0
Original Research: Lung Cancer |

Identification of Stage I Non-small Cell Lung Cancer Patients at High Risk for Local Recurrence Following Sublobar ResectionSublobar Resection at High Risk for Recurrence FREE TO VIEW

John M. Varlotto, MD; Laura N. Medford-Davis, MD; Abram Recht, MD; John Flickinger, MD; Nengliang Yao, PhD; Clayton Hess, MD; Michael F. Reed, MD, FCCP; Jennifer Toth, MD; Dani S. Zander, MD, FCCP; Malcolm M. DeCamp, MD, FCCP
Author and Funding Information

From the Penn State Hershey Cancer Institute (Dr Varlotto); Pennsylvania State College of Medicine (Dr Hess); Penn State Heart and Vascular Institute (Dr Reed); Penn State Hershey Pulmonary Medicine (Dr Toth); and Penn State Hershey Medical Center (Dr Zander), Hershey, PA; Harvard Medical School (Drs Medford-Davis and Recht), Boston, MA; Department of Radiation Oncology (Dr Recht), Beth Israel Deaconess Medical Center, Boston, MA; Department of Radiation Oncology (Dr Flickinger), Pittsburgh Cancer Institute, Pittsburgh, PA; Department of Health Policy and Administration (Dr Yao), Pennsylvania State University, University Park, PA; and Division of Thoracic Surgery (Dr DeCamp), Department of Surgery, Northwestern Memorial Hospital, Chicago, IL.

Correspondence to: John M. Varlotto, MD, Penn State Hershey Cancer Institute, Radiation Oncology – CH63, 500 University Dr, PO Box 850, Hershey, PA 17033-0850; e-mail: jvarlotto@hmc.psu.edu


Funding/Support: The authors have reported to CHEST that no funding was received for this article.

Reproduction of this article is prohibited without written permission from the American College of Chest Physicians. See online for more details.


Chest. 2013;143(5):1365-1377. doi:10.1378/chest.12-0710
Text Size: A A A
Published online

Objective:  An increasing proportion of patients with stage I non-small cell lung cancer (NSCLC) is undergoing sublobar resection (L−). However, there is little information about the risks and correlates of local recurrence (LR) after such surgery, especially compared with patients undergoing lobectomy (L+).

Methods:  Ninety-three and 318 consecutive patients with stage I NSCLC underwent L− and L+, respectively, from 2000 to 2006. Median follow-up was 34 months.

Results:  In the L− group, the LR rates at 2, 3, and 5 years were 13%, 24%, and 40%, respectively. The risk of LR was significantly associated with tumor grade, tumor size, and T stage. The crude risk of LR was 33.8% (21 of 62) for patients whose tumors were grade ≥ 2. In the L+ group, the LR rates at 2, 3, and 5 years were 14%, 19%, and 24%, respectively. The risk of LR significantly increased with increasing tumor size, length of hospital stay, and the presence of diabetes. The L− group experienced a significant increase in failure in the bronchial stump/staple line compared with the L+ group (10% vs 3%; P = .04) and nonsignificant trends toward increased ipsilateral hilar and subcarinal failure rates.

Conclusions:  Patients with stage I NSCLC who undergo L− have an increased risk of LR compared with patients undergoing L+, particularly when they have tumors grade ≥ 2 or tumor size > 2 cm. If L− is considered, additional local therapy should be considered to reduce this risk of LR, especially with tumors grade ≥ 2 or size > 2 cm.

Figures in this Article

In 1995, the Lung Cancer Study Group (LCSG) reported the results of the only randomized trial comparing sublobar resection (L−) and lobectomy (L+) resections in patients with stage IA non-small cell lung cancer (NSCLC).1 The principle findings in this study were an increased recurrence rate in patients undergoing L− (0.094/person/y vs 0.058/person/y in the L+ group) with no significant effects on perioperative morbidity, mortality, late pulmonary function, or lung cancer deaths. However, the authors did not report whether L− was associated with different failure patterns, nor did they report if there were different factors associated with LR in the surgical groups.

Our investigation had several purposes. First, we wanted to assess whether the incidence of L− is rising in the United States and whether the proportion of patients with stage I NSCLC undergoing this surgical procedure was affected by the results from the LCSG. We also wanted to assess whether the recurrence patterns and recurrence factors were different between the L− and L+ groups. Furthermore, we wanted to see if we could identify a low-risk group for which a L− can be recommended. Additionally, we examined our results in the context of the current literature on the subject of L− since the publication of the LCSG trial in 1995. Particularly, we investigated how tumor factors (size, location [peripheral vs proximal], percentage with bronchioloalveolar carcinoma [BAC], percentage with lymphatic vascular invasion [LVI]) and treatment factors (percentage wedge and segmental resections, and extent of lymphadenectomy), as well as definition of local recurrence (LR) affected overall survival (OS), recurrence, and recommendations for L− vs L+. We emphasize how the restricted definitions of LR, patient selection, and investigative methodologies may impact the recurrence rates and the recommendations for L− or L+.

This retrospective study was conducted with approval of the institutional review boards (IRBs) of the Penn State Hershey Cancer Institute (Penn State Hershey IRB, protocol #29311), Beth Israel Deaconess Medical Center (Committee on Clinical Investigations, protocol #2006-P-000453/2), and the Denver and Boston Department of Veterans Affairs (VA) hospitals (Boston VA Healthcare IRB, protocol #2008). Hospital and departmental tumor registries were used to identify all patients undergoing potentially curative resection of stage I-IIIA NSCLC from 2000 to 2006. Patients were included in the study population if they had postoperative follow-up of at least 3 months, no second primary cancer diagnosed within 5 years of the index lung cancer, and no neoadjuvant or adjuvant radiotherapy. Medical records were reviewed for each patient to ensure eligibility. Of the 541 patients accrued, 428 had stage I NSCLC. Patients with stage I NSCLC undergoing a L− (wedge resection [WR] or segmentectomy [SMR], n = 93 patients) or L+ or bilobectomy (n = 318 patients) were selected for this study.

The main purposes of our retrospective database for this study were to assess the differences between L− and L+ in terms of Kaplan-Meier rates of LR and distal recurrence (DR), patterns of LR and DR, and factors associated with LR, DR, and survival. Multiple factors potentially related to outcome were abstracted from the available medical records. Patient-related and staging factors were the following: age; weight loss; preoperative hemoglobin level; BMI; presenting symptoms (cough, hemoptysis, dyspnea, or none); pulmonary function test results (FEV1, FEV1% predicted, and carbon monoxide diffusing capacity percent [Dlco% predicted]); smoking status (never, quit > 2 years, quit 1-2 years, quit 6 months to 1 year, quit 30 days to 6 months prior to surgery, or smoking at present); drinking of alcoholic beverages (never, quit > 2 years ago, quit 1-2 years ago, quit 6 months to 1 year ago, quit 30 days to 6 months ago, drinking at present); number of drinks per week; chronic use of steroids; history of intercurrent illnesses (diabetes, hypertension, myocardial infarction, coronary artery disease, cardiac rhythm disturbance, coronary artery bypass graft, cerebrovascular accidents, renal failure, thromboembolic disease); performance of staging PET-CT scan at diagnosis; medication use at diagnosis (aspirin, statins, and nonsteroidal antiinflammatory drugs); and hemoglobin and albumin values within 30 days prior to surgery.

Surgery-related factors recorded included operative time; fluid balances during surgery (urine output, estimated blood loss, fluids infused, total excess intraoperative fluids [assuming one unit of packed RBCs to be 300 mL]); transfusions (total number of units required within 90 days of surgery, and number of units required within the first day, within 2-4 days, and after 4 days); type of resection (WR, SMR, L+/bilobectomy, pneumonectomy); type of lymph node staging procedure; intraoperative and postoperative complications (pulmonary infection or pneumonia, ARDS, arrhythmias and their type [supraventricular tachycardia, ventricular tachycardia, or nonspecific], fistula formation, stump leak, prolonged air leak, anastomosis leak, mediastinitis/empyema, myocardial infarction, death, pulmonary embolus, DVT, and brachial plexus injury); length of hospital stay (LOS); and surgeon. Histopathologic factors recorded included tumor size; tumor grade (categorized as grade 1.5 when recorded as well to moderately differentiated and grade 2.5 when recorded as moderately to poorly differentiated); performance/positivity of previous fine-needle aspiration biopsy; LVI; perineural invasion; margin status; extracapsular lymph node involvement; details of node resection (number resected, lymph node level resected, number of N1 nodes and N2 nodes resected, number of nodal stations sampled at mediastinoscopy and during resection); histology (NSCLC not otherwise specified, squamous cell, adenocarcinoma, large cell carcinoma, neuroendocrine carcinoma, BAC, adenocarcinoma with bronchioloalveolar features); tumor lobe location; and invasion of the visceral pleura. Details of chemotherapy recorded included type of chemotherapy (carboplatin and paclitaxel, cisplatin and navelbine, cisplatin and etoposide, other cisplatin-containing regimens, gemcitabine-containing combinations, other regimen); treatment sequencing (giving chemotherapy preoperatively, postoperatively, or both); and number of cycles.

Patterns of failure were determined using physicians’ clinical assessments, radiographic reports, and/or review of imaging studies as previously reported.2,3 LR was defined as those occurring in the ipsilateral lungs and the N1-N3 nodal basins; all other failures were defined as DR. Patients were scored as having a nodal recurrence when a new or enlarging node measured > 1 cm along the short axis on follow-up CT scans. For patients who underwent PET-CT scanning at the time of recurrence, all sites of abnormal uptake that correlated with lymph node or soft-tissue mass were scored, regardless of size. When available, data from bronchoscopy, biopsy specimen, or mediastinoscopy were used to supplement the radiographic findings. Only the initial sites of recurrence were scored. Patients with simultaneous LR and DR were scored as having both types of failure. Information on failure sites was reviewed by two of the authors (J. M. V. and L. M. D.) for consistency among the radiologic, bronchoscopic, and/or pathologic studies. CT scan studies from the Boston VA hospital, Penn State Hershey Cancer Institute, and Beth Israel Deaconess Medical Center were available for review. CT or PET-CT scans were performed at > 98% of all follow-up visits at these specialized thoracic-oncology programs. In general, at all centers, patients were followed every 3 to 6 months during the first 2 postoperative years, and then every 6 months for the next 3 years.

The SEER (Surveillance Epidemiology and End Results) database-184 was accessed on May 9, 2012, for a case selection analysis during the years 1988-2009 to assess the proportion of L− in patients with stage I disease who underwent definitive surgical resection. A multivariate analysis was performed and assessed the following factors for their impact on the performance of a L−: age, race (non-Hispanic white, non-Hispanic black, Hispanic black, Hispanic white, Asian/Pacific Islander, and American Indian/Alaska Native), sex, year, history of lung cancer, and histology (adenocarcinoma, BAC, squamous cell carcinoma, large cell carcinoma, and other).

The median follow-up was 34 months (range, 3-98 months) for the entire group, and was 34 months (range, 3-84 months) and 34 months (range, 3-98 months) in the L− and L+ groups, respectively. Recurrence rates at 2, 3, and 5 years were calculated by the Kaplan-Meier method.5 Univariate and multivariate analyses using Cox proportional hazards models were performed to assess factors associated with LR in L− and L+ groups separately.6 Factors analyzed for the univariate analyses included LVI, diabetes, chemotherapy, tumor size, number of nodes resected, tumor grade, BMI, tumor stage, LOS, age, PET scan performed, FEV1, FEV1% predicted, histology, and tumor location. Differences in the L− and L+ groups regarding categorical variables were compared using the Fisher exact test7; comparisons between continuous variables were calculated by the Wilcoxon rank-sum test.8 The percentages of patients from each study site were as follows: Beth Israel Deaconess Medical Center, 49.9%; Boston VA hospital, 16.4%; Denver VA hospital, 14.2%; and Penn State Cancer Institute, 18.7%.

A literature review was performed using the terms “sub-lobar resection,” “lung cancer,” and “recurrence” on PubMed.com to find pertinent publications regarding L− (n ≥ 30) since the publication of the LCSG trial in 1995.1 Tables were assembled into studies for and against L− to examine how tumor factors (ie, size, location [peripheral vs proximal], percentage of BAC, percentage with LVI) and treatment factors (percentage of WR and SMR, and extent of lymphadenectomy), as well as definition of LR affected OS, recurrence, and recommendations for L− vs L+.

Presenting characteristics, treatment factors, and histopathologic characteristics for the L− and L+ surgical groups are reported in Tables 1, 2, and 3. Patients in the L− group were more likely to be older (median age, 71 years vs 68 years; P = .01), and have worse lung function (median Dlco% predicted, 48.5% vs 74%, P < .0001; median FEV1, 1.5 L vs 2.3 L, P < .0001), fewer nodes resected (median, 0 vs 7; P < .0001), smaller tumor size (median, 2 cm vs 2.5 cm; P < .0001), lower T stage (T1, 80% vs 60%; P = .0005), shorter LOS (4 days vs 5 days; P = .0004), and less chemotherapy (1.1% vs 9.8%; P = .0035).

Table Graphic Jump Location
Table 1 —Patient Characteristics by Group

Data given as No. (%) unless otherwise indicated. CABG = coronary artery bypass graft; CAD = coronary artery disease; Dlco = diffusing capacity of the lung for carbon monoxide; L+ = lobectomy; L− = sublobar resection; MI = myocardial infarction.

Table Graphic Jump Location
Table 2 —Treatment Factors by Group

Data given as No. (%) unless otherwise indicated. LOS = length of stay. See Table 1 legend for expansion of abbreviations.

Table Graphic Jump Location
Table 3 —Histopathologic Characteristics by Group

Data given as No. (%) unless otherwise indicated. BAC = bronchioloarterial carcinoma; LLL = left lower lobe; LUL = left upper lobe; LVI = lymphatic vascular invasion; MB = main bronchus; NOS = not otherwise specified; NSCLC = non-small cell lung cancer; RLL = right lower lobe; RUL = right upper lobe. See Table 1 legend for expansion of other abbreviations.

LR rates at 2, 3, and 5 years were 13.7%, 23.9%, and 39.5%, respectively, for the L− group, and 14.1%, 18.7%, and 24.6%, respectively, for the L+ group (Fig 1). There was a nonsignificant higher risk for a greater LR rate in the L− group (P = .30). The LR rate started to differ at around 3 years after resection.

Results of univariate analysis of LR in the L− group (22 of 93 patients) indicated three statistically significant variables: tumor size (hazard ratio [HR], 1.709; P = .0141), tumor grade (HR, 9.40; P = .0288), and T1 stage (HR, 0.397; P = 0.0382). A full list of variables in the univariate analysis is presented in Table 4. LR rates by tumor size (≤ 2 cm vs > 2 cm) and grade (≤ 2 vs > 2) are reported in Figures 2A and 2B. Only one of the 24 (4%) patients whose tumors were well or well to moderately differentiated experienced a LR (this was in the bronchial stump), whereas 21 of 62 (33.8%) patients with tumor grade ≥ 2 experienced a LR. None of the four patients with tumors > 2 cm and well or well to moderately differentiated tumors had a LR. LR rates at 1 year in the L− group by selected covariates are presented in e-Table 1.

Table Graphic Jump Location
Table 4 —Univariate Analysis of Local Recurrence in the L− Group

See Table 3 legend for expansion of abbreviations.

a 

Statistically significant at P < .05.

Univariate analysis for LR in the L+ group (62 of 318 patients) revealed five statistically significant variables: tumor size (HR, 1.226; P = .0001), LVI (HR, 1.905; P = .0453), diabetes (HR, 2.039; P = .0169), LOS (HR, 1.033; P = .0073), and right upper lobe location (HR, 0.347; P = .0102). A full list of variables is presented in e-Table 2. Multivariate analysis of LR for the L+ group (Table 5) was conducted using all significant factors from the univariate analysis. The following three variables remained statistically significant: tumor size (HR, 1.233; P = .0003), diabetes (HR, 1.982; P = .0294), and LOS (HR, 1.031; P = .0209). LR rates at 1 year for the L+ group by selected covariates are given in e-Table 3.

Table Graphic Jump Location
Table 5 —Multivariate Analysis for Local Recurrence in the L+ Group

See Table 1, 2, and 3 legends for expansion of abbreviations.

a 

Statistically significant at P < .05.

LR occurred significantly more often in the bronchial stump/staple line in the L− group (40.9% vs 17.7%; P = .04). The LR patterns in our analysis demonstrated no difference in the percentage of failures in the ipsilateral lung, but showed trends toward higher recurrence in the ipsilateral hilum (27.3% vs 16.1%; P = .34) and subcarinal area (22.7% vs 11.3%; P = .28), and significantly higher percentage of recurrences in the bronchial stump/anastomoses/staple line (40.9% vs 17.7%; P = .04). Patterns of LR for all sites are presented in Table 6.

Table Graphic Jump Location
Table 6 —Sites of Local Failure Between Groups

Data given as % unless otherwise indicated. See Table 1 legend for expansion of abbreviations.

a 

Statistically significant at P < .05.

There was no difference in DR (P = .41) (Fig 3) or failure patterns (e-Table 4) between the L− and L+ groups. Only eight patients in the L− group experienced DR, and hence no univariate or multivariate analysis was conducted. In the L+ group, 39 patients had DR and univariate analysis demonstrated that only tumor grade was significant for DR (HR, 1.662; P = .0199). Univariate analysis results of all factors considered for DR in the L+ group are reported in e-Table 5.

Survival rates at 2,3, and 5 years were 84.5%, 79.5%, and 54.5%, respectively, for the L− group, and 90.5%, 80.7%, 64.5%, respectively, for the L+ group. The crude death rate was 36.6% in the l− group (n = 34) and the 25.2% in the L+ group (n = 80). Univariate analysis revealed three statistically significant variables related to survival for the L− group: diabetes (HR, 0.373; P = .0484), tumor grade (HR, 2.365; P = .0043), and age (HR, 1.043; P = .0387). e-Table 6 presents the P values and HRs for all factors analyzed in the univariate analysis for survival in the L− group. No multivariate analysis was conducted for survival in the L− group. Univariate analysis of survival for the L+ group revealed seven statistically significant variables: LVI (HR, 1.832; P = .0352), diabetes (HR, 2.265; P = .0012), tumor grade (HR, 1.570; P = .0043), LOS (HR, 1.043; P = .0005), age (HR, 1.036; P = .0043), non-squamous cell carcinoma (HR; 0.488; P = .0021), and squamous cell carcinoma (HR, 2.26; P = .0040). e-Table 7 contains all factors in the univariate analysis for survival in the L+ group. Multivariate analysis of survival in the L+ group, using all significant factors from the univariate analysis, yielded the following results: LVI (HR, 1.732; P = .0802), diabetes (HR, 2.027; P = .0085), tumor grade (HR, 1.555; P = .0118), age (HR, 1.033; P = .0192), LOS (HR, 1.046; P = .0006), squamous cell carcinoma (HR, 1.449; P = .2229), and adenocarcinoma (HR, 1.066; P = .8356).

The percentages of patients with stage I NSCLC who were treated by L−, L+ or bilobectomy, and pneumonectomy during the years 1988 to 2009 using the SEER-18 Database are reported in Figure 4A. The percentage of patients who were treated by L− ranged from 14% to 17% from 1988 to 2001, but steadily increased during the years 2002 to 2009 from 18% to 24%. L+ (including bilobectomy) ranged from 78% to 84% from 1988 to 2002, but decreased to 74% to 75% from 2003-2009. The percentage of patients with stage I NSCLC treated by pneumonectomy steadily declined from 1988 to 2009, decreasing from 8% to 5% during the period 1988-1997, and ranging from 0% to 3% from 1998-2009. Similar data for the years 1998-2009 are presented in Figure 4B. During these years, the percentage of patients receiving a WR increased steadily from 12% to 19%, while the percentage of patients receiving an SMR (3%-5%) or L− not otherwise specified (0%-2%) remained consistent. A multivariate analysis was performed to assess the following factors for their association with L−: age, race (non-Hispanic white, non-Hispanic black, Hispanic black, Hispanic white, Asian/Pacific Islander, and American Indian/Alaska Native), sex, year, history of lung cancer, and histology (adenocarcinoma, BAC, squamous cell carcinoma, large cell carcinoma, and other). Significantly more L− procedures were performed from 2002-2009 than in 1988 (HR range, 1.24-1.58; all P values < .0001). Other factors that were associated with significantly more L− procedures (all P < .0001) were non-Hispanic black vs non-Hispanic white race (HR = 1.18), non-Hispanic white (HR = 1.28) vs Asian/Pacific Islander race, those with a past history of lung cancer (HR = 1.85), female sex (HR = 1.25), adenocarcinomas vs BAC (HR = 1.149), and other histology vs adenocarcinoma (HR = 1.07). Multivariate analysis of all factors associated with the performance of L− are presented in e-Table 8. Although the median tumor size for resected stage I NSCLCs remained stable throughout the study (22-25 mm), the percentage of tumors < 2 cm remained stable from 37% to 39% from 1988 to 2002, but increased sharply from 40% to 46% during the years 2003-2009. The proportion of tumors that ranged from 21 to 40 mm (39%-46%) and 41 to 60 mm (9%-15%) remained relatively stable from 1988 to 2009 (Fig 4C).

One of the purposes of our investigation was to assess whether stage I NSCLC is increasing in frequency in the general lung cancer population and among those undergoing surgery. Furthermore, we wanted to assess whether the proportion of L− procedures was affected by the LCSG trial.1 Other aims of our investigation were to assess whether the failure patterns and risk factors for LR differ between patients receiving a L− or L+ for stage I NSCLC. There have been multiple studies that refuted1,912 or supported1322 the use of L− as compared with L+, but that was not the purpose of our study (Tables 7, 8).

Table Graphic Jump Location
Table 7 —Studies That Support L+ Over L−

BP = bronchopulmonary nodal area; H = ipsilateral hilar node; HR = hazard ratio; LAD = lymphadenectomy; LR = local-regional recurrence; LR, DEF = definition of local recurrence that was used in the study; M = mediastinal nodes; NR = not reported; OS = overall survival; SMR = segmentectomy; WR = wedge resection. See Table 1 and 3 legends for expansion of other abbreviations.

a 

Corrected data did not list local recurrence rates, but rather rate/person/y.

b 

Corrected data did not list OS, but rather death rate/person/y.

c 

Crude statistical calculations.

d 

Kaplan-Meier statistics.

Table Graphic Jump Location
Table 8 —Studies Supporting the Efficacy of L−

As shown by our analysis of the SEER-18 data, the proportion of L− is rising in the stage I NSCLC population undergoing surgery (Figs 4A, 4B). It is surprising to see that the percentage of L− did not decrease after the results of the LCSG trial1 were first published in 1995. This prospective, randomized trial demonstrated that limited resections (67.2% SMR and 32.8% WR) had significantly worse local control and were associated with strong trends for worse overall death and cancer death rates than L+. Although the percentage of L− procedures remained relatively stable from 1988-2001, there was a large increase in L− from 2002-2009. We speculate that the rise in L− may have been due to the 2001 Okada et al15 publication of the prospective, multicenter, Japanese experience with SMR. Patient eligibility for SMR was defined by tumor size < 2 cm on CT scan and tolerance of a L+ by cardiopulmonary function. This trial revealed that the 70-patient SMR group had no LR and had a similar 5-year survival rate (87.1%) as compared with patients undergoing L+ (87.7%) during the same period. However, the percentage of SMRs that were performed after this publication did not increase, whereas that of WRs did. This increase in WRs seemed to parallel the increased percentage of patients who underwent resection with tumor sizes ≤ 2 cm (Fig 4C).

In our retrospective analysis, patients receiving L− had smaller tumors (median tumor size, 2.0 cm vs 2.5 cm), worse lung function, fewer nodes resected, less use of chemotherapy, and were older than patients receiving L+. Despite having smaller median tumor size, the L− group showed a trend toward a higher LR rate. Interestingly, like the phase 3 LCSG trial,1 the curves for LR started to deviate around 3 years. Because of the delayed LR in the L− group, perhaps L− resections could be performed in an older group of patients with poor prognosis without a survival decrement, as suggested by the SEER database review by Mery et al.10

It should be noted that varying LR rates can be confounded by the definition of LR, classification of simultaneous LR and DR (recording only one recurrence instead of both), patient inclusion criteria, nodal sampling/dissection technique, and method of LR calculation (crude vs Kaplan-Meier or actuarial). Table 7, which lists investigations favoring L+, and Table 8, which lists investigations supporting L−, were created to show the number of possible confounding factors and the difficulty of literature interpretation when assessing the value of L− or L+ in particular patients. Whether supporting or refuting L−, the definition of LR was not stated in many of the studies in Tables 7911 and 8.14,17,19,20 Furthermore, LVI has been shown to be associated with recurrence and survival,23,24 but this factor was only mentioned in our study (8.6% of the L− and 12.9% of the L+ groups) and in the investigation by Kwiatkowski et al9 in table 7 in which LVI was present in 36% of all patients. Moreover, only a few of the trials in either table mention the median number of nodes removed. Although nodal involvement may not be a risk factor for LR,2 one study has shown that there is an OS and disease-free survival benefit to lymphadenectomy, even in patients with stage I NSCLC.25 Clearly, most studies supporting the benefit of L− were associated with a more thorough node dissection.

Furthermore, unlike our investigation, none of these studies mentioned if simultaneous LR and DR were marked as having LR or mentioned how LR was distinguished from second, primary lung cancers. Of the studies in Table 8 that support the benefit of L−, it should be noted that the reports by Okada et al15 and El-Sherif et al21 defined LR as being in the ipsilateral lung or same lung lobe and N1 nodes, respectively. Also, Koike et al16 defined intrapulmonary metastases as being a DR. Because the definitions of LR in these three studies may not have been inclusive of all three areas associated with a strong trend or higher risk of LR in our study (the ipsilateral hilum, subcarina, and the bronchial stump/staple line), these studies may have underestimated LR and may have falsely supported the efficacy of L− based upon definition of LR and not efficacy of the surgical resection. On the other hand, three studies that indicated L− was inferior to L+ did not report LR, but only stated OS.911 Those studies did not report the extent of lymphadenectomy10 or they revealed that the L− group had fewer nodes removed.9,11 Therefore, the inferior survival associated with L− in these studies may have been due to less extensive nodal dissection, which resulted in a poor survival in the L− groups due to understaging.

It should also be noted that most studies supporting the benefits of L− were in Japanese patients1417,19,20,22 and the relevancy of this relatively homogenous population to that of a more heterogeneous Westernized population can be questioned, especially when there are known biologic differences.26 LRs were calculated crudely in all the studies listed in Tables 7 and 8, and, therefore, the estimated rate of LR may have been underestimated as compared with actuarial or Kaplan Meier methodology.

In the L− group, LR was significantly associated with tumor size, T stage, and tumor grade. Significant factors for LR in the L+ group were tumor size, presence of diabetes at diagnosis, and LOS. Thus, other than tumor size, the factors associated with LR are different based on the type of surgical resection. In Table 8, all investigations in Japanese populations were prospective and used a maximum CT scan based tumor size of 2 cm for eligibility.1417,19,20,22 However some of these investigations were further restricted by including only tumors with ground-glass opacities,17 or peripheral tumor location.14,16,22 The policy of using tumor size < 2 cm and the association of larger sizes with recurrence in L− was largely based upon speculation.15 Currently, the Cancer and Leukemia Group B study 140503 (A Phase III Randomized Trial of Lobectomy versus Sublobar Resection for Small [< 2 cm] Peripheral Non-small Cell Lung Cancer) is randomizing patients with peripheral, small (< 2 cm) tumors to L− or L+ to assess whether these same selection criteria will result in a low rate of LR in a Western population undergoing L−. Nevertheless, we demonstrated that tumor size > 2 cm was associated with a higher risk of LR in the L− group, and the estimated 5-year recurrences were high in both surgical groups for tumors ≤ 2 cm (31% L+, 45% L−). Our results demonstrated that tumor grade < 2 in the L− group was more predictive of LR. Of the 24 patients with grade < 2 tumors in the L− group, only one experienced LR. Furthermore, none of the four patients with well-differentiated tumors > 2 cm experienced LR. Therefore, we feel that the selection of patients for L− based upon size criteria alone may result in improper selection. Additionally, the selection of patients for L− based upon appearance of T1 disease on CT scan was associated with upgrading to stage ≥ T2 in 28% of patients in a multiinstitution Cancer and Leukemia Group B study.27 Because recent studies have shown that tumor grading can be adequately obtained from cytologic specimens,28 we feel that a biopsy procedure should be performed prior to the consideration of elective L−. If the biopsy specimen is suggestive of moderately or poorly differentiated disease, then the physicians should strongly consider L+ as the therapeutic procedure of choice, especially when tumors approach 2 cm on CT scan.

In the L+ group, we found a unique factor associated with worse LR and survival: LOS. We decided to analyze this factor for recurrence because this group had demonstrated that pneumonectomy (and possibly surgical trauma) was associated with DR.2 Indeed, another group found that LOS was associated with worse survival in patients with stage I NSCLC who were undergoing surgical resection, but there was no mention of this factor’s effect upon LR.29 We postulate that LOS may be a surrogate for a patient’s comorbid medical conditions and/or performance status or possible temporary immunosuppression related to surgical stress.30,31

In terms of DR, there was no difference in the failure rates or failure pattern between the two surgical groups. Because of the small number of DRs in the L− group, assessment of factors associated with DR could not be estimated. In the L+ group, DR was significantly associated with tumor grade.

The percentage of stage I NSCLC is rising in the surgical population, and there is an increasing portion of such patients undergoing L−. Patients undergoing L− are more likely to fail in the bronchial stump/staple line than patients undergoing L+. Both tumor size and grade were associated with LR in the L− group, with the latter factor being more predictive than the former. Future studies are clearly needed to explore whether adjunctive therapy, such as interstitial brachytherapy,32 would be effective in lowering the LR in L− patients, especially for those tumors > 2 cm and/or grade ≥ 2. External beam maybe too toxic following L−26 in compromised patients, but since it was applied successfully in patients with resected stage I tumors following L+,33 the toxicity may have also been technique related.

Figure Jump LinkFigure 1. Local-regional recurrence rates for patients in the L− and L+ groups. L+ = lobectomy; L− = sublobar resection.Grahic Jump Location
Figure Jump LinkFigure 2. A, B, Local-regional recurrence rates by (A) tumor size and (B) tumor grade in patients in the L− group. See Figure 1 legend for expansion of abbreviations.Grahic Jump Location
Figure Jump LinkFigure 3. Distal recurrence rates for patients in the L− and L+ groups. See Figure 1 legend for expansion of abbreviations.Grahic Jump Location
Figure Jump LinkFigure 4. Percentage of patients with stage I non-small cell lung cancer (NSCLC) who were treated by various types of surgical resection during the years A, 1988-2009; and B, 1998-2009. C, Median tumor size and percentage of tumors 0-60 mm in patients with stage I NSCLC, 1998-2009. NOS = not otherwise specified.Grahic Jump Location

Author contributions: Dr Varlotto had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Dr Varlotto: contributed to data acquisition and analysis, writing the manuscript, approval of the final manuscript, and served as principal author.

Dr Medford-Davis: contributed to contributed to data acquisition and analysis, writing the manuscript, and approval of the final manuscript.

Dr Recht: contributed to data analysis, writing the manuscript, and approval of the final manuscript.

Dr Flickinger: contributed to data analysis, writing the manuscript, and approval of the final manuscript.

Dr Yao: contributed to data analysis, writing the manuscript, and approval of the final manuscript.

Dr Hess: contributed to data acquisition and analysis, writing the manuscript, and approval of the final manuscript.

Dr Reed: contributed to writing the manuscript, and approval of the final manuscript.

Dr Toth: contributed to writing the manuscript, and approval of the final manuscript.

Dr Zander: contributed to writing the manuscript, and approval of the final manuscript.

Dr DeCamp: contributed to data analysis, writing the manuscript, and approval of the final manuscript.

Financial/nonfinancial disclosures: The authors have reported to CHEST that no potential conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article.

Additional information: The e-Tables can be found in the “Supplemental Materials” area of the online article.

BAC

bronchioloalveolar carcinoma

Dlco

diffusing capacity of the lung for cabon monoxide

DR

distal recurrence

HR

hazard ratio

IRB

institutional review board

L+

lobectomy

L−

sublobar resection

LCSG

Lung Cancer Study Group

LOS

length of stay

LR

local recurrence

LVI

lymphatic vascular invasion

NSCLC

non-small cell lung cancer

OS

overall survival

SEER

Surveillance Epidemiology and End Results

SMR

segmentectomy

WR

wedge resection

Lederle FA. Lobectomy versus limited resection in T1 N0 lung cancer. Ann Thorac Surg. 1996;62(4):1249-1250. [CrossRef] [PubMed]
 
Varlotto JM, Recht A, Flickinger JC, Medford-Davis LN, Dyer AM, Decamp MM. Factors associated with local and distant recurrence and survival in patients with resected nonsmall cell lung cancer. Cancer. 2009;115(5):1059-1069. [CrossRef] [PubMed]
 
Varlotto JM, Medford-Davis LN, Recht A, Flickinger JC, Schaefer E, DeCamp MM. Failure rates and patterns of recurrence in patients with resected N1 non-small-cell lung cancer. Int J Radiat Oncol Biol Phys. 2011;81(2):353-359. [CrossRef] [PubMed]
 
Surveillance Research Program, National Cancer Institute. SEER data, 1973-2009. National Cancer Institute websitehttp://seer.cancer.gov/publicdata/. Accessed May 9, 2012.
 
Kaplan ES, Meier P. Non-parametric estimation from incomplete observation. JAmer Statist Assoc. 1958;53(282):457-481. [CrossRef]
 
Cox DR. Regression models and life tables. J R Stat Soc [Ser A]. 1972;34(2):187-220.
 
Fisher RA. On the interpretation of χ2from contingency tables, and the calculation of P. J R Stat Soc. 1922;85(1):87-94. [CrossRef]
 
Wilcoxon F. Individual comparisons of grouped data by ranking methods. J Econ Entomol. 1946;39:269. [PubMed]
 
Kwiatkowski DJ, Harpole DH Jr, Godleski J, et al. Molecular pathologic substaging in 244 stage I non-small-cell lung cancer patients: clinical implications. J Clin Oncol. 1998;16(7):2468-2477. [PubMed]
 
Mery CM, Pappas AN, Bueno R, et al. Similar long-term survival of elderly patients with non-small cell lung cancer treated with lobectomy or wedge resection within the surveillance, epidemiology, and end results database. Chest. 2005;128(1):237-245. [CrossRef] [PubMed]
 
Kraev A, Rassias D, Vetto J, et al. Wedge resection vs lobectomy: 10-year survival in stage I primary lung cancer. Chest. 2007;131(1):136-140. [CrossRef] [PubMed]
 
Sienel W, Stremmel C, Kirschbaum A, et al. Frequency of local recurrence following segmentectomy of stage IA non-small cell lung cancer is influenced by segment localisation and width of resection margins—implications for patient selection for segmentectomy. Eur J Cardiothorac Surg. 2007;31(3):522-527. [CrossRef] [PubMed]
 
Landreneau RJ, Sugarbaker DJ, Mack MJ, et al. Wedge resection versus lobectomy for stage I (T1 N0 M0) non-small-cell lung cancer. J Thorac Cardiovasc Surg. 1997;113(4):691-698. [CrossRef] [PubMed]
 
Tsubota N, Ayabe K, Doi O, et al. Ongoing prospective study of segmentectomy for small lung tumors. Study Group of Extended Segmentectomy for Small Lung Tumor. Ann Thorac Surg. 1998;66(5):1787-1790. [CrossRef] [PubMed]
 
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(3):956-960.- [CrossRef] [PubMed]
 
Koike T, Yamato Y, Yoshiya K, Shimoyama T, Suzuki R. Intentional limited pulmonary resection for peripheral T1 N0 M0 small-sized lung cancer. J Thorac Cardiovasc Surg. 2003;125(4):924-928. [CrossRef] [PubMed]
 
Nakata M, Sawada S, Saeki H, et al. Prospective study of thoracoscopic limited resection for ground-glass opacity selected by computed tomography. Ann Thorac Surg. 2003;75(5):1601-1605. [CrossRef] [PubMed]
 
Keenan RJ, Landreneau RJ, Maley RH Jr, et al. Segmental resection spares pulmonary function in patients with stage I lung cancer. Ann Thorac Surg. 2004;78(1):228-233. [CrossRef] [PubMed]
 
Nakamura H, Saji H, Ogata A, Saijo T, Okada S, Kato H. Lung cancer patients showing pure ground-glass opacity on computed tomography are good candidates for wedge resection. Lung Cancer. 2004;44(1):61-68. [CrossRef] [PubMed]
 
Watanabe T, Okada A, Imakiire T, Koike T, Hirono T. Intentional limited resection for small peripheral lung cancer based on intraoperative pathologic exploration. Jpn J Thorac Cardiovasc Surg. 2005;53(1):29-35. [CrossRef] [PubMed]
 
El-Sherif A, Gooding WE, Santos R, et al. Outcomes of sublobar resection versus lobectomy for stage I non-small cell lung cancer: a 13-year analysis. Ann Thorac Surg. 2006;82(2):408-415. [CrossRef] [PubMed]
 
Okada M, Koike T, Higashiyama M, Yamato Y, Kodama K, Tsubota N. Radical sublobar resection for small-sized non-small cell lung cancer: a multicenter study. J Thorac Cardiovasc Surg. 2006;132(4):769-775. [CrossRef] [PubMed]
 
Shields TW. Prognostic significance of parenchymal lymphatic vessel and blood vessel invasion in carcinoma of the lung. Surg Gynecol Obstet. 1983;157(2):185-190. [PubMed]
 
Bréchot JM, Chevret S, Charpentier MC, et al. Blood vessel and lymphatic vessel invasion in resected nonsmall cell lung carcinoma. Correlation with TNM stage and disease free and overall survival. Cancer. 1996;78(10):2111-2118. [CrossRef] [PubMed]
 
Varlotto JM, Recht A, Nikolov M, Flickinger JC, Decamp MM. Extent of lymphadenectomy and outcome for patients with stage I nonsmall cell lung cancer. Cancer. 2009;115(4):851-858. [CrossRef] [PubMed]
 
Fukuoka M, Yano S, Giaccone G, et al. Multi-institutional randomized phase II trial of gefitinib for previously treated patients with advanced non-small-cell lung cancer (The IDEAL 1 Trial) [corrected]. J Clin Oncol. 2003;21(12):2237-2246. [CrossRef] [PubMed]
 
Shennib H, Bogart J, Herndon JE, et al; Cancer and Leukemia Group B; Eastern Cooperative Oncology Group. Video-assisted wedge resection and local radiotherapy for peripheral lung cancer in high-risk patients: the Cancer and Leukemia Group B (CALGB) 9335, a phase II, multi-institutional cooperative group study. J Thorac Cardiovasc Surg. 2005;129(4):813-818. [CrossRef] [PubMed]
 
Sigel CS, Rudomina DE, Sima CS, et al. Predicting pulmonary adenocarcinoma outcome based on a cytology grading system. Cancer Cytopathol. 2012;120(1):35-43. [CrossRef] [PubMed]
 
Kilic A, Schuchert MJ, Landreneau RJ, et al. Impact of length of stay of hospitalization following surgical resection of stage I non-small cell lung cancer on long-term survival. J Clin Oncol. 2009;27(15s): abstr7584.
 
Kono K, Takahashi A, Iizuka H, Fujii H, Sekikawa T, Matsumoto Y. Effect of oesophagectomy on monocyte-induced apoptosis of peripheral blood T lymphocytes. Br J Surg. 2001;88(8):1110-1116. [CrossRef] [PubMed]
 
van Sandick JW, Gisbertz SS, ten Berge IJ, et al. Immune responses and prediction of major infection in patients undergoing transhiatal or transthoracic esophagectomy for cancer. Ann Surg. 2003;237(1):35-43. [CrossRef] [PubMed]
 
Colonias A, Betler J, Trombetta M, et al. Mature follow-up for high-risk stage I non-small-cell lung carcinoma treated with sublobar resection and intraoperative iodine-125 brachytherapy. Int J Radiat Oncol Biol Phys. 2011;79(1):105-109. [CrossRef] [PubMed]
 
Trodella L, Granone P, Valente S, et al. Adjuvant radiotherapy in non-small cell lung cancer with pathological stage I: definitive results of a phase III randomized trial. Radiother Oncol. 2002;62(1):11-19. [CrossRef] [PubMed]
 

Figures

Figure Jump LinkFigure 1. Local-regional recurrence rates for patients in the L− and L+ groups. L+ = lobectomy; L− = sublobar resection.Grahic Jump Location
Figure Jump LinkFigure 2. A, B, Local-regional recurrence rates by (A) tumor size and (B) tumor grade in patients in the L− group. See Figure 1 legend for expansion of abbreviations.Grahic Jump Location
Figure Jump LinkFigure 3. Distal recurrence rates for patients in the L− and L+ groups. See Figure 1 legend for expansion of abbreviations.Grahic Jump Location
Figure Jump LinkFigure 4. Percentage of patients with stage I non-small cell lung cancer (NSCLC) who were treated by various types of surgical resection during the years A, 1988-2009; and B, 1998-2009. C, Median tumor size and percentage of tumors 0-60 mm in patients with stage I NSCLC, 1998-2009. NOS = not otherwise specified.Grahic Jump Location

Tables

Table Graphic Jump Location
Table 1 —Patient Characteristics by Group

Data given as No. (%) unless otherwise indicated. CABG = coronary artery bypass graft; CAD = coronary artery disease; Dlco = diffusing capacity of the lung for carbon monoxide; L+ = lobectomy; L− = sublobar resection; MI = myocardial infarction.

Table Graphic Jump Location
Table 2 —Treatment Factors by Group

Data given as No. (%) unless otherwise indicated. LOS = length of stay. See Table 1 legend for expansion of abbreviations.

Table Graphic Jump Location
Table 3 —Histopathologic Characteristics by Group

Data given as No. (%) unless otherwise indicated. BAC = bronchioloarterial carcinoma; LLL = left lower lobe; LUL = left upper lobe; LVI = lymphatic vascular invasion; MB = main bronchus; NOS = not otherwise specified; NSCLC = non-small cell lung cancer; RLL = right lower lobe; RUL = right upper lobe. See Table 1 legend for expansion of other abbreviations.

Table Graphic Jump Location
Table 4 —Univariate Analysis of Local Recurrence in the L− Group

See Table 3 legend for expansion of abbreviations.

a 

Statistically significant at P < .05.

Table Graphic Jump Location
Table 5 —Multivariate Analysis for Local Recurrence in the L+ Group

See Table 1, 2, and 3 legends for expansion of abbreviations.

a 

Statistically significant at P < .05.

Table Graphic Jump Location
Table 6 —Sites of Local Failure Between Groups

Data given as % unless otherwise indicated. See Table 1 legend for expansion of abbreviations.

a 

Statistically significant at P < .05.

Table Graphic Jump Location
Table 7 —Studies That Support L+ Over L−

BP = bronchopulmonary nodal area; H = ipsilateral hilar node; HR = hazard ratio; LAD = lymphadenectomy; LR = local-regional recurrence; LR, DEF = definition of local recurrence that was used in the study; M = mediastinal nodes; NR = not reported; OS = overall survival; SMR = segmentectomy; WR = wedge resection. See Table 1 and 3 legends for expansion of other abbreviations.

a 

Corrected data did not list local recurrence rates, but rather rate/person/y.

b 

Corrected data did not list OS, but rather death rate/person/y.

c 

Crude statistical calculations.

d 

Kaplan-Meier statistics.

Table Graphic Jump Location
Table 8 —Studies Supporting the Efficacy of L−

References

Lederle FA. Lobectomy versus limited resection in T1 N0 lung cancer. Ann Thorac Surg. 1996;62(4):1249-1250. [CrossRef] [PubMed]
 
Varlotto JM, Recht A, Flickinger JC, Medford-Davis LN, Dyer AM, Decamp MM. Factors associated with local and distant recurrence and survival in patients with resected nonsmall cell lung cancer. Cancer. 2009;115(5):1059-1069. [CrossRef] [PubMed]
 
Varlotto JM, Medford-Davis LN, Recht A, Flickinger JC, Schaefer E, DeCamp MM. Failure rates and patterns of recurrence in patients with resected N1 non-small-cell lung cancer. Int J Radiat Oncol Biol Phys. 2011;81(2):353-359. [CrossRef] [PubMed]
 
Surveillance Research Program, National Cancer Institute. SEER data, 1973-2009. National Cancer Institute websitehttp://seer.cancer.gov/publicdata/. Accessed May 9, 2012.
 
Kaplan ES, Meier P. Non-parametric estimation from incomplete observation. JAmer Statist Assoc. 1958;53(282):457-481. [CrossRef]
 
Cox DR. Regression models and life tables. J R Stat Soc [Ser A]. 1972;34(2):187-220.
 
Fisher RA. On the interpretation of χ2from contingency tables, and the calculation of P. J R Stat Soc. 1922;85(1):87-94. [CrossRef]
 
Wilcoxon F. Individual comparisons of grouped data by ranking methods. J Econ Entomol. 1946;39:269. [PubMed]
 
Kwiatkowski DJ, Harpole DH Jr, Godleski J, et al. Molecular pathologic substaging in 244 stage I non-small-cell lung cancer patients: clinical implications. J Clin Oncol. 1998;16(7):2468-2477. [PubMed]
 
Mery CM, Pappas AN, Bueno R, et al. Similar long-term survival of elderly patients with non-small cell lung cancer treated with lobectomy or wedge resection within the surveillance, epidemiology, and end results database. Chest. 2005;128(1):237-245. [CrossRef] [PubMed]
 
Kraev A, Rassias D, Vetto J, et al. Wedge resection vs lobectomy: 10-year survival in stage I primary lung cancer. Chest. 2007;131(1):136-140. [CrossRef] [PubMed]
 
Sienel W, Stremmel C, Kirschbaum A, et al. Frequency of local recurrence following segmentectomy of stage IA non-small cell lung cancer is influenced by segment localisation and width of resection margins—implications for patient selection for segmentectomy. Eur J Cardiothorac Surg. 2007;31(3):522-527. [CrossRef] [PubMed]
 
Landreneau RJ, Sugarbaker DJ, Mack MJ, et al. Wedge resection versus lobectomy for stage I (T1 N0 M0) non-small-cell lung cancer. J Thorac Cardiovasc Surg. 1997;113(4):691-698. [CrossRef] [PubMed]
 
Tsubota N, Ayabe K, Doi O, et al. Ongoing prospective study of segmentectomy for small lung tumors. Study Group of Extended Segmentectomy for Small Lung Tumor. Ann Thorac Surg. 1998;66(5):1787-1790. [CrossRef] [PubMed]
 
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(3):956-960.- [CrossRef] [PubMed]
 
Koike T, Yamato Y, Yoshiya K, Shimoyama T, Suzuki R. Intentional limited pulmonary resection for peripheral T1 N0 M0 small-sized lung cancer. J Thorac Cardiovasc Surg. 2003;125(4):924-928. [CrossRef] [PubMed]
 
Nakata M, Sawada S, Saeki H, et al. Prospective study of thoracoscopic limited resection for ground-glass opacity selected by computed tomography. Ann Thorac Surg. 2003;75(5):1601-1605. [CrossRef] [PubMed]
 
Keenan RJ, Landreneau RJ, Maley RH Jr, et al. Segmental resection spares pulmonary function in patients with stage I lung cancer. Ann Thorac Surg. 2004;78(1):228-233. [CrossRef] [PubMed]
 
Nakamura H, Saji H, Ogata A, Saijo T, Okada S, Kato H. Lung cancer patients showing pure ground-glass opacity on computed tomography are good candidates for wedge resection. Lung Cancer. 2004;44(1):61-68. [CrossRef] [PubMed]
 
Watanabe T, Okada A, Imakiire T, Koike T, Hirono T. Intentional limited resection for small peripheral lung cancer based on intraoperative pathologic exploration. Jpn J Thorac Cardiovasc Surg. 2005;53(1):29-35. [CrossRef] [PubMed]
 
El-Sherif A, Gooding WE, Santos R, et al. Outcomes of sublobar resection versus lobectomy for stage I non-small cell lung cancer: a 13-year analysis. Ann Thorac Surg. 2006;82(2):408-415. [CrossRef] [PubMed]
 
Okada M, Koike T, Higashiyama M, Yamato Y, Kodama K, Tsubota N. Radical sublobar resection for small-sized non-small cell lung cancer: a multicenter study. J Thorac Cardiovasc Surg. 2006;132(4):769-775. [CrossRef] [PubMed]
 
Shields TW. Prognostic significance of parenchymal lymphatic vessel and blood vessel invasion in carcinoma of the lung. Surg Gynecol Obstet. 1983;157(2):185-190. [PubMed]
 
Bréchot JM, Chevret S, Charpentier MC, et al. Blood vessel and lymphatic vessel invasion in resected nonsmall cell lung carcinoma. Correlation with TNM stage and disease free and overall survival. Cancer. 1996;78(10):2111-2118. [CrossRef] [PubMed]
 
Varlotto JM, Recht A, Nikolov M, Flickinger JC, Decamp MM. Extent of lymphadenectomy and outcome for patients with stage I nonsmall cell lung cancer. Cancer. 2009;115(4):851-858. [CrossRef] [PubMed]
 
Fukuoka M, Yano S, Giaccone G, et al. Multi-institutional randomized phase II trial of gefitinib for previously treated patients with advanced non-small-cell lung cancer (The IDEAL 1 Trial) [corrected]. J Clin Oncol. 2003;21(12):2237-2246. [CrossRef] [PubMed]
 
Shennib H, Bogart J, Herndon JE, et al; Cancer and Leukemia Group B; Eastern Cooperative Oncology Group. Video-assisted wedge resection and local radiotherapy for peripheral lung cancer in high-risk patients: the Cancer and Leukemia Group B (CALGB) 9335, a phase II, multi-institutional cooperative group study. J Thorac Cardiovasc Surg. 2005;129(4):813-818. [CrossRef] [PubMed]
 
Sigel CS, Rudomina DE, Sima CS, et al. Predicting pulmonary adenocarcinoma outcome based on a cytology grading system. Cancer Cytopathol. 2012;120(1):35-43. [CrossRef] [PubMed]
 
Kilic A, Schuchert MJ, Landreneau RJ, et al. Impact of length of stay of hospitalization following surgical resection of stage I non-small cell lung cancer on long-term survival. J Clin Oncol. 2009;27(15s): abstr7584.
 
Kono K, Takahashi A, Iizuka H, Fujii H, Sekikawa T, Matsumoto Y. Effect of oesophagectomy on monocyte-induced apoptosis of peripheral blood T lymphocytes. Br J Surg. 2001;88(8):1110-1116. [CrossRef] [PubMed]
 
van Sandick JW, Gisbertz SS, ten Berge IJ, et al. Immune responses and prediction of major infection in patients undergoing transhiatal or transthoracic esophagectomy for cancer. Ann Surg. 2003;237(1):35-43. [CrossRef] [PubMed]
 
Colonias A, Betler J, Trombetta M, et al. Mature follow-up for high-risk stage I non-small-cell lung carcinoma treated with sublobar resection and intraoperative iodine-125 brachytherapy. Int J Radiat Oncol Biol Phys. 2011;79(1):105-109. [CrossRef] [PubMed]
 
Trodella L, Granone P, Valente S, et al. Adjuvant radiotherapy in non-small cell lung cancer with pathological stage I: definitive results of a phase III randomized trial. Radiother Oncol. 2002;62(1):11-19. [CrossRef] [PubMed]
 
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).
Supporting Data

Online Supplement

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Find Similar Articles
CHEST Journal Articles
  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543