Clinical Characterization of Node-Negative Lung Adenocarcinoma: Results of a Prospective Investigation

ORIGINAL ARTICLE

Clinical Characterization of Node-Negative Lung Adenocarcinoma: Results of a Prospective Investigation Ichiro Yoshino, MD, PhD, Yukito Ichinose, MD, PhD, Akira Nagashima, MD, PhD, Sadanori Takeo, MD, PhD, Akira Motohiro, MD, PhD, Tokujiro Yano, MD, PhD, Hideki Yokoyama, MD, PhD, Hitoshi Ueda, MD, PhD, Kenji Sugio, MD, PhD, Teruyoshi Ishida, MD, PhD, Kosei Yasumoto, MD, PhD, and Yoshihiko Maehara, MD, PhD, on behalf of the Kyushu Lung Cancer Surgery Cooperative Group

Objectives: In clinical stage IA lung adenocarcinoma, the clinical features of a truly node-negative population were prospectively investigated by means of a prospective investigational study. Methods: The clinical data and radiographic features of highresolution computed tomography (HR-CT) were registered before operation in 169 clinical stage IA lung cancer patients who were scheduled to undergo a standard lobectomy and systemic mediastinal node dissection. The nodal metastasis was pathologically defined and the clinical factors associated with the presence of the nodal metastasis were evaluated. Results: In 114 assessable cases with adenocarcinomas, 15 (13.1%) were node-positive. The serum carcinoembryonic antigen (CEA), retraction sign, and intratumoral air-bronchogram on HR-CT were suggested to be predictive factors for lymph node metastasis, with hazard ratios of 12.44 (p ⫽ 0.0003), 6.53 (p ⫽ 0.0533), and 0.17 (p ⫽ 0.0073), respectively. In combination with the radiologic features and serum CEA, cases with elevated serum CEA or presence of retraction sign included 15.6% of node metastasis-positive, whereas all cases with normal CEA and absence of retraction sign showed no nodal metastasis. Cases with elevated serum CEA or absence of intratumoral air-bronchogram included 24.5% of node metastasis, whereas cases with normal CEA and presence of airbronchogram showed 4.6% of node metastasis. The tumor size and *Department of Surgery and Science, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan; †Department of Thoracic Oncology, National Kyushu Cancer Center, Fukuoka, Japan; ‡Department of Chest Surgery, Kitakyushu Municipal Medical Center, Kitakyushu, Japan; §Department of General Thoracic Surgery, National Kyushu Medical Center, Fukuoka, Japan;
Department of Surgery, National Fukuoka Hospital, Fukuoka, Japan; ¶Department of Surgery, Saga Prefectural Hospital Koseikan, Saga, Japan; #Department of Chest Surgery, National Beppu Medical Center, Beppu, Japan; **Department of Chest Surgery, Matsuyama Red Cross Hospital, Japan; ††Department of Surgery, Hiroshima Red Cross Atomic Hospital, Hiroshima, Japan; and ‡‡Second Department of Surgery, Faculty of Medicine, University of Occupational and Environmental Health, Kitakyushu, Japan. Address for correspondence: Ichiro Yoshino, MD, PhD, Department of Surgery and Science, Graduate School of Medical Sciences, Kyushu-University, 3-1-1 Maidashi, Fukuoka 812-8582, Japan. E-mail: iyoshino@surg2. med.kyushu-u.ac.jp Copyright © 2006 by the International Association for the Study of Lung Cancer ISSN: 1556-0864/06/0108-0825

the proportion of ground-glass attenuation were not associated with the incidence of nodal metastasis. Conclusions: The serum CEA and HR-CT features thus allowed us to identify node-negative lung adenocarcinomas measuring 3 cm or less in size. Key Words: Non-small cell lung cancer, Limited resection, Tumor, node, metastasis (TNM) stage, Lymph node metastasis, High-resolution computed tomography. (J Thorac Oncol. 2006;1: 825–831)

B

ecause of recent advances in such diagnostic tools as computed tomography (CT), the chance of detecting and thus performing surgery on peripheral small-sized lung cancers has increased.1 A limited resection for peripheral type non-small cell lung cancer (NSCLC) is well indicated in patients with a compromised cardiopulmonary reserve, although its clinical outcome is relatively unfavorable in comparison with a standard lobectomy.2,3 The application of an intentional limited resection for every peripheral small-sized NSCLC remains controversial. A prospective, randomized trial conducted by the Lung Cancer Study Group showed a significantly higher frequency of local relapse in the limited resection group than in the standard lobectomy group.4 The selection criteria for patients with indications to undergo a limited resection therefore have to be established. The largest issue to be solved is the preoperative evaluation criteria for nodal metastasis. In the clinical staging of NSCLC, an examination of the nodal factors by CT scanning is still unsatisfactory, and more than 20% of clinically defined stage IA cases proved to have nodal metastasis by pathologic examination of surgical specimens.5,6 Until now, several authors have reported a positive relationship between tumor size and regional lymph node metastasis. Asamura et al.5 reported that 16.3% of cases of NSCLC with a primary tumor measuring 3 cm or less in diameter showed hilar or mediastinal node metastasis, whereas even in cases with tumors measuring 2 cm or less in size, 21% of adenocarcinomas showed node metastasis and 6% of squamous cell carcinomas showed nodal metastasis. Ohta et al.7 demonstrated tumor size to be significantly associated with nodal metastasis in squamous

Journal of Thoracic Oncology • Volume 1, Number 8, October 2006

825

I. Yoshino et al.

Journal of Thoracic Oncology • Volume 1, Number 8, October 2006

cell carcinoma but not in adenocarcinoma in clinical stage I NSCLC. Therefore, the tumor size may not be a definitive condition for a lesser resection. To establish the criteria for indicating an intentional limited resection without any nodal dissection of peripheral small-sized NSCLC, the preoperative selection of “true stage IA” (node-negative) is required. In this study, we prospectively investigated the features of node-negative adenocarcinoma of the lung.

PATIENTS AND METHODS Registration of Patients The protocol of this study was permitted by the institutional review boards of all cooperative institutes. The patients were enrolled between September 1, 2003, and August 31, 2005. High-resolution (HR) CT was used for clinical intrathoracic staging and for characterization of the radiographic features. Positron emission tomography was not used for evaluation for preoperative staging. The inclusion criteria of this study were as follows: (1) pathologically or cytologically proven lung cancer, or strong suspicion of lung cancer by radiographic examinations; (2) with a larger diameter of a primary tumor of less than 3 cm; (3) more than 3 cm between the proximal margin of the tumor and hilar vessels; (4) no enlarged regional lymph nodes; (5) the ability to tolerate a standard lobectomy; (6) no prior anticancer treatment; and (7) written informed consent. The exclusion criteria were as follows: (1) the presence of metachronous or synchronous multiple lung cancer or thoracic malignancies; (2) a history of thoracic radiation; and (3) the presence of uncontrollable complications.

embryonic antigen (CEA) detected by enzyme-linked immunosorbent assay (the cutoff value was decided according to each institutional laboratory), and (3) the histologic type based on the criteria of the World Health Organization.8 The HR-CT features of the primary lesion were (1) the lobe location; (2) a retraction sign such as pleural retraction or vascular convergence; (3) the intratumoral air-bronchogram; (4) the intratumoral cavity; (5) notching; (6) an obstructive sign such as peripheral atelectasis, hyperinflation, or pneumonia; and (7) the proportion of ground-glass attenuation (GGA) (GGA ratio, calculated as [the largest diameter – diameter of non-GGA portion]/the largest diameter). The definition of GGA was determined to be a hazy increase in lung attenuation without obscuration of the underlying vascular marking based on the HR-CT scan findings, mainly, as follows: tube voltage, 120 kV; tube current, 250 mA; 2-mm collimation; table speed, 2 mm/second, one rotation per second; pitch, 1; and reconstruction interval, 2 mm using a bone algorithm. The images were photographed with a window level of – 650 HU and a window width of 1800 HU. In the final analysis, patients undergoing an operation smaller than a lobectomy or an incomplete nodal dissection, or patients who were revealed to have diseases other than NSCLC by postsurgical pathologic examinations, were excluded. The stage of all patients was pathologically defined according to the revised international staging system.9

Statistical Analysis Statistical difference in rates of nodal metastasis among subgroups divided by each clinical factor was analyzed by means of the ␹2 test. A multivariate analysis for factors affecting the incidence of nodal metastasis was analyzed by means of a logistic regression model.

Study Design The study was performed in a multi-institutional setting. The clinical data and radiographic features of HR-CT described below were registered at the entry of this study before operation. Next, the pathologic data of a surgical specimen of a standard operation (lobectomy and systemic mediastinal node dissection) were collected, and the relationship between the presence or absence of nodal metastasis and the clinical and radiographic factors was investigated. A systemic node dissection was performed uniformly among all institutions, and the policy was as follows: when a primary tumor was located at the right upper or middle lobe, lymph nodes at stations 1, 2, 3, 4, 7, 10, 11, and 12 and intralobar nodes were dissected. When a primary tumor was located at the right lower lobe, lymph nodes at stations 1, 2, 3, 4, 7, 8, 9, 10, 11, and 12 were dissected in addition to intrapulmonary nodes. When a primary tumor was located at the left upper lobe, lymph nodes at stations 4, 5, 6, 7, 10, 11, and 12 were dissected. When a primary tumor was located at the left lower lobe, lymph nodes at stations 4, 5, 6, 7, 8, 9, 10, 11, and 12 were dissected.

Preoperative Parameters Clinical data and radiographic features of HR-CT were registered before the operation. The clinical data collected included the (1) smoking status, (2) serum level of carcino-

826

RESULTS Profile of Analyzed Patients During this period, 169 patients were registered from 12 institutes (see Appendix). Of them, 14 were eliminated because of incomplete information. Of the 155 remaining patients, 140 had NSCLC, three had small cell lung cancer, and 12 had inflammatory lesions (organizing pneumonia and tuberculoma). Of the 140 patients with NSCLC, 130 underwent a standard lobectomy or bilobectomy with a systemic mediastinal nodal dissection and 10 underwent a lesser resection or an omission of mediastinal node dissection by the surgeon’s decision. Of them, 114 patients with adenocarciTABLE 1. Stage IA IB IIA IIIA IIIB

Profile of All Patients* Pathologic Stage

No. (%)

T1N0M0 T2N0M0 T1N1M0 T1N2M0 T4N0M0

88 (77.2) 10 (8.8) 8 (7.0) 7 (6.1) 1 (0.9)

*There were 56 male patients and 58 female patients, with a mean age of 64.7 yr (range, 40 – 85 yr).

Copyright © 2006 by the International Association for the Study of Lung Cancer

Copyright © 2006 by the International Association for the Study of Lung Cancer

M F F F F M F M F F F M M M F F M F F F F M F M F M

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

58 66 68 81 64 60 58 45 74 79 50 69 66 55 76 69 68 77 74 51 63 66 69 64 61 69

Age (yr)

Non-BAC Non-BAC Non-BAC Non-BAC Non-BAC BAC Non-BAC With BAC Non-BAC Non-BAC Non-BAC Non-BAC Non-BAC Non-BAC Non-BAC Non-BAC With BAC Non-BAC With BAC Non-BAC Non-BAC Non-BAC Non-BAC With BAC With BAC Non-BAC

Subtype RML LLL RUL LUL LUL RUL LUL RUL RUL LUL LLL LUL RUL RUL LUL RLL RLL LUL RUL LLL RUL RUL RUL RUL RUL RLL

Location 20 31 32 35 31 31 16 28 23 27 18 29 24 20 18 26 14 19 28 12 12 20 28 22 27 25

Size (mm) – – – – – – – – – – – – ⫹ – – ⫹ – ⫹ ⫹ ⫹ ⫹ – ⫹ – – –

Serum CEA Retraction – ⫹ ⫹ ⫹ ⫹ ⫹ – ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ – ⫹ ⫹ ⫹

GGA Ratio ⬍0.5 ⬍0.5 ⬍0.5 ⱖ0.5 ⬍0.5 ⱖ0.5 ⱖ0.5 ⬍0.5 ⬍0.5 ⱖ0.5 ⬍0.5 ⱖ0.5 ⱖ0.5 ⬍0.5 ⬍0.5 ⱖ0.5 ⱖ0.5 ⱖ0.5 ⱖ0.5 ⬍0.5 ⬍0.5 ⬍0.5 ⱖ0.5 ⱖ0.5 ⬍0.5 ⱖ0.5 – – ⫹ ⫹ – ⫹ – ⫹ – ⫹ ⫹ ⫹ ⫹ – – – – ⫹ ⫹ – ⫹ ⫹ – – – ⫹

AirBronchogram

Profile of the Patients with Up-Staged Lesions Based on Pathologic Examination

IB IB IB IB IB IB IB IB IB IB IIA IIA IIA IIA IIA IIA IIA IIA IIIA IIIA IIIA IIIA IIIA IIIA IIIA IIIB

P Stage

RML, right middle lobe; LLL, left lower lobe; RUL, right upper lobe; LUL, left upper lobe; RLL, right lower lobe; BAC, bronchioloalveolar carcinoma.

Sex

Case

TABLE 2.

T2-p2 T2-size T2-size T2-size T2-size T2-size T2-p2 T2-p3 T2-p2 T2-p2 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T4-pm

T Factor 2 0 0 0 0 0 2 3 (interlobe) 2 2 1 1 1 1 0 0 1 1 1 1 1 1 1 1 1 0

Pleural Factor

N0 N0 N0 N0 N0 N0 N0 N0 N0 N0 N1 N1 N1 N1 N1 N1 N1 N1 N2 N2 N2 N2 N2 N2 N2 N0

Node Factor

13 10 12 13 12 10 13 13 1, 11 4, 9 3 3 1, 2, 3, 7, 10 3, 4, 10 3

Station of Node Metastasis

Journal of Thoracic Oncology • Volume 1, Number 8, October 2006 Preoperative Detection of Early Lung Adenocarcinoma

827

Journal of Thoracic Oncology • Volume 1, Number 8, October 2006

I. Yoshino et al.

TABLE 3. Analysis of the Clinicopathologic Factors for Lymph Node Metastasis in Lung Adenocarcinoma

Gender Male Female Smoking Yes No CEA Elevated Normal Size ⬍20 mm ⬍20 mm GGA ratio ⬎0.5 ⬍0.5 Retraction sign Presence Absence Air-bronchogram Presence Absence Notching Presence Absence

No. of Patients

N0

N1/2

N1–2 (%)

56 58

50 49

4/5 4/2

15.5 10.7

46 68

39 60

3/4 5/3

15.2 11.8

17 97

10 89

3/4 5/3

41.2 8.2

59 55

51 48

5/3 3/4

13.6 12.7

57 57

49 50

5/3 3/4

14.0 12.3

93 21

79 20

8/6 0/1

15.0 4.8

78 36

71 28

4/3 4/4

9.0 22.2

30 84

25 74

3/2 5/5

16.7 11.9

p Value

NPV (%)

0.4467

0.5948

0.0012 58.8 91.8 0.8955

0.7817

0.1636 84.9 95.2 0.0601 91.0 77.8 0.5167

NPV, negative predictive value for node metastasis.

noma were subjected to the analysis. The profiles of the 114 patients are summarized in Table 1. The mean age was 64.7 years, ranging from 40 to 85 years (there were 56 male patients and 58 female patients). The pathologic lymph node metastasis was recognized in 15 patients (13.2%). Pathologic stages were T1N0M0 in 88 patients (77.2%), T2N0M0 in eight (7.0%), T1N1M0 in eight (7.0%), T1N2M0 in seven (6.1%), and T4N0M0 in one (0.9%). The detailed information for the 26 up-staged patients (22.8%) is summarized in Table 2. The reasons for T2 in a total of 10 patients were pleural invasion in five and a diameter of more than 3 cm in five. The reason for T4 in one patient was intrapulmonary metastasis. The location of the nodal metastases was the intrapulmonary node in four patients, the hilar node in four, and the mediastinal node in seven. Of the seven patients with mediastinal node metastasis, three had single and skip metastases from the upper lobes.

Clinicopathologic Analysis for Lymph Node Metastasis in Adenocarcinoma Patients The relationship between the incidence of nodal metastasis and the radiographic features in HR-CT in adenocarcinomas was examined (Table 3). Of these factors, only the elevated serum CEA was significantly associated with lymph node metastasis. In 17 patients with elevated CEA, seven (41.2%) had nodal metastasis, whereas only eight of 97

828

patients (8.2%) with normal CEA had nodal metastasis (p ⫽ 0.0012). The presence of intratumoral air-bronchogram tended to predict nodal metastasis. In 78 patients with presence of air-bronchogram, seven (9.0%) had nodal metastasis, whereas eight of 36 patients (22.2%) with absence of airbronchogram had nodal metastasis (p ⫽ 0.0601). Although no statistical significance was observed, only one of 21 patients (4.8%) with absence of retraction sign had nodal metastasis. The negative-predictive value for lymph node metastasis was 91.8% in normal serum CEA, 95.1% in the absence of retraction sign, and 91.0% in the presence of air-bronchogram. The tumor size and the GGA ratio and notching were not associated with the incidence of nodal metastasis. A logistic multivariate analysis revealed elevated serum CEA, presence of retraction sign, and absence of intratumoral air-bronchogram to be independent factors for predicting nodal metastasis, with hazard ratios of 12.44 ((p ⫽ 0.0003), 6.53 (p ⫽ 0.0533), and 0.17 ((p ⫽ 0.0073), respectively (Table 4). When the relationship between each size of the tumor and a lymph node dissection was analyzed, lymph node metastasis was not observed in adenocarcinoma cases with tumors measuring 10 mm or less in diameter (Table 5). In a detailed analysis of the relationship between the GGA ratio and the incidence of lymph node metastasis, lymph node metastasis was observed in each grade of the GGA ratio

Copyright © 2006 by the International Association for the Study of Lung Cancer

Journal of Thoracic Oncology • Volume 1, Number 8, October 2006

TABLE 4. Multivariate Analysis of the Clinical Predictive Factors for Lymph Node Metastasis in Lung Adenocarcinoma

Serum CEA Elevated/normal Retraction Presence/absence Air-bronchogram Presence/absence

␹2 Value

Hazard Ratio

p Value

12.92

12.44

0.0003

3.74

6.53

0.0533

7.19

0.17

0.0073

TABLE 5. Tumor Size and Incidence of Lymph Node Metastasis in Lung Adenocarcinoma Incidence of Lymph Node Metastasis (%) Tumor Size (mm) ⱕ10 11 to ⱕ20 20 to ⱕ30 ⬎30

All Cases

Adenocarcinoma

0/6* (0.0) 10/65 (15.4) 8/53 (15.1) 0/6 (0.0)

0/6 (0.0) 8/53 (15.1) 7/50 (14.0) 0/5 (0.0)

TABLE 6. GGA Ratio and the Incidence of Lymph Node Metastasis in Lung Adenocarcinoma Incidence of Lymph Node Metastasis (%)

0–0.25 0.25–0.5 0.5–0.75 0.75–1.0 1.0

bronchogram or retraction sign were compared for the incidence of node metastasis (Table 7). None of the patients with both features of normal serum CEA and absence of retraction sign had node metastasis, and the difference in the rates of node metastasis among the subgroups divided by serum CEA (normal or elevated) and retraction sign (presence or absence) was significant (p ⫽ 0.0015). The combination of normal serum CEA and absence of retraction sign might predict truly negative for node metastasis in clinical stage IA NSCLC (negative predictive value, 100%). The rate of nodal metastasis in patients with a normal serum CEA and presence of air-bronchogram (24.5%) was significantly higher than that of the other cases (4.6%) (p ⫽ 0.0001), and the negative predictive value of this categorization for node metastasis was 95.4%.

DISCUSSION

*One of the six cases showed a positive serum CEA level.

GGA Ratio

Preoperative Detection of Early Lung Adenocarcinoma

6/36 (16.7) 1/21 (4.8) 2/15 (13.3) 5/35 (14.3) 1/7 (7.1)

(Table 6). Even in seven cases with tumors demonstrating pure GGA, one showed lymph node metastasis. Next, the subgroups categorized by the factors of serum CEA level and HR-CT features such as intratumoral air-

It is well known that the size of the primary tumor significantly correlates with lymphatic permeation.10 Several clinicopathologic studies by thoracic surgeons5–7,11 have revealed that tumor size is associated with the incidence of nodal metastasis in resected NSCLC, and that a lesser resection for small-sized NSCLC was proposed. In a series of 89 cases that were scheduled to undergo a segmentectomy for clinical stage IA (⬍20 mm) NSCLC,11 however, 12 cases were converted to other procedures because of nodal involvement and seven were converted because of intralobar invasion or tumor location, and one of 70 who underwent a segmentectomy was revealed to have intralobar nodal metastasis in a postoperative pathologic examination. Therefore, the size of the tumor was found to be a predictive factor for the nodal status in squamous cell carcinoma but not in adenocarcinoma of the lung. In our prospective series, one of 11 squamous cell carcinomas measuring 20 mm or less and eight of 59 adenocarcinomas measuring 20 mm or less showed nodal metastasis. Six (5.2%) adenocarcinomas measuring less than 10 mm in diameter showed no lymph node metastasis, and these findings correlate with a previous retrospective study performed at our institute.12 In cases with

TABLE 7. Analysis of the Serum CEA and Radiologic Features as Predictive Clinicopathologic Factors of Node-Negative Adenocarcinoma Factors

No. of Patients

N0

N1/2

N1-2(%)

NPV(%)

18 79 4 13

18 71 2 8

0/0 5/3 1/1 2/3

0.0 10.1 50.0 38.5

100* 89.9 50.0 61.5

65 32 13 4

62 27 9 1

2/1 3/2 2/2 1/2

4.6 15.6 30.8 75.0

CEA/retraction Normal/absence Normal/presence Elevated/absence Elevated/presence CEA/air-bronchogram Normal/presence Normal/absence Elevated/presence Elevated/absence

p Value 0.0015

0.0001 95.4† 74.6 69.2 25.0

*Calculated as 18/(18 ⫹ 0). †Calculated as 62/(62 ⫹ 3). NPV, negative predictive value.

Copyright © 2006 by the International Association for the Study of Lung Cancer

829

I. Yoshino et al.

Journal of Thoracic Oncology • Volume 1, Number 8, October 2006

tumors larger than 1 cm, however, tumor size did not correlate with the incidence of nodal metastasis. In this study, two of the node-positive cases exhibited tumor sizes of 12 mm and 13 mm, and both were positive for serum CEA. Based on the uncertainty regarding the clinical measurement of the tumor diameter, errors ranging from 1 to 3 mm are considered to be unavoidable. Therefore, the size criteria may be inappropriate for clinically predicting nodal metastasis. Noguchi et al.13 classified six subgroups of small-sized adenocarcinomas according to their pathologic features. Localized bronchioloalveolar carcinoma (BAC) without any invasive growth, referred to as Noguchi type A or B, showed no nodal metastasis and no relapse following curative surgery. The Noguchi types A and B adenocarcinoma might be “early lung cancer”; however, such types can hardly be distinguished preoperatively. In recent retrospective studies, the presence of GGA in tumors on HR-CT could help identify node-negative adenocarcinomas because intratumoral GGA is a reflection of a lining growth pattern under alveolar septa, which is a feature of Noguchi types A and B. Matsuguma et al.14 advocated that adenocarcinomas measuring 20 mm or less in size were negative for lymph node metastasis when the diameter of the GGA region occupied 50% or more of the total diameter of the tumor. In this study, we also investigated whether the proportion of the GGA region in the tumors was associated with nodal metastasis or not; however, no correlation was observed. Recently, a pathologic classification of lung adenocarcinoma in special reference to a component of BAC and to invasiveness has been proposed.15 Most likely, the amount of the GGA portion is not positively related to invasiveness. In this study, the presence of air-bronchogram and retraction sign was effective for predicting nodal metastasis in adenocarcinomas. The presence of an air-bronchogram may reflect the preserved alveolar and bronchiole space in the tumor proliferating in the lining pattern. Such a growth pattern is a feature of BAC. In contrast, the retraction of the neighboring structure by pulmonary adenocarcinoma is considered to result from collapse of involving parenchyma and intratumoral scar formation. Scar-like fibrosis in the central portion is a feature of invasive lung adenocarcinoma,16 which is categorized as type C according to the Noguchi classification13 or as “adenocarcinoma with invasiveness” according to the classification of the Memorial Sloan-Kettering Cancer Center.15 The radiographic features of the presence of an intratumoral air-bronchogram and the absence of pleural retraction have recently been reported to be significantly associated with the prognosis of lung adenocarcinoma17,18; therefore, those radiologic features were included in this study as clinical factors to be investigated. A significant association between serum CEA and nodal metastasis was recognized in this prospective study. Several authors have reported that an elevated serum CEA level suggested spread of the disease to the regional lymph nodes or to distant organs in cancer even if conventional clinical staging indicated them to be resectable.19,20 In our present data, 41% of the elevated CEA cases showed nodal metastasis; therefore, special caution may be required at

830

surgery for cases with elevated CEA even if no apparent nodal swelling is observed on preoperative CT scans. However, 8.2% of the cases with normal CEA levels had nodal metastasis. To select node-negative adenocarcinomas, the combination of serum CEA and radiographic features was thus found to be significantly effective. Even in a limited number of cases, all 18 adenocarcinoma cases with normal CEA levels and absence of retraction sign exhibited no lymph node metastasis in adenocarcinoma even though they constituted 15.5% of all adenocarcinomas (n ⫽ 114) and 18% of the node-negative clinical stage IA adenocarcinomas (n ⫽ 99).

CONCLUSIONS The combination of serum CEA level and HR-CT features may therefore be useful for predicting the presence of node-negative lung adenocarcinoma, although the sensitivity of this diagnostic modality is still low. The present information might help to decide whether limited resection, stereotactic radiotherapy,21 or radiofrequency ablation therapy is indicated22 for peripheral small-sized lung adenocarcinomas.

APPENDIX The principal investigators of institutions that participated in this study included the following: K. Yamazaki, T. Tagawa, T. Kameyama, and T. Yohena (Kyushu University Hospital); R. Maruyama, T. Okamoto, and R. Suemitsu (National Kyushu Cancer Center, Fukuoka, Japan); M. Takenoyama (Kitakyushu Municipal Medial Center, Kitakyushu, Japan); T. Kometani (National Kyushu Medical Center, Fukuoka, Japan); K. Sakada, M. Tomiyasu, and A. Osoegawa (National Fukuoka Hospital, Fukuoka, Japan); C. Ushijima (National Beppu Medical Center, Beppu, Japan); F. Shoji (Saga Prefectural Hospital Koseikan, Saga, Japan); M. Hamatake (Fukuoka Municipal Hospital, Fukuoka, Japan); K. Ondo and S. Fukuyama (Matsuyama Red Cross Hospital, Matsuyama, Japan); Y. Fukuyama (Nakatsu Municipal Hospital, Nakatsu, Japan); M. Yamaguchi (National Fukuoka Higashi Medical Center, Koga, Japan); and S. Tsukamoto (Hiroshima Red Cross Hospital, Hiroshima, Japan). REFERENCES 1. Yoshino I, Baba H, Fukuyama F, et al. A time trend of profile and surgical results in 1123 patients with non-small cell lung cancer. Surgery 2002;131:S242–S248. 2. Errett LE, Wilson J, Chiu RC, Munro DD. Wedge resection as an alternative procedure for peripheral bronchogenic lung carcinoma in poor-risk patients. J Thorac Cardiovasc Surg 1985;90:656–661. 3. Yano T, Yokoyama H, Yoshino I, et al. Results of a limited resection for compromised patients with clinical stage I non-small cell lung carcinoma of the lung. J Am Coll Surg 1995;181:33–37. 4. Lung Cancer Study Group. Randomized trial of lobectomy versus limited resection for T1N0 non-small cell lung cancer. Ann Thorac Surg 1995;60:615–623. 5. Asamura H, Nakayama H, Kondo H, Tsuchiya R, Shimosato Y, Naruke T. Lymph node involvement, recurrence, and prognosis in resected small, peripheral, non-small cell lung carcinomas: Are these carcinomas candidates for video-assisted lobectomy? J Thorac Cardiovasc Surg 1996;111:1125–1134. 6. Suzuki K, Nagai K, Yoshida J, Nishimura M, Nishikawa Y. Predictors of lymph node and intrapulmonary metastasis in clinical stage IA non-small cell lung carcinoma. Ann Thorac Surg 2001;72:352–356.

Copyright © 2006 by the International Association for the Study of Lung Cancer

Journal of Thoracic Oncology • Volume 1, Number 8, October 2006

7. Ohta Y, Oda M, Wu J, et al. Can tumor size be a guide for limited surgical intervention in patients with peripheral non-small cell lung cancer? Assessment from the point of view of nodal micrometastasis. J Thorac Cardiovasc Surg 2001;122:900–906. 8. World Health Organization. Histological Typing of Lung and Pleural Tumours, 3rd Ed. Geneva, Switzerland: Springer-Verlag, 1999. 9. Mountain CF. Revisions in the international system for staging lung cancer. Chest 1997;111:1710–1717. 10. Ichinose Y, Yano T, Yokoyama H, Inoue T, Asoh H, Katsuda Y. The correlation between tumor size and lymphatic vessel invasion in resected peripheral stage-I non-small-cell lung-cancer: A potential risk of limited resection. J Thorac Cardiovasc Surg 1994;108:684–686. 11. Okada M, Yoshikawa K, Hatta T, Tsubota N. Is segmentectomy with lymph node assessment an alternative to lobectomy for non-small cell lung cancer of 2 cm or smaller. Ann Thorac Surg 2001;71:956–961. 12. Ishida T, Yano T, Maeda K, Kaneko S, Tateishi M, Sugimachi K. Strategy for lymphadenectomy in lung cancer three centimeters or less in diameter. Ann Thorac Surg 1990;50:708–713. 13. Noguchi M, Morikawa A, Kawasaki M, et al. Small adenocarcinoma of the lung: Histologic characteristics and prognosis. Cancer 1995;75: 2844–2852. 14. Matsuguma H, Yokoi K, Anraku M, et al. Proportion of ground-glass opacity on high-resolution computed tomography in clinical T1N0M0 adenocarcinoma of the lung: A predictor of lymph node metastasis. J Thorac Cardiovasc Surg 2002;124:278–284.

Preoperative Detection of Early Lung Adenocarcinoma

15. Ebright MI, Zakowski MF, Martin J, et al. Clinical pattern and pathologic stage but not histologic features predict outcome for bronchioloalveolar carcinoma. Ann Thorac Surg 2002;74:1640–1646. 16. Edwards C, Carlie A. Scar adenocarcinoma of the lung: A light and electron microscopic study. J Clin Pathol 1986;39:423–427. 17. Yoshino I, Nakanishi R, Kodate M, et al. Pleural retraction and intratumoral air-bronchogram as prognostic factors for stage I pulmonary adenocarcinoma following complete resection. Int Surg 2000;85:1–8. 18. Kondo T, Yamada K, Noda K, Nakayama H, Kameda Y. Radiologicprognostic correlation in patients with small pulmonary adenocarcinomas. Lung Cancer 2002;36:49–57. 19. Icard P, Regnard JF, Essomba A, Panebianco V, Magdeleinat P, Levasseur P. Preoperative carcinoembryonic antigen level as a prognostic indicator in resected primary lung cancer. Ann Thorac Surg 1994;58:811–814. 20. Okada M, Nishio W, Sakamoto T, et al. Prognostic significance of perioperative serum carcinoembryonic antigen in non-small cell lung cancer: Analysis of 1,000 consecutive resections for clinical stage I disease. Ann Thorac Surg 2004;78:216–221. 21. Onishi H, Araki T, Shirato H, et al. Stereotactic hypofractionated high-dose irradiation for stage I non-small cell lung carcinoma: Clinical outcomes in 245 subjects in Japanese multiinstitutional study. Cancer 2004;101:1623–1631. 22. Suh R, Reckamp K, Zeidler M, Cameron R. Radiofrequency ablation in lung cancer: Promising results in safety and efficacy. Oncology 2006; 19:12–21.

Copyright © 2006 by the International Association for the Study of Lung Cancer

831