Asian Cardiovasc Thorac Ann 2008;16:375-380
© 2008 Asia Publishing EXchange Ltd
Prognostic Variability in Completely Resected pN1 Non-Small-Cell Lung Cancer
Alessandro Gonfiotti, MD,
Emanuele Crocetti, MD1,
Andrea Lopes Pegna, MD2,
Eugenio Paci, MD1,
Alberto Janni, MD
Thoracic Surgery Unit University Hospital Careggi
1 Clinical Epidemiology Center for Study and Prevention of Cancer
2 Pneumology Unit, University Hospital Careggi, Florence, Italy
For reprint information contact: Alessandro Gonfiotti, MD Tel: 39 055 7947935 7740 Fax: 39 055 79479357740 Email: agonfiotti{at}hotmail.com, Thoracic Surgery Unit, Azienda Ospedaliero-Universitaria Careggi, Viale Pieraccini n17, 50100, Firenze, Italy.
 |
ABSTRACT
|
|---|
We used the Tuscan Cancer Registry archives to retrieve records of 2,896 patients with a histological diagnosis of lung tumor from January 1996 to December 2000. Of 2,410 patients with non-small-cell lung cancer, 767 (31.8%) underwent complete resection. The following variables were analyzed for their influence on survival in the 157 patients with pathologic N1 status: sex, age, cell type, pathologic tumor status, number and level of involved lymph nodes, tumor grade, and type of surgery. Overall 5-year survival rates were 43.9% for 417 patients with pN0 disease, 10.8% for 176 with pN2 disease, and 31.6% for those with pN1 disease. In pN1 disease, the overall 5-year survival rates for patients with hilar and non-hilar lymph node involvement were 27.4% and 39.6%, respectively. Univariate analysis demonstrated that pathological T status and level of N1 involvement weresignificant prognostic factors. Cox proportional hazards analysis indicated that hilar lymph node involvement was an independent prognostic factor. N1 lymph node status was identified as an independent prognostic factor in a combination of subgroups with different prognoses.
|
INTRODUCTION
|
|---|
The status of lymph node metastasis (pN stage) is one of the most important prognostic factors in non-small-cell lung cancer (NSCLC).1,2 N1 disease represents a heterogeneous group with different lymph node involvement and variable survival rates. It includes patients with early stage disease and those with locally advanced disease, and the 5-year survival rates vary between 45% and 60% when associated with T1 disease, and 35% 50% with T2 disease.3,4 Although the survival rate in N1 disease is intermediate between those of N0 and N2, the prognostic significance of N1 disease has not yet been defined.5,6 Recently, it was suggested that the prognosis for pN1 patients may differ according to the level of lymph node involvement.4,7 We used the Tuscan Cancer Registry to evaluate retrospectively the prognostic significance of N1 disease with hilar lymph node involvement, compared to N1 disease without hilar lymph node involvement.
|
PATIENTS AND METHODS
|
|---|
The Tuscan Cancer Registry is a population-based registry (approximately 1,150,000 inhabitants), active since 1984 in the provinces of Florence and Prato, central Italy. The records of 2,896 patients who had lung tumors diagnosed from January 1996 to December 2000 were retrieved. We excluded 480 patients with small-cell carcinoma and 6 with non-epithelial tumors. Among 2,410 patients with a histological diagnosis of NSCLC, 767 (31.8%) underwent complete resection of the tumor (defined as resection of all macroscopic tumor tissue with resection margins free of disease at microscopic examination). All patients had apparently resectable disease at preoperative staging and thoracotomy. They were staged before surgery according to the 1997 TNM classification, and all dissected lymph nodes were examined pathologically and classified according to the Mountain-Dresler regional lymph node classification.8,9 The pleural reflection was used between the N1 and N2 regions. Patients diagnosed in 1996 were re-staged. All patients included in this study were operated on by the same surgical group (Thoracic Surgery Unit of Florence), and systematic nodal dissection was carried out in 756 (98.6%), using a well-standardized technique: on the right side as en-bloc resection of the upper mediastinal nodes (2R and 4R); the nodes in front of the superior vena cava and/or posterior to the trachea (3A and 3P); on the left side, we removed the subaortic (station 5), paraaortic (station 6), and inferior paratracheal (4L) nodes. Division of the ligamentum arteriosus and mobilization of aortic arch was performed only in cases of pneumonectomy. Subcarinal and lower mediastinal nodes (stations 7, 8, 9) were always removed. As well as mediastinal nodes, the hilar (stations 10, 11) and intrapulmonary lymph nodes were also dissected. After pathological examination of the lymphadenectomy specimen, the number of involved lymph nodes and nodal stations were recorded.
Follow-up was until the end of December 2005, or death if earlier. The Kaplan-Meier product-limit estimate was used for computing overall and variable-specific observed survival probabilities. The influences of the following factors on survival were analyzed: age (< 65 or 
65 years); sex; histologic type (squamous cell carcinoma, others); pathologic T status, type of resection (lobectomy, pneumonectomy); and for pN1, the lymph nodes involved and the number of involved nodes (1, 1+). Lymph node stations in the N1 region were defined according to the Mountain-Dresler regional lymph node classification that distinguishes hilar, interlobar, lobar, and parenchymal nodes.8,9 N1 lymph nodes were further defined as: pulmonary, including lobar (12) and parenchymal (13) nodes; and hilar, including interlobar (11) and main bronchial (10) nodes. A medical oncologist recommended adjuvant chemotherapy in 11 patients and adjuvant radiotherapy in 3; one patient with a Pancoast tumor received radiotherapy before surgery. The log-rank test was used to assess the survival function across groups within each variable.10 For each variable, the hazard of death was evaluated in comparison to the reference class (hazard ratio = 1). All variables considered in the univariate analysis were included in a multivariate Cox proportional hazards model to evaluate their independent effects.
|
RESULTS
|
|---|
Among 767 cases of complete surgical resection, the pathological T status was available for 764: there were 170 (22.1%) pT1, 472 (61.4%) pT2, 105 (13.6%) pT3, and 17 (2.2%) pT4. Analysis of lymph nodes revealed 417 (55.1%) pN0, 157 (20.8%) pN1, 176 (23.2%) pN2, as the anatomic border and 6 (0.7%) pN3.The ages of the 157 patients with pN1 disease ranged from 42.9 to 81.1 years (mean 64.88 ± 8.14 years). By microscopic examination, 82 of these cases were classified as squamous cell and 75 as non-squamous cell carcinoma (Table 1
); the latter included 55 adenocarcinomas, 15 adenosquamous carcinomas, and 5 large cell carcinomas. Of the 157 patients with N1 status, 120 (76.4%) died during the follow-up period; 113 from cancer-related causes (107 lung cancer), 6 from non-cancer-related causes, and 1 died abroad from an unknown cause. The overall 5-year survival rate for pN1 patients was 31.8% (95% confidence interval [CI], 0.24–0.39), and the cause-specific 5-year survival rate was 30.5% (95% CI, 0.23–0.38). The 5-year observed survival rates for the analyzed variables are shown in Table 1. There was no significant difference in survival based on age, sex, number of metastatic nodes, cell type, or surgical method. A significant effect was observed for pathological T status and the level of metastatic nodes (hilar vs pulmonary nodes; Figure 1), while the 5-year survival rate of interlobar (station 11) and hilar (station 10) nodes separately were 30.0% and 26.7%, respectively ( p = 0.07). In Table 2, the univariate prognostic effect of each variable is shown in comparison to the reference categories. There was a significantly increased risk of death with hilar compared to pulmonary lymph node involvement. Multivariate Cox regression analysis revealed an independent prognostic effect for lymph node involvement. Hilar node involvement had an increased risk of death compared to pulmonary involvement, even when adjusted for the other variables (Table 2). The 5-year survival rate for the 417 patients with pN0 disease was 45.0% (95% CI, 0.40–0.49), and for the 176 patients with pN2 disease it was 11.9% (95% CI, 0.07–0.17). The survival rates of pN0, pN1 pulmonary, pN1 hilar and pN2 cases were compared (Figure 2). No differences were found between pulmonary N1 and N0 cases, but there were significant differences in survival between hilar N1 and N2 cases.
View this table:
[in this window]
[in a new window]
|
Table 1. Log-rank Analysis of 5-Year Survival According to Selected Variables in 157 Patients with pN1 Lung Cancer
|
|
View larger version (11K):
[in this window]
[in a new window]
|
Figure 1. Tuscany Cancer Registry: 5-year survival in 157 patients with pN1 lung cancer, based on level of lymph node involvement.
|
|
View this table:
[in this window]
[in a new window]
|
Table 2. Cox Regression Analysis of 5-year Survival According to Selected Variables in 157 Patients with pN1 Lung Cancer
|
|
View larger version (17K):
[in this window]
[in a new window]
|
Figure 2. Tuscany Cancer Registry: 5-year survival in 157 patients with pN1 lung cancer, based on lymph node stations.
|
|
|
DISCUSSION
|
|---|
Staging of lymph node involvement has long been recognized as a key aspect of the management of NSCLC, essential for therapy assignment and prognosis, but it remains an open issue. Several studies in the last 20 years have evaluated the validity of the N descriptors and suggested that these could be refined to provide more accurate prognostic stratification by subdividing them according to specific anatomic locations. 11 The boundary between N1 and N2 lymph nodes remains an important problem, and a new agreement among specialists is necessary to clarify the modality of classification and the real prognosis of these patients. This boundary was initially disregarded by Naruke and colleagues12 and subsequently located distal to the mediastinal pleural reflection; been accepted as the anatomic border between N1 and N2 levels.8,9 Nevertheless, it is very difficult to define the exact site of the pleural reflectionand thus to distinguish between N1 and N2 stations.7 Management of nodes around the main bronchus is important because it is positioned partially within and partially outside the pleural envelope. Consequently, the main problem is to decide whether the main bronchus nodes belong to the N1 or N2 station in terms of prognosis as well as anatomy. Apart from the landmark of the take-off of the upper lobe bronchus from the main bronchus as a boundary between station 4 and 10, the demarcation between the following stations is hardly detectable: the transition zone between stations 4 and 10 at the tracheobronchial angle; and the border between station 7 in the subcarinal area and station nodes 10 on the medial side of the main stem bronchi. Some studies have shown that the prognosis of patients with hilar N1 disease and single-station N2 disease is similar, suggesting that the pleural reflection is not a useful anatomic border between N1 and N2 stations.3
Our analysis showed that the survival of patients with the pleuralreflection has N0 pathological status did not significantly differ from that of patients with N1 pulmonary involvement, while the survival of patients with N2 pathological status was significantly worse than that of patients with N1 hilar involvement, suggesting 3 subgroups of lymph nodes: pN0 pN1 pulmonary, pN1 hilar, and pN2. Recently, numerous studies have evaluated prognostic heterogeneity in patients with pN1 NSCLC, and different prognostic subgroups have been suggested.13 Some results have indicated that patients with hilar lymph node involvement may have a better prognosis than those with N2 disease, but no significant difference was found. 4,6 In other studies, the survival rate in patients with pulmonary lymph node involvement was significantly better than in those with hilar lymph node involvement.3,5 Marra and colleagues14 showed that patients with hilar lymph node metastasis had a 5-year survival rate of 30%. It has been suggested that hilar N1 disease can be regarded as an early form of N2 disease because the poor survival rate in this group compares closely with that in patients with single-station N2 disease.
The most salient message of our study on a population-based case series is that hilar lymph node involvement increases the risk of death compared to pulmonary lymph node involvement, even after multivariate analysis. This finding was suggested by previous studies, but not confirmed by the IASLC Lung Cancer Staging Project in the latest proposals for revision of the N descriptors; it was unable to identify differences in outcome for patients with peripheral vs hilar N1 disease, but found that survival was significantly worse in cases of multiple vs single levels of N1 nodal metastases.12 However, these results provided the impetus for further studies to validate these findings, and the rationale for stratifying patients according with these prognostic groups.
Treatment other than surgery may have a relevant prognostic effect. In our series, 14 patients received adjuvant therapy, and 1 had radiotherapy before surgery. Exclusion of these patients did not modify the results with regard to either the effect of pulmonary or hilar lymph node involvement on survival, or to the similar prognosis for patients with N0 and N1 pulmonary lymph node involvement and those with N2 and N1 hilar lymph node involvement (data not shown). These results may have important clinical implications. The possibility that patients with N1 hilar disease undergo adjuvant therapy in addition to radical resection, as patients with N2 disease, has been hypothesized in some studies.15 Several recent randomized trials of patients with completely resected stages II IIIA NSCLC have confirmed a modest survival benefit with cisplatin-based adjuvant chemotherapy, as demonstrated by the International Adjuvant Lung Cancer Trial Collaborative Group.16 The role of adjuvant chemotherapy for stage IB disease remains controversial, now even more so as the updated results from Cancer and Leukemia Group B are statistically negative.17 Considering the morbidity of adjuvant therapy, it would be important to identify subgroups more at risk of recurrence or with reduced long-term survival.
Our analysis confirms the more favorable prognosis of hilar lymph node disease compared to N2 disease.4,15,18 Further studies on survival according to lymph node pattern should be reported to guide future clinical trial development, treatment strategies, and eventually revision of the TNM staging system.11 In addition, no significant differences in survival rates were observed on the basis of tumor site, sex, age, type of resection, or number of N1 lymph nodes involved, while there was a significant difference on the basis of T classification. The finding pulmonary lymph node involvement had a better prognosis than those with hilar lymph node involvement suggests the presence of two N1 subgroups with different behavior. A redefinition of lymph node involvement, if confirmed by further studies, could have effects on the management of lung cancer patients, and might lead to the adoption of more effective therapeutic approaches.
|
REFERENCES
|
|---|
- Koike T, Tsuchiya R, Goya T, Sohara Y, Miyaoka E. Prognostic factors in 3315 completely resected cases of clinical stage I non-small cell lung cancer in Japan. J Thorac Oncol 2007;2: 408–13.[Medline]
- Watanabe Y. TNM classification for lung cancer. Ann Thorac Cardiovasc Surg 2003;9:343–50.[Medline]
- Okada M, Sakamoto T, Yuki T, Mimura T, Nitanda H, Miyoshi K, et al. Border between N1 and N2 stations in lung carcinoma: lessons from lymph node metastatic patterns of lower lobe tumors. J Thorac Cardiovasc Surg 2005;129:825–30.[Abstract/Free Full Text]
- Osaki T, Nagashima A, Yoshimatsu T, Tashima Y, Yasumoto K. Survival and characteristics of lymph node involvement in patients with N1 non-small cell lung cancer. Lung Cancer 2004;43:151–7.[Medline]
- Riquet M, Manac’h D, Le Pimpec-Barthes F, Dujon A, Chehab A. Prognostic significance of surgical-pathologic N1 disease in non-small cell carcinoma of the lung. Ann Thorac Surg 1999;67: 1572–6.[Abstract/Free Full Text]
- Asamura H, Suzuki K, Kondo H, Tsuchiya R. Where is the boundary between N1 and N2 stations in lung cancer? Ann Thorac Surg 2000;70:1839–46.[Abstract/Free Full Text]
- Rea F, Marulli G, Callegaro D, Zuin A, Gobbi T, Loy M, et al. Prognostic significance of main bronchial lymph nodes involvement in non-small cell lung carcinoma: N1 or N2? Lung Cancer 2004;45: 215–20.[Medline]
- Mountain CF, Dresler CM. Regional lymph node classification for lung cancer staging. Chest 1997;111:1718–23.[Medline]
- Mountain CF. Revisions in the International System for Staging Lung Cancer. Chest 1997;111:1710–7.[Medline]
- Mantel N. Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 1966;50:163–70.[Medline]
- Rusch V, Crowley J, Giroux D, Goldstraw P, Im JG, Tsuboi M, et al. The IASLC Lung Cancer Staging Project: proposals for the revision of the N descriptors in the forthcoming seventh edition of the TNM classification for lung cancer. J Thorac Oncol 2007;2: 603–12.[Medline]
- Naruke T, Suemasu K, Ishikawa S. Lymph node mapping and curability at various levels of metastasis in resected lung cancer. J Thorac Cardiovasc Surg 1978;76:832–9.[Abstract]
- Fujimoto T, Cassivi S, Yang P, Barnes S, Nichols F, Deschamps C, et al. Completely resected N1 non-small cell lung cancer: factors affecting recurrence and long-term survival. J Thorac Cardiovasc Surg 2006;132:499–506.[Abstract/Free Full Text]
- Marra A, Hillejan L, Zaboura G, Fujimoto T, Greschuchna D, Stamatis G. Pathologic N1 non-small cell lung cancer: correlation between pattern of lymphatic spread and prognosis. J Thorac Cardiovasc Surg 2003;125:543–53.[Abstract/Free Full Text]
- Yano T, Yokoyama H, Inoue T, Asoh H, Tayama K, Ichinose Y. Surgical results and prognostic factors of pathologic N1 disease in non-small-cell carcinoma of the lung. Significance of N1 level: lobar or hilar nodes. J Thorac Cardiovascular Surg 1994;107: 1398–402.[Abstract/Free Full Text]
- Arriagada R, Bergman B, Dunant A, Le Chevalier T, Pignon JP, Vansteenkiste J, et al. Cisplatin-based adjuvant chemotherapy in patients with completely resected non-small cell lung cancer. N Engl J Med 2004;350:351–60.[Abstract/Free Full Text]
- Strauss GM, Herndon JE II, Maddaus MA, Johnstone DW, Johnson EA, Watson DM, et al. Adjuvant chemotherapy in stage IB non-small cell lung cancer (NSCLC): Update of Cancer and Leukemia Group B (CALGB) protocol 9633. ASCO Annual Meeting Proceedings. J Clin Oncol 2006;24(Suppl):7007. Available at: http://meeting.ascopubs.org/cgi/content/abstract/24/18_suppl/7007. Accessed February 29, 2008.
- van Velzen E, Snijder RJ, Brutel de la Riviere A, Elbers HJ, van den Bosch JM. Type of lymph node involvement influences survival rates in T1N1M0 non-small cell lung carcinoma. Chest 1996;110:1469–73.[Medline]
TOP
ABSTRACT
INTRODUCTION
