(Chest. 2004;126:342-346.)
© 2004
American College of Chest Physicians
Chest CT and Serum Vascular Endothelial Growth Factor-C Level To Diagnose Lymph Node Metastasis in Patients With Primary Non-small Cell Lung Cancer*
Masaya Tamura, MD;
Makoto Oda, MD;
Yoshio Tsunezuka, MD;
Isao Matsumoto, MD;
Kazuyuki Kawakami, MD;
Yasuhiko Ohta, MD and
Go Watanabe, MD
* From the Department of General and Cardiothoracic Surgery (Drs. Tamura, Oda, Tsunezuka, Matsumoto, Kawakami, and Watanabe), Kanazawa University School of Medicine, Kanazawa; and Department of Thoracic Surgery (Dr. Ohta), Ishikawa Prefectural Central Hospital, Kanazawa, Japan.
Correspondence to: Masaya Tamura, MD, Department of General and Cardiothoracic Surgery, Kanazawa University School of Medicine, Takara-machi 13-1, Kanazawa, 920-8640, Japan; e-mail: m-tamura{at}sf.m.kanazawa-u.ac.jp
 |
Abstract
|
|---|
Study objective: Accurate tumor staging is essential for choosing the appropriate treatment strategy for lung cancer. CT of the chest is the most commonly used noninvasive staging method of the lymph node metastasis, but it is far from satisfying. We evaluated whether circulating vascular endothelial growth factor (VEGF)-C could give additional information for diagnosing lymph node metastasis in patients with lung cancer.
Patients and method: Serum samples were obtained from 116 patients with primary non-small cell lung cancer (NSCLC). All patients underwent preoperative CT of the thorax. Clinical T and N stages were compared to the final T and N stages obtained from pathologic findings. Serum VEGF-C concentration was assayed by commercially available sandwich enzyme-linked immunosorbent assay. We evaluated the utility of serum VEGF-C level as a marker for nodal metastasis comparing the utility of CT.
Results: Preoperative and final T categories completely agreed in 82.8%. Regarding nodal metastasis, the accuracy of CT was 68.1%. Patients with lymph node metastasis showed higher serum VEGF-C concentrations than those without lymph node metastasis (p = 0.0007). Serum VEGF-C reached the highest sensitivity and specificity in diagnosing lymph node metastasis when a cut-off value of 1,850.6 pg/mL was applied (sensitivity, 70.0%; specificity, 77.3%). Serum VEGF-C visually correlated with CT scan in the detection of lymph node metastasis (sensitivity, 74.0%; specificity, 80.3%; positive predictive value, 74.0%; negative predictive value, 80.3%; accuracy, 77.6%). When the cases were limited to adenocarcinoma, better results could be obtained.
Conclusions: Serum VEGF-C is a reliable marker for lymph node metastasis in NSCLC. Serum VEGF-C evaluation and CT examination are complementary to each other for accurate lymph node staging in NSCLC.
Key Words: enzyme-linked-immunosorbent assay • lymph node • metastasis • non-small cell lung cancer • vascular endothelial growth factor-C
|
Introduction
|
|---|
Lung cancer is a common disease with a dismal prognosis.1 Even after a complete resection, over two thirds will die of a relapse within 5 years. The first important decision concerns whether or not to operate, in order to avoid surgical risks without benefit. The second question concerns whether or not to provide neoadjuvant therapy. Both questions must be answered relying on preoperative staging, presently based on the TNM classification. The only CT imaging criteria for tumor involvement are morphologic; that is, the criteria rely on the size and shape of the lymph nodes. We have previously reported that vascular endothelial growth factor (VEGF)-C expression in non-small cell lung cancer (NSCLC) was significantly associated with lymph node metastasis, lymphatic vessel invasion and,2 furthermore, nodal microdissemination.3 In this study, we compared the relative accuracies of CT and serum VEGF-C level for evaluating regional lymph nodes in patients with NSCLC, and evaluated whether circulating VEGF-C could give additional information for discriminating between the absence and presence of lymph node metastasis in patients with lung cancer.
|
Materials and Methods
|
|---|
Patients
Serum samples were obtained from 116 patients with primary NSCLC who received curative operations with routine systematic nodal dissection of both the hilar and the mediastinal lymph nodes in Kanazawa University Hospital and Ishikawa Prefectural Central Hospital between April 2002 and March 2003. The pathologic stage was classified according to the Japanese Lung Cancer Society classification. The 81 men and 35 women had a median age of 69 years (range, 40 to 82 years). The pathologic types were 77 adenocarcinomas, 29 squamous cell carcinomas, 6 adenosquamous carcinomas, and 4 large cell carcinomas. Patients with malignant pleuritis or a medical condition contraindicating thoracotomy were excluded.
Methods
All patients underwent preoperative chest radiography and CT of the thorax. Distant metastasis was ruled out preoperatively by brain MRI, radionuclide bone scan, and abdominal ultrasonography. Native and IV contrast bolus-enhanced CT scans were routine. The mediastinum was scanned at 5- to 8-mm intervals. The nodal stage was determined by CT, and the mediastinal lymph node levels with enlarged nodes were noted. Mediastinal lymph nodes 
1.0 cm at their minimum cross-sectional diameter were considered to be metastatic. The median delay between CT and surgery was 36 days (range, 5 to 97 days). Clinical T and N stages were compared to the final pathologic T and N stages. Blood samples were drawn by venous puncture preoperatively, and divided into tubes without anticoagulant for serum. Within an hour of collection, blood samples were centrifuged at 2,000 revolutions per minute for 10 min, and aliquots were frozen at – 80°C for later analysis. Informed consent was obtained from all of the patients. The VEGF-C was assayed by commercially available sandwich enzyme-linked immunosorbent assay (code No. 17741; IBL; Fujioka; Gunma, Japan). The limits of sensitivity of the VEGF-C assay were 46.9 pg/mL. The coefficient of variation was < 5.0%.
Statistical Analysis
Because of their skewed distribution, median value (range) is given to describe VEGF-C level. Peripheral levels of angiogenic factors between different groups were compared using the Mann-Whitney U test. To calculate sensitivity and specificity of peripheral levels of angiogenic factors at serial cut-off points, a receiver operating characteristics curve was plotted using a statistical software package (version 4; SAS Institute; Cary, NC).
|
Results
|
|---|
The correlation between the clinicopathologic findings and serum VEGF-C levels were shown in Table 1
. There was no association between serum VEGF-C level and patient gender, lung and lobe involved, histology, and pathologic T factor. As to preoperative lymph node staging, the distribution of the results of CT scan, serum VEGF-C, and serum VEGF-C correlated with CT scan, as shown in Table 2
. Enlarged lymph nodes on CT were found in 45 patients; only 29 of them had lymph node metastasis at pathologic diagnosis. Seventy-one patients did not have enlarged lymph nodes; unexpected metastasis was found in 21 of them. The median serum VEGF-C concentration (25 to 75% quartile) was 1,768.2 pg/mL (1,180.2 to 2,782.9 pg/mL) in patients with lung cancer. The median serum VEGF-C concentration gradually increased in correspondence with pathologic T stage categories, but a statistically significant difference could not be detected: T1, 1,453.1 pg/mL (1,098.9 to 2,133.9 pg/mL); T2, 1,780.2 pg/mL (1,255.3 to 2209.4 pg/mL); T3, 1,820.2 pg/mL (1208.3 to 2387.3 pg/mL) [p = 0.337]. Patients with lymph node metastasis revealed higher serum VEGF-C concentrations than those without: 1,465.5 pg/mL (1,110.5 to 1,903.5 pg/mL) vs 2009.2 pg/mL (1,100.5 to 2,987.0 pg/mL) [p = 0.0007; Fig 1 ]. No statistical difference was found between N1 and N2 category: 1,920.4 pg/mL (1,214.5 to 2350.9 pg/mL) vs 2,041.2 pg/mL (1,675.0 to 2476.5 pg/mL), respectively (p = 0.6143). Serum VEGF-C reached the highest sensitivity and specificity in diagnosing lymph node metastasis when a cut-off value of 1,850.6 pg/mL was applied.
View larger version (14K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1.. Boxplot-presentation of serum VEGF-C levels in patients with lung cancer according to lymph node metastasis. Patients with lymph node metastasis (Meta) revealed higher serum VEGF-C concentrations than those without.
|
|
As to the diagnosis of lymph node metastasis using serum VEGF-C, there were 50 cases positive for lymph nodes metastasis, and this was correct in 35 cases. In 66 cases, there was no suspicion of lymph nodes metastasis on serum VEGF-C levels; 15 of these cases had lymph nodes metastasis at pathologic diagnosis (Table 2). All the cases were classified into four groups by the serum VEGF-C levels, ie, < 25% quartile (group 1), 25 to 50% quartile (group 2), 50 to 75% quartile (group 3), and > 75% quartile (group 4). In preoperative staging, negative lymph nodes were diagnosed in all patients in group 1, and positive lymph nodes were diagnosed in all patients in group 4. The patients in group 2 were deemed to have positive lymph nodes if the CT showed the enlarged lymph nodes. However, patients in group 3 had negative lymph nodes if the lymph nodes were not swollen on the CT. Diagnostic values of the different examinations are given in Table 2 and Figure 2
. Serum VEGF-C visually correlated with CT scan in the detection of lymph node metastasis, the sensitivity was 74.0% (37 of 50 cases), the specificity was 80.3% (53 of 66 cases), the positive predictive value (PPV) was 74.0% (37 of 50 cases), the negative predictive value was 80.3% (53 of 66 cases), and the accuracy was 77.6% (90 of 116 cases). When the patients were limited to the adenocarcinoma, better results could be obtained: 77.8%, 82.9%, 80.0%, 81.0%, and 80.5%, respectively.
|
Comment
|
|---|
The involvement of lymph nodes metastasis is a very important prognostic factor in patients with potentially resectable NSCLC. CT of the chest is the most commonly used noninvasive staging method of lymph node metastasis, but it is far from satisfying and less accurate than mediastinoscopy. Regarding the presence or absence of lymph node metastasis, the accuracy of CT was 68.1%. These data were in agreement with the 51.4 to 83.0% reported previously.4567 Surgical techniques like mediastinoscopy are widely regarded as the most useful methods for mediastinal staging.89 Several reports1011 have described fluorodeoxyglucose positron emission tomography (PET) as advantageous for diagnosing the nodal staging of NSCLC. A noninvasive, accurate and accessible technique for nodal staging is urgently needed because mediastinoscopy is invasive and PET is performed only at limited numbers of facilities. Evaluating serum VEGF-C concentrations would be accessible to hospitals where there is no access to PET scanning; furthermore, it would be a noninvasive and inexpensive examination.
Although not statistically significant, serum VEGF-C levels correlated with T status (Table 1). The important question is whether increased VEGF-C expression is correlated with tumor enlargement or lymph node involvement. To answer this question, we compared serum VEGF-C levels of the patients with pathologic T1N0 stage (n = 41) and pathologic T2N0 stage (n = 24), and no statistically significant differences could be found between them (p = 0.256). However, a statistically significant difference could be found between the patients in pathologic T1N0 stage (n = 41) and those in pathologic T1N1 or T1N2 stages (n = 36) [p = 0.045]. We can say that the high expression of serum VEGF-C represents lymph node involvement more precisely than tumor enlargement.
In general, metastasis of normal-sized regional lymph nodes is revealed as adenocarcinoma by histologic examination. However, nodal enlargement can be due to reactive hyperplasia or other nonmalignant conditions in squamous cell cancer.11 In this study, 12 of 18 cases (66.7%) diagnosed as node positive by CT and as negative by VEGF-C revealed no lymph nodes metastasis pathologically. Ten of these 12 cases were squamous cell carcinoma. Histologic typing of primary tumor should be considered because most of these false-positive cases were reactive hyperplasia. In the beginning, this study was performed for a larger patient group including various histologic types. A predictive value of CT in diagnosing lymph node suffers biases by their histology, so we reanalyzed the data on selected category of patients. A better PPV (80%) could be obtained when limited to patients with adenocarcinoma, compared to all cases (74%). This may be due to the high sensitivity of serum VEGF-C in adenocarcinoma. Among the 21 patients with node-negative CT and node-positive VEGF-C diagnoses, 16 patients (76.2%) received a node-positive pathologic diagnosis. Furthermore, all of these 16 cases were adenocarcinoma.
All the patients were classified into four groups by the serum VEGF-C levels (see "Results"). Two patients in group 2 (n = 29; 25 to 50% quartile) revealed enlarged lymph nodes by CT, and 2 patients in group 3 (n = 29; 50 to 75% quartile) revealed normal-sized lymph nodes. The predictive value of preoperative lymph node metastasis improved when applying the CT findings to the patients in group 2 and group 3. In group 1 patients (n = 29), 24 patients revealed no lymph node swelling on CT, and 91.7% (22 of 24 patients) received a node-negative pathologic diagnosis. In group 4 patients (n = 29), 20 patients showed enlarged lymph nodes, and 85.0% (17 of 20 patients) received a node- positive diagnosis.
A combination diagnosis by serum VEGF-C and CT is relatively accurate, but there are false-positive as well as false-negative cases. It may be dangerous to start induction chemotherapy solely relying on CT and VEGF-C levels. Invasive staging such as mediastinoscopy cannot be omitted, but we believe that the presented combination diagnosis provides helpful information to select patients who need or do not need mediastinoscopy. In the future, in order to examine the diagnostic value of combination assay using other markers, a more detailed study is required.
In conclusion, our results show the value of serum VEGF-C level as a marker for nodal metastasis compared to use of CT. Evaluating serum VEGF-C concentrations in patients with primary NSCLC preoperatively is noninvasive, easily performed, and inexpensive. Combination diagnosis of CT and serum VEGF-C evaluation is a more reliable marker for lymph node metastasis in NSCLC compared to a single diagnosis by either CT or serum VEGF-C alone.
|
Footnotes
|
|---|
Abbreviations: NSCLC = non-small cell lung cancer; PET = positron emission tomography; PPV = positive predictive value; VEGF = vascular endothelial growth factor
Received for publication August 19, 2003.
Accepted for publication April 28, 2004.
|
References
|
|---|
- Thonos, S, Ritva, J, Pekka, K, et al (2003) Comparison of computed tomography and systematic lymph node dissection in determining TNM and stage in non-small cell lung cancer. Eur J Cardiothorac Surg 23,403-408[Abstract/Free Full Text]
- Kajita, T, Ohta, Y, Kimura, K, et al The expression of vascular endothelial growth factor C and its receptors in non-small cell lung cancer. Br J Cancer 2001;20,255-260
- Ohta, Y, Nozawa, H, Tanaka, Y, et al Increased vascular endothelial growth factor and vascular endothelial growth factor-c and decreased nm23 expression associated with microdissemination in the lymph nodes in stage I non-small cell lung cancer. J Thorac Cardiovasc Surg 2000;119,804-813[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[CrossRef][ISI][Medline]
- Izbicki, J, Thetter, O, Karg, O, et al Accuracy of computed tomographic scan and surgical assessment for staging of bronchial carcinoma: a prospective study. J Thorac Cardiovasc Surg 1992;104,413-420[Abstract]
- Gdeedo, A, Van Schil, P, Corthouts, B, et al Comparison of imaging TNM and pathological TNM in staging bronchogenic carcinoma. Eur J Cardiothorac Surg 1997;12,224-227[Abstract]
- Daly, BDT, Mueller, JD, Faling, JL, et al N2 lung cancer: outcome in patients with false-negative computed tomographic scans of the chest. J Thorac Cardiovasc Surg 1993;105,904-911[Abstract]
- Carlens, E Mediastinotomy: a method for inspection and tissue biopsy in the superior mediastinum. Dis Chest 1959;36,343-347[ISI][Medline]
- McNeil, TM, Chamberlain, JM Diagnostic anterior mediastinotomy. Ann Thorac Surg 1966;2,532-539[Medline]
- Kernstine, KH, McLaughlin, KA, Menda, RY, et al Can FDG-PET reduce the need for mediastinoscopy in potentially resectable non-small cell lung cancer? Ann Thorac Surg 2002;73,394-402[Abstract/Free Full Text]
- Hans, CS, Markus, H, Florin, A, et al Non-small cell lung cancer: nodal staging with FDG-PET versus CT with correlative lymph node mapping and sampling. Radiology 1997;202,441-446[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
M. Tamura, M. Oda, I. Matsumoto, Y. Tsunezuka, K. Kawakami, Y. Ohta, and G. Watanabe
The Combination Assay with Circulating Vascular Endothelial Growth Factor (VEGF)-C, Matrix Metalloproteinase-9, and VEGF for Diagnosing Lymph Node Metastasis in Patients With Non-Small Cell Lung Cancer
Ann. Surg. Oncol.,
October 1, 2004;
11(10):
928 - 933.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
Abstract
Introduction
