Japanese Journal of Clinical Oncology 30:523 (2000)
© 2000 Foundation for Promotion of Cancer Research
Letters to the Editor |
Author’s Reply
To the Editor:
Watine and Friedberg are concerned about the laboratory tests of carcinoembryonic antigen (CEA) in our study without our fully specifying the methods involved. In our study we retrospectively examined 218 patients with resectable colorectal cancer between 1991 and 1994. All patients enrolled were diagnosed and treated in a single institute [Veterans General Hospital–Taipei (VGH–Taipei)] and their blood samples for CEA testing were collected and examined in a single department (Department of Nuclear Medicine of VGH–Taipei). Furthermore, all CEA values were measured using the same method [radioimmunoassay (RIA)] according to the directions in the manual and the kits for CEA testing were purchased from a single institute (CIS Bio International, France). Therefore, the errors due to different methods of measurement should be negligible.
Smokers were reported as having a slightly higher circulating CEA concentration than non-smokers (1). In our study, a CEA value of more than 5 ng/ml was considered abnormal. According to the manual for the CEA–RIA test from CIS Bio International, 100% of non-smokers and 95% of smokers fit the cut-off level of 5 ng/ml. Only 5% of healthy smokers showed CEA 5–7 ng/ml, which were slightly higher than the cut-off level of 5 ng/ml. However, the authors agree that smoking indeed has an influence on CEA and different cut-off values should preferably be used according to the patient’s smoking status.
The time of sampling is one of the preanalytical variations because it is related to the well-known CEA circadian rhythm (2). However, in the study conducted by Focan et al. (3), a circadian rhythm of CEA was demonstrated in individuals not suffering from cancer. CEA exhibited a diurnal peak with a nocturnal dip in healthy individuals but it disappeared in cancer groups (3).
We must admit that many clinicians and even some laboratory specialists are not aware of that preanalytical conditions can significantly affect all laboratory variables. Factors of variation, such as the sample transport or storage conditions before analysis, freezing and thawing of plasma or serum, therapies administered, etc., may cumulate in daily clinical practice (4). We agree with Watine and Friedberg that the quality of data in clinical trials is very important and really can be improved by well-organized collecting of information on methods of measurement. In the future, both the analytical and preanalytical methods should be clearly described in studies published in this field.
Wei-Shu Wang
Division of Medical Oncology, Department of Medicine, Veterans General Hospital–Taipei, Taipei, Taiwan
REFERENCES
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.
2 Touitou Y, Levi F, Bogdan A, Benavides M, Bailleul F, Misset JL. Rhythm alteration in patients with metastatic breast cancer and poor prognostic factors. J Cancer Res Clin Oncol 1995;121:181–8.[Web of Science][Medline]
3 Focan C, Focan-Henrard D, Collette J, Mecnkouri M, Levi F, Hrushesky W, et al. Cancer-associated alteration of circadian rhythms in carcinoembryonic antigen (CEA) and alpha-fetoprotein (AFO) in humans. Anticancer Res 1986;6:1137–44.[Web of Science][Medline]
4 Watine J, Miedouge M. Preanalytical variation of laboratory variables. Clin Chem Lab Med 2000;38:263.[Web of Science][Medline]
; .
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||