoma
43.2 ± 9.0 50
41.5 ± 7.2 53.9
45.3 ± 10.8 45.5
0.48 1.00
6.5 (1–18) 6 (25%) 1 (1–7)
8 (1–18) 5 (38.5%) 1 (1–7)
6 (3–14) 1 (9.1%) 2 (1–3)
0.19
12 (50%) 12 (50%)
9 (69%) 4 (31%)
3 (27%) 8 (73%)
0.58 0.99
10 9 8 Number of patients
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Windsor on 07/10/14 For personal use only.
*p-Values are from conditional logistic regression analysis. Abbreviations: CF = Cystic fibrosis; SD = Standard deviation.
7 6
1 polyps
5
2 polyps
4
3 polyps
3
4 polyps
2
7 polyps
1 0 Cystic fibrosis
Controls
Number of polyps Figure 1. Number of colorectal adenomas per patient is shown. In total, 50 polyps were identified among the 24 patients. There was no difference in the number of polyps in the CF group versus the controls.
multiplicity or early invasive cancer), including 16 (32%) adenomas among CF patients and 6 (6%) adenomas among controls. Advanced adenoma was approximately seven times more likely to occur in the CF population when compared to controls (OR = 6.95, 95% CI: 2.30–21.01; p = 0.001). Bowel preparation The subjective quality of the bowel preparation was significantly worse in the CF group with 22% having a poor preparation compared to 7% in controls (p = 0.001). When compared to controls there was a fourfold increased likelihood (OR = 4.49, 95% CI: 1.4–14.39; p = 0.01) of a poor compared to average-good preparation in CF patients (Figure 2). Discussion The main finding of this exploratory case–control study is the demonstration that patients with CF are at a significantly higher risk of colorectal neoplasia,
including CRC, advanced adenomas, and adenomatous polyps compared to age, gender, and colonoscopy-indication matched controls. Specifically, we found that CF is associated with a 10-fold increased risk of CRC, threefold higher risk of adenomas, and approximately sevenfold increased risk of advanced adenomas compared to carefully matched controls. Moreover, our study might have even underestimated the risk of colonic neoplasia in CF as we found a significantly increased frequency of poor bowel preparation in the CF group that could have reduced our adenoma detection rates, particularly for small lesions. These important data add to the accumulating body of evidence from other cohort studies in demonstrating that subjects with CF are at increased risk of digestive tract malignancy and, in particular, colon cancer. In addition, we highlight the significant risk of associated colorectal adenomas in CF. Several large studies have examined the cancer risk in CF patients (Table IV) [10,11,21-23]. In the largest of these, Maisonneuve et al. performed a retrospective cohort study of more than 41,000 North
Colorectal cancer in patients with cystic fibrosis
5
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Windsor on 07/10/14 For personal use only.
Number of patients (%)
100
80
60 p < 0.001 CF 40
Controls
20
0 Good
Average
Poor
Quality of bowel preparation Figure 2. Bowel preparation in CF patients versus controls is shown. Compared to controls, patients with CF had significantly worse bowel preparation (22% vs. 7%, p < 0.001). CF patients were four times more likely to have poor or average preparation when compared to controls (OR = 4.49, 95% CI: 1.4–14.39; p = 0.01). Abbreviations: CF = Cystic fibrosis; CI = Confidence interval; OR = Odds ratio.
American CF patients and while observing no overall increase in cancer, they found that digestive tract cancers, including small bowel, colon, and biliary tract, were 3.5 times more likely to develop in CF, and that the cancer risk was particularly high in the transplanted population [11]. Indeed, they found an approximately sevenfold increased risk of developing colon cancer in non-transplanted CF subjects that increased to 30-fold among transplant recipients when compared to the general population [11]. Meyer et al. reported similar findings, observing a significantly higher incidence of colon cancer in transplanted CF subjects compared to non-CF transplant recipients [9]. In contrast, Johannesson et al. in a Swedish study of 884 CF patients observed a fivefold increased risk of digestive tract cancer in CF, yet reported no significant increase in the CRC risk when compared to the general population [22]. Nevertheless, the above data combined with ours strongly suggest that the adult CF population is at risk of CRC and, therefore, should be considered for enrollment into CRC surveillance programs. Of the five CF patients in our study diagnosed with CRC, four were men with the mean age of 49.2 years (range 46–60 years). This age is significantly lower than the mean age of 69 years at diagnosis of CRC
among the general population in Australia [24]All six had known CF mutations, five being DF508 homozygous and one being DF508/G5F42X. One patient had a family history of CRC; however, this was his indication for endoscopy and as such was controlled for. Otherwise, none of the patients had known risk factors other than CF for the development of digestive tract cancer with none having been transplanted, exposed to immunosuppression, or a history of inflammatory bowel disease. The predominant indication for colonoscopy in the five patients with CRC was bleeding (n = 4), either overt (n = 1) or occult (n = 3). The indication in the patient with the ileal adenocarcinoma was abdominal pain. Multivariate analysis to identify predictors for the development of adenocarcinoma in CF could not be performed due to the small number of patients with adenocarcinoma in this study. In our study, we found a sevenfold increased risk of developing advanced adenomas in CF patients. This composite group (i.e. adenomas >10 mm or with adverse histological features or adenocarcinoma) is important since it is well established that adenomas of this type indicate a risk of metachronous advanced adenomas and cancers and an 8% risk of developing colorectal malignancy within 10 years in the general
Table IV. Previous cohort studies examining the risk of malignancy in CF. Author (year) Maisonneuve et al. (2012) Johannesson et al. (2009) Neglia et al. (1995)
Number of patients 41,188 884 28,511
Overall cancer risk increased
Digestive tract cancer risk increased
Colorectal cancer risk increased
NS 3.2 (SIR) NS
3.5 (SIR) NS 6.5 (OR)
6.2 (SIR) NS NE
Abbreviations: NS = Not statistically significant; NE = Not examined; SIR = Standard incidence ratio; OR = Odds ratio.
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Windsor on 07/10/14 For personal use only.
6
I. Gory et al.
population [25,26]. With the improved survival of CF patients, the adenoma-carcinoma sequence is more likely to occur in this aging population and as such identifying these high-risk lesions becomes more imperative. When advanced adenomas are identified, current guidelines recommend colonoscopic surveillance at 3-yearly intervals initially because of the increased risk of developing CRC [27]. Subjective assessment of the quality of bowel preparation in our study confirmed that patients with CF have a significantly higher frequency of a poor result compared to non-CF controls. This finding is not unexpected given that constipation, distal intestinal obstruction syndrome, and slow gastrointestinal transit time are well-documented complications of CF [28]. Poor bowel preparation may cause underreporting of polyps and other lesions due to poor visualization of the colonic mucosa, meaning that the risk of colorectal neoplasia in CF may in fact be higher than documented in this study. Further, we now recommend that CF patients undergo a more intensive bowel preparation regime before colonoscopy, particularly in view of their increased risk of colorectal lesions. While the mechanism(s) underlying the increased risk of digestive tract adenocarcinoma and colorectal neoplasia in CF is(are) unknown, there are several factors present in CF that could contribute. These include persistent pathological disease processes such as intestinal stasis [29], perturbations in gut microflora (e.g. bacterial overgrowth) [30], and chronic inflammation [31] that might increase cell proliferation and turnover, as well as malabsorption-related micronutrient deficiencies (e.g. selenium, vitamin E) that potentially play a role in cancer prevention [32,33]. In addition, failure to thrive and malnutrition that are commonly present in CF are associated with a higher risk of developing digestive tract cancers [11]. More specifically, a non-functioning CFTR has been linked to reduced serum- and membrane-bound levels of the 3-omega fatty acid, docosahexaenoic acid, in CF, which in epidemiological studies is associated with a reduced risk of colon cancer [34,35]. A further possible mechanism may follow from acid dumping into the lower gastrointestinal tract causing metaplastic change analogous to Barrett’s esophagus. We have previously documented hyper-acidification in CF secondary to gastroesophageal reflux disease in CF [36]. In addition, patients with CF have pancreatic failure with limited capacity to neutralize pH with bicarbonate, potentiating the acid environment. It should be noted that all five CRC subjects in this study were homozygous for the class V mutation F508del, associated with the severest form of pancreatic insufficiency.
While the strengths of this study include the evaluation of both CRC and adenomas, the assessment of histopathology, and inclusion of family history as a risk factor, the study has its limitations including small sample size compared with published CF registries. However, our single center study is from the largest CF center in Australia and, despite its size, it was able to show a significant increase in risk of CRC and advanced adenomas in CF compared to matched controls. In addition, information was not routinely collected on other confounders that may increase the risk of CRC such as smoking and obesity. Still, these risk factors are less likely to be present in the CF population, particularly smoking, because of the associated lung disease and need for compliance with Australian eligibility criteria requirements for lung transplantation [37]. Moreover, obesity is uncommon among adult patients with CF with most having a normal-to-low body mass index [38] often due to comorbidities, including bronchiectasis and malabsorption. In conclusion, our study demonstrates that patients with CF are at an increased risk of CRC, advanced adenomas, and adenomatous polyps. While CRC remains an uncommon but well-reported event in young CF patients [7,18,19], these data suggest that the CF population may benefit from enrollment in CRC screening surveillance programs, particularly as their survival continues to improve and the age-related CRC risk increases. Surveillance strategies could include periodic colonoscopy and/or fecal occult blood testing in all CF patients. Further prospective data are required for determining the importance of specific risk factors and for determining the most effective screening tool(s) for this high-risk population. Declaration of interest: None of the authors have any disclosures to declare for this study. No financial support was received to conduct this study. References [1] Cystic Fibrosis Australia,Inc. [online]. 2012. Available from http:www.cysticfibrosisaustralia.org.au. Accessed October 2012. [2] Cohn JA, Strong TV, Picciotto MR, Nairn AC, Collins FS, Fitz JG. Localization of the cystic fibrosis transmembrane conductance regulator in human bile dict epithelial cells. Gastroenterology 1993;105:1857–64. [3] FitzSimmons SC. The changing epidemiology of cystic fibrosis. J Pediatr 1993;122:1–9. [4] Reid DW, Blizzard CL, Shugg DM, Flowers C, Cash C, Greville HM. Changes in cystic fibrosis mortality in Australia, 1979-2005. Med J Aust 2011;195:392–5.
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Windsor on 07/10/14 For personal use only.
Colorectal cancer in patients with cystic fibrosis [5] Dodge JA, Lewis PA, Stanton M, Wilsher J. Cystic fibrosis mortality and survival in the UK: 1947-2003. Eur Respir J 2007;29:522–6. [6] Bellis G, Cazes MH, Parant A, Gaimard M, Travers C, Le Roux E, et al. Cystic fibrosis mortality trends in France. J Cyst Fibros 2007;6:179–86. [7] Ramalle-Gomara E, Perucha M, González MA, Quiñones C, Andrés J, Posada M. Cystic fibrosis mortality trends in Spain among infants and young children: 1981-2004. Eur J Epidemiol 2008;23:523–9. [8] Hernandez-Jimenez I, Fischman D, Cheriyath P. Colon cancer in cystic fibrosis patients: is this a growing problem? J Cyst Fibros 2008;7:343–6. [9] Meyer KC, Francois ML, Radford KL, Hawes DS, Mack TL, Cornwell RD. Colon cancer in lung transplant recipients with CF: increased risk and results of screening. J Cyst Fibros 2011;10:366–9. [10] Neglia JP, Fitzsimons SC, Maisonneuve P, Schöni MH, Schöni-Affolter F, Corey M, et al. The Cystic Fibrosis and Cancer Study Group. The risk of cancer among patients with cystic fibrosis. N Engl J Med 1995;332:494–9. [11] Maisonneve P, Marshall BC, Knapp EA, Lowenfels AB. Cancer risk in cystic fibrosis patients: a 20-year nationwide study. J Natl Cancer Inst 2013;105:122–9. [12] Abdul-Karim FW, King TA, Dahms BB, Gauderer MW, Boat TF. Carcinoma of extrahepatic biliary system in a patient with cystic fibrosis. Gastroenterology 1982;82:758–62. [13] Naderi AS, Farsian FN, Lee WM. Cholangiocarcinoma after lung transplantation in a patient with cystic fibrosis. Eur J Gastroenterol Hepatol 2008;20:1115–17. [14] Prost DJ, Hubert D, Gaudric M, Scatton O, Soubrane O. Pancreatic mucinous cystadenoma in an adult with cystic fibrosis. Clin Res Hepatol Gastroenterol 2011;35:759–61. [15] Roberts JA, Tullett WM, Thomas JS, Galloway D, Stack BH. Bowel adenocarcinoma in a patient with cystic fibrosis. Scott Med J 1986;31:109. [16] McKenna PB, Mulcahy E, Waldron D. Early onset of colonic adenocarcinoma associated with cystic fibrosis – a case report. Ir Med J 2006;99:310–11. [17] Alexander CL, Urbanski SJ, Hilsden R, Rabin H, MacNaughton WK, Beck PL. The risk of gastrointestinal malignancies in cystic fibrosis: case report of a patient with a near obstructing villous adenoma found on colon cancer screening and Barrett’s esophagus. J Cyst Fibros 2008;7:1–6. [18] Ibele AR, Koplin SA, Slaughenhoupt BL, Kryger JV, Friedl A, Lund DP. Colonic adenocarcinoma in a 13 year old with cystic fibrosis. J Pediatr Surg 2007;42:E1–3. [19] Chaun H, Paty B, Nakielna EM, Schmidt N, Holden JK, Melosky B. Colonic Cancer in two adult CF patients. Can J Gastroenterol 1996;10:440–2. [20] Sheldon CD, Hodson ME, Carpenter LM, Swerdlow AJ. A cohort study of cystic fibrosis and malignancy. Br J Cancer 1993;68:1025–8. [21] Schoni MH, Maisonneuve P, Schoni-Affolter F, Lowenfels AB. Cancer risk in patients with cystic fibrosis: the European Data. J R Soc Med 1996;89:38–43. [22] Johannesson M, Askling J, Montgomery SM, Ekbom A, Bahmanyar S. Cancer risk among patients with cystic fibrosis
[23] [24]
[25]
[26]
[27]
[28] [29]
[30]
[31]
[32]
[33]
[34] [35]
[36]
[37]
[38]
7
and their first degree relatives. Int J Cancer 2009;125: 2953–6. Neglia JP, Wielinski CL, Warwick WJ. Cancer risk among patients with cystic fibrosis. J Pediatr 1991;119:764–6. Australian Institute of Health and Welfare. Cancer in Australia 2010: an overview. Cancer series no. 60. Cat no. CAN 56 Canberra: AIHW; 2010. Stryker SJ, Wolff BG, Culp CE, et al. Natural history of untreated colonic polyps. Gastroenterology 1987;93: 1009–13. Otchy DP, Ransonhoff DF, Wolff BG, Weaver A, Ilstrup D, Carlson H, et al. Metachronous colon cancer in persons who have a large adenomatous polyp. Am J Gastroenterol 1996; 91:448–54. Levin B, Lieberman DA, McFarland B, Andrews KS, Brooks D, Bond J, et al. Screening and surveillance for the early detection of colorectal cancer and adenomatous polyps, 2008: a joint guideline from the American Cancer Society, the US Multi-Society Task Force on Colorectal Cancer, and the American College of Radiology. Gastroenterology 2008;134:1570–95. Modolell I, Guarner L, Malagelada JR. Digestive system involvement in cystic fibrosis. Pancreatology 2002;2:12–16. Bali A, Stableforth DE, Asquith P. Prolonged small-intestinal transit time in cystic fibrosis. Br Med J (Clin Res Ed) 1983; 287:1011–13. Fridge JL, Conrad C, Gerson L, Castillo RO, Cox K. Risk factors for small bowel bacterial overgrowth in cystic fibrosis. J Pediatr Gastroenterol Nutr 2007;44:212–18. Werlin SL, Benuri-Silbiger I, Kerem E, Adler SN, Goldin E, Zimmerman J, et al. Evidence of intestinal inflammation in patients with cystic fibrosis. J Pediatr Gastroenterol Nutr 2010;51:304–8. Stead RJ, Redington AN, Hinks IJ, Clayton BE, Hodson ME, Batten JC. Selenium deficiency and possible increased risk of carcinoma in adults with cystic fibrosis. Lancet 1985;2:862–3. Kune G, Watson L. Colorectal cancer protective effects and the dietary micronutrients folate, methionine, vitamins B6, B12, C, E, selenium, and lycopene. Nutr Cancer 2006;56: 11–21. Rose DP, Connolly JM. Omega-3 fatty acids as cancer chemopreventive agents. Pharmacol Ther 1999;83:217–44. Strandvik B, Gronowitz E, Enlund F, Martinsson T, Wahlström J. Essential fatty acid deficiency in relation to genotype in patients with cystic fibrosis. J Pediatr 2001;139: 650–5. Button BM, Roberts S, Kotsimbos TC, Levvey BJ, Williams TJ, Bailey M, et al. Gastroesophageal reflux (symptomatic and silent): a potentially significant problem in patients with cystic fibrosis before and after lung transplantation. J Heart Lung Transplant 2005;24:1522–9. The Transplantation Society of Australia and New Zealand [online]. 2012. Available from www.tsanz.com.au/organallocationprotocols/lungprotocol.asp. Accessed November 2012. Gordon CM, Anderson EJ, Herlyn K, Hubbard JL, Pizzo A, Gelbard R, et al. Nutrient status of adults with cystic fibrosis. J Am Diet Assoc 2007;107:2114–19.
ORIGINAL ARTICLE
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Windsor on 07/10/14 For personal use only.
Increased risk of colorectal neoplasia in adult patients with cystic fibrosis: a matched case–control study ILANA GORY1, GREGOR BROWN1, JOHN WILSON2, WILLIAM KEMP1, ELDHO PAUL3 & STUART K ROBERTS1 1
Department of Gastroenterology, The Alfred Hospital, Melbourne, Australia, 2Department of Respiratory Medicine, The Alfred Hospital, Melbourne, Australia, and 3Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Australia
Abstract Objective. As the lifespan of patients with cystic fibrosis (CF) increases, concerns regarding the occurrence of bowel cancer have arisen. However, previous cohort studies have yielded contradictory results as to the risk of colorectal cancer (CRC) in the CF population. The aim of this study was to determine whether CF is associated with an increased risk of colorectal neoplasia. Methods. We performed a matched case–control study at a statewide CF center. Cases were adult CF patients undergoing colonoscopy in a 5-year period from 2007 to 2012. Controls were matched in a 2:1 ratio for age, gender, and colonoscopy indication. Results. A total of 50 patients with CF who underwent colonoscopy were identified. Among CF patients, there were 5 (10%) cases with CRC, 1 (2%) with ileal adenocarcinoma, 13 (26%) with adenomas, and 16(32%) with advanced adenomas. In contrast, there was 1 (1%) case with CRC, 11 (11%) with adenomas, and 6 (6%) with advanced adenomas identified among controls. Compared to controls, CF was associated with a 10-fold increased risk of CRC (odds ratio [OR] = 10.0, 95% confidence interval [CI]: 1.2–85.6; p = 0.03), threefold increased risk of adenomas (OR = 3.34, 95%CI: 1.23–9.08; p = 0.018], and sevenfold increased risk of advanced adenomas (OR = 6.95, 95% CI: 2.30–21.01; p = 0.001). Moreover, the quality of bowel preparation was significantly worse in CF patients with a fourfold (p = 0.01) increased likelihood of a poor preparation. Conclusions. Our data suggest that CF patients are at significantly increased risk of CRC compared to age, sex, and colonoscopy-indication matched controls. Consideration should be given to the introduction of a CRC surveillance program in the CF population.
Key Words: adenomas, case–control, colorectal cancer, cystic fibrosis
Introduction Cystic fibrosis (CF) is the most common life-limiting genetic condition in Caucasians with a gene carrier frequency of 1 in 25 and affecting 1 in every 2500 live births [1]. It is a multisystem disorder characterized by a defect in the CF transmembrane conductance regulator (CFTR), thereby affecting several organs including the airways, pancreatic ducts, bile ducts and intestine [2]. Over the past three decades, the median survival of CF patients in developed countries has more than doubled due to advances in the early diagnosis and
treatment of CF including improvements in pulmonary, nutritional, and lung transplantation care [3-7]. Consequently, concerns surrounding age-related conditions such as colorectal cancer (CRC) in the CF population have heightened in recent times [8,9]. Increasing data suggest that CF predisposes to cancer, and in particular gastrointestinal malignancy [10,11]. Since the first report in 1982 of a cholangiocarcinoma in a CF patient [12], multiple case reports have been published of neoplasms in CF involving the digestive tract, mainly involving the pancreas, small bowel, and colon [8,13-19]. In 1993, Sheldon et al.
Correspondence: Stuart Roberts, Department of Gastroenterology, The Alfred Hospital, Commercial Rd, Melbourne 3004, Australia. Tel: +61 3 9076 3375. Fax: +61 3 9076 2194. E-mail: [email protected]
(Received 1 June 2014; revised 13 June 2014; accepted 13 June 2014) ISSN 0036-5521 print/ISSN 1502-7708 online 2014 Informa Healthcare DOI: 10.3109/00365521.2014.936035
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Windsor on 07/10/14 For personal use only.
2
I. Gory et al.
published a cohort study that described a possible association between adenocarcinoma and CF, particularly involving the ileum and pancreas [20]. Several larger studies have further examined the risk of cancer in CF patients reporting an increased risk of between three- to six-fold in the CF population [10,11,21-23]. However these studies have yielded inconsistent results and have been mostly focused on the composite target of digestive tract malignancies. Moreover, in some of these studies limited histological data were available and information on genotype and family history of malignancy was incomplete. Nevertheless, a growing body of evidence mostly derived from case reports [8,15-18], limited case series [19], and lung transplant cohorts [9,11] strongly suggest that there is an increased risk of CRC in CF. In this context, we observed anecdotally that CRC was occurring in CF at a relatively young age at our statewide CF referral center. Hence, we conducted an exploratory case–control study to evaluate the relative risk of colorectal neoplasia among adult patients with CF.
Patients and methods Study subjects and design This is a retrospective exploratory case–control study of patients with the diagnosis of CF from The Alfred Hospital, which is a statewide adult CF referral center (catchment 3 million) and the largest CF center in Australia with over 300 patients undergoing assessment at least three times annually by a respiratory physician. Eligible case patients included adults ‡18 years with a confirmed diagnosis of CF based on clinical, sweat, and genetic testing who underwent colonoscopy for any indication between June 2007 and June 2012. The control group consisted of patients who underwent colonoscopy during the same period without the diagnosis of CF.
Identification of cases All patients who undergo colonoscopy at The Alfred Hospital are recorded on the endoscopy database. Using a combination of this system, medical records, and the hospital coding data, we identified all CF patients who had undergone a colonoscopy during the 5-year period from June 1 2007 to June 1 2012. Demographical and epidemiological data was obtained by examination of hospital medical records.
Identification of controls For each case, we randomly selected two controls from the endoscopy database blinded to the results of the colonoscopy findings. Using this system we matched for age, gender, and indication of colonoscopy. For each case subject, two controls who met all the above matching criteria could be found. Controls were not matched for lung function or nutritional status. Colonoscopy indications and findings Indications for colonoscopy were recorded as follows: family history of malignancy, rectal bleeding, irondeficiency anemia, abdominal pain, previous polyps, diarrhea, constipation, and abnormal imaging. Possible confounders were also recorded. These include personal or family history of cancer and a history of lung transplantation. Endoscopy findings of colorectal and intestinal malignancy, and size, number, and location of polyps were recorded. Hospital electronic medical records were then examined to confirm histological diagnosis of adenocarcinoma and adenomas and their anatomical location were identified. In addition, the quality of the bowel preparation as subjectively reported by the endoscopist on the endoscopy report (i.e., poor, average, good) was recorded. Poor preparation was defined as solid or adherent stool that was unable to be washed away and therefore the preparation was considered inadequate to exclude colonic lesions. Average preparation was defined as opaque fluid and some adherent stool but was able to be washed away and provide reasonable views to exclude lesions. Good preparation was defined as clear or opaque fluid only with good views of the entire colon. Definitions Colorectal adenocarcinoma and adenomas were defined histopathologically using tissues retrieved at surgery or colonoscopic polypectomy. Advanced adenomas were defined as adenomas that were either >10 mm, demonstrated a villous component, and had high-grade dysplasia, multiplicity, or early invasive cancer. Statistical analysis Statistical analysis was conducted using SAS software version 9.2 (SAS Institute, Cary, NC, USA). Data were summarized using mean ± standard deviation (SD) or median (range) for continuous variables and number (%) for categorical variables. Risk factors for
Colorectal cancer in patients with cystic fibrosis
20% of the cohort. Other common indications included abdominal pain (16%), loss of weight (10%), and family history of CRC (10%) (Table I).
Table I. Baseline demographics and indication for colonoscopy.
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Windsor on 07/10/14 For personal use only.
Characteristic
CF (n = 50)
Age (years), mean ± SD 37.6 ± 10.9 Male sex % 54 Indication for colonoscopy Bleeding 22 (45%) Rectal bleeding 10 (20%) Fecal occult blood test 7 (14%) Iron-deficiency anemia 5 (10%) Abdominal pain 8 (16%) Family history 5 (10%) Loss of weight 5 (10%) Diarrhea 4 (8%) Constipation 3 (6%) Polyp follow up 1 (2%) Abnormal imaging 1 (2%)
3
Control (n = 100) 37.6 ± 10.8 54
Colorectal adenocarcinoma 44(45%) 20 (20%) 14 (14%) 10 (10%) 16 (16%) 10 (10%) 10 (10%) 8 (8%) 6 (6%) 2 (2%) 2 (2%)
Among the 50 CF patients, there were 5 (10%) cases of colorectal malignancy and 1 (2%) case of ileal adenocarcinoma. In contrast, there was only 1 (1%) case of CRC among the 100 controls. The location and size of the tumors are detailed in Table II. When compared to controls, a diagnosis of CF was associated with a 12-fold increased risk of malignancy identified at colonoscopy (OR = 12.0, 95% CI: 1.45–99.67; p = 0.02]. CF was still associated with a 10-fold increased risk of CRC after exclusion of the single case of ileal adenocarcinoma (OR = 10.0, 95% CI: 1.2–85.6; p = 0.03).
Abbreviations: CF = Cystic fibrosis; SD = Standard deviation.
the development of CRC in CF patients were assessed using conditional logistic regression, taking into account the matched design, with results reported as odds ratios (OR) and 95% confidence intervals (CI). Subgroup analysis was performed in patients with colorectal adenomas to compare histology, number, and size of polyps identified. Statistical significance was set at a two-sided p-Value of 0.05.
Colorectal adenomas There were 13 (26%) cases with adenomatous polyps identified among the 50 CF patients compared with 11 (11%) among the 100 controls. In comparison with controls, CF was associated with a threefold increased risk of adenoma development (OR = 3.34, 95% CI: 1.23–9.08; p = 0.018). On histopathology, CF patients were more likely to have a villous component compared with controls (69% vs. 27%); however this was not statistically significant (p = 0.99). Other characteristics including size and number of adenomas were similar between the groups (Table III). In total, 50 polyps were identified among the 24 patients (Figure 1). The histology of adenomas was predominantly tubulovillous (low-grade dysplasia) among CF patients and tubular adenoma among controls (Table III).
Study conduct This study was conducted in accordance with the Helsinki Declaration of 1975 and approved by the Ethics Committee of the Alfred Hospital, Melbourne, Australia, prior to commencement of the study. Results Between 2007 and 2012, 50 patients with CF who had colonoscopies were identified of which 54% were male with a mean age of 37.6 (range 20–70) years. The clinical characteristics of the CF and control groups are shown in Table I. The major indication for colonoscopy was rectal bleeding, being present in
Advanced adenomas There were 22 patients with advanced adenomas (i.e. >10 mm, villous component, high-grade dysplasia,
Table II. Summary of findings in patients with adenocarcinoma. Age CF 46 48 46 46 50 33 Control 70
Gender
Indication
Site
Size (mm)
CF gene
Male Female Female Male Male Male
Family history Rectal bleeding Positive FOBT Positive FOBT Iron-deficiency anemia Abdominal pain
Sigmoid Rectal Sigmoid Cecum Ascending Ileum
15 45 30 40 60 50
DF508/DF508 DF508/DF508 DF508/DF508 DF508/DF508 DF508/DF508 DF508/g542x
No No No No No No
Female
Iron-deficiency anemia
Ascending
50
-
No
Abbreviations: CF = Cystic fibrosis; FOBT = Fecal occult blood test.
Transplant
4
I. Gory et al.
Table III. Summary of findings in patients with colorectal adenomas. Characteristic
Overall (n = 24)
CF (n = 13)
Control (n = 11)
p-Value*
Age (years), mean ± SD Male sex % Size (mm) Median (range) >10 mm (%) Number of polyps, median (range) Histology Villous component Tubular adenoma
43.2 ± 9.0 50
41.5 ± 7.2 53.9
45.3 ± 10.8 45.5
0.48 1.00
6.5 (1–18) 6 (25%) 1 (1–7)
8 (1–18) 5 (38.5%) 1 (1–7)
6 (3–14) 1 (9.1%) 2 (1–3)
0.19
12 (50%) 12 (50%)
9 (69%) 4 (31%)
3 (27%) 8 (73%)
0.58 0.99
10 9 8 Number of patients
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Windsor on 07/10/14 For personal use only.
*p-Values are from conditional logistic regression analysis. Abbreviations: CF = Cystic fibrosis; SD = Standard deviation.
7 6
1 polyps
5
2 polyps
4
3 polyps
3
4 polyps
2
7 polyps
1 0 Cystic fibrosis
Controls
Number of polyps Figure 1. Number of colorectal adenomas per patient is shown. In total, 50 polyps were identified among the 24 patients. There was no difference in the number of polyps in the CF group versus the controls.
multiplicity or early invasive cancer), including 16 (32%) adenomas among CF patients and 6 (6%) adenomas among controls. Advanced adenoma was approximately seven times more likely to occur in the CF population when compared to controls (OR = 6.95, 95% CI: 2.30–21.01; p = 0.001). Bowel preparation The subjective quality of the bowel preparation was significantly worse in the CF group with 22% having a poor preparation compared to 7% in controls (p = 0.001). When compared to controls there was a fourfold increased likelihood (OR = 4.49, 95% CI: 1.4–14.39; p = 0.01) of a poor compared to average-good preparation in CF patients (Figure 2). Discussion The main finding of this exploratory case–control study is the demonstration that patients with CF are at a significantly higher risk of colorectal neoplasia,
including CRC, advanced adenomas, and adenomatous polyps compared to age, gender, and colonoscopy-indication matched controls. Specifically, we found that CF is associated with a 10-fold increased risk of CRC, threefold higher risk of adenomas, and approximately sevenfold increased risk of advanced adenomas compared to carefully matched controls. Moreover, our study might have even underestimated the risk of colonic neoplasia in CF as we found a significantly increased frequency of poor bowel preparation in the CF group that could have reduced our adenoma detection rates, particularly for small lesions. These important data add to the accumulating body of evidence from other cohort studies in demonstrating that subjects with CF are at increased risk of digestive tract malignancy and, in particular, colon cancer. In addition, we highlight the significant risk of associated colorectal adenomas in CF. Several large studies have examined the cancer risk in CF patients (Table IV) [10,11,21-23]. In the largest of these, Maisonneuve et al. performed a retrospective cohort study of more than 41,000 North
Colorectal cancer in patients with cystic fibrosis
5
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Windsor on 07/10/14 For personal use only.
Number of patients (%)
100
80
60 p < 0.001 CF 40
Controls
20
0 Good
Average
Poor
Quality of bowel preparation Figure 2. Bowel preparation in CF patients versus controls is shown. Compared to controls, patients with CF had significantly worse bowel preparation (22% vs. 7%, p < 0.001). CF patients were four times more likely to have poor or average preparation when compared to controls (OR = 4.49, 95% CI: 1.4–14.39; p = 0.01). Abbreviations: CF = Cystic fibrosis; CI = Confidence interval; OR = Odds ratio.
American CF patients and while observing no overall increase in cancer, they found that digestive tract cancers, including small bowel, colon, and biliary tract, were 3.5 times more likely to develop in CF, and that the cancer risk was particularly high in the transplanted population [11]. Indeed, they found an approximately sevenfold increased risk of developing colon cancer in non-transplanted CF subjects that increased to 30-fold among transplant recipients when compared to the general population [11]. Meyer et al. reported similar findings, observing a significantly higher incidence of colon cancer in transplanted CF subjects compared to non-CF transplant recipients [9]. In contrast, Johannesson et al. in a Swedish study of 884 CF patients observed a fivefold increased risk of digestive tract cancer in CF, yet reported no significant increase in the CRC risk when compared to the general population [22]. Nevertheless, the above data combined with ours strongly suggest that the adult CF population is at risk of CRC and, therefore, should be considered for enrollment into CRC surveillance programs. Of the five CF patients in our study diagnosed with CRC, four were men with the mean age of 49.2 years (range 46–60 years). This age is significantly lower than the mean age of 69 years at diagnosis of CRC
among the general population in Australia [24]All six had known CF mutations, five being DF508 homozygous and one being DF508/G5F42X. One patient had a family history of CRC; however, this was his indication for endoscopy and as such was controlled for. Otherwise, none of the patients had known risk factors other than CF for the development of digestive tract cancer with none having been transplanted, exposed to immunosuppression, or a history of inflammatory bowel disease. The predominant indication for colonoscopy in the five patients with CRC was bleeding (n = 4), either overt (n = 1) or occult (n = 3). The indication in the patient with the ileal adenocarcinoma was abdominal pain. Multivariate analysis to identify predictors for the development of adenocarcinoma in CF could not be performed due to the small number of patients with adenocarcinoma in this study. In our study, we found a sevenfold increased risk of developing advanced adenomas in CF patients. This composite group (i.e. adenomas >10 mm or with adverse histological features or adenocarcinoma) is important since it is well established that adenomas of this type indicate a risk of metachronous advanced adenomas and cancers and an 8% risk of developing colorectal malignancy within 10 years in the general
Table IV. Previous cohort studies examining the risk of malignancy in CF. Author (year) Maisonneuve et al. (2012) Johannesson et al. (2009) Neglia et al. (1995)
Number of patients 41,188 884 28,511
Overall cancer risk increased
Digestive tract cancer risk increased
Colorectal cancer risk increased
NS 3.2 (SIR) NS
3.5 (SIR) NS 6.5 (OR)
6.2 (SIR) NS NE
Abbreviations: NS = Not statistically significant; NE = Not examined; SIR = Standard incidence ratio; OR = Odds ratio.
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Windsor on 07/10/14 For personal use only.
6
I. Gory et al.
population [25,26]. With the improved survival of CF patients, the adenoma-carcinoma sequence is more likely to occur in this aging population and as such identifying these high-risk lesions becomes more imperative. When advanced adenomas are identified, current guidelines recommend colonoscopic surveillance at 3-yearly intervals initially because of the increased risk of developing CRC [27]. Subjective assessment of the quality of bowel preparation in our study confirmed that patients with CF have a significantly higher frequency of a poor result compared to non-CF controls. This finding is not unexpected given that constipation, distal intestinal obstruction syndrome, and slow gastrointestinal transit time are well-documented complications of CF [28]. Poor bowel preparation may cause underreporting of polyps and other lesions due to poor visualization of the colonic mucosa, meaning that the risk of colorectal neoplasia in CF may in fact be higher than documented in this study. Further, we now recommend that CF patients undergo a more intensive bowel preparation regime before colonoscopy, particularly in view of their increased risk of colorectal lesions. While the mechanism(s) underlying the increased risk of digestive tract adenocarcinoma and colorectal neoplasia in CF is(are) unknown, there are several factors present in CF that could contribute. These include persistent pathological disease processes such as intestinal stasis [29], perturbations in gut microflora (e.g. bacterial overgrowth) [30], and chronic inflammation [31] that might increase cell proliferation and turnover, as well as malabsorption-related micronutrient deficiencies (e.g. selenium, vitamin E) that potentially play a role in cancer prevention [32,33]. In addition, failure to thrive and malnutrition that are commonly present in CF are associated with a higher risk of developing digestive tract cancers [11]. More specifically, a non-functioning CFTR has been linked to reduced serum- and membrane-bound levels of the 3-omega fatty acid, docosahexaenoic acid, in CF, which in epidemiological studies is associated with a reduced risk of colon cancer [34,35]. A further possible mechanism may follow from acid dumping into the lower gastrointestinal tract causing metaplastic change analogous to Barrett’s esophagus. We have previously documented hyper-acidification in CF secondary to gastroesophageal reflux disease in CF [36]. In addition, patients with CF have pancreatic failure with limited capacity to neutralize pH with bicarbonate, potentiating the acid environment. It should be noted that all five CRC subjects in this study were homozygous for the class V mutation F508del, associated with the severest form of pancreatic insufficiency.
While the strengths of this study include the evaluation of both CRC and adenomas, the assessment of histopathology, and inclusion of family history as a risk factor, the study has its limitations including small sample size compared with published CF registries. However, our single center study is from the largest CF center in Australia and, despite its size, it was able to show a significant increase in risk of CRC and advanced adenomas in CF compared to matched controls. In addition, information was not routinely collected on other confounders that may increase the risk of CRC such as smoking and obesity. Still, these risk factors are less likely to be present in the CF population, particularly smoking, because of the associated lung disease and need for compliance with Australian eligibility criteria requirements for lung transplantation [37]. Moreover, obesity is uncommon among adult patients with CF with most having a normal-to-low body mass index [38] often due to comorbidities, including bronchiectasis and malabsorption. In conclusion, our study demonstrates that patients with CF are at an increased risk of CRC, advanced adenomas, and adenomatous polyps. While CRC remains an uncommon but well-reported event in young CF patients [7,18,19], these data suggest that the CF population may benefit from enrollment in CRC screening surveillance programs, particularly as their survival continues to improve and the age-related CRC risk increases. Surveillance strategies could include periodic colonoscopy and/or fecal occult blood testing in all CF patients. Further prospective data are required for determining the importance of specific risk factors and for determining the most effective screening tool(s) for this high-risk population. Declaration of interest: None of the authors have any disclosures to declare for this study. No financial support was received to conduct this study. References [1] Cystic Fibrosis Australia,Inc. [online]. 2012. Available from http:www.cysticfibrosisaustralia.org.au. Accessed October 2012. [2] Cohn JA, Strong TV, Picciotto MR, Nairn AC, Collins FS, Fitz JG. Localization of the cystic fibrosis transmembrane conductance regulator in human bile dict epithelial cells. Gastroenterology 1993;105:1857–64. [3] FitzSimmons SC. The changing epidemiology of cystic fibrosis. J Pediatr 1993;122:1–9. [4] Reid DW, Blizzard CL, Shugg DM, Flowers C, Cash C, Greville HM. Changes in cystic fibrosis mortality in Australia, 1979-2005. Med J Aust 2011;195:392–5.
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Windsor on 07/10/14 For personal use only.
Colorectal cancer in patients with cystic fibrosis [5] Dodge JA, Lewis PA, Stanton M, Wilsher J. Cystic fibrosis mortality and survival in the UK: 1947-2003. Eur Respir J 2007;29:522–6. [6] Bellis G, Cazes MH, Parant A, Gaimard M, Travers C, Le Roux E, et al. Cystic fibrosis mortality trends in France. J Cyst Fibros 2007;6:179–86. [7] Ramalle-Gomara E, Perucha M, González MA, Quiñones C, Andrés J, Posada M. Cystic fibrosis mortality trends in Spain among infants and young children: 1981-2004. Eur J Epidemiol 2008;23:523–9. [8] Hernandez-Jimenez I, Fischman D, Cheriyath P. Colon cancer in cystic fibrosis patients: is this a growing problem? J Cyst Fibros 2008;7:343–6. [9] Meyer KC, Francois ML, Radford KL, Hawes DS, Mack TL, Cornwell RD. Colon cancer in lung transplant recipients with CF: increased risk and results of screening. J Cyst Fibros 2011;10:366–9. [10] Neglia JP, Fitzsimons SC, Maisonneuve P, Schöni MH, Schöni-Affolter F, Corey M, et al. The Cystic Fibrosis and Cancer Study Group. The risk of cancer among patients with cystic fibrosis. N Engl J Med 1995;332:494–9. [11] Maisonneve P, Marshall BC, Knapp EA, Lowenfels AB. Cancer risk in cystic fibrosis patients: a 20-year nationwide study. J Natl Cancer Inst 2013;105:122–9. [12] Abdul-Karim FW, King TA, Dahms BB, Gauderer MW, Boat TF. Carcinoma of extrahepatic biliary system in a patient with cystic fibrosis. Gastroenterology 1982;82:758–62. [13] Naderi AS, Farsian FN, Lee WM. Cholangiocarcinoma after lung transplantation in a patient with cystic fibrosis. Eur J Gastroenterol Hepatol 2008;20:1115–17. [14] Prost DJ, Hubert D, Gaudric M, Scatton O, Soubrane O. Pancreatic mucinous cystadenoma in an adult with cystic fibrosis. Clin Res Hepatol Gastroenterol 2011;35:759–61. [15] Roberts JA, Tullett WM, Thomas JS, Galloway D, Stack BH. Bowel adenocarcinoma in a patient with cystic fibrosis. Scott Med J 1986;31:109. [16] McKenna PB, Mulcahy E, Waldron D. Early onset of colonic adenocarcinoma associated with cystic fibrosis – a case report. Ir Med J 2006;99:310–11. [17] Alexander CL, Urbanski SJ, Hilsden R, Rabin H, MacNaughton WK, Beck PL. The risk of gastrointestinal malignancies in cystic fibrosis: case report of a patient with a near obstructing villous adenoma found on colon cancer screening and Barrett’s esophagus. J Cyst Fibros 2008;7:1–6. [18] Ibele AR, Koplin SA, Slaughenhoupt BL, Kryger JV, Friedl A, Lund DP. Colonic adenocarcinoma in a 13 year old with cystic fibrosis. J Pediatr Surg 2007;42:E1–3. [19] Chaun H, Paty B, Nakielna EM, Schmidt N, Holden JK, Melosky B. Colonic Cancer in two adult CF patients. Can J Gastroenterol 1996;10:440–2. [20] Sheldon CD, Hodson ME, Carpenter LM, Swerdlow AJ. A cohort study of cystic fibrosis and malignancy. Br J Cancer 1993;68:1025–8. [21] Schoni MH, Maisonneuve P, Schoni-Affolter F, Lowenfels AB. Cancer risk in patients with cystic fibrosis: the European Data. J R Soc Med 1996;89:38–43. [22] Johannesson M, Askling J, Montgomery SM, Ekbom A, Bahmanyar S. Cancer risk among patients with cystic fibrosis
[23] [24]
[25]
[26]
[27]
[28] [29]
[30]
[31]
[32]
[33]
[34] [35]
[36]
[37]
[38]
7
and their first degree relatives. Int J Cancer 2009;125: 2953–6. Neglia JP, Wielinski CL, Warwick WJ. Cancer risk among patients with cystic fibrosis. J Pediatr 1991;119:764–6. Australian Institute of Health and Welfare. Cancer in Australia 2010: an overview. Cancer series no. 60. Cat no. CAN 56 Canberra: AIHW; 2010. Stryker SJ, Wolff BG, Culp CE, et al. Natural history of untreated colonic polyps. Gastroenterology 1987;93: 1009–13. Otchy DP, Ransonhoff DF, Wolff BG, Weaver A, Ilstrup D, Carlson H, et al. Metachronous colon cancer in persons who have a large adenomatous polyp. Am J Gastroenterol 1996; 91:448–54. Levin B, Lieberman DA, McFarland B, Andrews KS, Brooks D, Bond J, et al. Screening and surveillance for the early detection of colorectal cancer and adenomatous polyps, 2008: a joint guideline from the American Cancer Society, the US Multi-Society Task Force on Colorectal Cancer, and the American College of Radiology. Gastroenterology 2008;134:1570–95. Modolell I, Guarner L, Malagelada JR. Digestive system involvement in cystic fibrosis. Pancreatology 2002;2:12–16. Bali A, Stableforth DE, Asquith P. Prolonged small-intestinal transit time in cystic fibrosis. Br Med J (Clin Res Ed) 1983; 287:1011–13. Fridge JL, Conrad C, Gerson L, Castillo RO, Cox K. Risk factors for small bowel bacterial overgrowth in cystic fibrosis. J Pediatr Gastroenterol Nutr 2007;44:212–18. Werlin SL, Benuri-Silbiger I, Kerem E, Adler SN, Goldin E, Zimmerman J, et al. Evidence of intestinal inflammation in patients with cystic fibrosis. J Pediatr Gastroenterol Nutr 2010;51:304–8. Stead RJ, Redington AN, Hinks IJ, Clayton BE, Hodson ME, Batten JC. Selenium deficiency and possible increased risk of carcinoma in adults with cystic fibrosis. Lancet 1985;2:862–3. Kune G, Watson L. Colorectal cancer protective effects and the dietary micronutrients folate, methionine, vitamins B6, B12, C, E, selenium, and lycopene. Nutr Cancer 2006;56: 11–21. Rose DP, Connolly JM. Omega-3 fatty acids as cancer chemopreventive agents. Pharmacol Ther 1999;83:217–44. Strandvik B, Gronowitz E, Enlund F, Martinsson T, Wahlström J. Essential fatty acid deficiency in relation to genotype in patients with cystic fibrosis. J Pediatr 2001;139: 650–5. Button BM, Roberts S, Kotsimbos TC, Levvey BJ, Williams TJ, Bailey M, et al. Gastroesophageal reflux (symptomatic and silent): a potentially significant problem in patients with cystic fibrosis before and after lung transplantation. J Heart Lung Transplant 2005;24:1522–9. The Transplantation Society of Australia and New Zealand [online]. 2012. Available from www.tsanz.com.au/organallocationprotocols/lungprotocol.asp. Accessed November 2012. Gordon CM, Anderson EJ, Herlyn K, Hubbard JL, Pizzo A, Gelbard R, et al. Nutrient status of adults with cystic fibrosis. J Am Diet Assoc 2007;107:2114–19.
