Interact CardioVasc Thorac Surg 2009;9:199-202. doi:10.1510/icvts.2009.206698
© 2009 European Association of Cardio-Thoracic Surgery
Institutional report - Thoracic oncologic |
Clinical significance of preoperative carcinoembryonic antigen level for clinical stage I non-small cell lung cancer: can preoperative carcinoembryonic antigen level predict pathological stage?
Riken Kawachi*,
Yohko Nakazato,
Hidefumi Takei,
Yoshihiko Koshi-ishi and
Tomoyuki Goya
Thoracic Surgery Division, Kyorin University Hospital, 6-20-2 Shinkawa, Mitaka-shi, Tokyo 181-0004, Japan
Received 6 March 2009;
received in revised form 3 May 2009;
accepted 5 May 2009
*Corresponding author. Tel.: +81-3-3542-2511; fax: +81-3-3542-3815.
E-mail address: rkawachi{at}kpe.biglobe.ne.jp (R. Kawachi).
 |
Abstract
|
|---|
The purpose of the present study was to retrospectively analyze the clinicopathological characteristics and clarify whether or not the preoperative carcinoembryonic antigen (CEA) level could be used as a decision-making factor as an adjunct to the TNM staging system in patients with clinical stage I non-small cell lung cancer (NSCLC). Between 1993 and 2006, 815 patients who had clinical stage I NSCLC were analyzed retrospectively. The CEA level was defined as being either normal (CEA 5 ng/ml), high (5<CEA30 ng/ml) and very high (CEA>30 ng/ml) sub-groups. The rate of patients with an elevated CEA level was 33.6%. The five-year disease-free survival rates for patients with normal, high and very high CEA levels were 76.7, 60.0 and 31.3%, respectively. The survival curve for patients with a normal CEA level almost overlapped that for p-stage I, that for a high CEA level nearly overlapped that for p-stage II, and that for a very high CEA level nearly overlapped that for p-stage III. The present study demonstrated that the preoperative CEA level was a very good predictor of the pathological stage. These findings suggest that the preoperative CEA level may be useful as an adjunct to the TNM staging system.
Key Words: Lung cancer; Carcinoembryonic antigen; TNM classification
|
1. Introduction
|
|---|
The carcinoembryonic antigen (CEA) level has been reported to be a prognostic factor and an indicator of recurrence after surgical resection for non-small cell lung cancer (NSCLC) [1–3]. However, several guidelines for preoperative workup have not included measurement of the CEA level [4], and the preoperative CEA level has not been included in the present TNM classification [5, 6]. Consequently, the CEA level is not routinely measured in Europe and the USA. Even if the preoperative CEA level is high, preoperative staging will usually be performed based only on radiological procedures. We hypothesized that the CEA level may correspond to the extent of disease progression. This hypothesis led us to conduct a retrospective study aimed at investigating whether the preoperative CEA level can predict the pathological stage in patients with clinical stage I disease and to identify whether the CEA level can be used as an adjunct to the TNM staging system.
|
2. Patients and methods
|
|---|
2.1. Patients
From May 1993 to December 2007, 1197 patients underwent surgery for primary lung cancer at Kyorin University Hospital, Tokyo, Japan. Among them, the CEA level was examined preoperatively and clinical stage I NSCLC was observed in 815 patients. Preoperative staging routinely included chest X-ray (CXR) and chest and abdominal computed tomography (CT). Positron emission tomography (PET) was performed only when further examination was required. All patients were staged according to the International Union Against Cancer TNM classification system [5]. The histology of the tumor was described according to the World Health Organization classification [7]. Postoperative follow-up consisted of CXR every 3 months and chest and abdominal CT every 6 months for 5 years after surgical resection. In symptomatic patients, additional procedures, such as MRI, bone scintigraphy, or PET, were performed. The serum CEA concentration was measured using the chemiluminescent enzyme immunoassay technique.
2.2. Grouping
A preoperative CEA level of >5 ng/ml was considered to be elevated. In the survival analysis, the CEA level was classified as either normal (CEA5 ng/ml) or elevated (CEA>5 ng/ml), and the latter was further classified into high (5<CEA30 ng/ml) and very high (CEA>30 ng/ml) sub-groups. These groups and pathological stages were compared with regard to survival.
2.3. Statistical analysis
Survival was calculated using the Kaplan–Meier method, and differences in survival were determined by a log-rank analysis. Multivariate analysis was performed using the Cox proportional hazard model. P<0.05 were considered statistically significant.
|
3. Results
|
|---|
3.1. Clinicopathological findings
The clinical characteristics according to the preoperative CEA level are shown in Table 1. The incidence of an elevated CEA level (5 ng/ml) was 33.6%. Patients with an elevated CEA level were more common among those in an advanced status: vascular invasion, lymphatic permeation, pulmonary metastasis, and higher stages.
3.2. Survival
The median follow-up time for patients was 65.1 months. The 5-year disease-free survival rates were 70.2, 76.7, and 56.6% for the overall, normal, and elevated groups, respectively (Fig. 1a). The 5-year disease-free survival rates were 84.5% for the normal CEA group and 72.7% for the elevated CEA group among patients with pathological stage I (Fig. 1b). In clinical and pathological stage I, a difference was observed between patients with normal CEA and elevated CEA. The results of a univariate survival analysis are shown in Table 2.
View larger version (11K):
[in this window]
[in a new window]
|
Fig. 1. Disease-free survival curves for clinical stage I (a) and for pathological stage I (b) according to the carcinoembryonic antigen (CEA) level.
|
|
3.3. Multivariate analysis
As a result of the multivariate analysis for disease-free survival, smoking history, CEA, clinical T factor, histologic type, pleural invasion, lymphatic permeation were independent prognostic factors (Table 3).
3.4. Comparison of pathological stage and CEA level according to survival
The 5-year disease-free survival rates for the high and very high CEA groups were 60.0 and 31.3%, respectively (Fig. 2). The survival curve for pathological stage I was almost the same as that for the normal CEA group (P=0.11) (Fig. 3a), and the survival curve for pathological stage II overlapped that of the high CEA group (P=0.14) (Fig. 3b). Furthermore, the survival curve for p-stage III was superimposed on that of the very high CEA group (P=0.50) (Fig. 3c).
View larger version (14K):
[in this window]
[in a new window]
|
Fig. 3. Disease-free survival curves for pathological stage I and normal CEA group (a), pathological stage II and high CEA group (b), and for pathological stage III and very high CEA group (c).
|
|
|
4. Comment
|
|---|
The preoperative CEA level has been shown to be an independent prognostic factor in patients who are surgically treated for clinical stage I NSCLC [1–3]. As in previous reports, our univariate and multivariate analyses showed that the preoperative CEA level was correlated with the prognosis. Moreover, the present study demonstrated the combination of preoperative CEA level and clinical TNM staging could predict more exact survival. The survival curves for every CEA level group corresponded to those for every pathological group in clinical stage I: the survival curve for the normal CEA group was almost the same as that for p-stage I, that for the high CEA group (5<CEA level30 ng/ml) was similar to that for p-stage II, and that for the very high CEA group (CEA>30 ng/ml) matched that for p-stage III. These findings indicate that a high CEA level suggests N1 disease and a very high CEA level suggests N2 disease. Thus, the preoperative CEA level can predict latent lymph node metastases that cannot be detected by radio logical imaging such as CT. Inoue et al. reported that CEA could be used to predict subclinical nodal involvement [1], which was expected to be one of the causes of the poor prognosis in patients with a high CEA level. Therefore, patients with a very high CEA level should receive special attention and careful staging, including mediastinoscopy or PET.
Several problems that have been already discussed should be emphasized regarding CEA when biologic marker would be adopted in staging system. First, different measuring kits have been reported to give different CEA levels [8]. Second, does CEA reflect any specific histologic type of tumor in NSCLC? Most recent reports have noted that the relationship between the CEA level and histologic type remained obscure [1–3]. In the present study, significant differences were found between the histologic types with regard to survival, and histologic type was not shown to be a prognostic factor. There were few patients with a histologic type other than adenocarcinoma and this issue remains unclear. Third, several investigators have reported that cigarette smoking could be related to an elevated CEA level [9]. Thus, this point must be clarified for inclusion in the TNM staging system.
The present study demonstrated that the preoperative CEA level could be a kind of biological staging model. A biological staging model using molecular biology was proposed, and the staging model could predict prognosis well [10]. However, the 7th edition of TNM staging system has already been proposed, and biologic staging has not been adopted [6]. The survival curves for every CEA level group corresponded to those for every pathological group, and the preoperative CEA level should be used as an adjunct to the TNM staging system in patients with clinical stage I NSCLC. Even so, a CEA level by itself cannot be used to determine the need for surgical intervention and further examination including PET scan and mediastinoscopy should be required. Moreover, some treatment other than surgical resection could be a therapeutic option in such patients who have a poor performance status or complications.
|
References
|
|---|
- Inoue M, Minami M, Shiono H, Sawabata N, Ideguchi K, Okumura M. Clinicopathologic study of resected, peripheral, small-sized, non-small cell lung cancer tumors of 2 cm or less in diameter: pleural invasion and increase of serum carcinoembryonic antigen level as predictors of nodal involvement. J Thorac Cardiovasc Surg 2006;131:988–993.[Abstract/Free Full Text]
- Matsuoka K, Sumitomo S, Nakashima N, Nakajima D, Misaki N. Prognostic value of carcinoembryonic antigen and CYFRA21-1 in patients with pathological stage I non-small cell lung cancer. Eur J Cardiothorac Surg 2007;32:435–439.[Abstract/Free Full Text]
- Matsuguma H, Nakahara R, Igarashi S, Ishikawa Y, Suzuki H, Miyazawa N, Honjo S, Yokoi K. Pathologic stage I non-small cell lung cancer with high levels of preoperative serum carcinoembryonic antigen: clinicopathologic characteristics and prognosis. J Thorac Cardiovasc Surg 2008;135:44–49.[Abstract/Free Full Text]
- Scott WJ, Howington J, Feigenberg S, Movsas B, Pisters K, American College of Chest Physicians. Treatment of non-small cell lung cancer stage I and stage II: ACCP evidence-based clinical practice guidelines (2nd edition). Chest 2007;132(3 Suppl):234S–242S.[Abstract/Free Full Text]
- International Union Against Cancer, Lung tumours. In Sobin LH, Wittekind CH, TNM classification of malignant tumours, 6th edition. New York: Wiley-Liss; 2002.
- Rami-Porta R, Crowley JJ, Goldstraw P. The revised TNM staging system for lung cancer. Ann Thorac Cardiovasc Surg 2009;15:4–9.[Medline]
- Travis WD, Colby TV, Corrin B, Shimosato Y, Brambilla E. Histological typing of lung and pleural tumors, World Health Organization International Histological Classification of Tumors. Berlin: Springer; 1999.
- Sawabata N. Discrepancy in serum carcinoembryonic antigen levels by measuring kits. Ann Thorac Surg 2003;75:973–980.[Abstract/Free Full Text]
- Stevens DP, Mackay IR. Increased carcinoembryonic antigen in heavy cigarette smokers. Lancet 1973;2:1238–1239.[Medline]
- Cox G, Jones JL, Andi A, Waller DA, O'Byrne KJ. A biological staging model for operable non-small cell lung cancer. Thorax 2001;56:561–566.[Abstract/Free Full Text]
|
|
Top
Abstract
1. Introduction