This study was to evaluate the frequency of colorectal neoplasia in renal transplant recipients and to investigate the association with Epstein-Barr virus (EBV) and cytomegalovirus (CMV) infection. We compared the frequency of colorectal neoplasia among renal transplant recipients with that of the healthy subjects. Specimens of colorectal neoplasia were examined for EBV and CMV using in situ hybridization and immunohistochemistry, respectively. Of 796 renal transplantation cohorts, 315 were enrolled. The frequency of colorectal neoplasia among the patients was 22.9%. Compared with the healthy subjects, the odds ratio (OR) for advanced adenoma was 3.32 (95% CI, 1.81–6.10). The frequency of cancer among the patients was 1.9% (OR, 12.0; 95% CI, 1.45–99.7). A long interval between transplantation and colonoscopy was a significant factor in the development of advanced colorectal neoplasia. EBV positivity was detected in 30.6% of colorectal neoplasia specimens from renal transplant recipients, which was higher than that for the controls (p = 0.002). CMV was not detected in any lesions of patients or controls. In conclusion, renal transplant recipients have a significantly increased risk of advanced colorectal neoplasia. EBV was more frequently found in specimens of advanced colorectal neoplasm obtained from the renal transplant recipients.
Several studies have shown that transplant recipients have a three to five times higher risk of developing malignancies (1,2). Although some studies have identified an increased risk of colorectal neoplasia among these patients (3–6), the topic remains controversial.
Immunosuppression impairs normal host cancer surveillance mechanisms, thereby increasing the risk of malignancy (7). Immunosuppressive medications decrease immune surveillance, enabling reactivation of potentially oncogenic viruses such as human papillomavirus, human cytomegalovirus (CMV), Epstein-Barr virus (EBV) and human herpes virus-8 (8). EBV is lymphotrophic and enhances the transformation of normal resting B cells into indefinitely proliferating lymphoblastoid cell lines (9). EBV is a causative factor in some gastrointestinal malignancies, especially gastric cancer (10–12). As lymphoepithelial carcinoma has been observed in colorectal lesions, their pathogenesis may involve virologic factors (13–17). Human CMV also may be involved in the carcinogenesis and progression of colorectal carcinoma. CMV nucleic acids and proteins specifically localized to neoplastic cells in human colorectal polyps and adenocarcinomas have been reported (18).
The hypothesis of our study was that long-term immune suppression and associated oncogenic viral reactivation increase the risk for colorectal malignancies. The aims of this study were to evaluate the frequency of colorectal neoplasia in renal transplant recipients who underwent surveillance colonoscopy and to investigate the association of CMV or EBV infection in colorectal neoplasia.
A total of 1657 patients received kidney transplantations at Kangnam St. Mary's Hospital, Seoul, Korea, since 1969. Among them, 307 patients had graft failure, 281 died and 209 were lost to follow-up. By 2004, 860 cohorts were registered at the transplantation center. We invited all the transplant cohorts under the age of 75 to take part in a surveillance program for colorectal neoplasia.
Exclusion criteria were as follows: (1) unwillingness to participate (N = 398), (2) under 18 years of age (N = 7), (3) a serious medical condition that interferes with colonoscopy (N = 46), (4) intolerance of cleansing solution intake for colon preparation (N = 25), (5) history of colorectal neoplasia (N = 5) and (6) patients who had colonoscopic examination within 12 months after transplantation were excluded because a colorectal malignancy diagnosed at this period would probably not relate to transplantation (N = 64). Between February 2004 and January 2006, 315 renal transplant recipients agreed to participate in our study.
Prior to this study, screening of colonic neoplasia in the renal transplant recipients had not been fully installed due to insufficient medical resources in our hospital. Colonoscopic examination or barium study had been performed only in 22 cases with alarm symptoms of colonic diseases, and thereby, only 22 patients had been evaluated 2 years before this study. Among them, 1 patient had colon adenoma, and 5 were diagnosed with colorectal cancer.
The usual protocol for immunosuppression in the renal transplantation unit was as follows: the initial immunosuppressive regimen was cyclosporine A and steroid; FK506 or mycophenolate mofetile was added as a primary immunosuppressant for patients who exhibited clinical or pathological evidence of rejection.
Control subjects were enrolled from consecutive healthy subjects aged 20–75 years who underwent colonoscopic colorectal cancer surveillance or a routine health checkup during the study period. Each renal transplantation patient was matched for age and gender with 2 control patients who were randomly selected from the group of healthy subjects. The protocol was approved by the Institutional Review Board of Kangnam St. Mary's Hospital, Seoul, Korea.
A complete colonoscopic examination was performed. The patients were put under conscious sedation with an intravenous dose of midazolam. Colonoscopy was performed using an Olympus series Q240 or Q260 colonoscope. Colon preparation was done using polyethylene glycol according to protocol.
All polyps retrieved during the colonoscopy examination were fixed and embedded in paraffin. All polyps that were described as adenomatous were classified as either tubular, tubulovillous or villous and according to size and the presence of dysplasia. An adenoma was defined as advanced if it was reported as containing villous features (villous or tubulovillous adenomas), was large (≥1 cm as estimated by the endoscopist during colonoscopy) or had severe dysplasia. Carcinoma in situ was classified as severe dysplasia. Cancer was defined as the invasion of malignant cells beyond the muscularis mucosa. Advanced colorectal neoplasia was defined as advanced adenoma or invasive cancer.
In situhybridization of EBV
We conducted in situ hybridization to detect EBV in adenoma, advanced adenoma and cancer tissue. Two different nucleic acid probes were used to determine the presence of EBV in colorectal neoplasm. The DNA probes included the EBV-encoded small RNA (EBER) and Not I tandem DNA repeat regions. The probe was complementary to nucleic acids 10–33 of the EBER RNA sequence. EBER is an abundant RNA transcript localized in 107 copies within the nucleus of latently infected cells. This sequence was also commercially synthesized and biotin labeled. The Not I region (internal repeat 2) is present as an average of 12.6 copies within the EBV genome and in situ hybridization with an oligonucleotide probe specific for this region has been shown to identify the amplified EBV genome in productive EBV infection.
In situ hybridization of EBER was performed according to a modification of a previously described cDNA-mRNA in situ hybridization method using the manual Microbe System (Fisher Scientific, Pittsburgh, PA) (19). Briefly, formalin-fixed, paraffin-embedded tissues were placed on the distal one-third of Probe-On positively charged slides. The specimens were deparaffinized with Hemode (xylene substitute) at a ratio of 3:1 for 5 min, placed in 100% alcohol for 45 s and rehydrated with 1 Automation Buffer (Biomeda, Foster City, CA). The specimens were digested with pepsin (pH 2.0) for 6 min at 110°C and immediately washed in distilled water for 30 s. The biotinylated cDNA oligonucleotide probes, 5′-CCCTAGCAAAACCTCTAGGGCAGC-3′ (Research Genetics, Huntsville, AL) for EBER-mRNA and 5′-TGGGCCGCTGCCCCGCTCCGGGT-3′ for Not I tandem DNA repeats, were reconstituted to 1 g/L in 10 mM Tris (pH 7.5), diluted to a concentration of 200 ng/mL in the probe cocktail containing 12.5% (v/v) formamide, 0.5% (w/v) chondroitin sulphate, 675 mM sodium chloride, 138 mM sodium citrate, 12 mM sodium phosphate monobasic, 63 mM sodium dibasic, 0.1% (v/v) Tween 20 (JT Baker, Phillipsburg, NJ) and 1:100 000-diluted random 9-mers (Research Genetics). This probe mixture was applied for 10 min at 92°C and 30 min at 45°C to hybridize with target nucleic acids in the specimens. After hybridization, the specimens were washed in 5 × SSC (3 M NaCl plus 0.3 M sodium citrate [pH 7.0]) for 3 min and 2 × SSC for 3 min at room temperature. Detection of hybridized biotin probes was performed with avidin–alkaline phosphatase (Dako, Carpinteria, CA) diluted 1:200 in Tris buffer (pH 8.0), and signal localization was done using naphthol AS-MX phosphate/Fast Red TR chromogen solution. The slides were washed with distilled water, counterstained with hematoxylin for 5 s and cover slips were applied with Universal Mount (Research Genetics). Positive and negative control slides were included in each batch of slides. The positive control slide was a known case of EBV-positive gastric cancer to which was added the specific EBER or Not I oligonucleotide probe. The negative control slide was another section of the same case of known EBV-positive gastric cancer to which was added a random probe consisting of a biotin-labeled oligonucleotide cocktail. Using this random probe as a negative background control alongside the EBV probe contributes to the validation of staining obtained using the EBV probe.
Immunohistochemical staining of CMV
Immunohistochemical staining was performed using sections 4 mm thick obtained from the formalin-fixed, paraffin-embedded colorectal neoplastic tissues. We digested each section with pepsin (37°C, 4 min; BioGenex, San Ramon, CA), retrieved the antigen with citrate buffer (pH 6·0, 37°C, 12 h; BioGenex) and blocked nonspecific peroxidase activity (3% H2O2, 12 min) and antibody binding (Fc-receptor blocker, 10 min, 20°C; Innovex Sciences, Richmond, CA). We incubated each section with monoclonal antibody directed against CMV immediate early antigen (E13, Accurate Chemicals, San Diego, CA; 1:100 dilution) or without antibody in Tris-buffered saline (pH 7·6) plus 0·05% Tween 20 (TBST; 4°C, 12 h). The negative control for E13 immunohistochemistry consisted of elimination of antibody from the reaction, substitution of antibody with an equivalent concentration of anti-CD34 isotype-identical monoclonal antibody, which is specific for endothelial cells of blood vessels, or both. We rinsed the slides with TBST and colorimetric determination was done using a three-step horseradish peroxidase detection system (BioGenex) with the chromogen, diaminobenzidine (Innovex Sciences).
We compared the frequency of colorectal neoplasia between renal transplant recipients and age- and gender-matched healthy subjects. The frequency of EBV or CMV positivity in colorectal neoplasia samples was compared between patients and controls. All comparisons were subanalyzed according to age group (less or more than 50 years) because 50 is the recommended age for screening for colorectal neoplasia in Korea.
Database management and all statistical analyses were performed using SAS software (SAS Institute, Cary, NC). Rates and proportions were calculated for categorical data, and means and standard deviation for continuous data. Kaplan–Meier analysis was used to estimate the probability of the study end points. In addition, standard logistic regression methods were used to calculate odds ratios (ORs) for advanced colorectal neoplasia with 95% confidential intervals. In all procedures, p < 0.05 was considered the level of significance.
Characteristics of subjects
Among 796 cohorts who underwent renal transplantation, 315 (39.6%) patients participated in our study. Five patients (3 male and 2 female) who were diagnosed before this study as having colorectal neoplasia after undergoing renal transplantation were excluded. The interval between transplantation and the diagnosis of colorectal cancer in these patients ranged from 46 to 125 months (mean, 105 months). Two of these patients had extensive liver metastasis, two underwent surgery and one underwent an endoscopic removal.
The mean age of the 315 renal transplantation recipients was 50.1 ± 10.2 years (range, 20–75 years) and 48.9% were older than 50 years. Male patients comprised 59.4% of the patients. The interval between transplantation and colonoscopic examination was 134.9 ± 66.4 months (range, 13–362 months). The mean interval between transplantation and colonoscopic examination for patients less than 50 years old was 128.0 ± 62.3 months and that for patients aged 50 years or older was 142.1 ± 70.0 months (p = 0.06). Two hundred and fifty-seven (81.6%) patients received a kidney from a living donor and the others received a kidney from a deceased donor.
The mean age of the 630 healthy subjects was 50.2 ± 10.1 years (range, 20–75 years) and 48.9% were older than 50 years. Male patients comprised 59.3% of the patients.
Frequencies of colorectal lesions
The frequencies of colorectal lesions in patients and controls are shown in Table 1. Of the renal transplant recipients, 31.7% had colorectal lesions.
Table 1. Frequencies of colorectal pathologic lesions
Renal transplant recipients (N = 315)
Healthy subjects (N = 630)
P < 0.01 between the renal transplant recipients and the healthy subjects.
1Two cases of inflammation.
210 cases of aphthous ulcers and 7 cases of chronic inflammation.
Carcinoma in situ
Villous or tubulovillous adenoma
Seventy-two (22.9%) renal transplant recipients had colorectal neoplasia (including adenoma, advanced adenoma and cancer); 22 (7.0%) patients had advanced adenomas and 6 (1.9%) patients had cancer. The frequency of total colorectal neoplasia was significantly higher in the patient group than the control group (p = 0.005). The OR of advanced adenoma was 3.52 (95% confidence interval [CI], 1.90–6.53) between renal transplant recipients and controls (p < 0.001, Figure 1).
Analysis limited to colorectal cancer revealed that the renal transplant recipients had a greater risk of developing cancer compared with the control subjects (OR, 12.0; CI, 1.45–99.7; p = 0.021).
The frequency of colorectal neoplasia increased with age (Table 2). Differences in the frequency of colorectal neoplasia between control and transplantation recipients were significant for the age groups of over 60 years and 40–49 years (linear-by-linear association, p < 0.001 and p = 0.019, respectively). Frequencies of colorectal neoplasia, advanced adenoma and cancer did not differ between male and female patients in the patient and control groups.
Table 2. Colorectal neoplasia according to age
Age group (years)
*Linear-by-linear association between transplantation recipients and healthy subjects.
Values are number of patients with percentages in parentheses.
Duration of immune suppression and advanced colorectal neoplasia in renal transplant recipients
The frequency of colorectal neoplasia increased as the interval between transplantation and colonoscopy increased, which coincides an increase with the duration of immunosuppression (linear-by-linear association, p = 0.005; Table 3 and Figure 2). This association was statistically significant when the analysis was restricted to advanced adenoma and cancer (linear-by linear association, p < 0.01).
Table 3. Frequency of colorectal neoplasia according to the interval between transplantation and colonoscopy
Age Interval between KT and CFS (years)
<50 years (N = 161)
≥50 years (N = 154)
Total (N = 315)
CFS, colonoscopy; KT, kidney transplantation
Frequency of advanced colorectal neoplasia 15 years after transplantation was significantly higher than that of shorter intervals after transplantation (p = 0.04).
The results of logistic regression analysis adjusted for age and gender are shown in Table 4. The frequency of advanced colorectal neoplasia 15 years after transplantation was significantly higher than that of shorter intervals after transplantation (p = 0.04).
Table 4. Linear regression model of the development of advanced colorectal neoplasia
Lower limit of 95% CI
Upper limit of 95% CI
OR, odds ratio; CI, confidence interval.
Advanced colorectal neoplasia cases included those with advanced adenoma and cancer.
Interval after transplantation, 1–4 years
Interval after transplantation, 5–9 years
Interval after transplantation, 10–14 years
Interval after transplantation, ≥15 years
Age <50 years
Age ≥50 years
Colorectal neoplasia and EBV
As shown in Figure 3, for all types of colorectal neoplasia, the frequency of EBER positivity was significantly higher among patients than controls (30.6% and 4.2%, respectively; p < 0.001; OR, 10.23; 95% CI, 3.34–31.34). This difference was also observed between patients and controls with advanced neoplasia (42.9% and 5.6%, respectively; p = 0.008; OR, 12.00; 95% CI, 1.39–103.48). The cancer tissues of the patients showed EBER positivity in five out of six lesions.
Tissues that exhibited EBER positivity were analyzed using in situ hybridization of Not I, a marker of EBV reactivation. Not I was detected in 3 transplant patients, whereas it was not detected in control subjects with colorectal neoplasia.
The EBER-1-positive cells consisted of lymphocytes in the area of colorectal neoplasia in all patients except one. EBER-1 was detected in the colonic cancer cells of 1 patient (Figure 4).
Results of CMV immunohistochemical staining of colorectal neoplasia tissue
There was no positive result for the patient or the control group.
Our study revealed that renal transplant recipients have a significantly increased risk of colorectal adenocarcinoma or advanced adenoma. Advanced colorectal neoplasia, including cancer and advanced adenoma was found in 8.9% of renal transplant recipients. The incidence of colorectal adenocarcinoma among our cohort of renal transplant recipients was 1.9%, and the risk of colorectal adenocarcinoma among renal transplant recipients was 12-fold higher than that of the controls. The presence of EBV positivity among the renal transplant recipients suggests that EBV is involved in the pathogenesis of colorectal neoplasia.
This study was conducted using a large cohort data from a single center in which at-risk populations were specified. There have been several studies for the risk of colorectal malignancies in patients with transplantation, almost of which had used registry data. We think this study can provide a better insight for the relative risk of colorectal neoplasia in postrenal transplantation than other previous studies using registry data. Registry data may have inaccuracy due to the tendency of underreporting and nonuniform reporting of posttransplant tumors, because patients are to be included in the analysis only if the data were confirmed in writing by the participating transplant center. Considering the invasive nature of colonoscopy, the strongest point of our study is the systematic colonoscopic examination performed to all medically eligible patients.
The high risk of colorectal neoplasia among renal transplant recipients may be associated with long-term immunosuppression. Chronic disease states requiring immunosuppression, such as rheumatoid arthritis and systemic lupus erythematosis, are also associated with high rates of malignancy (20). The risk of colorectal cancer among transplantation patients has been the subject of debate. European studies, including Swedish and Danish studies, reported an increased risk of colorectal cancer among solid-organ recipients, especially among liver transplant recipients (3,4). A 30-year follow-up study of Australian and New Zealand transplant recipients reported that the risk ratio of colorectal cancer was 2.6 for renal transplant recipients (21). However, in a study of American patients, only 3 of 556 renal recipients developed colorectal cancer (6).
Considering that we excluded five previously diagnosed cases of colorectal cancers from our study, the actual incidence of colorectal neoplasia among transplantation patients may be higher than our study indicated. In our study, colon cancer was detected in one of the patients less than 50 years old, whereas advanced adenoma was detected in 6.7% of patients of the fifth decade of life. The frequency of advanced adenoma in patients of the fifth decade was higher than those of healthy subjects of the sixth or seventh decades. In Korea, the peak incidence of colorectal cancer is 0.15% in the sixth decade of life (22). Therefore, renal transplant recipients should be regarded as having a high risk of developing colorectal cancer, and screening for colorectal neoplasia among these patients should be started at younger age.
The risk of advanced colorectal neoplasia tended to increase with an increase in the interval after transplantation. When adjusted for age and gender, a high incidence of advanced colorectal neoplasia was still evident for patients with a long interval between renal transplantation and colonoscopy. This association may be caused by an extended period of immunosuppression, which may increase the rate of development and aggressiveness of neoplastic cells.
It remains unclear whether EBV plays a significant role in the carcinogenesis of colorectal cancer. Several cases of lymphoepithelial carcinoma in colorectal lesions have been reported, suggesting that EBV may play a role in the pathogenesis of colorectal cancer (13–15,23). In our study, we detected one case of lymphoepithelial carcinoma with EBER positivity.
In the present study, infiltration of EBV-positive lymphocytes into 30.6% of colorectal neoplasia specimens collected from renal transplant recipients may indicate that EBV plays a role in colorectal carcinogenesis. In lymphomas and nasopharyngeal carcinoma, EBV proteins incorporated in these cells may mimic the actions of growth factors, transcription factors and antiapoptotic factors that interfere with control of the cellular pathways, which would promote the development of malignancy (24). EBV infection itself results in the production of cytokines such as interleukin (IL)-8, IL-10, tumor necrosis factor (TNF)-α and TNF-β (25,26). The involvement of these cytokines in the pathogenesis of colorectal neoplasia in transplantation recipients is plausible in view of the intimate relationship between infected lymphocytes of the lamina propria and adjacent colonocytes. EBV may be activated in the gut as a consequence of immunosuppression and trigger or accelerate an inflammatory response, resulting in progression of an adenoma to a carcinoma or acceleration of the development of advanced adenoma or cancer in this group of patients. However, chance variation related to the small sample size could cause variation in the comparison of the frequency of EBV positivity between the controls and the patients. This is a limitation with regard to the interpretation of the EBV data in our study.
Our results suggest that EBV reactivation of the recipients themselves may be related with the development of colorectal neoplasia. There is also a possibility that colorectal neoplasia may be caused by EBV transfer from donor. Opelz et al. reported that EBV seronegative kidney transplant recipients had increased risk in non-Hodgkin lymphoma when they were exposed to EBV seropositive donors (27). The authors suggested the importance of confirming EBV serostatus before kidney transplantation. However we did not study the EBV serostatus before and after transplantation. This also can be a limitation in interpreting our EBV tissue results whether the EBV activation resulted from the recipients or from donor. Still, there has been no report whether EBV serology is correlated with the presence of EBV in colon tissue.
Immunosuppressive agents themselves may be associated with colorectal malignancies. Some studies on humans have suggested a link between cancers and immunosuppressive agents such as cyclosporin A or OKT3 (28,29). This link was also supported by a study that demonstrated that cyclosporin A had a tumor growth-promoting effect in immunodeficient mice (30). However, an epidemiological study conducted in the Netherlands did not find that the risk of malignancies was increased among patients treated with cyclosporin A (31). The association between immunosuppressive agents and colorectal malignancy observed in our study remains to be clarified.
We did not detect positive staining for CMV in colorectal neoplasia specimens from patients or in the controls. Although one study found that human CMV infection might play a role in colorectal cancer progression (18), another study reported found no association between CMV infection and tumor progression or metastasis formation in colorectal cancer (32).
In conclusion, our results provide a new perspective for understanding the increased susceptibility to colorectal neoplasia in renal transplant recipients and may be useful for future screening guidelines and treatment strategies for this population. The risk of developing colorectal neoplasia is high among renal transplant recipients and is associated with the duration of immune suppression, irrespective of age. Therefore, surveillance for colorectal neoplasia is strongly recommended for renal transplant recipients who have long history of immunosuppression. The association of EBV with colorectal neoplasia warrants further study to determine the underlying molecular and virological mechanisms.
The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.
The authors thank Clinical Research Coordinating Center of Catholic Medical Center for statistical assistance.
Funding source: This work was partly supported by a 2006 research grant from the Catholic Cancer Center, Korea and by 2009 Seoul St. Mary's Clinical Medicine Research Program.