Preoperative Carcinoembryonic Antigen Level as an Independent ...

Jpn J Clin Oncol 2000;30(1)12–16

Preoperative Carcinoembryonic Antigen Level as an Independent Prognostic Factor in Colorectal Cancer: Taiwan Experience Wei-Shu Wang1, Jen-Kou Lin2, Tzeon-Jye Chiou1, Jin-Hwang Liu1, Frank S. Fan1, Chueh-Chuan Yen1, Tzu-Chen Lin2, Jeng-Kae Jiang2, Shung-Haur Yang2, Huann-Sheng Wang2 and Po-Min Chen1 1Division

of Medical Oncology, Department of Medicine and 2Division of Colorectal Surgery, Department of Surgery, Veterans General Hospital–Taipei and National Yang-Ming University School of Medicine, Taipei, Taiwan

Received August 9, 1999; accepted October 7, 1999

Background: Preoperative carcinoembryonic antigen (CEA) level is considered as a factor predictive of survival in colorectal cancer patients. Patients with normal (5 ng/ml was considered abnormal. All patients were followed up until death and the duration from the CEA determination until death was recorded. STATISTICAL ANALYSIS Survival curves were constructed by the Kaplan–Meier product limit method and the data were analyzed by a log-rank test using a microcomputer (6,7). A multivariate analysis using Cox’s proportional hazards regression model was then performed to determine the most important predictors of survival among all of the possible variables (8).

RESULTS The incidence of abnormal preoperative CEA (>5 ng/ml) in patients with different Dukes’ stages are shown in Table 1. Of

13

C2 (52) Overall (218)

Number

2-year p(log-rank survival (%) test)

>5

13 (25%)

98

5

10 (23%)

78

5

16 (50%)

82

5

21 (55%)

32

5

43 (83%)

10

5

103 (47%)

44

5 ng/ml in Dukes’ A, B, C and D diseases were 3, 25, 45 and 65%, respectively (10). The current study confirmed this observation. In our study, the incidence of preoperative CEA >5 ng/ml in Dukes’ A, B and C diseases were 25, 39 and 71%, respectively. The Dukes’ stage is the most important predictor of survival in colorectal cancer and the incidence of abnormal preoperative CEA levels is higher in advanced colorectal cancer (Dukes’ C) than in early-

Jpn J Clin Oncol 2000;30(1)

Figure 4. Survival curves according to preoperative (pre-OP) CEA level of patients with Dukes’ A disease. •, Pre-OP CEA >5 ng/ml (n = 13); {cir}, preOP CEA 5 ng/ml (n = 10); {cir}, preOP CEA 5 ng/ml. Wolmark et al. analyzed 706 Dukes’ B and C patients who entered into the clinical trials of the National Surgical Adjuvant Breast and Bowel Project (12). They found that the relative risk of treatment failure for patients with a preoperative CEA value >10 ng/ml compared with those with a CEA value 5 ng/ml (n = 16); {cir}, preOP CEA 10 ng/ml was associated with a significantly decreased 5-year survival rate (p < 0.05). In contrast, after reviewing 563 colorectal cancer patients, Staab et al. (16) reached a different conclusion. They reported that the differences between survival curves based on preoperative CEA ranges of 5 ng/ml are significant for patients with TNM stage II tumors (p < 0.02) but not for patients with lymph node metastasis (p = 0.1). A study by the Gastrointestinal Tumor Study Group concluded that only in patients with one to four lymph nodes involving preoperative CEA values of >5 ng/ml are associated with a significantly greater risk of recurrence (42 vs 14%, p < 0.02) (17). Moertal et al. studied 272 patients who had undergone resection at the Mayo Clinic. They argued that preoperative CEA was a significant and independent prognostic determinant only in patients with involvement of four or more lymph nodes. In this group (n = 29), the 5-year survival rates

16

Preoperative CEA in colorectal cancer

for patients with CEA levels 10 ng/ml (p = 0.028) (18). However, in a later study of the same patient group using Cox’s multivariate analysis model, Scott et al. reported a preoperative CEA level of >10 ng/ml to be an independent prognostic variable (p = 0.027) (19). There are several reasons that may account for such confusing and even contradictory results. First, some reports with negative results did not have a large enough sample size after stratification. Second, the different statistical methods used may have resulted in different conclusions. Third, different definitions of abnormal CEA levels or different CEA kits were used. The current study of 218 patients treated at a single institution confirmed that an elevated preoperative CEA level of >5 ng/ml is associated with a poorer prognosis only in Dukes’ C colorectal cancer. Besides malignant diseases, elevated serum CEA levels >5 ng/ml are frequently found in other benign disorders. For example, smokers have been reported to have a higher circulating CEA concentration than non-smokers (20). Furthermore, because the liver is the major site for clearance of CEA, moderate to significant elevation of serum CEA levels can be observed in a variety of benign liver diseases, including hepatitis, cirrhosis, cholelithiasis, obstructive jaundice and cholangitis (20). There is a high prevalence of hepatitis B virus infection in Taiwan. Whether an elevated preoperative CEA level is an independent prognostic factor in colorectal cancer or a factor associated with hepatobiliary disorders is of interest to us. In our study, survival analysis was carried out to identify the influence on survival of impaired liver function tests, hepatobiliary tract diseases and smoking habit. Using univariate analysis, the above-mentioned factors were not identified as having significant influence on survival (Table 2). In our study, besides preoperative CEA levels, the existence of penetration of the bowel wall was identified as another independent prognostic factor of colorectal cancer. The Dukes’ classification scheme proposed in 1932 failed to subdivide patients with positive nodes on the basis of the depth of tumor penetration. At the time of formation of the Dukes’ classification, it was commonly believed that lymph node metastases occurred only when there was full thickness penetration of the bowel wall and once nodal metastases were in evidence, it was theorized that the depth of tumor penetration was of little prognostic significance. By introducing a subdivision of the Dukes’ C category, Astler and Coller clearly found that tumors with penetration of the bowel wall had a poor prognosis that was independent of the presence of nodal metastases (21). According to their definition, a ‘C1’ lesion referred to a tumor that had not penetrated the entire thickness of the bowel wall, whereas a ‘C2’ tumor manifested full thickness penetration. In our study, by multivariate analysis, patients with penetration of the bowel wall survived significantly shorter than those without bowel wall penetration (p = 0.0001). In summary, the data from our study indicate that preoperative CEA level is an independent prognostic factor in colo-

rectal cancer. The relationship is especially evident in Dukes’ C disease. In our opinion, preoperative CEA levels could serve as appropriate modifications of the initial Dukes’ scheme. It is recommended that stratification for further clinical trials in colorectal cancer patients should be carried out according to preoperative CEA levels.

Acknowledgments This work was supported by a grant from the Yen Tjing-Ling Medical Foundation.

References 1. Gold P, Freedman SO. Specific carcinoembryonic antigen of the human digestive system. J Exp Med 1965;122:467–81. 2. National Institutes of Health Consensus Development Conference. Carcinoembryonic antigen: its role as a marker in the management of cancer. Cancer Res 1981;41:2017–8. 3. Fletcher RH. Carcinoembryonic antigen. Ann Intern Med 1986;104:66–73. 4. Dukes CE. The classification of cancer of the rectum. J Pathol 1932;35:323–32. 5. Astler VA, Coller FA. The prognostic significance of direct extension of carcinoma of the colon and rectum. Ann Surg 1954;139:846–52. 6. Kaplan EL, Meier P. Non-parametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457–81. 7. Peto R, Pike MD. Conservatism of the approximation in the log-rank test for survival data or tumor incidence data. Biometrics 1973;29:579–84. 8. Cox DR: Regression models and life tables. J R Stat Soc Br 1972;34:187– 220. 9. Wang JY, Tang R, Chiang JM. Value of carcinoembryonic antigen in the management of colorectal cancer. Dis Colon Rectum 1994;37:272–7. 10. Ladenson JH, McDonald JM. Colorectal carcinoma and carcinoembryonic antigen (CEA). Clin Chem 1980;26:1213–20. 11. Wanebo HJ, Rao B, Pinsky CM, Hoffman RG, Stearns M, Schwartz MK, et al. Preoperative carcinoembryonic antigen level as a prognostic indicator in colorectal cancer. N Engl J Med 1978;299:448–51. 12. Wolmark N, Fisher B, Wieand HS, Henry RS, Lerner H, Legault-Poisson S, et al. The prognostic significance of preoperative carcinoembryonic antigen levels in colorectal cancer – results from NSABP clinical trials. Ann Surg 1984;199:375–81. 13. Chu DZ, Erickson CA, Russell MP, Thompson C, Lang NP, Broadwater RJ, et al. Prognostic significance of carcinoembryonic antigen in colorectal carcinoma. Arch Surg 1991;126:3:314–6. 14. Goslin R, Steele G Jr, MacIntyre J, Mayer R, Sugarbaker P, Cleghorn K, et al. The use of preoperative plasma CEA levels for stratification of patients after curative resection of colorectal cancers. Ann Surg 1980;192:747–51. 15. Lewi H, Blumgart LH, Carter DC, Gillis CR, Hole D, Ratcliffe JG, et al. Preoperative carcinoembryonic antigen and survival in patients with colorectal cancer. Br J Surg 1984;71:206–8. 16. Staab HJ, Anderer FA, Brummendorf T, Strumpf E, Fischer R. Prognostic value of preoperative serum CEA level compared to clinical staging. Br J Cancer 1981;44:652–62. 17. Steel G Jr, Ellenberg S, Ramming K, O’Connell M, Moertel C, Lessner H, et al. Carcinoembryonic antigen monitoring among patients in multi-institutional adjuvant G.I. therapy protocols. Ann Surg 1982;196:162–9. 18. Moertal CG, O’Fallon JR, Go VLW, O’Connell MJ, Thynne GS. The preoperative carcinoembryonic antigen test in the diagnosis, staging and prognosis of colorectal cancer. Cancer 1986;58:603–10. 19. Scott NA, Wieand HS, Moertel CG, Cha SS, Beart RW, Lieber MM. Colorectal cancer: Dukes’ stage, tumor site, preoperative plasma CEA level and patient prognosis related to tumor DNA ploidy pattern. Arch Surg 1987;122:1375–9. 20. American Society of Clinical Oncology. Clinical practice guidelines for the use of tumor markers in breast and colorectal cancer. J Clin Oncol 1996;14:2843–77. 21. Astler VA, Coller FA. The prognostic significance of direct extension of carcinoma of the colon and rectum. Ann Surg 1954;139:846–52.