SEARCH

SEARCH BY CITATION

Keywords:

  • alternative splicing;
  • biomarker;
  • colorectal cancer;
  • exon-level gene expression;
  • SLC39A14

Abstract

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

An alternative transcript variant of SLC39A14, caused by pre-mRNA splicing, was recently suggested as a biomarker for colorectal cancer (CRC). In our study, we have validated the cancer-specific splicing pattern of the mutually exclusive exons 4A and 4B in altogether 244 colorectal tissue samples. Exon-specific quantitative RT-PCR analyses across 136 Stage I–IV CRC samples and 44 normal colonic mucosa samples showed complete cancer-specificity, as well as 94% sensitivity of SLC39A14-exon4B relative to SLC39A14-exon4A expression. However, across 20 samples from a range of healthy tissues, 18 expressed the CRC variant. This was true also for ten benign lymph nodes, demonstrating that the cancer-specificity is mainly confined to the colon and rectum. Hence, clinical use of SLC39A14-exon4B as a detection marker for CRC other than in samples taken from the bowel wall is diminished. Prognostic value by detection of metastasis to lymph nodes is also abated, elucidating an important pitfall to biomarker discovery. However, analyses of ten nondysplastic biopsies from patients with active inflammatory bowel disease showed negative results in seven samples and only weakly positive results in three samples, suggesting value of SLC39A14-exon4B as a marker to distinguish CRC from other pathologic conditions of the colon. In conclusion, the SLC39A14-exon4B transcript variant is a CRC biomarker with high sensitivity and organ-confined specificity. Further use of the transcript and its encoded protein isoform should be explored in an organ-confined context.

Colorectal cancer (CRC) is one of the most prevalent and lethal cancers worldwide.1 The only biomarker approved for routine clinical use in CRC is serum levels of carcinoembryonic antigen to aid in postoperative monitoring of Stage II and III disease, as well as metastatic disease during systemic therapy.2 Recently, solute carrier family 39 (zinc transporter), member 14 (SLC39A14) was proposed as a potential biomarker for CRC, with indications of applicability also in gastric and lung cancers.3 Exons 4A and 4B of this gene are mutually exclusive and cancer-specific alternative splicing results in an increased exon 4B/4A ratio in both carcinomas and adenomas. Colorectal, gastric and lung cancers are known to have an aberrantly activated WNT-signaling pathway.4–6 Hence, SLC39A14-exon4B was proposed as a detection biomarker in cancer types found to have an activated WNT-signaling pathway.3 For a biomarker to be considered for clinical use, its applicability should be carefully validated in independent datasets by independent researchers.7 No such validation studies have been reported for SLC39A14.

In our study, we have confirmed cancer-specific alternative splicing of SLC39A14, i.e., high relative expression of SLC39A14-exon4B compared to SLC39A14-exon4A, in CRC by exon microarray and quantitative reverse transcription PCR (qRT-PCR) analyses. However, we found this specificity to be organ-confined, as the biomarker was positive also for healthy tissues from other organs. Its potential to serve as detection marker other than in colorectal tissue biopsies, or prognostic marker for metastasis to lymph nodes, is thus diminished. However, we present data suggesting relative SLC39A14-exon4B expression as a marker to distinguish CRC from nondysplastic tissue in patients with inflammatory bowel disease (IBD).

Material and Methods

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Material

One hundred and ninety-four CRC tissue samples from two independent clinical series collected from patients treated surgically at hospitals in the Oslo region were analyzed in the study (Table 1). There were 10, 97, 76 and 11 Stage I, II, III and IV samples, respectively. Forty-two of the samples showed microsatellite instability, and 139 were microsatellite stable (13 samples not analyzed).8, 9 Additionally, 59 nonmalignant colorectal biopsies were analyzed, including normal colonic mucosa taken from disease free areas of the colon of 43 of the CRC patients, six adenomas from patients enrolled in a multihospital study for young age onset of CRC and ten nondysplastic biopsies from inflammatory regions of patients with IBD (five with ulcerative colitis and five with Crohn's disease), all collected at hospitals in the South East region of Norway. Furthermore, ten benign lymph nodes from IBD patients, as well as ten lymph nodes with metastasis (evaluated by light-microscopy of hematoxylin and eosin (HE) stained sections of adjacent tissue) and corresponding tumors from eight CRC patients treated at Stavanger University Hospital, Norway, were included. For a more general assessment of the biomarker, we analyzed 88 cell lines from altogether 16 types of cancer (Supporting Information Table 1), and a panel of 21 normal tissues from different organs and tissue types (FirstChoice® Human Normal Tissue Total RNA, each a pool of RNA from at least three individuals, with the exception of an individual sample from the stomach; Ambion, Applied Biosystems by Life Technologies, Carlsbad, CA). From the adenomas, CRC samples, cell lines, IBD biopsies and lymph nodes, RNA was extracted using the Qiagen AllPrep DNA/RNA Mini Kit (Qiagen GmbH, Hilden, Germany). The Ambion RiboPure™ kit (Applied Biosystems) was used to obtain RNA from the normal colonic mucosa samples. Both procedures were performed according to the manufacturers' protocols.

Table 1. Samples included in the study
inline image

The research biobanks have been registered according to national legislation, and the study has been approved by the Regional Committee for Medical Research Ethics (numbers 2781 and 236-2005-16141). The identity of all cell lines has been authenticated by DNA profiling.

Five publicly available exon microarray datasets were included in the study for comparison to the CRC and normal colonic mucosa samples. These datasets included cancer and normal tissues from the colon, lung and stomach, as well as 33 samples from a panel of 11 different normal tissues (available from the Gene Expression Omnibus (GEO) at the National Centre for Biotechnology Information and the Affymetrix web pages; Table 1).

Exon microarray analysis

We have previously analyzed 77 Stage II and III CRC samples and 13 adjacent normal colonic mucosa samples by Affymetrix GeneChip Human Exon 1.0 ST arrays10 and deposited the raw data to GEO (GSE24550). Cell intensity (CEL)-files from the altogether 90 samples were background corrected, inter-chip quantile normalized and summarized at the exon level by the robust multiarray average algorithm11 implemented in the Affymetrix Expression Console 1.1 software, using the HuEx-1_0-st-v2.r2 exon-core library files (available from the Affymetrix web page). For plotting purposes, exons were numbered relative to the transcript variants of SLC39A14 described. Exons 1, 4B and 9 are presented as the median expression signals of multiple probe sets.

The publicly available exon microarray datasets (Table 1) were preprocessed in the same manner (normalized together with the 90 colorectal tissue samples).

Quantitative RT-PCR

Reverse transcription of 2 μg total RNA was done using the High Capacity cDNA Reverse Transcription kit (Applied Biosystems), with Multiscribe reverse transcriptase and random primers, according to the manufacturer's protocol. Amplification of both the SLC39A14-exon4A and SLC39A14-exon4B transcripts was achieved using oligonucleotide primers targeting exons three (GGCCAAGCGCTGTTGAAG) and five (TCTTCCAGAGGGTTGAAACCAA). Both exons three and five are expressed in CRC and normal colonic mucosa (as indicated in Ensembl transcript annotation and by in-house unpublished RNA deep sequencing data). The primers were designed using the Primer3 software.12 Exon-specific probes were manually designed based on parameters given by the Primer Express Software v3.0 (Applied Biosystems). The probe targeting exon 4A (CTCACTGATTAACCTGGCC) was labeled at the 5′ end with the fluorescent dye 6-FAM, and the exon 4B-specific probe (ACCGTCATCTCCCTCTG) was labeled with VIC. Both probes were modified at the 3′ end with a nonfluorescent quencher suppressing the fluorescence of the respective dyes before primer extension, and a minor groove binder raising the melting temperature of the probes to ensure its hybridization to the target before primer annealing and extension. Primers and probes were custom manufactured by MedProbe (Oslo, Norway) and Applied Biosystems, respectively. The assays were run in triplicates with real-time detection on an ABI 7900HT Fast Real-Time PCR System (TaqMan; Applied Biosystems). Expression levels were reported as the median threshold cycle (CT) of the triplicates. A maximum threshold value was set at CT = 35, indicating miniscule expression amounts of the target transcript (there were 162/332 and 2/332 samples exceeding this threshold for the SLC39A14-exon4A and SLC39A14-exon4B assay, respectively). In both assays, all samples with CT-values lower than this threshold had standard deviations of the triplicate runs < ±2% (except for two samples with standard deviations of 2.2 and 6.1%). Sample-wise relative expression of SLC39A14-exon4A and SLC39A14-exon4B are reported as the difference in CT-values between the two assays (CT(SLC39A14-exon4A) − CT(SLC39A14-exon4B)), with positive values indicating higher relative expression of SLC39A14-exon4B. A CT-value difference of one corresponds approximately to twice the expression of SLC39A14-exon4B compared to SLC39A14-exon4A (efficiency of PCR-amplification >80%, as measured by a standard curve of Universal Human Reference RNA, Stratagene, Agilent Technologies, Santa Clara, CA; data not shown).

Results

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Exon microarray analysis of SLC39A14

Exon microarray data for the colorectal samples indicated cancer-specific alternative splicing of the two mutually exclusive exons 4A and 4B of SLC39A14 (Figs. 1a and 1b). The cancer-specificity of exon 4B vs. 4A expression was more strongly indicated in microarray data from colorectal tissues than gastric and lung tissues (Fig. 1c). Surprisingly, microarray profiles from nine of 11 normal tissue types (breast, cerebellum, heart, muscle, pancreas, prostate, spleen, testes and thyroid) indicated increased relative expression of exon 4B, the same exon that was cancer-specific within colorectal tissues (Fig. 1d). Only tissues from normal liver and kidney had indications of higher relative expression of exon 4A than 4B, the same as for normal colonic mucosa.

thumbnail image

Figure 1. Exon-level expression of SLC39A14 in microarray datasets. (a) The two exons 4A and 4B in SLC39A14 are mutually exclusive. (b) Exon microarray profiles of CRC and normal colonic mucosa indicate cancer-specific splicing of these exons, i.e., higher relative expression levels of exon 4A in normal samples and exon 4B in cancer samples. (c) Compared to colorectal tissues, the cancer-specific inclusion of exon 4B vs. 4A is not as strongly indicated in the microarray profiles of gastric and lung tissues. (d) Microarray profiles for a panel of normal tissues from 11 different organs indicate that nine have expression patterns similar to CRC (breast, cerebellum, heart, muscle, pancreas, prostate, spleen, testes and thyroid). Only in normal kidney and liver was exon 4A more highly expressed than 4B. In panels c and d, the exon-wise log2 expression levels are centered on the median gene level expression of SLC39A14 within each sample group. The intensity level for the probeset targeting exon 5 was below the background threshold in all sample groups, and therefore not included in the plots.

Download figure to PowerPoint

Quantitative RT-PCR analysis of the SLC39A14-exon4B and SLC39A14-exon4A transcripts

Individual qRT-PCR assays for the SLC39A14-exon4A and SLC39A14-exon4B transcripts (Fig. 2a) were applied to a total of 332 samples from various tissue types (Table 1). In the following, results are reported as expression of SLC39A14-exon4B relative to SLC39A14-exon4A (as represented by differences in CT-values). Samples with higher relative expression of SLC39A14-exon4B have CT-value differences above zero and are considered positive for the biomarker. Contrarily, samples with higher relative expression of the SLC39A14-exon4A transcript are considered negative for the biomarker. All normal colonic mucosa samples (n = 44) were negative for the biomarker (CT(SLC39A14-exon4A) − CT(SLC39A14-exon4B) range −2.5 to −0.1; Fig. 2b). Hence, the cancer-specificity of the splicing event was complete (100%). Contrarily, and corresponding with the microarray data, most CRC samples (tissue biopsies and cell lines) and adenomas were positive for the biomarker (n = 147/155 and 4/6 samples, respectively). Among Stage I–IV CRCs (from two independent series; n = 136) and adenomas (n = 6), the sensitivities were 94 and 67%, respectively. The few CRC samples that were negative for the biomarker (n = 8) were taken from tumors of all four stages (n = 1, 3, 2 and 2 for Stage I–IV CRC, respectively) and with both microsatellite stability and instability (n = 6 and 2, respectively). Nineteen of the Stage II and III CRC samples were also analyzed by exon microarrays, and all were positive for SLC39A14-exon4B, in accordance with the microarray data.

thumbnail image

Figure 2. Quantitative RT-PCR analysis of SLC39A14-exon4A and SLC39A14-exon4B. (a) Two qRT-PCR assays were designed for specific expression analysis of the SLC39A14-exon4A and SLC39A14-exon4B transcripts (the common flanking PCR primers and exon-specific TaqMan probes are indicated). (b) Relative SLC39A14-exon4B expression (relative to SLC39A14-exon4A; CT(SLC39A14-exon4A) − CT(SLC39A14-exon4B)) was negative in all normal colonic mucosa samples. Four of six colorectal adenomas, 128 of 136 CRCs, and all CRC cell lines were positive for SLC39A14-exon4B. (c) As opposed to the expression pattern observed within colorectal tissues, 18 normal samples from 20 different tissues (including the lung and stomach; Table 1) showed positive relative expression of SLC39A14-exon4B. Only tissue samples from the small intestine and liver were negative, similar to the situation for normal colonic mucosa. Cell lines from various cancer tissues also showed positive relative SLC39A14-exon4B expression. (d) Both lymph nodes with and without metastasis showed positive SLC39A14-exon4B expression, comparable to the corresponding tumors taken from patients with the affected lymph nodes (corresponding samples are plotted next to each other). Seven of ten nondysplastic biopsies from patients with active IBD had negative relative SLC39A14-exon4B expression, similar to normal colonic mucosa.

Download figure to PowerPoint

To compare the cancer-specificity of high relative SLC39A14-exon4B expression with tissues outside of the colon and rectum, we analyzed samples from a number of other healthy organs, as well as cancer cell lines by qRT-PCR (Fig. 2c). Most normal tissues (18 of 20) had higher relative expression of the SLC39A14-exon4B transcript and were accordingly positive for the biomarker (relative expression ranging from 1.6 to 8.8). Negative biomarker results were seen only in normal tissue from the small intestine and liver (relative expression −1.2 and −0.6, respectively). Eighty-eight cell lines from 16 different cancer types (Supporting Information Table 1) were all positive for higher relative SLC39A14-exon4B expression. As indicated by the exon microarray data, there was only weak differential splicing between normal tissue and cancer cell lines from the lung and stomach, and all samples were positive for SLC39A14-exon4B (relative expression in the normal lung tissue sample was 5.1, and ranging from 6.5 to 8.5 in the four lung cancer cell lines; relative expression in the normal stomach sample was 6.4, and ranging from 8.1 to 9.8 in the four gastric cancer cell lines).

To investigate the value of the cancer-specific splicing of SLC39A14 as a biomarker for lymph node metastasis of CRC, we applied the two exon-specific qRT-PCR assays to lymph nodes with known metastasis (n = 10) and their corresponding CRC tumors, as well as on benign lymph nodes from patients with IBD (n = 10; Fig. 2d). In general, the lymph nodes with metastatic CRC cells exhibited expression patterns similar to their corresponding CRC samples (high relative SLC39A14-exon4B expression; range 0.2 to 8.1). However, positive biomarker results were also found for all ten benign lymph nodes from cancer free individuals (relative expression ranging from 5.0 to 6.6).

To further explore the value of SLC39A14-exon4B expression as a biomarker for cancer in tissue within the colon and rectum, we analyzed ten nondysplastic biopsies from patients with active IBD by exon-specific qRT-PCR (Fig. 2d). Seven of the ten biopsies were negative for higher relative SLC39A14-exon4B expression (relative expression ranging from −2.3 to −0.6). The three remaining biopsies had only weakly positive biomarker results (two from patients with ulcerative colitis and one with Crohn's disease) but with lower relative expression (range 0.1 to 1.3) than 84% (114 of 136) of the CRC tissue samples.

Discussion

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Differentially expressed genes and proteins are commonly suggested as biomarkers for cancer. This notion is complicated by alternative splicing of pre-mRNA, a process which is commonly deregulated in cancer,13 allowing for the use of cancer-specific RNA transcript variants and their resulting protein isoforms as biomarkers. However, the cancer-specificity of these biomarkers, and thereby also their clinical use, is challenged by the fact that their expression is not organ-confined.

An alternative transcript variant of SLC39A14 was recently reported as a potential exon-level biomarker for CRC and precancerous lesions.3 By analyses of publicly available exon microarray data, this alternative splicing event was further suggested to have potential usefulness also in gastric and lung tissues. In our study, we have validated the cancer-specific, mutually exclusive expression pattern of exons 4A and 4B in colorectal tissues. By qRT-PCR analyses, we have showed high cancer-specificity and sensitivity (100 and 94%, respectively) of increased relative expression of transcripts including exon 4B, SLC39A14-exon4B, relative to transcripts including exon 4A. Individual biomarkers often suffer from limited sensitivity and/or specificity, and improvements are achieved with the assembly of multiple markers into marker panels.14 On the other hand, relative expression of SLC39A14-exon4B has now been shown to be a strong individual biomarker in two independent studies. However, this cancer-specificity was found to be tissue-confined. By analyses of a panel of 20 tissues from organs outside of the colon and rectum, we show that relative SLC39A14-exon4B expression does not discriminate CRC from healthy tissues in other organs, with exceptions of the liver and small intestine. Hence, we regard the specific splicing pattern of SLC39A14 as an appropriate biomarker only within the colon and rectum. Cancer specific biomarkers have great potential for diagnosis of cancer in a noninvasive manner, e.g., by detection of circulating tumor cells in the blood, and for CRC, also by analysis of fecal samples. Although such samples have not been analyzed here, our detection of SLC39A14-exon4B in various healthy tissues does not support clinical value in such noninvasive settings. Also, we did not find supportive results for use of the biomarker as a prognostic marker for lymph node involvement in CRC (with a cautionary note that the benign lymph nodes analyzed here were collected from cancer free patients only, not from patients with CRC). This elucidates another pitfall to biomarker discovery and emphasizes the high requirements cancer biomarkers have to fulfill.

Interestingly, SLC39A14-exon4B expression discriminates between CRC and healthy liver. The liver is often the first site of distant metastasis from CRC.15 Early detection of metastatic lesions may influence the treatment strategy and improve the patient's prognosis.16 Currently, detection is mainly based on imaging techniques.17 Metastatic lesions were not analyzed in our study, and further analyses are required to address the potential of SLC39A14-exon4B expression to detect CRC metastases in biopsies from hepatic lesions.

When complying with the limitation of tissue-confined specificity, SLC39A14-exon4B expression can be further explored as a biomarker for discrimination between cancer and other pathologic conditions of the colon and rectum. We demonstrate this by showing negative biomarker results in the majority of nondysplastic biopsies from patients with active IBD. Patients with IBD (ulcerative colitis and Crohn's disease) are faced with an increased risk of developing CRC,18, 19 and current recommendations for primary prevention of CRC among IBD patients involve periodical surveillance by colonoscopy.20 Detecting early dysplastic lesions from HE stained sections of biopsies from such screening remains a difficult task, and dysplasia can occur in flat, normal appearing mucosa.21 We were not able to get hold of RNA from dysplastic IBD biopsies for testing and can therefore not conclude whether SLC39A14-exon4B is a marker for dysplasia in IBD, but results from the nondysplastic biopsies warrant testing of this potential.

In conclusion, the SLC39A14-exon4B transcript variant is a highly specific and sensitive cancer biomarker for colorectal tissue biopsies. However, the cancer-specificity is restricted to the colon and rectum, and further exploration of the biomarker is appropriate only in an organ-confined context.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Cholangiocarcinoma cell lines included in the study were kindly provided by Dr. Gregory Gores, Mayo Clinic, MN, USA, and Dr. Alexander Knuth, University Hospital Zurich, Switzerland. Gall bladder cancer cell lines were also kindly provided by Dr. Alexander Knuth. A.S. has a PhD grant from the Research Council at Rikshospitalet-Radiumhospitalet Health Enterprise (R.A.L.). T.H.Å. has a PhD grant from the Norwegian Cancer Society. The study has been financed by grants from the Norwegian Cancer Society (PR-2008-0163; G. E. Lind, PR-2006-0442; R.A.L., PR-2007-0166; R.I.S.) and South-Eastern Norway Regional Health Authority (G. E. Lind).

References

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information
  • 1
    Ferlay J, Shin HR, Bray F, Forman D, Mathers C, Parkin DM. Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008. Int J Cancer 2010; 127: 2893917.
  • 2
    Locker GY, Hamilton S, Harris J, Jessup JM, Kemeny N, Macdonald JS, Somerfield MR, Hayes DF, Bast RC. ASCO 2006 update of recommendations for the use of tumor markers in gastrointestinal cancer. J Oncol Pract 2006; 24: 531327.
  • 3
    Thorsen K, Mansilla F, Schepeler T, Øster B, Rasmussen MH, Dyrskjøt L, Karni R, Akerman M, Krainer AR, Laurberg S, Andersen CL, Ørntoft TF. Alternative splicing of SLC39A14 in colorectal cancer is regulated by the Wnt pathway. Mol Cell Proteomics 2011; 10: M110.002998.
  • 4
    Bienz M, Clevers H. Linking colorectal cancer to Wnt signaling. Cell 2000; 103: 31120.
  • 5
    Clements WM, Wang J, Sarnaik A, Kim OJ, MacDonald J, Fenoglio-Preiser C, Groden J, Lowy AM. beta-Catenin mutation is a frequent cause of Wnt pathway activation in gastric cancer. Cancer Res 2002; 62: 35036.
  • 6
    Yue W, Sun Q, Dacic S, Landreneau RJ, Siegfried JM, Yu J, Zhang L. Downregulation of Dkk3 activates beta-catenin/TCF-4 signaling in lung cancer. Carcinogenesis 2008; 29: 8492.
  • 7
    Diamandis EP. Cancer biomarkers: can we turn recent failures into success? J Natl Cancer Inst 2010; 102: 14627.
  • 8
    Berg M, Danielsen SA, Ahlquist T, Merok MA, Agesen TH, Vatn MH, Mala T, Sjo OH, Bakka A, Moberg I, Fetveit T, Mathisen O, et al. DNA sequence profiles of the colorectal cancer critical gene set KRAS-BRAF-PIK3CA-PTEN-TP53 related to age at disease onset. PLoS One 2010; 5: e13978.
  • 9
    Thorstensen L, Diep CB, Meling GI, Aagesen TH, Ahrens CH, Rognum TO, Lothe RA. WNT1 inducible signaling pathway protein 3, WISP-3, a novel target gene in colorectal carcinomas with microsatellite instability. Gastroenterology 2001; 121: 127580.
  • 10
    Sveen A, Agesen TH, Nesbakken A, Rognum TO, Lothe RA, Skotheim RI. Transcriptome instability in colorectal cancer identified by exon microarray analyses: associations with splicing factor expression levels and patient survival. Genome Med 2011; 3: 32.
  • 11
    Irizarry RA, Hobbs B, Collin F, Beazer-Barclay YD, Antonellis KJ, Scherf U, Speed TP. Exploration, normalization, and summaries of high density oligonucleotide array probe level data. Biostatistics 2003; 4: 24964.
  • 12
    Rozen S, Skaletsky HJ. Primer 3. 1998. Available at: http://biotools.umassmed.edu/bioapps/primer3_www.cgi. Accessed on 1 Dec 2009.
  • 13
    Skotheim RI, Nees M. Alternative splicing in cancer: noise, functional, or systematic? Int J Biochem Cell Biol 2007; 39: 143249.
  • 14
    Ransohoff DF. Cancer. Developing molecular biomarkers for cancer. Science 2003; 299: 167980.
  • 15
    Weiss L, Grundmann E, Torhorst J, Hartveit F, Moberg I, Eder M, Fenoglio-Preiser CM, Napier J, Horne CH, Lopez MJ. Haematogenous metastatic patterns in colonic carcinoma: an analysis of 1541 necropsies. J Pathol 1986; 150: 195203.
  • 16
    Sahani DV, Kalva SP, Fischman AJ, Kadavigere R, Blake M, Hahn PF, Saini S. Detection of liver metastases from adenocarcinoma of the colon and pancreas: comparison of mangafodipir trisodium-enhanced liver MRI and whole-body FDG PET. AJR Am J Roentgenol 2005; 185: 23946.
  • 17
    Bipat S, van Leeuwen MS, Comans EF, Pijl ME, Bossuyt PM, Zwinderman AH, Stoker J. Colorectal liver metastases: CT, MR imaging, and PET for diagnosis—meta-analysis. Radiology 2005; 237: 12331.
  • 18
    Eaden JA, Abrams KR, Mayberry JF. The risk of colorectal cancer in ulcerative colitis: a meta-analysis. Gut 2001; 48: 52635.
  • 19
    Jess T, Gamborg M, Matzen P, Munkholm P, Sorensen TI. Increased risk of intestinal cancer in Crohn's disease: a meta-analysis of population-based cohort studies. Am J Gastroenterol 2005; 100: 27249.
    Direct Link:
  • 20
    Itzkowitz SH, Present DH. Consensus conference: colorectal cancer screening and surveillance in inflammatory bowel disease. Inflamm Bowel Dis 2005; 11: 31421.
  • 21
    Mattar MC, Lough D, Pishvaian MJ, Charabaty A. Current management of inflammatory bowel disease and colorectal cancer. Gastrointest Cancer Res 2011; 4: 5361.

Supporting Information

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
IJC_27399_sm_suppinfo.doc36KSupporting Informaion Supplementary Table 1. Cell lines included in the study

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.