Ann Thorac Surg 2007;83:216-221
© 2007 The Society of Thoracic Surgeons
Original Articles: General Thoracic
Clinical Value of Serum Cytokeratin 19 Fragment and Sialyl-Lewis X in Non-Small Cell Lung Cancer
Shinjiro Mizuguchi, MD, PhDa,b,*,
Noritoshi Nishiyama, MD, PhDa,
Takashi Iwata, MD, PhDa,
Tatsuya Nishida, MD, PhDa,
Nobuhiro Izumi, MD, PhDa,
Takuma Tsukioka, MDa,
Kiyotoshi Inoue, MD, PhDa,b,
Masao Kameyama, MD, PhDb,
Shigefumi Suehiro, MD, PhDc
a Department of Thoracic Surgery, Osaka City University Hospital, Osaka, Japan
b Department of Thoracic Surgery, Bell Land General Hospital, Osaka, Japan
c Department of Cardiovascular Surgery, Osaka City University Graduate School of Medicine, Osaka, Japan
Accepted for publication August 23, 2006.
* Address correspondence to Dr Mizuguchi, Department of Thoracic Surgery, Osaka City University Medical School, 1-4-3 Asahimachi, Abeno-ku, Osaka 545-8585, Japan (Email: m1293795{at}msic.med.osaka-cu.ac.jp).
 |
Abstract
|
|---|
BACKGROUND: This study aimed to establish the clinical significance of preoperative serum cytokeratin 19 fragment (CYFRA21-1) and sialyl-Lewis x (SLex) as prognostic markers.
METHODS: The study involved 272 patients (181 male, 91 female; median age 69 years; range, 32 to 92) with non-small cell lung cancer (NSCLC) who underwent pulmonary resection with mediastinal lymph node dissection. Tumor markers carcinoembryonic antigen (CEA), squamous cell carcinoma antigen (SCC), CYFRA21-1, and SLex were examined.
RESULTS: A log-rank test revealed that age, gender, performance status, CEA, SCC, CYFRA21-1, and SLex were associated with the survival rate. By multivariate analysis, age, gender, performance status, CYFRA21-1 (risk ratio, 2.42) and SLex (risk ratio, 6.18) were independent prognostic factors. For patients positive for both markers, the relative risk was 6.10 compared with patients negative for both markers. The patients were divided into three groups: negative for both CYFRA21-1 and SLex (n = 97); positive for either marker (n = 136); and positive for both markers (n = 39). The 1-, 3-, and 5-year survival rates were the following: 98%, 82%, and 75% in the first group; 90%, 63%, and 49% in the second group; and 62%, 31%, and 25% in the third group (p < 0.001). Sixty-four percent of patients positive for both markers were histologic stage III/IV, and 68% of patients negative for both markers were stage I.
CONCLUSIONS: Serum CYFRA21-1 and SLex were prognostic markers for NSCLC. Their combination should contribute to the classification of NSCLC patients. Preoperative staging should be carefully performed in patients positive for both tumor markers.
 |
Introduction
|
|---|
It is generally accepted that primary lung cancer represents one of the most aggressive solid tumors, and its prognosis is still poor despite improvements in pulmonary surgery, chemotherapy, and radiation therapy. In previous reports, neoadjuvant therapy has been shown to improve the postoperative outcome of patients with advanced stage N2 disease [14]. Therefore, it is important to diagnose patients with advance disease correctly. In spite of careful preoperative staging, including the use of imaging studies (positron emission tomography [PET] or high-resolution computed tomography [CT]), mediastinoscopy, or both, metastases to mediastinal lymph nodes have been detected histologically in approximately 25% of patients diagnosed preoperatively with N2-negative non-small cell lung cancer (NSCLC) [5, 6]. On the other hand, tumor markers are regarded as conventional and less sensitive tools in the decision-making process for lung cancer and their value appears limited. None of the available possible tumor markers is actually considered when deciding treatment options.
Recently, cytokeratin 19 fragment (CYFRA21-1) was found to be significantly more sensitive than established markers like carcinoembryonic antigen (CEA) and squamous cell carcinoma-related antigen (SCC) in evaluating NSCLC. High serum concentrations of CFYRA21-1 are mainly related to tumor burden (T factor) [711], and indicate a poor prognosis [712].
On the other hand, adhesion molecules, such as carbohydrate antigens, including sialyl-Lewis x (SLex), have been used as markers for carcinomas from many organs, including the lung. In gastric and colon cancer, it has been reported that high serum concentrations of SLex are related to lymph node and distant metastases [1319]. Several authors have determined that increased expression of SLex or other carbohydrate antigens in malignant tissue, based on immunohistochemical analysis, correlates with tumor progression and poor outcome in patients with NSCLC [20, 21]. It has been reported that high serum concentrations of SLex predict a poor outcome associated with lymph node metastasis in NSCLC (N factor) [7, 22]. In this study, we hypothesize that CYFRA21-1 and SLex considered together could have a better prognostic value in NSCLC than each marker separately.
 |
Patients and Methods
|
|---|
Patients
From January 1998 to December 2003, 349 patients with NSCLC underwent pulmonary resection at our institution. The final collection of survival data was in December 2005. We excluded patients with any other cancers and those who underwent neoadjuvant therapy; thus, the subjects in the current study were 272 patients with NSCLC who underwent pulmonary resection with mediastinal lymph node dissection. The study was approved by the local Institution Review Board after informed consent was obtained from all patients.
Criteria of respectability were assessed by CT (chest, abdominal, and brain) scan, bronchoscopy, and bone scanning. Whenever a mediastinal lymph node could be detected (short-axis over 10 mm), mediastinal staging procedure (video-assisted thoracoscopy, fine-needle aspiration through bronchoscopy, or PET) was performed. Histologic classification was performed according to the World Health Organization histologic typing of lung tumors [23]. Postoperative staging of the patients was based on the international TNM classification for lung cancer [24]. Mediastinal lymph node dissection in this study involved radical en bloc mediastinal lymphadenectomy, according to previously described methods [25, 26].
Methods
The serum concentrations of CEA, CYFRA21-1, SCC, and SLex in the peripheral blood were measured retrospectively in the 272 patients between January 1998 and December 2003. Serum samples were collected just before surgery and were stored at 80°C until the time of analysis. We measured SLex once every month using a commercially available radioimmunoassay kit (Fh-6 "Otsuka"; Otsuka Assay Lab, Tokushima, Japan). For comparison, chemiluminescence immunoassays or immunoradiometric assays were used to measure conventional tumor markers (CEA, CYFRA21-1, and SCC). According to the manufacturers instructions, the diagnostic cut-off values for each marker were 38 U/mL for SLex, 6.5 ng/mL for CEA, 2.0 ng/mL for CYFRA21-1, and 1.5 ng/mL for SCC.
For correlations between SLex and CYFRA21-1, the Pearson correlation coefficient was used, with 0.4 to 0.4 considered as having no correlation. Data are reported as the median with 25th and 75th percentiles for marker concentrations. The Kaplan-Meier method and the log-rank test were used for survival analysis. For correlations between tumor markers and the histologic stage of disease, simple linear regression was used. Multivariate analysis was performed using a Cox regression model with forward stepwise selection. Multivariate risk ratios with 95% confidence intervals (CIs) were computed. A p value less than 0.05 was considered statistically significant. Statistical analysis was performed using the Stat View 5.0J software package (SAS Institute, Cary, NC).
 |
Results
|
|---|
Patient Characteristics and Tumor Marker Sensitivity
The 272 patients included 181 males and 91 females, with a median age of 69 years (range, 32 to 92 years). One operative death occurred. One hundred and sixty four patients had adenocarcinoma, 99 had squamous cell carcinoma, five had large cell carcinoma, and four had adenosquamous carcinoma. In clinical staging, 186 patients were in stage I, 48 in stage II, 35 in stage III, and three in stage IV (pulmonary metastases to the ipsilateral other lobe). In histologic staging, 140 patients were in stage I, 45 in stage II, 73 in stage III, and 14 in stage IV (pulmonary metastases to the other lobe). The median and interquartile ranges of serum CEA, CYFRA21-1, SCC, and SLex concentrations were 6.1 (3.5 to 11.1) ng/mL, 2.3 (1.3 to 4.4) ng/mL, 0.9 (0.5 to 1.7) ng/mL, and 28.5 (23.3 to 36.7) U/mL, respectively. In this study, the most sensitive marker in all patients was CYFRA21-1 (55%); the sensitivity of CEA was 47%, that of SCC was 28%, and that of SLex was 24%.
Multivariate Analysis of Tumor Markers and Clinical Functions
A log-rank test revealed that age, gender, performance status, CEA, CYFRA21-1, SCC, SLex, clinical stage, clinical T factor, and clinical N factor were associated with a significant survival rate (Table 1). By simple linear regressions, CEA (p = 0.042), CYFRA21-1 (p = 0.048), and SLex (p < 0.001) were significantly correlated with histologic stage of disease. We performed multivariate analysis, exclude imaging studies (clinical stage, T factor, and N factor), which often introduce subjective bias. By multivariate analysis, age, gender, performance status, CYFRA21-1 (risk ratio, 2.42; p = 0.015) and SLex (risk ratio, 6.18; p < 0.001) were independent factors associated with survival (Table 2). For patients positive for both CYFRA21-1 and SLex, the relative risk was 6.10 (p < 0.001) compared with patients who were negative for both markers.
Postoperative Outcome of Patients Based on the Combination of Serum CYFRA21-1 and SLex
No correlations between CYFRA21-1 and SLex were found (coefficient of correlation: SLex vs CYFRA, 0.084). Figure 1
shows the survival rates for patients with NSCLC, based on the combination of serum CYFRA21-1 and SLex. The mean follow-up period for the entire study population was 1,098 days (range, 32 to 2,894 days). The patients were divided into three groups with regard to the two tumor markers: negative for both (n = 97); positive for either (n = 136; 110 positive for CYFRA21-1 and 26 positive for SLex); and positive for both markers (n = 39). The 1-, 3-, and 5-year survival rates were the following: 98%, 82%, and 75% in the group negative for both markers; 90%, 63%, and 49% for the group positive for one of the markers; and 62%, 31%, and 25% for the group positive for both markers (p < 0.001). There was a significant difference between the two groups, and prognostic classification using the tumor markers was possible (negative for both markers versus positive for one marker, p < 0.001; and positive for both markers versus positive for one marker, p < 0.001). Moreover, Figure 2
shows the survival rates for patients with NSCLC, based on histologic stage. The 1-, 3-, and 5-year survival rates in patients with histologic stage I, stage II, and stage III/IV were the following: 97%, 82%, and 68%; 89, 64%, and 52%; and 74%, 37%, and 34%, respectively (p < 0.001). There was a significant difference between the two groups (stage I versus stage II, p = 0.006; and stage II versus stage III/IV, p = 0.023).
On the other hand, Figure 3
shows the survival rates for patients with NSCLC, based on clinical stage. The 1-, 3-, and 5-year survival rates in patients in stage I, stage II, and stage III/IV were the following: 92%, 70%, and 61%; 92%, 60%, and 43%; and 66%, 49%, and 42%, respectively (p < 0.001). According to prognostic classification, there was no significant difference between the two groups (stage I versus stage II, p = 0.051; and stage II versus stage III/IV, p = 0.178).
There was also a significant correlation between the combination of serum tumor markers (CYFRA21-1 and SLex) and postoperative stage (Table 3). Sixty-eight percent of patients negative for both tumor markers were histological stage I. On the other hand, 64% of patients positive for both tumor markers were histologic stage III/IV.
 |
Comment
|
|---|
Recently, CYFRA21-1 was found to be significantly more sensitive than established tumor markers (CEA and SCC) in predicting outcome. In the current study, the serum concentration of CYFRA21-1 had the highest sensitivity (55%) in all NSCLC patients, which has previously been reported as being 49% to 65% [1012, 27]. It is believed that high concentrations of CFYRA21-1 in serum are mainly related to tumor burden and are associated with poor outcome [812]. High concentrations of CYFRA21-1 were also related to poor outcome in our study (p = 0.001). According to combinations of tumor markers, some investigators have suggested the usefulness of tumor marker combinations in the clinical care of NSCLC patients, and most have selected CYFRA21-1. With the combination of CYFRA21-1 and CEA [8, 2830], CYFRA21-1 and tissue polypeptide specific antigen [3133] had a better prognostic value in NSCLC than each of the markers on their own. Moreover, the combination of more than three tumor markers has also been reported, with the combination of CYFRA21-1, CEA, and neuron-specific enolase [34], and CYFRA21-1, CEA, and cancer antigen 125 [28] being reported as useful.
In contrast, the sensitivity of SLex, an adhesion molecule-like carbohydrate antigen, was only 24% to 32% in a previous report [7, 35], and therefore does not appear acceptable as a screening tumor marker. Sialyl-Lewis x has been used as a marker for carcinomas from many organs [36], including the lungs. High serum concentrations of SLex are associated with tumor stage, lymph node metastasis, depth of invasion, liver metastasis, peritoneal dissemination, and distant metastasis in gastric cancer [13, 14], and also with tumor stage, liver metastasis, and early recurrence in colon cancer [1619]. In relation to lung cancer, several authors have demonstrated that high serum concentrations of SLex in advanced cancer [7, 22, 37], or increased immunohistochemical expression of SLex in tumor tissues [20, 21], are associated with poor outcome. We have reported that high serum concentrations of SLex are predictive of poor outcome associated with lymph node metastasis (N factor). Especially, the sensitivity of SLex for predicting multilevel N2 or N3 disease is extremely high (71%) [7]. Thus, it is reasonable to use the combination of CYFRA21-1 (associated T factor) and SLex (associated N factor), which were both independent prognostic factors in patients with NSCLC.
In this study, there was no significant difference based on clinical stage, which included the risk of underestimating lymph node or pulmonary metastases. Improvements in imaging techniques such as helical CT and PET have made possible the early detection of NSCLC. The preoperative staging of NSCLCs has also been improved by these imaging techniques. In spite of careful preoperative staging, including the use of imaging studies or a mediastinal biopsy, metastases to mediastinal lymph nodes were detected preoperatively in approximately 75% of patients with histologic N2 [5, 6]. Other markers, such as a biomolecular one, have been expected to aid preoperative diagnosis to reduce the risk of underestimations. There was also a significant correlation between the combination of serum tumor markers (CYFRA21-1 and SLex) and the histologic stage in the current study. We estimated clinical stage I in 54% of patients positive for both tumor markers. In terms of histologic stage, only 18% of patients positive for both tumor markers were histologic stage I. We consider that the underestimation in imaging studies or biomolecular markers, including tumor markers, was the reason for the discrepancy between clinical and histological staging. Moreover, 64% of patients positive for both tumor markers were histologic stage III/IV. The 5-year survival rates in patients positive for both tumor markers was only 25%; therefore, we propose that preoperative staging should be carefully performed in such patients.
In conclusion, serum CYFRA21-1 and SLex were prognostic markers in patients with NSCLC. Their combination should significantly contribute to the classification of NSCLC patients. We need to carefully perform preoperative staging and consider adjuvant and neoadjuvant therapies to improve prognosis in patients positive for both tumor markers.
 |
References
|
|---|
- Robinson LA, Wagner Jr H, Ruckdeschel JC. Treatment of stage IIIA non-small cell lung cancer Chest 2003;123:202-220.[Medline]
- De Leyn P, Vansteenkiste J, Deneffe G, Van Raemdonck D, Coosemans W, Lerut T. Result of induction chemotherapy followed by surgery in patients with stage IIIA N2 NSCLC: importance of pre-treatment mediastinoscopy Eur J Cardiothorac Surg 1999;15:608-614.[Abstract/Free Full Text]
- Friedel G, Hruska D, Budach W, et al. Neoadjuvant chemoradiotherapy of stage III non-small-cell lung cancer Lung Cancer 2000;30:175-185.[Medline]
- Trodella L, Granone P, Valente S, et al. Neoadjuvant concurrent radiochemotherapy in locally advanced (IIIA-IIIB) non-small-cell lung cancer: long-term results according to downstaging Ann Oncol 2004;15:389-398.[Abstract/Free Full Text]
- Fernando HC, Goldstraw P. The accuracy of clinical evaluative intrathoracic staging in lung cancer as assessed by postsurgical pathologic staging Cancer 1990;65:2503-2506.[Medline]
- Goldstraw P, Mannam GC, Kaplan DK, Michail P. Surgical management of non-small-cell lung cancer with ipsilateral mediastinal node metastasis (N2 disease) J Thorac Cardiovasc Surg 1994;107:19-27.[Abstract/Free Full Text]
- Mizuguchi S, Inoue K, Iwata T, et al. High serum concentrations of sialyl Lewis(x) predict multilevel N2 disease in non-small-cell lung cancer Ann Surg Oncol 2006;13:1010-1018.[Medline]
- Moro D, Villemain D, Vuillez JP, Delord CA, Brambilla C. CEA, CYFRA21-1 and SCC in non-small cell lung cancer Lung Cancer 1995;13:169-176.[Medline]
- Pujol JL, Molinier O, Ebert W, et al. CYFRA 21-1 is a prognostic determinant in non-small-cell lung cancer: results of a meta-analysis in 2063 patients Br J Cancer 2004;90:2097-2105.[Medline]
- Stieber P, Hasholzner U, Bodenmuller H, et al. CYFRA 21-1A new marker in lung cancer. Cancer 1993;72:707-713.[Medline]
- Sugama Y, Kitamura S, Kawai T, et al. Clinical usefulness of CYFRA assay in diagnosing lung cancer: measurement of serum cytokeratin fragment Jpn J Cancer Res 1994;85:1178-1184.
- Brechot JM, Chevret S, Nataf J, et al. Diagnostic and prognostic value of Cyfra21-1 compared with other tumour markers in patients with non-small cell lung cancer: a prospective study of 116 patients Eur J Cancer 1997;33:385-391.[Medline]
- Imada T, Rino Y, Takahashi M, et al. Serum CA 19-9, SLX, STN and CEA levels of the peripheral and the draining venous blood in gastric cancer Hepatogastroenterology 1999;46:2086-2090.[Medline]
- Nakagoe T, Sawai T, Tsuji T, et al. Predictive factors for preoperative serum levels of sialy Lewis(x), sialyl Lewis(a) and sialyl Tn antigens in gastric cancer patients Anticancer Res 2002;22:451-458.[Medline]
- Ikeda Y, Mori M, Kajiyama K, Haraguchi Y, Sasaki O, Sugimachi K. Immunohistochemical expression of sialyl Tn, sialyl Lewis a, sialyl Lewis a-b-, and sialyl Lewis x in primary tumor and metastatic lymph nodes in human gastric cancer J Surg Oncol 1996;62:171-176.[Medline]
- Nakagoe T, Kusano H, Hirota M, et al. Serological and immunohistochemical studies on sialylated carbohydrate antigens in colorectal carcinoma Gastroenterol Jpn 1991;26:303-311.[Medline]
- Nakagoe T, Fukushima K, Hirota M, et al. Immunohistochemical expression of sialyl Lex antigen in relation to survival of patients with colorectal carcinoma Cancer 1993;72:2323-2330.[Medline]
- Tanaka K, Togo S, Nanko M, et al. Sialyl Lewis X expression in vascular permeating lesions as a factor for predicting colorectal cancer metastasis Hepatogastroenterology 1999;46:875-882.[Medline]
- Yamaguchi A, Goi T, Seki K, et al. Clinical significance of combined immunohistochemical detection of CD44v and sialyl LeX expression for colorectal cancer patients undergoing curative resection Oncology 1998;55:400-403.[Medline]
- Ogawa J, Tsurumi T, Yamada S, Koide S, Shohtsu A. Blood vessel invasion and expression of sialyl Lewisx and proliferating cell nuclear antigen in stage I non-small cell lung cancerRelation to postoperative recurrence. Cancer 1994;73:1177-1183.[Medline]
- Tsumatori G, Ozeki Y, Takagi K, Ogata T, Tanaka S. Relation between the serum E-selectin level and the survival rate of patients with resected non-small cell lung cancers Jpn J Cancer Res 1999;90:301-307.
- Satoh H, Ishikawa H, Yamashita YT, et al. Predictive value of preoperative serum sialyl Lewis X-i antigen levels in non-small cell lung cancer Anticancer Res 1998;18:2865-2868.[Medline]
- WHO Histological typing of lung and pleural tumoursIn: Travis WD, Colby TV, Corrin B, Shimosato Y, Brambilla E, editors. In collaboration with pathologists in 14 countries. 3rd edition. Geneva: World Heath Organization; 1999.
- Mountain CF. Revisions in the international system for staging lung cancer Chest 1997;111:1710-1717.[Medline]
- Naruke T, Suemasu K, Ishikawa S. Surgical treatment for lung cancer with metastasis to mediastinal lymph nodes J Thorac Cardiovasc Surg 1976;71:279-285.[Abstract]
- Martini N, Flehinger BJ. The role of surgery in N2 lung cancer Surg Clin North Am 1987;67:1037-1049.[Medline]
- Takada M, Masuda N, Matsuura E, et al. Measurement of cytokeratin 19 fragments as a marker of lung cancer by CYFRA 21-1 enzyme immunoassay Br J Cancer 1995;71:160-165.[Medline]
- Ando S, Kimura H, Iwai N, Shima M, Ando M, Kuriyama T. Optimal combination of seven tumour markers in prediction of advanced stage at first examination of patients with non-small cell lung cancer Anticancer Res 2001;21:3085-3092.[Medline]
- Koga H, Eguchi K, Shinkai T, et al. Preliminary evaluation of the new tumor marker, CYFRA 21-1, in lung cancer patients Jpn J Clin Oncol 1994;24:263-268.[Abstract/Free Full Text]
- Huang MS, Jong SB, Tsai MS, et al. Comparison of cytokeratin fragment 19 (CYFRA 21-1), tissue polypeptide antigen (TPA) and carcinoembryonic antigen (CEA) as tumour markers in bronchogenic carcinoma Respir Med 1997;91:135-142.[Medline]
- Nisman B, Lafair J, Heching N, et al. Evaluation of tissue polypeptide specific antigen, CYFRA 21-1, and carcinoembryonic antigen in nonsmall cell lung carcinoma: does the combined use of cytokeratin markers give any additional information? Cancer 1998;82:1850-1859.[Medline]
- Kulpa J, Wojcik E, Radkowski A, Kolodziejski L, Stasik Z. CYFRA 21-1, TPA-M, TPS, SCC-Ag and CEA in patients with squamous cell lung cancer and in chemical industry workers as a reference group Anticancer Res 2000;20:5035-5040.[Medline]
- Plebani M, Basso D, Navaglia F, De Paoli M, Tommasini A, Cipriani A. Clinical evaluation of seven tumour markers in lung cancer diagnosis: can any combination improve the results? Br J Cancer 1995;72:170-173.[Medline]
- Barlesi F, Gimenez C, Torre JP, et al. Prognostic value of combination of Cyfra 21-1, CEA and NSE in patients with advanced non-small cell lung cancer Respir Med 2004;98:357-362.[Medline]
- Oyama T, Kawamoto T, Matsuno K, et al. A case-case study comparing the usefulness of serum trace elements (Cu, Zn and Se) and tumor markers (CEA, SCC and SLX) in non-small cell lung cancer patients Anticancer Res 2003;23:605-612.[Medline]
- Nishida K, Yamamoto H, Ohtsuki T, et al. Elevated tissue concentrations of sialyl Lex-i in cancerous tissues compared with those in noncancerous tissues of various organs Cancer 1991;68:111-117.[Medline]
- Kawai T, Suzuki M, Kase K, Ozeki Y. Expression of carbohydrate antigens in human pulmonary adenocarcinoma Cancer 1993;72:1581-1587.[Medline]
Related Article
-
Invited commentary
- Thomas DAmico
Ann. Thorac. Surg. 2007 83: 221-222.
[Extract]
[Full Text]
[PDF]
This article has been cited by other articles:

|
 |

|
 |
 
H.-J. Haussmann
Smoking and Lung Cancer: Future Research Directions
International Journal of Toxicology,
July 1, 2007;
26(4):
353 - 364.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
T. D'Amico
Invited commentary
Ann. Thorac. Surg.,
January 1, 2007;
83(1):
221 - 222.
[Full Text]
[PDF]
|
 |
|