274

Chin Med J 2013;126 (2)

Original article
Impact of chronic kidney disease on serum tumor markers concentrations
TONG Hong-li, DONG Zhen-nan, WEN Xin-yu, GAO Jing, WANG Bo and TIAN Ya-ping Keywords: retrospective study; kidney insufficiency; tumor marker
Background Serum tumor markers have always been of clinical importance in the diagnosis, monitoring disease progression and therapy efficacy for patients with malignant diseases. However, elevated serum tumor markers are found in some benign conditions, especially in chronic kidney disease (CKD). The elevation of them in CKD might cause confusion and misuse of these tumor markers. We conducted this retrospective study to investigate which of the five widely used tumor markers including carcinoembryonic antigen (CEA), alpha-fetoprotein (AFP), cytokeratin 19 fragment antigen 21-1 (Cyfra21-1), squamous cell carcinoma antigen (SCC) and neuron specific enolase (NSE) are affected markedly by CKD, in order to use them more effectively. Methods Serum tumor marker concentrations, biochemical, hematological parameters, and urinalysis were measured in CKD patients and healthy controls. The positive rate and median tumor markers level in CKD patients and controls, and those in CKD patients stratified by CKD grade were compared using nonparametric rank tests. Correlation analysis of serum tumor markers and other parameters in CKD patients were performed using the Spearman correlation coefficient. Multivariate Logistic regression analysis was used to estimate the important variables that caused elevated serum concentrations of these markers in CKD patients. Results The overall positive rates and serum concentrations of Cyfra21-1, SCC, CEA in CKD group were significantly higher than those in control group. Positive rate and serum concentrations of those tumor markers increased as kidney function decreased. Both univariate analysis and multivariate regression analysis showed that the elevations of those tumor markers were not only associated with kidney function, but also with nutritional status. Conclusions Serum concentrations of Cyfra21-1, SCC, CEA are significantly influenced by kidney function, as well as nutritional status. Therefore, in clinical work, the indices of kidney function and nutritional status could be simultaneously measured to improve interpretation of the results of those tumor marker concentrations. Chin Med J 2013;126 (2): 274-279

umor markers are biochemical indicators for the presence of tumor. They are used as important tools for early diagnosis, monitoring disease progression and therapeutic efficacy for patients with cancer in selected sites.1-4 In clinical practice, elevated serum tumor marker levels are also found in some benign conditions, especially in chronic kidney disease (CKD). The elevation of serum tumor markers in CKD might cause confusion and misuse of these tumor markers. This problem has been marked in the past several decades, because the incidence of CKD has sharply increased worldwide.5-10 Furthermore, most patients with cancer are elderly and have various degrees of kidney insufficiency, and their kidney function sometimes becomes worse during therapy.11 Therefore, it is important to determine the relations between the concentration of tumor markers and the extent of kidney insufficiency to make the tumor markers be used more effectively. At present, there are a number of published reports that suggest kidney function influences the blood concentrations of tumor markers.12-16 However, the effect of CKD on serum tumor marker levels is not well understood. We conducted this retrospective observational study to explore whether and to what extent kidney insufficiency influences the serum concentrations of five widely used tumor markers

(carcinoembryonic antigen (CEA), alpha-fetoprotein (AFP), cytokeratin 19 fragment antigen 21-1 (Cyfra21-1), squamous cell carcinoma antigen (SCC) and neuron specific enolase (NSE)), in order to make better use of these tumor markers in clinical practice. METHODS Study population The study population consisted of 539 non-dialysis CKD patients with varying degrees of kidney insufficiency, and 223 healthy individuals served as controls. They are all Chinese. CKD patients were inpatients of the Department of Nephrology, Chinese Peoples Liberation Army
DOI: 10.3760/cma.j.issn.0366-6999.20121589 Department of Clinical Biochemistry, Chinese Peoples Liberation Army General Hospital, Beijing 100853, China (Tong HL, Dong ZN, Wen XY, Gao J, Wang B and Tian YP) Correspondence to: Dr. TIAN Ya-ping, Department of Clinical Biochemistry, Chinese Peoples Liberation Army General Hospital, Beijing 100853, China (Tel: 86-10-66939374. Fax: 86-10-88217385. Email: tianyp61@gmail.com) This study was supported by a grant from the National Science and Technology Infrastructure Program of China (No. 2009BAI86B05). Conflict of interests: None.

Chinese Medical Journal 2013;126 (2)

275

General Hospital, from January 2008 to January 2009. CKD was diagnosed and classified according to Kidney Disease: Improving Global Outcomes (KDIGO) clinical practice guidelines.17 CKD was defined as either kidney damage or estimated glomerular filtration rate (eGFR) <60 ml/min per 1.73 m2 for 3 or more months. eGFR was calculated from pre-dialyzed serum creatinine using the modified glomerular filtration rate estimating equation for Chinese:18 eGFR=175 (serum creatinine)1.234 (age)0.179 0.79 (if the individual is female). CKD was classified according to eGFR in ml/min per 1.73 m2 of stage I 90, stage II 6089, stage III 3059, stage IV 1529, and stage V <15. Kidney damage was defined as an albumin-to-creatinine ratio >30 mg/g in two of three spot urine specimens. The patients who met the diagnosis criteria for CKD, and did not accept dialysis therapy are eligible for inclusion. After systematic health examination, patients with acute kidney function deterioration, malignancy, pleural effusion or ascites, heart failure, hepatocirrhosis or liver failure were excluded. Healthy controls were healthy individuals without known major disease who visited Chinese Peoples Liberation Army General Hospital for a routine physical examination. Measurement of serum tumor markers, biochemical, hematological parameters and urinalysis In the morning, fasting venous blood and urine samples of patients and control were collected and labelled by nurses and sent to clinical laboratories for measurement in time. For each patient and control, 3 ml whole blood with the coagulant EDTAK2 was used for hematological parameters, 5 ml blood without coagulant was used for detection of tumor markers and biochemical parameters, and 3 ml urine was used for urinalysis. Chemiluminescence immunoassays were used to measure serum tumor markers (CEA, AFP, Cyfra21-1, NSE) using a Roche E170 automatic immune analyzer (Roche Diagnostics, Shanghai, China). A microparticle enzyme immunoassay was used to measure serum SCC levels using an Abbott IMX immunoassay analyzer (Abbott Diagnostics Inc., Chicago, USA). Serum biochemical parameters such as alanine aminotransferase (ALT), aspartate aminotransferase (AST), protein, albumin, urea, creatinine were measured using a Hitachi 7600 automatic biochemistry analyzer (Hitachi Ltd., Tokyo, Japan). Hematological parameters were measured using a Sysmex SE2100 Hematology Analyzer (Sysmex, Kobe, Japan). Urinalysis was performed using the Max AX4280 (ARKRAY, Kyoto, Japan). All tests were performed according to the manufacturers instructions of the reagents and equipments within twenty-four hours. The maximum value of reference range for serum tumor markers were defined as the cut-off values. The cut-off values for each tumor markers were 5 g/L for CEA, 20 g/L for AFP, 4 g/L for Cyfra21-1, 24 g/L for NSE and 1.5 g/L for SCC. Measured values greater than or equal to the cut-off value were defined as positive. Statistical analyses Data were analyzed using SPSS17.0 software (SPSS Inc.,

USA). Data were given as the mean standard deviation (SD) in case of normal distributions and as median with 25th and 75th percentiles in case of non-normal distributions. The data with normal distributions were compared using Students t test. The tumor marker levels of the study population stratified by CKD grade were compared by using nonparametric rank tests. Pearsons chi-square test was used to compare the positive rate of serum tumor markers between subgroups in the study populations. Multivariate Logistic regression analysis using a stepwise backward method was used to estimate the important variables that caused elevated serum concentrations of these markers in CKD patients. The Spearman rank correlation coefficient was used for correlations between serum tumor markers and biochemical and hematological parameters. A value of P <0.05 was considered statistically significant. RESULTS Clinical characteristics of CKD patients and controls The clinical characteristics of CKD patients and controls are presented in Table 1. There were no significant differences in age and gender distribution between CKD patients and controls (P >0.05). Compared to healthy controls, CKD patients had higher serum urea, creatinine levels, lower serum protein, albumin levels, lower hemoglobin concentration and decreased eGFR (P <0.001). Overall positive rate and median tumor markers levels in CKD patients and controls In control group, the serum levels of those five tumor markers were all within their reference ranges. The positive rates of those five tumor markers in the control group were all 0. The overall positive rates of Cyfra21-1 (30.80%), SCC (26.72%), CEA (9.65%) in CKD group were significantly higher than those in control group (P <0.01). There were no statistically significant differences in the positive rates of AFP, NSE between two groups (P >0.05, Table 2). The median serum Cyfra21-1, SCC, CEA,
Table 1. Clinical characteristics of CKD patients and controls
Characteristics Gender (n) Male Female Age (years) Serum creatinine (mg/dl) Serum urea (mmol/L) eGFR (ml/min per 1.73 m2) Hemoglobin (g/L) Serum protein (g/L) Serum albumin (g/L) CKD stage 1 2 3 4 5 CKD patients (n=539) 323 216 51.1214.64 2.472.67 11.338.44 68.9455.41 120.2828.05 59.4611.85 33.448.18 177 (32.84) 84 (15.58) 110 (20.41) 53 (9.83) 115 (21.34) Controls (n=223) 135 88 51.019.78 0.760.14 5.531.02 115.6721.68 151.7312.04 74.624.08 45.042.56 P values >0.05 >0.05 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001

276
Data expressed as mean standard deviation, or number (percentage).

Chin Med J 2013;126 (2) Table 2. Positive rate of serum tumor markers in CKD patients and controls (n (%))
Tumor markers Cyfra21-1 SCC CEA AFP NSE Controls (n=223) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) CKD patients (n=539) 166 (30.80) 144 (26.72) 52 (9.65) 4 (0.74) 0 (0) P values <0.001 <0.001 <0.001 0.46 1.00

NSE levels were higher in CKD group than those in control group (P <0.05). The median serum AFP level in the CKD group was lower than that in control group (P <0.01, Table 3). Positive rate and median tumor marker levels at different stages of CKD The positive rates of serum Cyfra21-1, SCC increased progressively along with the worsening CKD. CEA showed a tendency similar to that of SCC, but the tendency was weaker (Figure 1). The concentrations of Cyfra21-1, SCC, CEA increased with the worsening CKD (P <0.001). Serum AFP and NSE levels did not differ significantly among different stage of CKD (P >0.05, Table 4). Correlation of serum tumor markers and biochemical and hematological parameters and urinalysis in CKD patients The Spearman rank correlation analysis showed that serum Cyfra21-1, SCC positively correlated with age, CKD grade, serum urea, creatinine, whole blood leukocyte, urine erythrocyte, urine leukocyte, urine protein and urine SEC, negatively correlated with eGFR, serum protein, serum albumin, and hemoglobin (P <0.01). Serum CEA positively correlated with age, CKD grade, urea, creatinine, urine leukocyte, urine protein, urine SEC and whole blood leukocyte, negatively correlated with serum protein, albumin, hemoglobin, platelet, and eGFR (P <0.01). Multivariate Logistic regression analysis using Cyfra21-1 test status (positive or negative) as a dependent variable, showed that CKD grade (OR=1.876, 95% CI= 1.3062.696, P=0.001), urine protein (OR=1.328, 95% CI=1.0761.639, P=0.008) were significant risk factors of elevated serum Cyfra21-1 level, serum protein (OR=0.944, 95% CI=0.9060.983, P <0.001) was protective factor. Using SCC test status as a dependent variable, Logistic regression analysis showed that CKD grade (OR=3.543, 95% CI=2.2485.583, P <0.001), male (OR=2.111, 95% CI=1.1633.831, P=0.014) and urine protein (OR=1.269, 95% CI=1.0141.587, P=0.028) were risk factors of elevated serum SCC level, whereas serum albumin (OR=0.955, 95% CI=0.9120.965, P=0.023) was protective factor. Using CEA test status as a dependent variable, results showed that CKD grade (OR=1.640, 95% CI=1.0132.656, P=0.044), male (OR=2.466, 95% CI=
Tumor markers CEA AFP Cyfra21-1 NSE SCC CKD1 (n=177) 1.67 (1.16, 2.71) 2.43 (1.67, 3.41) 2.25 (1.58, 3.40) 9.24 (7.38, 11.19) 0.50 (0.20, 0.80) CKD2 (n=84) 1.85 (1.21, 2.91) 2.29 (1.57, 3.24) 2.77 (2.07, 4.10) 9.24 (7.54, 11.78) 0.70 (0.40, 1.00)

Table 3. Serum tumor marker levels in CKD patients and controls (g/L, median (25th, 75th percentile))
Tumor markers Cyfra21-1 SCC CEA AFP NSE Controls (n=223) 1.59 (1.20, 2.10) 0.30 (0.10, 0.50) 1.46 (1.02, 2.21) 2.49 (1.85, 3.69) 8.80 (7.70, 10.00) CKD patients (n=539) 2.94 (2.07, 4.32) 0.80 (0.50, 1.60) 2.05 (1.35, 3.24) 2.26 (1.47, 3.21) 9.23 (7.42, 11.77) P values <0.001 <0.001 <0.001 <0.001 0.014

Figure 1. Positive rate of serum Cyfra21-1, SCC and CEA stratified by CKD grade.

1.1275.399, P=0.024), age (OR=1.048, 95% CI=1.0481.075, P <0.001), serum urea (OR=1.108, 95% CI=1.0471.172, P <0.001), serum creatinine (OR=1.014, 95% CI=1.0041.025, P=0.006) were significant predictors of elevated serum CEA level, while eGFR (OR=0.997, 95% CI=0.9941.000, P=0.034), hemoglobin (OR=0.981, 95% CI=0.9650.998, P=0.025) were protective factors. DISCUSSION Serum tumor marker levels can be influenced by a variety of factors, the important one of which is kidney function.19 There have been a number of published reports that suggest kidney function may influence the serum concentrations of tumor markers.12-16 The results
CKD4 (n=53) 2.30 (1.65, 3.81) 2.19 (1.53, 3.00) 3.54 (2.67, 4.55) 9.34 (6.03, 11.92) 1.30 (0.70, 2.00) CKD5 (n=114) 2.88 (1.92, 4.22) 1.98 (1.17, 2.96) 4.13 (2.90, 5.01) 9.83 (8.03, 12.61) 1.90 (1.18, 2.80) P values <0.001 0.089 <0.001 0.148 <0.001

Table 4. Serum tumor marker levels stratified by CKD grade (g/L, median (25th, 75th percentile))
CKD3 (n=110) 1.96 (1.28, 2.90) 2.24 (1.53, 3.00) 2.72 (2.13, 4.24) 8.63 (7.40, 11.77) 0.90 (0.50, 1.40)

Chinese Medical Journal 2013;126 (2)

277

are not entirely consistent. We used eGFR calculated from serum creatinine using the modified glomerular filtration rate estimating equation for Chinese as predictor of kidney function, which is better than serum creatinine alone or creatinine clearance rate. The CKD patients with kidney insufficiency of varying degrees were classified according to eGFR, with range from 3.07 to 309.87 ml/min per 1.73 m2. First, we compared the overall positive rate and median tumor markers level in patients with CKD and healthy control. Then, we compared the positive rate and median tumor markers level in CKD patients stratified by CKD grade. The results showed that the serum concentrations of SCC, Cyfra21-1 and CEA were affected markedly by kidney function. The positive rates and concentrations of serum Cyfra21-1, SCC, CEA increased progressively as the worsening CKD, but the tendency was weaker for serum CEA. It is consistent to the work by Nomura et al16 which showed the mean concentration of CEA, SCC, Cyfra21-1 but NSE significantly increased with the severity of renal failure. In order to explore the possible mechanisms by which kidney function or other factors affected serum tumor markers concentration, we used Spearman rank correlation analysis and multivariate Logistic regression analysis to explore the relationship between serum concentrations of these markers and kidney function, and other biological parameters. The results showed that serum concentrations of Cyfra21-1, SCC, CEA were not only positively correlated with the indices of kidney function and kidney damage, but also inversely correlated with the indices of nutritional status. Multivariate Logistic regression analysis using Cyfra21-1, SCC, CEA test status (positive or negative) as a dependent variable showed similar results. What is the possible underlying mechanism to explain the observed link between serum tumor marker levels and indices of kidney function and other biological indices? CKD is set in 5 stages of increasing severity with a decrease in glomerular filtration rate. Direct injury, high metabolic demands, or stimuli from renal dysfunction activate tubular cells. They produce cytokines, support inflammatory responses, causing further pathologic changes in the renal parenchyma.20,21 SCC is a sub fraction of glycoprotein TA-4 isolated from squamous cell carcinoma of the uterine cervix.22 It is widespread in the epithelium, especially in squamous epithelial cells. The serum SCC level in healthy people is less than 1.5 g/L. Cyfra21-1, a soluble fragments of cytokeratin 19 identified by BM 12.21 and KS19.1 antibodies, is a new tumor marker of non-small cell lung cancer developed in recent years.23 It is also used as an independent predictor for definitive chemoradiotherapy sensitivity in esophageal squamous cell carcinoma.24 CK19 is a characteristic protein component of intermediate filaments of epithelial cell, and it exists in a variety of malignant tissues as well as normal epithelia. In healthy persons, the serum level of Cyfra21-1 is very low without

any influence of sex, age and smoking habits. 25 After cell death, proteolytic enzymes can degrade and solubilize cytokeratin. Cyfra21-1, the soluble fragment of CK19, can be released into the blood. Experiment in vitro showed that Cyfra21-1 can be abundantly released into the extracellular space during the intermediate stage of epithelial cell apoptosis.26 Kashiwabara et al27 reported that in patients with diabetic nephropathy, increased serum Cyfra21-1 levels attribute to metabolic abnormality in the kidney itself rather than the decreased urinary excretion per se. So we speculate that inflammatory response, dedifferentiation or dying of kidney epithelium cells may play important role in elevation of serum Cyfra21-1, SCC levels in CKD patients. Additionally, both SCC22 and Cyfra21-125 are metabolized in the kidney. Declined kidney function could reduce the clearance of SCC and Cyfra21-1 by kidney. Therefore, this could be another mechanism of how CKD affects serum SCC and Cyfra21-1 levels. Our data, the stable increased positive rate and serum SCC, Cyfra21-1 level with increasing CKD grade, the strong association between serum SCC, Cyfra21-1 levels and the indices of renal function and kidney damage, support this hypothesis. On the other hand, patients with CKD frequently have a compromised nutritional status. In our study, the serum protein level ((59.4611.85) vs. (74.624.08) g/L for the CKD and control groups, P <0.001), the serum albumin level ((33.448.18) vs. (45.042.56) g/L for the CKD and control groups, P <0.001) and hemoglobin concentration ((120.2828.05) vs. (151.7312.04) g/L for the CKD and control groups, P <0.001) were significantly lower than control. Studies have confirmed that albuminuria and eGFR associate with anemia, hypoalbuminemia, and other complications of CKD.28 And anemia is an independent risk factor for end-stage renal disease (ESRD).29,30 Albumin is not only an indice of nutritional status, but also an important transport protein for non-water-soluble protein-bound drugs and uraemic toxins. Its transport capacity is reduced in patients with CKD and unbound fractions of uraemic toxins are related to complications of CKD.31 Therefore, hemoglobin and albumin are important renal protective factors. In our study, univariate analysis showed positive associations between serum SCC, Cyfra21-1 levels and urine protein, negative associations between serum protein, albumin and hemoglobin, the associations remaining valid after multivariate regression analysis. These findings indicate that the elevation of serum SCC, Cyfra21-1 levels could also relate to anemia, hypoalbuminemia, albuminuria or proteinuria, in addition to renal function. CEA, an oncofetal glycoprotein is widely used as a tumor marker due to its high expression in adenocarcinomas, particularly in colorectal cancer.32 However, CEA levels can be affected by many factors, including chronic inflammatory disease, renal and hepatic insufficiency, aging, and smoking.33,34 In our study, the prevalence of

278

Chin Med J 2013;126 (2) Cases A, et al. SCC antigen measured in malignant and nonmalignant diseases. Clin Chem 1990; 36: 251-254. Gris JM, Xercavins J, Trillo L, Encabo G. Study of the squamous cell carcinoma (SCC) antigen in benign diseases. Med Clin (Barc) 1996; 106: 205-208. Filella X, Cases A, Molina R, Jo J, Bedini JL, Revert L, et al. Tumor markers in patients with chronic renal failure. Int J Biol Markers 1990; 5: 85-88. Xiaofang Y, Yue Z, Xialian X, Zhibin Y. Serum tumour markers in patients with chronic kidney disease. Scand J Clin Lab Invest 2007; 67: 661-667. Nomura F, Koyama A, Ishijima M, Takano S, Narita M, Nakai T. Serum levels of five tumor markers for lung cancer in patients with chronic renal failure. Oncol Rep 1998; 5: 389-392. Eknoyan G, Levin N. NKF-K/DOQI clinical practice guidelines: Update 2000. Foreword. Am J Kidney Dis 2001; 37 (1 Suppl 1): S5-S6. Ma YC, Zuo L, Chen JH, Luo Q, Yu XQ, Li Y, et al. Modified glomerular filtration rate estimating equation for Chinese patients with chronic kidney disease. J Am Soc Nephrol 2006; 17: 2937-2944. Trap J, Filella X, Alsina-Donadeu M, Juan-Pereira L, Bosch-Ferrer , Rigo-Bonnin R. Increased plasma concentrations of tumour markers in the absence of neoplasia. Clin Chem Lab Med 2011; 49: 1605-1620. Hodgkins KS, Schnaper HW. Tubulointerstitial injury and the progression of chronic kidney disease. Pediatr Nephrol 2012; 27: 901-909. Nangaku M. Chronic hypoxia and tubulointerstitial injury: a final common pathway to end-stage renal failure. J Am Soc Nephrol 2006; 17: 17-25. Kato H, Morioka H, Aramaki S, Torigoe T. Radioimmunoassay for tumor antigen of human cervical squamous cell carcinoma. Cell Mol Biol 1979; 25: 51-56. Grenier J, Pujol JL, Guilleux F, Daures JP, Pujol H, Michel FB. Cyfra 21-1, a new marker of lung cancer. Nucl Med Biol 1994; 21: 471-476. Yan HJ, Wang RB, Zhu KL, Jiang SM, Zhao W, Xu XQ, et al. Cytokeratin 19 fragment antigen 21-1 as an independent predictor for definitive chemoradiotherapy sensitivity in esophageal squamous cell carcinoma. Chin Med J 2012; 125: 1410-1415. Rastel D, Ramaioli A, Cornillie F, Thirion B. CYFRA 21-1, a sensitive and specific new tumour marker for squamous cell lung cancer. Report of the first European multicentre evaluation. CYFRA 21-1 Multicentre Study Group. Eur J Cancer 1994; 30A: 601-606. Sheard MA, Vojtesek B, Simickova M, Valik D. Release of cytokeratin-18 and -19 fragments (TPS and CYFRA 21-1) into the extracellular space during apoptosis. J Cell Biochem 2002; 85: 670-677. Kashiwabara K, Kishi K, Nakamura H, Yagyu H, Kobayashi K, Watanabe O, et al. Mechanism of increased serum cytokeratin 19 fragment levels in patients with diabetic nephropathy as a model of chronic renal failure. Intern Med 1998; 37: 917-921. Inker LA, Coresh J, Levey AS, Tonelli M, Muntner P. Estimated GFR, albuminuria, and complications of chronic

elevated CEA was 9.65%, significantly lower than that of Cyfra21-1 (30.80%), SCC (26.72%). The maximum value was 14.12 g/L, no more than two times above the upper reference limit. Elevated CEA levels in CKD patients were associated with not only renal function and nutritional status, but also age and male gender, and the associations were weaker. Because of without considering the primary disease and smoking status and other potential confounders, further studies are needed to validate those associations. In summary, serum Cyfra21-1, SCC, CEA levels were markedly influenced by kidney function. Elevated serum Cyfra21-1, SCC, CEA levels in CKD patients were not only associated with kidney function, but also nutritional status. Therefore, in clinical work, the indices of kidney function and nutritional status could be simultaneously measured to improve interpretation of the results of those tumor marker concentrations and to prevent diagnostic errors derived from false positive results.
REFERENCES

13.

14.

15.

16.

17.

18.

19. 1. Orntoft TF, Bech E. Circulating blood group related carbohydrate antigens as tumour markers. Glycoconj J 1995; 2: 200-205. Lai LC, Cheong SK, Goh KL, Leong CF, Loh CS, Lopez JB, et al. Clinical usefulness of tumour markers. Malays J Pathol 2003; 25: 83-105. Pandha HS, Waxman J. Tumour markers. QJ Med 1995; 88: 233-241. Virji MA, Mercer DW, Herberman RB. Tumor markers in cancer diagnosis and prognosis. CA Cancer J Clin 1988; 38: 104-126. Coresh J, Selvin E, Stevens LA, Manzi J, Kusek JW, Eggers P, et al. Prevalence of chronic kidney disease in the United States. JAMA 2007; 298: 2038-2047. Collins AJ, Foley R, Herzog C, Chavers B, Gilbertson D, Ishani A, et al. Excerpts from the United States Renal Data System 2007 annual data report. Am J Kidney Dis 2008; 51 (1 Suppl 1): S1-S320. Ninomiya T, Kiyohara Y, Kubo M, Tanizaki Y, Doi Y, Okubo K, et al. Chronic kidney disease and cardiovascular disease in a general Japanese population: the Hisayama Study. Kidney Int 2005; 68: 228-236. Li ZY, Xu GB, Xia TA, Wang HY. Prevalence of chronic kidney disease in a middle and old-aged population of Beijing. Clin Chim Acta 2006; 366: 209-215. Zhang L, Zuo L, Xu G, Wang F, Wang M, Wang S, et al. Community-based screening for chronic kidney disease among populations older than 40 years in Beijing. Nephrol Dial Transplant 2007; 22: 1093-1099. Chen W, Wang H, Dong XQ. Epidemiologic study of chronic kidney disease in Guangzhou urban area (in Chinese). Chin J Nephrol (Chin) 2007; 23: 147-151. Aapro M, Launay-Vacher V. Importance of monitoring renal function in patients with cancer. Cancer Treat Rev 2012; 38: 235-240. Molina R, Filella X, Torres MD, Ballesta AM, Mengual P,

2.

20.

3. 4.

21.

22.

5.

23.

6.

24.

7.

25.

8.

9.

26.

10.

27.

11.

28.

12.

Chinese Medical Journal 2013;126 (2) kidney disease. J Am Soc Nephrol 2011; 22: 2322-2331. Iseki K, Ikemiya Y, Iseki C, Takishita S. Haematocrit and the risk of developing end-stage renal disease. Nephrol Dial Transplant 2003; 18: 899-905. Mohanram A, Zhang Z, Shahinfar S, Keane WF, Brenner BM, Toto RD. Anemia and end-stage renal disease in patients with type 2 diabetes and nephropathy. Kidney Int 2004; 66: 1131-1138. Klammt S, Wojak HJ, Mitzner A, Koball S, Rychly J, Reisinger EC, et al. Albumin-binding capacity (ABiC) is reduced in patients with chronic kidney disease along with an accumulation of protein-bound uraemic toxins. Nephrol Dial Transplant 2012; 27: 2377-2383. Hammarstrom S. The carcinoembryonic antigen (CEA)

279 family: structures, suggested functions and expression in normal and malignant tissues. Semin Cancer Biol 1999; 9: 67-81. Ruibal Morell A. CEA serum levels in non-neoplastic disease. Int J Biol Markers 1992; 7: 160-166. Fukuda I, Yamakado M, Kiyose H. Influence of smoking on serum carcinoembryonic antigen levels in subjects who underwent multiphasic health testing and services. J Med Syst 1998; 22: 89-93.

29.

33. 34.

30.

31.

(Received June 13, 2012) Edited by WANG Mou-yue and LIU Huan

32.