Local Failure in Resected N1 Lung Cancer: Implications for Adjuvant Therapy

International Journal of

Radiation Oncology biology

physics

www.redjournal.org

Clinical Investigation: Thoracic Cancer

Local Failure in Resected N1 Lung Cancer: Implications for Adjuvant Therapy Kristin A. Higgins, M.D.,* Junzo P. Chino, M.D.,* Mark Berry, M.D.,y Neal Ready, M.D.,z Jessamy Boyd, M.D.,x David S. Yoo, M.D., Ph.D.,* and Chris R. Kelsey, M.D.* *Department of Radiation Oncology, yDepartment of Surgery, Division of Cardiovascular and Thoracic Surgery, and z Department of Medicine, Division of Medical Oncology, Duke University Medical Center, Durham, NC; and xUS Oncology, Dallas, TX Received May 19, 2011, and in revised form Jul 1, 2011. Accepted for publication Jul 22, 2011

Summary This study reports actuarial rates of local failure after surgical resection for patients with pathologic N1 non-small cell lung cancer. Local failure was 40% at five years. Multivariate analysis identified three predictive factors: visceral pleural invasion, increasing number of positive N1 lymph nodes, and VATS resection. Radiation fields are proposed based on observed patterns of failure in this study.

Purpose: To evaluate actuarial rates of local failure in patients with pathologic N1 nonesmallcell lung cancer and to identify clinical and pathologic factors associated with an increased risk of local failure after resection. Methods and Materials: All patients who underwent surgery for nonesmall-cell lung cancer with pathologically confirmed N1 disease at Duke University Medical Center from 1995 e2008 were identified. Patients receiving any preoperative therapy or postoperative radiotherapy or with positive surgical margins were excluded. Local failure was defined as disease recurrence within the ipsilateral hilum, mediastinum, or bronchial stump/staple line. Actuarial rates of local failure were calculated with the Kaplan-Meier method. A Cox multivariate analysis was used to identify factors independently associated with a higher risk of local recurrence. Results: Among 1,559 patients who underwent surgery during the time interval, 198 met the inclusion criteria. Of these patients, 50 (25%) received adjuvant chemotherapy. Actuarial (5year) rates of local failure, distant failure, and overall survival were 40%, 55%, and 33%, respectively. On multivariate analysis, factors associated with an increased risk of local failure included a video-assisted thoracoscopic surgery approach (hazard ratio [HR], 2.5; p Z 0.01), visceral pleural invasion (HR, 2.1; p Z 0.04), and increasing number of positive N1 lymph nodes (HR, 1.3 per involved lymph node; p Z 0.02). Chemotherapy was associated with a trend toward decreased risk of local failure that was not statistically significant (HR, 0.61; p Z 0.2). Conclusions: Actuarial rates of local failure in pN1 disease are high. Further investigation of conformal postoperative radiotherapy may be warranted. Ó 2012 Elsevier Inc. Keywords: Local failure, N1, Lung cancer, Postoperative radiotherapy

Reprint requests to: Kristin A. Higgins, M.D., Duke University Medical Center, DUMC Box 3085, Durham, NC 27710. Tel: (919) 668-1479; Fax: (919) 668-7345; E-mail: [email protected] Int J Radiation Oncol Biol Phys, Vol. 83, No. 2, pp. 727e733, 2012 0360-3016/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.ijrobp.2011.07.018

Conflict of interest: none.

728

Higgins et al.

Introduction Numerous randomized studies have been conducted examining the role of adjuvant chemotherapy and/or radiotherapy (RT) for resected nonesmall-cell lung cancer (NSCLC). Three recent trials have shown a survival benefit for adjuvant chemotherapy in Stage II and IIIA NSCLC based on the prior TNM staging system, where Stage II and Stage IIIA represented N1 and N2 disease, respectively (1e3). This was further supported by the LACE (Lung Adjuvant Cisplatin Evaluation) meta-analysis, which showed that adjuvant cisplatin-based chemotherapy improves overall survival by approximately 5.4% (4). Even with this modest improvement in survival with adjuvant chemotherapy, long-term survival remains suboptimal, largely because of local and distant disease recurrence. For patients with resected N2 NSCLC, where the tumor has spread to the mediastinal lymph nodes, adjuvant chemotherapy, often with postoperative RT, is routinely recommended. For patients with N1 disease, where hilar lymph node metastases are present but the mediastinal lymph nodes appear uninvolved, adjuvant chemotherapy is recommended without postoperative RT. The latter recommendation is largely based on findings from the PORT (Postoperative Radiotherapy) meta-analysis, in which postoperative RT was associated with inferior survival compared with surgery alone (5). Furthermore, the risk of local failure is generally considered to be low and overshadowed by the risk of distant metastases. Because the population of Stage II patients is small, relative rates and patterns of failure may not be well recognized. This study evaluates a large, single-institution experience managing patients with resected lung cancer to quantitate the actuarial risk of local failure and define clinical and pathologic risk factors for local failure within the N1 subset of patients. Ultimately, we aim to identify a population that is at particularly high risk and would be suitable for further study of postoperative RT.

Methods and Materials This institutional review boardeapproved study identified all patients who underwent surgery for T1e3N1 NSCLC at Duke University between 1995 and 2008. Patients were excluded if they received preoperative chemotherapy or preoperative or postoperative RT, had positive surgical margins, or had a history of lung cancer. Patients with positive surgical margins were excluded because the goal of this study was to determine risk of local failure imparted by N1 positivity. Positive margins would increase local failure substantially, thereby confounding results. Similarly, patients receiving postoperative RT were excluded because postoperative RT would be expected to decrease the overall risk of local failure and confound the analysis of risk factors. Medical records and pertinent radiologic imaging were reviewed to characterize each patient’s demographic information, obtain surgical and pathologic details, and score patterns of failure after surgery. The pathologic staging of patients in this cohort reflects the updated TNM staging for NSCLC (6). The primary objective of this study was to assess the risk of local recurrence after surgery for patients with pN1 disease, as well as to identify clinical, surgical, and pathologic features associated with a higher risk of local recurrence. Disease recurrence at the surgical resection margin, ipsilateral hilum, and/or mediastinum was considered a local (i.e., local/regional)

International Journal of Radiation Oncology  Biology  Physics recurrence. All other sites of failure, including the supraclavicular fossa and contralateral hilum, were considered distant recurrences. We also did not score ipsilateral parenchymal lung recurrences as local failures, unless they were at the surgical margin, because these are more likely to represent distant hematogenous metastases. Patterns of failure were assessed by means of follow-up imaging studies and information obtained from procedures such as computed tomography (CT)eguided transthoracic biopsies, bronchoscopy, endobronchial ultrasound, and mediastinoscopy. Nodal failures in the ipsilateral hilum and mediastinum were defined as a new or enlarging lymph node of 1 cm or greater on short-axis CT slices and/or hypermetabolic on positron emission tomography imaging, which in the patient’s subsequent clinical follow-up was consistent with disease progression. All cases of possible local and distant recurrence were reviewed by two authors (C.R.K. and J.B. or K.A.H.) to ensure accuracy. All patients had routine postsurgical surveillance with imaging studies, including chest CT, but frequency and choice of imaging modality were not standardized.

Patterns of local failure A secondary analysis was performed to evaluate whether more customized radiation fields could potentially replace traditional postoperative RT fields, which typically encompass the entire mediastinum. Assuming that the detrimental effect of RT in the PORT meta-analysis arose from normal tissue injury, smaller RT fields may improve the therapeutic ratio. On the basis of lymphoscintigraphy studies, mediastinoscopy evaluations, and patterns of failure (7e9), right-sided tumors primarily spread to ipsilateral mediastinal lymph node stations, only in frequently spreading in a contralateral manner. On the other hand, left-sided tumors, particularly lower lobe tumors, have a greater predilection to spread to the contralateral mediastinum. We hypothesized that for patients with right-sided pN1 disease, fields directed at the bronchial stump, ipsilateral hilum, and Levels 4R and 7 in the mediastinum would encompass the majority of failures. For patients with left-sided tumors, coverage of the bronchial stump, ipsilateral hilum, and Levels 5, 4L, 4R, and 7 was predicted to be sufficient.

Statistical analysis The Kaplan-Meier productelimit method (10, 11) was used to estimate 5-year recurrence probabilities and confidence intervals. Progression-free survival was calculated from the date of surgery to the date of treatment failure (defined as local and/or distant recurrence, development of a second primary lung cancer, or death from any cause). Time to local failure and time to distant failure were calculated from the date of surgery to the date of local recurrence and the date of distant recurrence, respectively. For these analyses, patients with development of what was believed to be a second primary lung cancer were censored on the date the second primary malignancy was diagnosed, because it is impractical to distinguish a local failure arising from the first vs. the second lung cancer. Local and distant recurrences were scored independently (i.e., patients with development of a distant recurrence were not censored for local failure but were assessed for local failure until the date of last follow-up or death). A Cox multivariate analysis was also performed to assess for clinical and pathologic factors associated with an increased risk

Volume 83  Number 2  2012 of local failure after resection. Variables were included in the multivariate model if they achieved p  0.2 on univariate analysis. Adjuvant chemotherapy was also added to the multivariate model, with the knowledge that the chemotherapy decreases the overall risk of disease recurrence in resected NSCLC. A Fisher exact test was used to make comparisons between failure patterns in patients undergoing video-assisted thoracoscopic surgery (VATS) resection compared those undergoing an open approach.

Results A total of 1,559 patients were identified who underwent initial surgery for a new lung cancer diagnosis between 1995 and 2008. Among these, 235 were pathologically deemed to have N1 tumors. Patients with positive surgical margins (n Z 16) or who received postoperative RT (n Z 21) were excluded, leaving 198 evaluable patients. Median follow-up was 24 months for all patients and 36 months for surviving patients. Table 1 summarizes patient characteristics. Most (73%) underwent open resection, but approximately a quarter of patients underwent a VATS resection, which was used with increasing frequency during the latter years of the study period. Nearly all patients (97%) underwent lobectomy or pneumonectomy. The median number of sampled N1 lymph nodes was 6 (range, 1e34). The median number of involved N1 lymph nodes was 1 (range, 1e11). Mediastinal lymph nodes were sampled or dissected in 93% of patients. The median number of mediastinal lymph node stations sampled was 4 (range, 0e8). The median number of mediastinal lymph nodes sampled for patients undergoing VATS resection was 4, whereas the median number of mediastinal lymph nodes sampled for patients undergoing open resection was 3. Postoperative death (within 30 days of surgery or during initial hospitalization) occurred in 9 patients (5%). For all patients, 3- and 5-year overall survival rates were 53% and 33%, respectively, whereas 3- and 5-year progression-free survival rates were 40% and 23%, respectively (Fig. 1). The most common pattern of failure was distant metastases. Considering only first failures, 56% of failures involved distant sites only, 19% involved distant plus local sites, and 25% were confined to local sites. The actuarial risk of distant metastases developing was 40% at 3 years and 55% at 5 years. The actuarial risk of local failure was 28% at 3 years and 40% at 5 years (Fig. 1). Table 2 shows the results of univariate and multivariate analyses evaluating factors associated with local failure in patients with pathologic N1 disease. Independent factors associated with local failure on multivariate analysis included VATS resection (hazard ratio [HR], 2.5; p Z 0.01), visceral pleural invasion (HR, 2.1; p Z 0.04), and increasing number of positive N1 nodes (HR, 1.3 per involved lymph node; p Z 0.02). Chemotherapy was not associated with decreased risk of local failure (HR, 0.61; p Z 0.2). As a result of the difficulty distinguishing ipsilateral hilar from bronchial stump recurrences radiographically, because these two sites are in close proximity, these failures were scored together. The most common site of failure was the mediastinum, with 53% of local failures involving the mediastinal lymph nodes alone and an additional 21% of failures involving the mediastinum and the bronchial stump/ipsilateral hilum. An additional 26% of local failures involved the ipsilateral hilum/bronchial stump without

Rates and risk factors of local recurrence in pN1 NSCLC Table 1

729

Patient characteristics

Characteristic Age (yr) Median Range Race African American White Other Sex Male Female Pathologic stage (TNM 7th edition) T1aN1 T1bN1 T2aN1 T2bN1 T3N1 TxN1 Histology Adenocarcinoma Squamous cell Large cell Bronchoalveolar Adenosquamous NSCLC NOS Surgical approach Open VATS Surgical procedure Wedge/segmentectomy Lobectomy* Pneumonectomy Mediastinal sampling/dissection Yes No Lymphovascular space invasion Yes No Unknown Visceral pleural invasion Yes No Unknown

Data 65 39e91 26 (13%) 168 (85%) 4 (2%) 131 (66%) 67 (34%) 26 33 92 19 27 1

(13%) (17%) (47%) (10%) (14%) (.5%)

79 85 20 2 2 10

(40%) (43%) (10%) (1%) (1%) (5%)

144 (73%) 55 (28%) 6 (3%) 150 (76%) 42 (21%) 184 (93%) 14 (7%) 79 (40%) 107 (54%) 12 (6%) 63 (32%) 128 (65%) 7 (3%)

Abbreviations: NSCLC NOS Z nonesmall-cell carcinoma not otherwise specified; VATS Z video-assisted thoracoscopic surgery. * Including bilobectomy and sleeve lobectomy.

mediastinal involvement. When considering patterns of failure between patients undergoing VATS and open resection, we found no statistically significant differences (Table 3). CT and/or positron emission tomography images could be retrieved from archived storage to document station-specific mediastinal failures in 29 of the 38 patients with pN1 disease in whom a local recurrence developed. Patterns of failure are shown in Fig. 2, which also shows proposed lobe-specific postoperative RT fields based on patterns of lymphatic drainage and regional nodal spread. As shown, 25 of 29 local failures (86%) would have been encompassed by these more conformal radiation fields outlined in the “Methods and Materials” section.

730

International Journal of Radiation Oncology  Biology  Physics

Higgins et al. 1.0

1.0

0.9

0.9

Fraction Free of Recurrence

Fraction Surviving

0.8 Overall Survival Progression Free Survival

0.7 0.6 0.5 0.4 0.3 0.2

0.8 0.7 0.6 0.5 0.4 Locoregional Control Freedom from Distant Metastases

0.3 0.2 0.1

0.1

number at risk 198 151

105

78

54

39

0

1

2

3

4

5

number at risk 198 151

105

78

54

39

3

4

5

0.0

0.0

0

1

Years

2

Years

Fig. 1. Left, Kaplan-Meier curves showing overall survival and progression-free survival for entire cohort. Right, Kaplan-Meier curves showing local control and to freedom from distant metastases for the entire cohort.

Table 2

Univariate and multivariate analyses for local recurrence Univariate analysis Factors*

Surgical procedure Wedge/segmentectomy Lobectomy or more Lymphovascular invasion Yes No Age (per yr) Tumor size (per cm) Visceral pleural invasion Yes No Grade Moderate/high Low Histology Squamous or large cell Nonsquamous VATS approach Yes No Adjuvant chemotherapy Yes No Malesex Yes No No. of N1 LNs sampled (per node) No. of involved N1 LNs (per node) No. of N2 LN stations sampled (per station) No. of N2 LNs sampled (per node)

Multivariate analysis

Hazard ratio

95% CI

p Value

Hazard ratio

95% CI

p Value

1.23

0.17e8.95

0.73

1.12

0.55e1.30

0.74

1.02 0.88

0.99e1.06 0.73e1.06

0.23 0.18

0.89

0.72e1.09

0.27

1.65

0.86e3.16

0.13

2.08

1.03e4.19

0.04

0.93

0.65e2.31

0.95

1.22

0.65e2.32

0.53

2.63

1.39e4.99

<0.01

2.48

1.21e5.07

0.01

0.75

0.36e1.54

0.43

0.61

0.29e1.31

0.20

1.29

0.64e2.60

0.48

1.00 1.18 0.94 1.04

0.93e1.07 0.94e1.49 0.77e1.14 0.77e1.39

0.97 0.16 0.51 0.82

1.32

1.04e1.67

0.02

Abbreviations: CI Z confidence interval; LN Z lymph node; VATS Z video-assisted thoracoscopic surgery. * Second factor listed is the reference set.

Volume 83  Number 2  2012 Table 3

Rates and risk factors of local recurrence in pN1 NSCLC

731

Patterns of local failure Site of failure

All patients (n Z 198)

VATS (n Z 54)

Open (n Z 144)

p Value*

Mediastinum Ipsilateral hilum/bronchial stump Mediastinum and ipsilateral hilum/bronchial stump

20 (53%) 10 (26%) 8 (21%)

9 (50%) 5 (28%) 4 (22%)

11 (55%) 5 (25%) 4 (20%)

>0.99 >0.99 >0.99

Abbreviation: VATS Z video-assisted thoracoscopic surgery. * Comparison between VATS and open lobectomy.

Discussion Adjuvant chemotherapy is a standard of care in fit patients with pathologic N1 NSCLC after resection (1e4), given a proven overall survival benefit in multiple randomized trials. Despite administration of adjuvant chemotherapy, disease-free survival is suboptimal, with local failures contributing significantly to overall high rates of recurrence.

Fig. 2. Patterns of local/regional failure in 29 patients with pN1 nonesmall-cell lung cancer. Proposed radiation fields are shown based on lobe-specific patterns of lymphatic spread. Top, Fields for right lung; bottom, fields for left lung. The asterisks indicate a prevascular (Level 3) failure. It should be noted that the total number of failures is greater than number of patients with recurrence because some patients had failures at multiple local/regional sites.

Most randomized trials evaluating chemotherapy have not reported patterns of failure. Thus the true risk of local failure after surgical resection of N1 NSCLC is poorly understood. Differing definitions of local failure only add to this general ambiguity (12). Our study clearly defines local failure as those that would have been encompassed in a conventional postoperative RT field (i.e., the surgical margin, ipsilateral hilum, and/or mediastinum). In this study, which is the largest evaluating local recurrences in the subset of patients with N1 disease, the actuarial risk of local failure was 40% at 5 years. Prior studies evaluating local failure after surgery for pathologic N1 disease show rates ranging from 20% to 46% (13e18). These studies are summarized in Table 4. To further clarify which patients may be at greatest risk of local failure, a multivariate analysis was performed, showing an association between increasing number of positive N1 lymph nodes, visceral pleural invasion, and a VATS approach with increased risk of local failure. Thus patients with these risk factors may be at particularly high risk of local recurrence after resection. Visceral pleural invasion of the primary tumor has been shown in multiple studies to be a poor prognostic factor for both recurrence-free and overall survival (19e21) and is recognized as such by the present TNM staging system. Of the prior studies evaluating the N1 subset, only Luzzi et al. (16) and Varlotto et al. (22) evaluated visceral pleural invasion as a potential prognostic factor, which in their experience was not independently associated with local failure. In addition, in our analysis increasing number of involved N1 nodes was associated with a higher risk of local failure in our experience. This finding is not surprising because it reflects an overall larger tumor burden present, and multi-station vs. single-station N1 involvement has been shown to be a poor prognostic factor for overall survival in a prior study (23). However, this has not been observed as an independent factor for increased risk of local failure in other studies, including those of Varlotto et al., Sawyer et al. (13), and Luzzi et al. Other studies summarized in Table 4 have identified potential prognostic factors for local recurrence. Sublobar resections are a well-recognized risk factor for local recurrence (24). We did not observe this in our series, but only 3% of our patients (n Z 6) underwent a wedge resection or segmentectomy. Mediastinal lymph node sampling has been associated with a higher risk of local recurrence compared with dissection (14) in N1 patients. We were unable to evaluate this optimally, because it was often difficult to retrospectively confirm the extent of mediastinal lymph node clearance. We observed no statistically significant differences in local failure with the use of chemotherapy. Another study noted an increased risk of local failure with the use of chemotherapy (22). Differing definitions of local failure and manner of scoring failures may be explanatory. In general, the available data regarding adverse prognostic factors for

732

International Journal of Radiation Oncology  Biology  Physics

Higgins et al.

Table 4

Rates of local recurrence after resection in pN1 NSCLC Study

Type

n

Local failure

Adverse prognostic factors

Sawyer et al. (13)

Retrospective

107

38% (5-yr actuarial)

Luzzi et al. (16) Kim et al. (17) Lardinois et al. (14) Immerman et al. (15) Feng et al. (18) Varlotto et al. (22) Our study

Retrospective Retrospective Retrospective Retrospective Prospective Retrospective Retrospective

116 98 23 22 82 60 198

25% 20% 35% 41% 29% 46% 40%

Negative bronchoscopy, tumor >3 cm, 10 N1 nodes dissected Lobectomy (vs. pneumonectomy) Sleeve resection N2 sampling vs. dissection NS NS Chemotherapy VATS, visceral pleural invasion, increasing No. of involved N1 nodes

(crude) (crude) (crude) (crude) (crude) (5-yr actuarial) (5-yr actuarial)

Abbreviations: NSCLC Z nonesmall-cell lung cancer; NS Z not stated; VATS Z video-assisted thoracoscopic surgery.

local recurrence in the N1 subset of patients are conflicting. However, this point becomes less relevant when one considers that the actuarial risk of local failure is very high (40% at 5 years) for the entire N1 subset in our study. Although we observed an association between VATS resection and local recurrence in patients with N1 disease, when all 1,559 patients were analyzed, VATS resection was not independently associated with local recurrence (data not shown). Prior studies evaluating VATS resection have been favorable, showing decreased intraoperative and postoperative morbidity with VATS resection relative to open lobectomy with similar survival (25). Whether there is an oncologic rationale for why VATS may be less effective for patients with N1 disease is not clear. It is plausible that VATS resection could lead to less comprehensive lymph node sampling and thus a higher number of patients with residual regional disease. In our study, however, the median number of mediastinal lymph node stations sampled was higher in the VATS group; thus this is not a valid explanation for increased local failure. However, extent of sampling is unknown (i.e., sampling vs. dissection), and it is plausible that patients undergoing VATS resection had a more limited mediastinal resection. That said, patterns of local failure were not significantly different between patients undergoing VATS and those undergoing open resection in our study. Given the very high risk of local failure after surgery for N1 NSCLC, further evaluation of postoperative RT may be warranted, particularly in those who are considered to be at highest risk of local recurrence. Given the findings of the PORT meta-analysis showing a 21% relative reduction in survival in patients receiving RT after resection, as well as other retrospective studies showing a detriment in N1 disease (26, 27), it will be a challenge to enroll such patients in future studies. It should be remembered that a 1995 meta-analysis showed that adjuvant chemotherapy (longterm alkylating agents) was associated with a detriment in survival similar to that seen by postoperative RT. Both chemotherapy and RT have been optimized over the last 20 years. Nevertheless, postoperative RT has not been rigorously tested in a randomized setting with contemporary beam energies, conformal treatment planning, and use of strict dosimetric parameters that are routinely used today. Furthermore, our data suggest that smaller, more customized RT fields may be adequate in patients with N1 disease. This will need to be validated by use of other datasets. Evidence exists supporting the concept that improved local control leads to improved overall survival in NSCLC. Saunders et al. (28) showed improved local control and overall survival with

continuous, hyperfractionated accelerated thoracic RT for locally advanced NSCLC. Furthermore, a recent meta-analysis evaluating the effects of concurrent vs. sequential chemoradiotherapy in locally advanced NSCLC showed a 4.5% overall survival benefit to concurrent chemoradiotherapy at 5 years, presumably because of a 6.1% improvement in local control at 5 years. There were no differences in rates of distant metastases (29). Given the high risk of local recurrence in resected N1 NSCLC, despite adjuvant chemotherapy, it would seem rational to reexplore postoperative RT by use of modern radiation techniques, perhaps with the use of customized fields. In addition to the biases against postoperative RT in general, another obstacle includes the small number of patients with N1 disease. If high-risk subsets could be identified, this would enrich a population that would most likely benefit from postoperative RT but further reduce the potential pool of eligible patients. Clearly, multi-institutional or cooperative group studies would be required.

Conclusions In our series the risk of local recurrence after surgery for N1 NSCLC was 40%. Clinical and pathologic features associated with increased risk of local failure include visceral pleural invasion, increasing number of positive N1 lymph nodes, and VATS resection. Customized RT fields would encompass the majority of local failures observed in this study.

References 1. Arriagada R, Bergman B, Dunant A, et al. Cisplatin-based adjuvant chemotherapy in patients with completely resected non-small-cell lung cancer. N Engl J Med 2004;350:351e360. 2. Douillard JY, Rosell R, De Lena M, et al. Adjuvant vinorelbine plus cisplatin versus observation in patients with completely resected stage IB-IIIA non-small-cell lung cancer (Adjuvant Navelbine International Trialist Association [ANITA]): A randomised controlled trial. Lancet Oncol 2006;7:719e727. 3. Winton T, Livingston R, Johnson D, et al. Vinorelbine plus cisplatin vs. observation in resected non-small-cell lung cancer. N Engl J Med 2005;352:2589e2597. 4. Pignon JP, Tribodet H, Scagliotti GV, et al. Lung adjuvant cisplatin evaluation: A pooled analysis by the LACE Collaborative Group. J Clin Oncol 2008;26:3552e3559.

Volume 83  Number 2  2012 5. Postoperative radiotherapy in non-small-cell lung cancer: Systematic review and meta-analysis of individual patient data from nine randomised controlled trials. PORT Meta-analysis Trialists Group. Lancet 1998;352:257e263. 6. American Joint Committee on Cancer. Cancer staging manualdlung. 7th ed. New York: Springer; 2010. 7. Kelsey CR, Light KL, Marks LB. Patterns of failure after resection of non-small-cell lung cancer: Implications for postoperative radiation therapy volumes. Int J Radiat Oncol Biol Phys 2006;65:1097e1105. 8. Hata E, Hayakawa K, Miyamoto H, et al. Rationale for extended lymphadenectomy for lung cancer. Theoretical Surgery 1990;5:19e25. 9. Nohl-Oser HC. An investigation of the anatomy of the lymphatic drainage of the lungs as shown by the lymphatic spread of bronchial carcinoma. Ann R Coll Surg Engl 1972;51:157e176. 10. Collett D. Modeling survival data in medical research. 2nd Ed. London: Chapman & Hall; 2003. 11. Therneau TM, Grambsch PM. Modeling survival data. New York: Springer; 2000. 12. Kelsey CR, Marks LB, Wilson LD. Postoperative radiation therapy for lung cancer: Where do we stand? Oncology (Williston Park) 2008;22: 301e310. discussion 310, 314e305, 319. 13. Sawyer TE, Bonner JA, Gould PM, et al. Factors predicting patterns of recurrence after resection of N1 non-small cell lung carcinoma. Ann Thorac Surg 1999;68:1171e1176. 14. Lardinois D, Suter H, Hakki H, et al. Morbidity, survival, and site of recurrence after mediastinal lymph-node dissection versus systematic sampling after complete resection for non-small cell lung cancer. Ann Thorac Surg 2005;80:268e274. discussion 274e265. 15. Immerman SC, Vanecko RM, Fry WA, et al. Site of recurrence in patients with stages I and II carcinoma of the lung resected for cure. Ann Thorac Surg 1981;32:23e27. 16. Luzzi L, Voltolini L, Campione A, et al. Pneumonectomy vs lobectomy in the treatment of pathologic N1 NSCLC: Could the type of surgical resection dictate survival? J Cardiovasc Surg (Torino) 2003; 44:119e123. 17. Kim YT, Kang CH, Sung SW, et al. Local control of disease related to lymph node involvement in non-small cell lung cancer after sleeve lobectomy compared with pneumonectomy. Ann Thorac Surg 2005; 79:1153e1161. discussion 1153e1161. 18. Feng QF, Wang M, Wang LJ, et al. A study of postoperative radiotherapy in patients with non-small-cell lung cancer: A randomized trial. Int J Radiat Oncol Biol Phys 2000;47:925e929.

Rates and risk factors of local recurrence in pN1 NSCLC

733

19. Hsu CP, Hsia JY, Chang GC, et al. Surgical-pathologic factors affect long-term outcomes in stage IB (pT2 N0 M0) non-small cell lung cancer: A heterogeneous disease. J Thorac Cardiovasc Surg 2009;138: 426e433. 20. Yilmaz A, Duyar SS, Cakir E, et al. Clinical impact of visceral pleural, lymphovascular and perineural invasion in completely resected non-small cell lung cancer. Eur J Cardiothorac Surg 2011; 40:664e670. 21. Schuchert MJ, Schumacher L, Kilic A, et al. Impact of angiolymphatic and pleural invasion on surgical outcomes for stage I non-small cell lung cancer. Ann Thorac Surg 2011;91:1059e1065. 22. Varlotto JM, Medford-Davis LN, Recht A, et al. Failure rates and patterns of recurrence in patients with resected N1 non-small-cell lung cancer. Int J Radiat Oncol Biol Phys 2011;81:353e359. 23. Osaki T, Nagashima A, Yoshimatsu T, et al. Survival and characteristics of lymph node involvement in patients with N1 non-small cell lung cancer. Lung Cancer 2004;43:151e157. 24. Ginsberg RJ, Rubinstein LV. Randomized trial of lobectomy versus limited resection for T1 N0 non-small cell lung cancer. Lung Cancer Study Group. Ann Thorac Surg 1995;60:615e622. discussion 622e613. 25. Paul S, Altorki NK, Sheng S, et al. Thoracoscopic lobectomy is associated with lower morbidity than open lobectomy: A propensitymatched analysis from the STS database. J Thorac Cardiovasc Surg 2010;139:366e378. 26. Douillard JY, Rosell R, De Lena M, et al. Impact of postoperative radiation therapy on survival in patients with complete resection and stage I, II, or IIIA non-small-cell lung cancer treated with adjuvant chemotherapy: The adjuvant Navelbine International Trialist Association (ANITA) Randomized Trial. Int J Radiat Oncol Biol Phys 2008; 72:695e701. 27. Lally BE, Zelterman D, Colasanto JM, et al. Postoperative radiotherapy for stage II or III non-small-cell lung cancer using the surveillance, epidemiology, and end results database. J Clin Oncol 2006;24:2998e3006. 28. Saunders M, Dische S, Barrett A, et al. Continuous hyperfractionated accelerated radiotherapy (CHART) versus conventional radiotherapy in non-small-cell lung cancer: A randomised multicentre trial. CHART Steering Committee. Lancet 1997;350:161e165. 29. Auperin A, Le Pechoux C, Rolland E, et al. Meta-analysis of concomitant versus sequential radiochemotherapy in locally advanced non-small-cell lung cancer. J Clin Oncol 2010;28:2181e2190.