Prognostic implication of metastasis limited to segmental (level 13) and/or subsegmental (level 14) lymph nodes in patients with surgically resected nonsmall cell lung carcinoma and pathologic N1 lymph node status

Authors


Abstract

BACKGROUND:

In patients with nonsmall cell lung carcinoma (NSCLC) who have with pathologic N1 (pN1) lymph node status, the prognostic significance of segmental lymph node (level 13) metastasis and/or subsegmental lymph node (level 14) metastasis is unknown.

METHODS:

Lymph node metastasis patterns were analyzed in 230 patients with NSCLC who had pN1 status. Clinical outcomes were examined for 230 patients with pN1 status and 700 patients with pN0 status. The pN1 group was stratified into 4 subgroups according to the highest level of lymph node involvement.

RESULTS:

The 5-year disease-free survival (5DFS) rates for pN1 and pN0 patients were 50.1% and 90.5%, respectively. The highest level of lymph node involvement was a significant prognostic indicator; the 5DFS rates for patients with pN1 status who had level 13/14, lobar (level 12), interlobar (level 11), and hilar (level 10) lymph node metastasis were 69.4%, 46.4%, 46.7%, and 37%, respectively. Patient outcomes were significantly worse for those with pN1 status who had only level 13/14 lymph node metastasis than for patients with pN0 status (P = .0034), and outcomes were significantly worse for patients with pN1 status who had level 11/12 lymph node metastasis than for patients who had only level 13/14 lymph node metastasis (P = .021). The median number of level 13/14 lymph nodes examined was 3 (range, 0-22 level 13/14 lymph nodes), and metastases to these lymph nodes were detected in 61% of patients who had pN1 status. A single lymph node pN1 disease, single-level pN1 status, and squamous cell carcinoma histopathology also were indicators of a better patient outcome.

CONCLUSIONS:

The current results indicated that the highest level of lymph node involvement may be used to stratify the outcome of patients who have NSCLC with pN1 status. Patients with pN1 status who had only level 13/14 lymph node metastasis had an intermediate 5DFS rate between that of patients with pN0 status and other patients with pN1 status. Routine examination of level 13/14 lymph nodes is important for accurate pathologic staging and for the predicting clinical outcome of patients with NSCLC. Cancer 2012. © 2012 American Cancer Society.

INTRODUCTION

Accurate pathologic staging of surgically resected nonsmall cell lung carcinoma (NSCLC) is important to predict patient survival. In the lung cancer section of the TNM Classification of Malignant Tumors, seventh edition,1 the ipsilateral peribronchial, hilar, and intrapulmonary lymph nodes are defined as N1 lymph nodes; and histopathologically detected metastases to these N1 lymph nodes, including involvement by direct extension but no metastasis to N2 or N3 nodes, are classified as pathologic N1 (pN1). N1 nodes comprise 5 lymph node stations or levels: hilar (level 10), interlobar (level 11), lobar (level 12), segmental (level 13), and subsegmental (level 14). For a pN0 diagnosis, histopathologic examination of hilar and mediastinal lymphadenectomy specimens should include ≥6 lymph nodes, ≥3 of which should be mediastinal, including the subcarinal lymph nodes, and another ≥3 should be N1 lymph nodes.1 The International Association for the Study of Lung Cancer (IASLC) lung cancer staging project divided N1 lymph nodes into the hilar/interlobar zone and the peripheral zone.2 The hilar/interlobar zone includes 2 levels (levels 10 and 11), and the peripheral zone includes the other 3 levels (levels 12, 13, and 14).

pN1 tumors represent a heterogeneous group within NSCLC patients in terms of clinical behavior. The reported 5-year survival rates for patients with pN1 vary between 27% and 67%.2-22 The IASLC reports a calculated 5-year survival rate of 38% in these patients.2

Lobar pN1 disease (levels 12-14),5-7, 10, 13, 17, 20 single-zone pN1 disease,2, 20 single-level pN1 disease,16, 19, 20 a single lymph node pN1 disease,4, 15, 19 direct invasion,6, 18 and microscopic metastasis11, 15 have been reported as good prognostic factors for patients with pN1 NSCLC. The number of lymph node metastases21 and a positive lymph node ratio22 also have been reported as prognostic factors.

In several malignant tumors, such as breast cancer and malignant melanoma, subsequent lymph node dissection is determined by referring to sentinel lymph node status. Sentinel lymph node mapping also has been attempted in patients with lung cancer23-26; however, detailed information and the implications of level 13/14 lymph node status currently are insufficient. To our knowledge, no study to date has performed a detailed investigation of the metastatic status of intrapulmonary level 13/14 lymph nodes in patients with NSCLC. Therefore, in the current study, we examined the metastatic status of patients with N1 lymph nodes, especially intrapulmonary level 13/14 lymph nodes, in patients with pN1 NSCLC and demonstrated the clinical importance of accurate pathologic staging.

MATERIALS AND METHODS

Patients and Methods

Of 2399 patients who underwent surgery in the National Cancer Center Hospital, Tokyo, Japan between 2000 and 2004, we included 230 consecutive patients with pN1 NSCLC and 700 consecutive patients with pN0 NSCLC in the current study. Patients who had synchronous, multiple lung cancers were excluded. Of these 930 patients, 902 underwent lobectomy, and 28 underwent pneumonectomy. All of these patients also underwent dissection of ≥6 lymph nodes, including ≥3 N1 lymph nodes and ≥3 N2 lymph nodes, which included subcarinal (level 7) lymph nodes. Clinical information was obtained from patients' medical records. Pathologic TNM stage was determined according to the International Union Against Cancer staging system.1 Forty-nine of 700 patients (7%) with pN0 disease and 35 of 230 patients (11%) with pN1 disease received adjuvant chemotherapy. No patients received neoadjuvant therapy. Follow-up ranged from 1 month to 144 months (median follow-up, 61 months). Informed consent was obtained from each patient, and the study was approved by the local institutional review board.

In the 230 patients with pN1 NSCLC, the status of metastasis in N1 lymph nodes was analyzed. All resected N1 lymph nodes were assigned to 1 of 4 lymph node levels (level 10, 11, 12, or 13/14), and the levels of metastatic lymph nodes in each patient were examined. Although lymph node levels 13 and 14 are divided conceptually, they were regarded as a single level, because it was not possible to accurately differentiate between these 2 sites.

Level 10, 11, and 12 lymph nodes were resected by surgeons; whereas level 13/14 lymph nodes were dissected by pathologists. A diagnosis of pN1 was made when metastasis was histopathologically confirmed in ≥1 N1 lymph node but no metastasis was observed in N2 or N3 lymph nodes. When pN1 metastases were observed in multiple lymph node levels, the advanced level was considered maximal in level 10 metastasis, followed by level 11, level 12, and level 13/14, and latter was considered the least advanced level. Furthermore, the number of patients with lymph nodes present, the incidence of patients with positive lymph nodes, the number of examined lymph nodes, and the number of positive lymph nodes at each level also were examined on an individual patient basis.

Pathologic Examination

All lobectomy or pneumonectomy specimens were infused with 10% formalin by using the transbronchial method and were fixed overnight. The dissected lymph nodes (N1 lymph nodes other than level 13/14, N2 lymph nodes, and N3 lymph nodes) were fixed separately overnight in 10% formalin. The lobectomy/pneumonectomy specimens were cut at 5-mm to 10-mm intervals the next day. From each resected specimen, tissue blocks were cut routinely from the whole tumor area, including the maximum cut surface, several tumor areas of additional tumor sections, all available level 13 and/or level 14 lymph nodes, the cut surface of hilar bronchus, and peripheral lung distant from the main tumor (Fig. 1). On average, 15 tissue blocks per patient were prepared routinely from a lobectomy specimen, except for separately processed lymph nodes. One representative tissue block was created from each dissected lymph node. If a lymphoid tissue measured ≥3 mm in greatest dimension and had a fibrous capsule, then it was defined as a lymph node.

Figure 1.

(A) This photograph reveals the gross features of a tumor from a patient who had nonsmall cell lung carcinoma with segmental/subsegmental (level 13/14)-positive pathologic N1 (pN1) lymph node status. The red arrow indicates a metastasis-positive level 13/14 lymph node. (B) This is a loupe view of the positive level 13/14 lymph node (hematoxylin and eosin [H&E] staining). (C) This photograph reveals the gross features of a tumor from a patient who had level 13/14-negative pN1 status. Blue arrows indicate metastasis-negative level 13/14 lymph nodes. (D) This is a loupe view of the negative level 13/14 lymph nodes (H&E staining).

Statistical Analysis

Survival curves were drawn for the 700 patients with pN0 disease and the 230 patients with pN1 disease stratified according to metastasis-positive lymph node levels using the Kaplan-Meier method. The clinicopathologic parameters of patients with pN1 disease and patients with pN0 disease were compared using the chi-square test. Univariate analysis was performed using the log-rank test for clinicopathologic parameters. Differences were considered significant if P ≤ .05, and considered a trend if .05 < P < .1.

RESULTS

Clinicopathologic Differences Between Patients With Pathologic N0 and N1 Status

The patients' clinicopathologic characteristics are summarized in Table 1. No parameter except pathologic stage differed significantly between patients with pN1 status and patients with pN0 status. In the pN1 group, the median number of N1 lymph nodes examined was 10 (range, 3-35 N1 lymph nodes), and the median of N2 lymph nodes was 6 (range, 3-26 N2 lymph nodes). In the pN0 group, the median number of N1 lymph nodes examined was 7 (range, 3-41 N1 lymph nodes), and the median number of N2 lymph nodes was 8 (range, 3-28 N2 lymph nodes).

Table 1. Clinicopathologic Characteristics of Patients With Resected Nonsmall Cell Lung Carcinoma, Including 230 Patients With Pathologic N1 Status and 700 Patients With Pathologic N0 Status
 No. of Patients (%) 
ParameterpN1pN0Pa
  • Abbreviations: NS, not significant; pN, pathologic lymph node status.

  • a

    P values were determined with the chi-square test.

  • b

    P value for stage I disease versus stages II-IV disease.

  • c

    P value for squamous cell carcinoma versus adenocarcinoma.

Total no. of patients230700 
Age, y   
 ≤65134 (58)356 (51)NS
 ≥6596 (42)344 (49) 
Sex   
 Men164 (71)391 (56)NS
 Women66 (29)309 (44) 
Tumor size, cm   
 ≤3100 (43)496 (71)NS
 ≥3130 (57)204 (29) 
Pathologic stage   
 I0635 (91).015b
 II172 (75)34 (5) 
 III-IV58 (25)31 (4) 
Total no. of lymph nodes examined
 6-1595 (41)338 (48)NS
 ≥16135 (59)363 (52) 
No. of N1 lymph nodes examined
 3-767 (29)367 (52)NS
 ≥8163 (71)334 (48) 
No. of N2 lymph nodes examined
 3-7136 (59)345 (49)NS
 ≥894 (41)356 (51) 
Histology   
 Squamous cell carcinoma75 (33)109 (16)NSc
 Adenocarcinoma135 (59)552 (78) 
 Others20 (8)40 (6) 

Detailed Lymph Node Status of Patients With Pathologic N1 Diseases

The detailed lymph node status of patients with pN1 disease is provided in Tables 2 and 3. Among all 230 patients, the highest lymph node level was level 13/14 in 21% of patients. The number of patients with the presence of lymph node level was 185 patients (80%) for level 10, 185 patients (80%) for level 11, 189 patients (82%) for level 12, and 160 patients (70%) for level 13/14. The number of patients with the presence of level 13/14 lymph nodes was slightly less than the number with of other N1 lymph nodes. The incidence of patients with positive lymph nodes was 24% (45 of 185 patients) for level 10, 39% (72 of 185 patients) for level 11, 63% (120 of 189 patients) for level 12, and 61% (97 of 160 patients) for level 13/14. The positive lymph node rate was significantly higher for peripheral zone lymph nodes (levels 12 and 13/14) than that for hilar/interlobar zone lymph nodes (levels 10 and 11; P = .0028).

Table 2. Detailed Status of Patients With Pathologic N1 Status Stratified by 4 Lymph Node Stations
 No. of Patients (%)Median [Range]
ParameterWith LNs PresentWith Positive LNsaNo. of LNs Examined per PatientNo. of Positive LNs per Patient
  • Abbreviations: LN, lymph node; NS, not significant.

  • a

    The P value (determined with the chi-square test) was .0028 for hilar and interlobar lymph nodes versus lobar and segmental/subsegmental lymph nodes.

Total no. of patients230 (100)230 (100)10 [1-35]2 [1-9]
LN level    
 10: Hilar185 (80)45 (24)2 [0-21]1 [1-2]
 11: Interlobar185 (80)72 (39)3 [0-16]1 [1-7]
 12: Lobar189 (82)120 (63)3 [0-15]1 [1-6]
 13/14: Segmental/subsegmental160 (70)97 (61)3 [0-22]1 [1-7]
Table 3. Correlation Between Outcome, Lymph Node Status, and Histopathology in 230 Patients With Resected Nonsmall Cell Lung Carcinoma Who Had Pathologic N1 Status
ParameterNo. of Patients With pN1 Disease (%)5-Year DFS, %Pa
  • Abbreviations: DFS, disease-free survival; NS, not significant; pN, pathologic lymph node status.

  • a

    P values were determined with the log-rank test.

  • b

    P value for hilar versus interlobar and lobar.

  • c

    P value for interlobar and lobar versus segmental/subsegmental.

  • d

    P value for squamous cell carcinoma versus adenocarcinoma.

Total no. of patients23050.1 
Highest lymph node level   
 10: Hilar45 (20)37.064b
 11: Interlobar63 (27)46.7 
 12: Lobar74 (32)46.4.021c
 13/14: Segmental/subsegmental48 (21)69.4 
No. of involved lymph nodes
 1114 (50)55.7.036
 ≥2116 (50)44.3 
No. of involved lymph node stations
 1148 (64)56.3.0012
 2-482 (36)37.8 
Involved status   
 Metastasis226 (98)49.7NS
 Only direct invasion4 (2)75 
Histopathology   
 Squamous cell carcinoma75 (32)61.055d
 Adenocarcinoma135 (59)45.6 
 Others20 (9)44.3 

The median number of lymph nodes examined was 2 level-10 lymph nodes (range, 0-21 lymph nodes), 3 level-11 lymph nodes (range, 0-16 lymph nodes), 3 level 12-lymph nodes (range, 0-15 lymph nodes), and 3 level-13/14 lymph nodes (range, 0-22 lymph nodes). Among all these lymph nodes, the median number of metastasis-positive lymph nodes was 2 (range, 1-9 positive lymph nodes). The median number of metastasis-positive lymph nodes was 1 level-10 lymph node (range, 1-2 lymph nodes), 1 level-11 lymph node (range, 1-7 lymph nodes), 1 level-12 lymph node (range, 1-6 lymph nodes), and 1 level-13/14 lymph node (range, 1-7 lymph nodes). The numbers of examined lymph nodes and the numbers of metastasis-positive lymph nodes were approximately equal among lymph nodes at all 4 levels.

Outcome of Patients With Pathologic N1 Status

The 5-year disease-free survival (5DFS) rates of the pN1 and pN0 groups were 50.1% and 90.5%, respectively (P < .0001) (Fig. 2). Disease-free survival curves for the 4 subgroups of pN1 patients were compared with those of pN0 patients (Fig. 2). The 5DFS rate was 69.4% for 48 patients with pN1 disease who had only level 13/14 lymph node metastasis, 46.4% for 74 patients with pN1 disease who had level 12 lymph node metastasis as their highest level, 46.7% for 63 patients who had pN1 disease with level 11 lymph node metastasis as their highest level, and 37% for 45 patients who had pN1 disease with level 10 lymph node metastasis. Patients with pN0 disease had significantly better outcomes than patients with pN1 disease who had only level 13/14 lymph node metastasis (P = .0034). The patients with pN1 disease who had only level 13/14 lymph node metastasis had significantly better outcomes than the patients with pN1 disease who had level 11 or 12 metastasis as their highest level (P = .021). The patients with pN1 disease who had level 11 or 12 lymph node metastasis as their highest level tended to have better outcomes than the patients with pN1 disease who had level 10 lymph node metastasis (P = .064).

Figure 2.

These survival curves illustrate disease-free survival for 700 patients who had pathologic N0 (pN0) status and for 230 patients with pN1 status who underwent lobectomy or pneumonectomy and lymph node dissection for primary nonsmall cell lung carcinoma. Patients with pN1 status were stratified according to the highest lymph node level: level 13/14, segmental/subsegmental lymph nodes (n = 48); level 12, lobar lymph nodes (n = 74); level 11, interlobar lymph nodes (n = 63), and level 10, hilar lymph nodes (n = 45). The curves differed significantly for patients with pN0 status versus patients with pN1 status who had level 13/14 lymph node metastasis (P = .0034) and for patients with pN1 status who had level 13/14 lymph node metastasis versus patients with pN1 status who had level 11 or 12 lymph node metastasis (P = .021), and the curve for patients with pN1 status who had level 11 or 12 lymph node metastasis versus patients with pN1 status who had level 10 lymph node metastasis also trended toward statistical significance (P = .064).

One hundred fourteen patients with pN1 disease who had a single lymph node metastasis had better outcomes than 116 patients with pN1 disease who had multiple lymph node metastasis, irrespective of the lymph node level (P = .036). One hundred forty-eight patients with pN1 disease who had metastasis in at single lymph node level had better outcomes than 82 patients with pN1 disease who had metastasis at ≥2 lymph node levels, irrespective of the number of metastatic lymph nodes (P = .0012). The histopathology of NSCLC also affected outcome; pN1 patients with squamous cell carcinomas tended to have a better 5DFS rate than patients with adenocarcinomas (P = .055) (Table 3).

DISCUSSION

In the current study, the 5DFS rate for patients with pN1 NSCLC was 50.1%, which was within the range (27%-67%) of previously reported 5-year survival rates among patients with pN1 disease,2-22 including the rate calculated by the IASLC (38%).2 Compared with the 5DFS rate in patients with pN0 NSCLC (90.5%), the 5DFS rate in patients with pN1 disease was strikingly lower (50.1%; P < .0001). Furthermore, the current results clearly demonstrate that the outcome of patients with pN1 disease can be stratified according to the highest level of lymph node involvement; patients with pN1 disease who had only level 13/14 lymph node metastasis had a survival rate intermediate between that of patients with pN0 disease and patients with pN1 who had level 11 or 12 lymph node metastasis as their highest level of lymph node involvement. Likewise, the 5DFS rate in patients with pN1 disease who had level 10 lymph node metastasis tended to be lower than the 5DFS rate in patients with pN1 who had level 11 or 12 lymph node metastasis as their highest level. On the basis of these data, N1 lymph node levels were stratified into 3 clinically distinct groups: levels 13/14, 11/12, and 10.

Previous studies have suggested that patients with pN1 who have metastasis to only lymph node levels 12, 13, or 14 have a better prognosis than patients with pN1 who have metastasis to lymph node levels 10 or 11.5-7, 10, 13, 17, 20 However, in the current study, the prognosis did not differ between patients with pN1 disease who had level 11 lymph node metastasis as the highest level and patients who had level 12 lymph node metastasis as the highest level. It should be noted that the number of level 13/14 lymph nodes retrieved in this study may be larger than the number retrieved in previous studies, and the prognosis differed significantly between patients with pN1 disease who had only level 13/14 lymph node metastasis and those who had level 11 or 12 lymph node metastasis.

Previously, it was reported that patients with single-zone pN1 disease,2, 20 single-level pN1 disease,16, 19, 20 single-lymph node pN1 disease,4, 15, 19 direct invasion,6, 18 and microscopic metastasis11, 15 all had a good prognosis along with patients who had lobar N1 disease (levels 12-14).5-7, 10, 13, 17, 20 The number of lymph node metastasis21 and a positive lymph node ratio22 also were reported as prognostic factors. In the current study, we confirmed that single-lymph node pN1 disease and single-level pN1 disease are indicators of a better prognosis in addition to the highest level of lymph node involvement. We also demonstrated that patients with squamous cell carcinoma tend to have better prognosis than those with adenocarcinoma, a finding that, to our knowledge, has not previously been reported.

In this study, we examined in detail, level 13/14 lymph node status. Twenty-one percent of patients with pN1 disease were positive only for level 13/14 lymph nodes, and level 12 and level 13/14 lymph nodes had a higher positive rate of metastasis than level 10 and level 11 lymph nodes. Therefore, detailed cutting and examination of intrapulmonary lymph nodes from formalin-fixed lobectomy/pneumonectomy specimens is very important. We also observed that the median number of examined level 13/14 lymph nodes was 3 (range, 0-22 level 13/14 lymph nodes), which was almost equal to the number of level 10, level 11, and level 12 lymph nodes examined. We believe that the current data can contribute to recent attempts at sentinel lymph node mapping in patients with lung cancer.23-26

In conclusion, the highest level of involved lymph nodes can be used to stratify patients with pN1 NSCLC. The current results demonstrated that patients who had pN1 NSCLC with only level 13/14 lymph node metastasis had a 5DFS rate intermediate between that of patients with pN0 disease and that of patients with pN1 disease who had level 11 or level 12 lymph node metastasis. Therefore, routine examination of intrapulmonary level 13/14 lymph nodes is important for the accurate pathologic staging of this disease.

Acknowledgements

We thank Mr. S. Tamura for technical assistance.

FUNDING SOURCES

This study was supported by the Foundation for Promotion of Cancer Research in Japan.

CONFLICT OF INTEREST DISCLOSURES

The authors made no disclosures.

Ancillary