Satellite Pulmonary Nodule in the Same Lobe (T4N0) Should Not Be Staged as IIIB Non Small Cell Lung Cancer

Satellite Pulmonary Nodule in the Same Lobe (T4N0) Should Not Be Staged as IIIB Non Small Cell Lung Cancer Ayesha S. Bryant, MSPH, MD, Sara J. Pereira, MD, Daniel L. Miller, MD, and Robert James Cerfolio, MD, FACS Department of Epidemiology, University of Alabama at Birmingham School of Public Health and Emory University, Birmingham, Alabama, and Atlanta, Georgia, Division of Cardiothoracic Surgery, Section of General Thoracic Surgery, University of Alabama at Birmingham, Birmingham, Alabama, and Division of Cardiothoracic Surgery, Section of Thoracic Surgery, Emory University, Atlanta, Georgia Background. Treatment of non small cell lung cancer depends on stage. Patients with T4 lesions represent a heterogeneous group. Methods. A case-control study of patients with pathologically proven, node-negative T4 lesions (T4 N0 M0) was conducted. Patients with T4 disease were stratified as T4 from a satellite nodule (T4-satellite) or T4 from local invasion (T4-invasion). T4-satellite patients were matched 1:4 for sex and histology with resected control patients with stage IA, IB, and IIA non small cell lung cancer and matched 1:3 with stage II non small cell lung cancer. Survival and the maximal standardized uptake value on F-18 fluorodeoxyglucose-positron emission tomography scans were compared. Results. There were 337 patients, 26 patients with T4-satellite lesions, 25 with T4-invasion lesions, and 286 controls (104 patients with T1 N0 M0, 104 with T2 N0 M0, and 78 with T1 N1 M0 or T2 N1 M0 lesions). The two T4 groups were similar for age, race, sex, and neoadjuvant therapy rates. The 5-year survival was 80% for the T1 N0 M0 patients, 68% for T2 N0 M0, 57% for T4-satellite N0 M0, 45% for T1 N1 M0 or T2 N1 M0, and 30% for the T4-invasion N0 M0 patients (p 0.016). Multivariate analysis showed that only the type of T4 impacted survival (p 0.011). The median maximal standardized uptake values of the cancers were 4.2 for T1 N0 M0, 4.8 for T4-satellite, 5.4 for T2 N0 M0, 7.8 for T1 N1 M0 or T2 N1 M0, and 8.8 for the T4-invasion patients. Conclusions. Larger studies are needed; however, patients with T4-satellite non small cell lung cancer who undergo complete resection have survival and maximal standardized uptake values similar to patients with stage IB and stage IIA lesions. Their survival is significantly better than those with T4-invasion. Patients with T4- satellite N0 M0 lesions should not be classified as stage IIIB and should not be grouped with patients with T4-invasion, and resection should be considered. (Ann Thorac Surg 2006;82:1808 14) 2006 by The Society of Thoracic Surgeons The TNM staging system of lung cancer has been fine tuned over the years on the basis of the survival of patients with non small cell lung cancer (NSCLC). It is used as a guide for the prognosis of patients with NSCLC and as a tool for treatment strategies based on the stage. The classification of T4 tumors in the current TNM system [1] includes any size tumor that (1) invades the trachea, carina, heart, great vessels, vertebral body, esophagus, or mediastinum, (2) has one or more satellite tumor nodules within the same lobe with the same histology, or (3) has the presence of a malignant pleural effusion. Although all lesions are grouped as T4 and hence stage IIIB, studies that evaluate survival of these patients show different outcomes for these different Accepted for publication March 29, 2006. Presented at the Forty-second Annual Meeting of The Society of Thoracic Surgeons, Chicago, IL, Jan 30 Feb 1, 2006. Address correspondence to Dr Cerfolio, Section of General Thoracic Surgery, Division of Cardiothoracic Surgery, University of Alabama at Birmingham, 1900 University Blvd, THT 720, Birmingham, AL 35294; e-mail: rcerfolio@surg.uab.edu. groups of T4 [2 4]. This is probably related to the heterogeneity of the different types of T4. The objective of this study was to compare the 5-year survival of patients staged as T4 because of satellite pulmonary nodules within the same lobe (T4-satellite) with those staged with T4 because of local invasion (T4-invasion) in patients who were node negative. These results were compared with three sets of control patients who underwent pulmonary resection during the same period by the same surgeon using the same staging and treatment algorithms. Patients and Methods This is a case-control study of 337 patients with pathologically staged, completely resected NSCLC between January 1998 and November 2005. Patients with pathologically confirmed T4 N0 M0 disease were divided into two groups: those with a satellite pulmonary nodule of the same histology within the same lobe (T4-satellite) and those with T4 because of local invasion of mediasti- 2006 by The Society of Thoracic Surgeons 0003-4975/06/$32.00 Published by Elsevier Inc doi:10.1016/j.athoracsur.2006.03.123

Ann Thorac Surg BRYANT ET AL 2006;82:1808 14 SATELLITE PULMONARY NODULES 1809 Fig 1. Treatment algorithm used for patients in this study. Patients were excluded from this study if they had N2 or N3 disease documented by pathology. All patients had pathologic confirmation of disease. (bx biopsy; CT computed tomography; EUS-FNA endoscopic ultrasonography with fine-needle aspiration; FDG-PET fluorodeoxyglucose positron emission tomography; med mediastinoscopy; MRI magnetic resonance imaging; NSCLC non small cell lung cancer.) nal structures (T4-invasion). Patients with T4-satellite lesions were matched 1:4 with patients with T1 N0 M0 (stage IA) or T2 N0 M0 (stage IB) and 1:3 with patients with T1 N1 M0 or T2 N1 M0 (stage IIA and IIB). Control patients were selected from our prospective database compiled by one surgeon at one institution. Control patients were based on procedure type, histology, and sex. Because of a lack of suitable controls in patients with completed resected T1 N1 M0 or T2 N1 M0 disease (stage IIA and IIB NSCLC) this control group was matched 1:3 instead of 1:4. Patients with T4 disease were excluded from this study if they had N2 or N1 disease documented either before neoadjuvant therapy or after resection as shown in Figure 1. All patients underwent similar preoperative and intraoperative staging, and all resections were performed by one general thoracic surgeon (R.J.C.) using the same technique. Entry criteria included a computed tomography (CT) scan, fluorodeoxyglucose positron emission tomography (FDG-PET), thoracotomy, and attempted margin-negative resection (R0 resection) with complete thoracic lymphadenectomy by means of thoracotomy. Specific techniques for these tests, the calculation of the maximum standardized uptake values (maxsuvs), and the surgical techniques used were performed as previously described in detail [5]. Patients were excluded if they were younger than 19 years old, did not have pathologic confirmation of stage, or had metastatic disease in any N2 or N3 lymph nodes. Patients with malignant pleural effusions (another type of T4 disease) were also excluded from the study group. The University of Alabama at Birmingham institutional review board approved both the electronic prospective database used for this study as well as this study. Patient consent was

1810 BRYANT ET AL Ann Thorac Surg SATELLITE PULMONARY NODULES 2006;82:1808 14 obtained for the entry of data into the prospective database. Staging and Surgery Patients were carefully staged. All suspicious N2, N3, or M1 areas (maxsuv 2.5) were biopsied before pulmonary resection. Mediastinoscopy or endoscopic ultrasonography with fine-needle aspiration was used to perform a biopsy of suspicious lymph nodes in the paratracheal area (stations 2R, 4R, 2L, 4L, and the anterior-superior portion of the subcarinal lymph node station 7), and endoscopic transesophageal ultra-sound was used to perform a biopsy of suspicious posterior aortopulmonary window and inferior lymph nodes (stations 5, 7, 8, and 9). Patients with suspected M1 disease in the liver, adrenal, or contralateral lung underwent definitive biopsy to prove or disprove M1 cancer. If the bone or brain was suspected to harbor metastases, magnetic resonance imaging was considered the standard reference. If patients had biopsy proven N3 or M1 disease, the stage was recorded but they were not resected as shown in Figure 1. If there was no evidence of N2 or higher disease, patients underwent thoracotomy, pulmonary resection, and complete thoracic lymphadenectomy. Pathologic review was performed with standard techniques, and immunohistochemically staining was used when appropriate. The pathologic stage was assessed using the international staging system [1]. The presence of one or more tumor nodules with the same histology within the same lobe defined T4 from satellite nodules. An N1 node was differentiated from a T4 second pulmonary lesion by its location and by the pathology. Any size tumor that invades the trachea, carina, heart, great vessels, vertebral body, esophagus, or mediastinum was defined as T4 from invasion. Because a T4-satellite lesion has two pulmonary nodules we used the larger of the two maxsuvs of the pulmonary nodules for the maxsuv data collection and analysis. Use of Neoadjuvant and Adjuvant Therapy Patients who had T4-invasive disease known before resection (as shown in Fig 1) underwent mediastinoscopy and endoscopic ultrasonography with fine-needle aspiration (even if N2 nodal disease was not suggested by integrated PET/CT or CT scan). If they were N2 positive they were eliminated from this study, but if they were N2 negative (and without clinical evidence of N1 disease) they underwent neoadjuvant chemoradiotherapy using 6,000 Gy of radiation with concurrent cisplatin-based chemotherapy. They were restaged by repeat PET and repeat CT scans. Adjuvant chemotherapy was used for patients with T2, N1, or T4 disease after resection, and in general this started between 6 and 8 weeks postoperatively. Statistical Methods Data were stored using an Excel database (Microsoft Corp, Seattle, WA) and analyzed using SAS 9.0 (SAS, Cary, NC). Continuous data are presented as means, and Table 1. Characteristics of Patients With T4 N0 M0 Lesions Characteristic categorical data are presented as percentages. Characteristics of the patients and treatments were compared by Fisher s exact test or Pearson s 2 test for discrete data and by the Wilcoxon test for continuous data. Survival estimates were derived by Kaplan Meier analysis, and log-rank test was used to assess differences in survival among groups. Stratified log-rank analyses and Cox proportional-hazards modeling were used to investigate and adjust for major prognostic and stratification factors. A two-sided probability value of less than 0.05 was considered statistically significant. Patients alive at the end of the study period were censored for purposes of data analysis. Results T4-Satellite N0 M0 N 26 T4-Invasion N0 M0 N 25 Age, median (y) 67 68 (range) (31 73) (29 70) Sex Males 17 (65%) 18 (72%) Females 9 (35%) 7 (28%) Received neoadjuvant 4 (15%) 9 (36%) treatment Histology Adenocarcinoma 13 (50%) 7 (28%) Squamous cell 6 (23%) 15 (60%) Bronchoalveolar 5 (19%) 0 (0%) carcinoma Other NSCLC 2 (8%) 3 (12%) Received adjuvant therapy 19 (73%) 21 (84%) NSCLC non small cell lung cancer. Figure 1 shows the algorithm of patients in this study. There were a total of 51 patients with T4 N0 M0 lesions as shown in Table 1. Twenty-six had node-negative T4 satellite lesions (labeled T4-satellite lesions) and 25 had node-negative T4 invasion lesions (labeled T4-invasion). The structures invaded in those patients with T4- invasion lesions included the mediastinum in 8 patients, carina in 6, trachea in 4, the left atrium in 5, and esophagus in 2. Types of tumor histology varied among the two groups, as significantly more patients with T4 satellite lesions had adenocarcinoma (50%) compared with those with T4 invasion lesions (28%; p 0.03). Additionally, significantly more patients in the T4- satellite group were resected as compared with the T4-invasion group (p 0.01). The median size of the larger of the two nodules in the patients with T4 satellite lesions was 2.4 cm (range, 1.2 to 7.6 cm). Table 2 depicts the patient characteristics for the 104 control patients with T1 N0 M0, 104 control patients with T2 N0 M0, and the 78 control patients with T1 N1 M0 or T2 N1 M0 lesions. These patients served as controls only for the

Ann Thorac Surg BRYANT ET AL 2006;82:1808 14 SATELLITE PULMONARY NODULES 1811 Table 2. Patient Characteristics of the T4-Satellite N0 M0 Patients Compared With Control Groups (T1 N0 M0, T2 N0 M0, and T1 or T2 N1 M0) Characteristic Cases Controls T4 N0 M0 Satellite Node N 26 T1 N0 M0 Controls T2 N0 M0 Controls T1 or 2 N1 M0 Controls N 78 Age Median (y) 67 69 68 68 (range) (31 73) (24 81) (26 79) (30 77) Sex Males 17 (65%) 68 (65%) 68 (65%) 51 (65%) Females 9 (35%) 36 (35%) 36 (35%) 27 (35%) Neoadjuvant treatment 4 (15%) 0 (0%) 0 (0%) 20 (26%) Histology Adenocarcinoma 13 (50%) 52 (50%) 52 (50%) 46 (59%) Squamous cell 6 (23%) 24 (23%) 24 (23%) 24 (30%) Bronchoalveolar 5 (19%) 20 (19%) 20 (19%) 2 (3%) carcinoma Other NSCLC 2 (8%) 8 (8%) 8 (8%) 6 (8%) NSCLC non small cell lung cancer. T4-satellite patients and were matched to their characteristics. Survival Multivariate analysis showed that the type of T4 disease was an independent predictor of survival (p 0.013). Figure 2 depicts the overall survival for the five groups. Patients in the T4-satellite group and T4-invasion group had a 5-year survival of 57% and 30%, respectively (p 0.009). The 5-year survival was 80% for the T1 N0 M0 patients, 68% for the T2 N0 M0 patients, and 45% for the T1 N1 M0 or T2 N1 M0 patients. The survival curves show a statistically significant difference between the T4- satellite patients compared with the T4-invasion patients. This difference was still present when we controlled for histology. The survival for the T4-satellite patients was between those with stage IB (T2 N0 M0) lesions and those with stage II (T1 N1 M0 or T2 N1 M0) lesions. The median survival was 2.3 years (range, 0.5 to 5.6 years) for T4 satellite lesions and 1.9 years for T4 invasive lesions (range, 0.4 to 5.4 years). Maximum Standardized Uptake Values Table 3 shows that the median maxsuv of patients with T4 satellite nodules differed significantly from patients with T4 N0 M0 invasion (4.8 compared with 8.8, respectively, p 0.001). The median maxsuv of the cancer was 4.2 for T1 N0 M0, 4.8 for T4-satellite, 5.4 for T2 N0 M0, 7.8 for T1 N1 M0 or T2 N1 M0, and 8.8 for the T4-invasion patients. Fig 2. Kaplan Meier survival curves stratified by the two types of T4 lesions (T4-satellite [26 patients] and T4-invasion [24 patients]) as well as by the controls (T1 N0 M0 [104 patients], T2 N0 M0 [104 patients], and T1 N0 M0 or T2 N0 M0 [78 patients]). The patients at risk for each of the T4 groups are shown below. Comment The TNM staging system for lung cancer is an international system that has been developed and fine-tuned for several decades. It is used to help select the most appropriate treatment of patients on the basis of the prognosis and survival of patients with different stages of NSCLC. The American Joint Committee on Cancer adopted a revised staging protocol in 1997 [6], in which all patients with T4 lung cancer were classified as stage IIIB. However, there are several different types of T4 tumors: mediastinal invasion, satellite tumors (a separate nodule with the same histology in the same lobe), and a metastatic pleural effusion. The survival of these patients seemed to be varied in our experience, especially those who were pathologically proven to be N0, so we decided to study the results [7, 8]. Traditionally, stage IIIB patients with T4 cancers have not been offered surgical resection because these tumors are associated with a very low 5-year survival of 5%.

1812 BRYANT ET AL Ann Thorac Surg SATELLITE PULMONARY NODULES 2006;82:1808 14 Table 3. Procedures and Outcomes of Patients With T4 N0 M0 Non Small Cell Lung Cancer Compared With Control Patients T4-Satellite N0 M0 N 26 T4-Invasion N0 M0 N 25 T1 N0 M0 T2 N0 M0 T1 or T2 N1 M0 N 78 Median maxsuv of the 4.8 8.8 4.2 5.4 7.8 primary pulmonary nodule (range) (2.1 8.5) (0 11.9) (0 10.1) (0 9.3) (0 12.6) Underwent complete 23 (87%) 12 (52%) 88 (100%) 88 (100%) 78 (100%) resection Surgery (for those completely resected) Sementectomy 2 (9%) 1 (8%) 11 (13%) 8 (9%) 5 (6%) Lobectomy or bilobectomy 18 (78%) 10 (83%) 77 (87%) 79 (90%) 71 (91%) Pneumonectomy 3 (13%) 1 (8%) 0 (0%) 1 (1%) 2 (3%) max SUV maximal standardized uptake value. However, Osaki and associates in 2003 [9] concluded that a subset of these patients deserved consideration of aggressive surgical resection. We adopted an aggressive approach for patients with T4 disease who do not have T4 because of a malignant effusion and who are mediastinal lymph node (N2) negative. Our practice has been to perform mediastinoscopy and endoscopic fine-needle aspiration in an attempt to prove benign mediastinal (N2) lymph nodes before offering thoracotomy and resection. We further select those patients who are good surgical candidates, ie, have adequate performance status, good pulmonary reserve, and normal cardiac stress tests for resection. We used neoadjuvant therapy in those patients with known T4-invasive disease (based on endoscopic ultrasonography with fine-needle aspiration that shows esophageal invasion or magnetic resonance imaging or CT that proves invasion of the mediastinum or great vessels and heart) and then attempted resection in those patients who have not progressed and still have apparent resectable disease. These important factors led to the select group of patients with T4-invasion N0 M0 disease described in this unique series. These facts must be carefully considered when interpreting these data. In this study, which features careful preoperative as well as intraoperative staging, all performed consistently by one surgeon, the survival of patients with T4-satellite N0 NSCLC is between those with stage IB and IIA disease. Before one can conclude that this means that patients who are node negative and T4 from a satellite nodule should not be staged as IIIB and should be moved up, we must look at the use of neoadjuvant and adjuvant therapy in our series. Very few of the control patients in our series received adjuvant therapy because the standard of care for using it for resected T2 N0 lesions did not start toward the end of this 7-year study. Only 36% of our patients with T4-invasion status received neoadjuvant therapy because most had unsuspected T4 invasion found at the time of thoracotomy. Thus only 52% underwent completed resection. These facts may help explain the high 5-year survival rate for T4-satellite patients seen in our series and the lower rates for the T4-invasion group when compared with other studies [10]. The high 5-year survival for the control T1 N0 M0 and T2 N0 M0 patients is possibly explained by the fact that all patients underwent FGD-PET scanning and all had a thoracic lymphadenectomy. This suggests that these patients really were stage IA or IB, and the chance of missing N2 or M1 disease is probably less than in other reported series. Bronchoalveolar carcinoma seen in 5 of the 26 patients (19%) with T4 satellite lesions represents another subset of T4 that is often N0, and an aggressive surgical approach for these patients seems to be supported by our data. Deslauriers and colleagues in 1989 [11] reported an overall 5-year survival of 21.6% for patients with satellite nodules. However, in that report most were located in the same lobe (so some were M1 lesions located in other lobes), 37% had N1 disease, 12% had N2 disease, and only 95% were completely resected. There is no mention of the use of adjuvant therapy. In our series, all patients were node negative, 17% received neoadjuvant therapy, and 73% received adjuvant therapy. Thus the difference in the survival between the two studies is explainable. The study by Osaki and coworkers in 2003 [9] reported a 26.7% 5-year survival rate for T4-satellite patients. This worse survival could be related to the lower complete resection rate than in our series (61.8% compared with 87%) and the fact that 13 of the 36 patients in their series had N2 disease and 7 of the 36 had N1 disease. Finally, there is no mention of neoadjuvant or adjuvant therapy in that study as well. They reported that only the pathologic N status and the completeness of resection were significant independent predictors of a worse prognosis. Yano and associates in 1998 [12] evaluated 352 patients with T2 to T4 NSCLC who underwent complete resection. The reported 5-year survival in that series for patients with T4 satellite lesions was 33.3%. Again, this lower survival can be explained by the fact that 50% had N2 disease. It has been shown that the maxsuv of a tumor is a way to quantify the biologic aggressiveness of a cancer in a particular host. The maxsuv has actually been shown to be a better predictor of survival than the current TNM staging system [13]. Thus, if the theory is true that a patient with a resected T4-satellite N0 M0 cancer has similar survival to one with a T2 N0 M0 or a T1 N1 M0 or

Ann Thorac Surg BRYANT ET AL 2006;82:1808 14 SATELLITE PULMONARY NODULES 1813 T2 N1 M0 cancer then the maxsuvs should be similar. In this study we used the larger of the two maxsuvs for patients with T4-satelitte nodules. The median maxsuv of the T4-satellite patients was 4.8, which compares favorably with the 4.2 maxsuv for those patients with T1 N0 M0 lesions, the 5.4 value for those with T2 N0 M0, and the 7.8 value for those with T1 N1 M0 or T2 N1 M0 lesions. This lends further credence to the belief that the theory may be correct. We and others [13, 14] have found that a high maxsuv of an NSCLC (such as 9 or 10) is a predictor of poor survival. Therefore, it is not surprising to see the T4-invasion group, which has a very low survival rate, had a median maxsuv of almost 9. Strengths of this study include the case-control design and the fact that the survival rates were compared among a homogeneous group of patients selected from our electronic prospective database. However, our study is limited by the relatively small number of patients with T4-satellite N0 M0 lesions. The results are also confounded by the use of different neoadjuvant and adjuvant therapies and the selective use of immunohistochemical staining. We were also limited by the smaller number of stage IIA control patients that could only supply a 3:1 match instead of a 4:1 match like we used for the T1 N0 M0 and the T2 N0 M0 patients. In conclusion, our data suggest that select patients with T4-satellite N0 NSCLC have a favorable prognosis with aggressive therapy. This may be because of the use of neoadjuvant therapy, undergoing complete resection, and receiving adjuvant therapy. These patients fare better than those with T4-invasion N0 M0 lesions. Therefore, the current TNM classification system should consider the separation of these heterogeneous types of T4. More data from prospective, multiinstitutional studies are needed that can provide greater patient numbers to corroborate these data, which are only from a single institution and a single surgeon. References 1. Mountain CF. Revisions in the international staging system for lung cancer. Chest 1997;111:1710 7. 2. Watanabe A, Tabata A, Ingu A, et al. [The validity and problems of a new staging system for lung cancer from the point of view of surgical results]. Kyobu Geka 2000;53: 929 32. 3. Rice TW, Blackstone EH. Radical resections for T4 lung cancer. Surg Clin North Am 2002;82:573 87. 4. Albain KS, Crowley JJ, Turrisi AT, et al. Concurrent cisplatin, etoposide, and chest radiotherapy in pathologic stage IIIB non-small-cell lung cancer: a Southwest Oncology Group phase II study, SWOG 9019. J Clin Oncol 2002;20:3454 60. 5. Cerfolio RJ, Bryant AS, Buddhiwardhan O, et al. Improving the inaccuracies of clinical staging of patients with NSCLC: a prospective trial. Ann Thorac Surg 2005;80:1207 14. 6. Greene FL, Page DL, Fleming ID, et al. AJCC Cancer Staging Manual, 5th ed. Philadelphia: Lippincott-Raven, 1997. 7. Watanabe Y, Shimuzu J, Oda M, et al. Proposals regarding some deficiencies in the new international staging system for non small cell lung cancer. Jpn J Clin Oncol 1991;21:160 8. 8. Shimuzu N, Ando A, Date H, Teramoto S. Prognosis of undetected intrapulmonary metastases in resected lung cancer. Cancer 1993;71:3868 72. 9. Osaki T, Sugio K, Hanagiri T, et al. Survival and prognostic factors of surgically resected T4 non small cell lung cancer. Ann Thorac Surg 2003;75:1745 51. 10. Yoshino I, Nakanishi R, Osaki T, et al. Postoperative prognosis in patients with non small cell lung cancer with synchronous ipsilateral intrapulmonary metastasis. Ann Thorac Surg 1997;64:809 13. 11. Deslauriers J, Brisson J, Cartier R, et al. Carcinoma of the lung: evaluation of satellite nodules as a factor influencing prognosis after resection. J Thorac Cardiovasc Surg 1989;97: 504 12. 12. Yano M, Arai T, Inagaki K, et al. Intrapulmonary satellite nodule of lung cancer as a T factor. Chest 1998;114:1305 8. 13. Cerfolio RJ, Bryant AS, Ohja B. The maximum standardized uptake values on positron emission tomography of a nonsmall cell lung cancer predict stage, recurrence and survival. J Thorac Cardiovasc Surg 2005;130:151 9. 14. Downey RJ, Akhurst T, Gonen M, et al. Preoperative F-18 fluorodeoxyglucose-positron emission tomography maximal standardized uptake value predicts survival after lung cancer resection. J Clin Oncol 2004;22:3255 60. DISCUSSION DR DAVID R. JONES (Charlottesville, VA): Doctor Bryant, that was a very nice presentation. I have two questions for you. How many of these T4 satellite lesions were identified preoperatively versus a pathologic finding when the pathologist called and said it is a T4 lesion based on a histologic finding? Second, it appears that 15% of the T4 satellite lesions had some type of induction therapy. Is it really fair to compare those patients to the T4 lesions that were treated with surgery only when you look at overall survival? DR BRYANT: Thank you for your excellent questions. I don t know the number of patients who were identified with T4 satellite lesions intraoperatively compared to those discovered postoperatively, but the vast majority were known preoperatively. As far as the question on the induction therapy, this is a good point and we agree it does muddy the waters a bit, but we controlled for this the best we could and tried to analyze this factor as well. DR THOMAS A. D AMICO (Durham, NC): Doctor Bryant, I also admired your presentation. I have a couple of questions. Number one, did you stratify the patients with T4 satellite lesions according to what the T status would be for the mother lesion (of the mother-daughter pair), because you wouldn t expect someone who had a satellite lesion that was T1 to have the same outcome as someone whose primary lesion was a T3. So, among the 25 or 26 patients who had T4 satellite lesions, did you stratify according to the T status of the larger lesion? Secondly, you mentioned that patients with satellite lesions should not be staged as stage IIIB. How do you recommend staging these? What paradigm do you think we should use? DR BRYANT: Thank you for your provocative questions. No, we did not consider the stage using the mother lesion alone. With regard to your second question as to what stage do we think it should be classified in, the best academic answer is that we

1814 BRYANT ET AL Ann Thorac Surg SATELLITE PULMONARY NODULES 2006;82:1808 14 await better data from multiinstitutional studies, but from these data we suggest stage IIA. DR FRANK DETTERBECK (New Haven, CT): I have a little bit of a follow-up on that. I think that regardless of how we classify these satellite lesions, I think the way that we think about them should be very different from a T4 that has mediastinal invasion or a IIIB that has N3 disease. I think that the pathophysiology is different. In a lot of the other studies that have looked at satellite lesions, actually the ones that have reported this have shown that the satellite lesion was in general either very close by or more peripheral than the more dominant lesion. So I was wondering if you had actually had a chance to look at that in your series, and if you hadn t, if perhaps there would be an opportunity to go back and look at that. That has some potential implications for what is actually going on pathophysiologically in these lesions. DR BRYANT: Thank you for your question. That is definitely an interesting observation, and that is something we have not yet evaluated. It would be tough now to go back and ask the pathologist to measure the distance between the lesions but perhaps we could do that with the CT (computed tomography) scan. I am unsure of the significance of the distant between them, but I agree with your point that a T4 satellite lesion is quite different from a T4 invasion or N3 lesions. DR DETTERBECK: I have another follow-up question. In the PET (positron emission tomography) scan data, you said that the maxsuv (maximal standardized uptake value) was for the two lesions put together? I didn t quite understand that. DR BRYANT: No, we said we used the highest maxsuv of the two primary lesions. DR JOHN R. ROBERTS (Nashville, TN): I just have one question that s somewhat similar to Dr Detterbeck s. I think it s often difficult to know whether these satellite nodules are either satellite nodules, synchronous malignancies, or even mets into a nodal met. My pathologists have often had trouble determining whether this was a T1 N2 or a T4 secondary to a satellite lesion. I wonder if you had any of those similar problems or if you have any comments about it. DR BRYANT: I would like to defer that question to Dr Cerfolio. DR ROBERT J. CERFOLIO (Birmingham, AL): She has done so well, I don t know why she s deferring anything to me at this point. Yes, we have had the same problem. I think you said N2. I think you meant N1. I mean sometimes it s hard to tell a large intraparenchymal N1 node from a second primary, but I think when you get in the operating room, it s clear that if it is completely surrounded by pulmonary parenchyma and it is not contiguous with another pulmonary nodule then it is a separate nodule and not an N1 node. It s interesting how many times they actually have different histology. Of course, those patients were not in this series. That would be a separate, unrelated non small cell lung cancer. But yes, we have had the same difficulties in the same way you ve mentioned. DR WAYNE L. HOFSTETTER (Houston, TX): Doctor Bryant, I enjoyed your talk very much. My question is, we all know that lung lesions are a heterogeneous group, as Dr Roberts was mentioning, did you have a chance to look at the tumor histology to the effect that some of these may be multifocal BACs (bronchoalveolar carcinoma) that you re looking at or were they stratified? DR BRYANT: Thank you for your question. We did look at the pathology and of course some are BAC. DR JEMI OLAK (Park Ridge, IL): I have a question for you about the survival curves that you have published. With a total of 51 patients contributing to the curves, how many have actually made it out to 5 years, as your study spanned 2001 to 2006? DR BRYANT: Approximately 50%. DR ARA A. VAPORCIYAN (Houston, TX): I enjoyed your talk very much. This is an area that we often offer surgery. I have one question though. I assume enrollment in the database was if the patient was resected, so this implies that the patient had a decision to go forward with a resection. What are the reasons that you use to determine resectability in a patient who presents with a T4 satellite nodule like this? DR BRYANT: Eighty-seven percent of our patients with a T4 satellite lesion were resected and 52% with a T4 invasion. DR CERFOLIO: I think that comes down to two factors. One is the risk to the patient and two is the amount of lung that is going to be resected. If the patient has a low risk and it s going to be a lobectomy, we re very, very aggressive. Now, for the first 2 years of this study we did not have a prospective database for patients that I saw in the clinic but did not operate on it was a very complete database but only on those that I operated on. I wasn t that smart. But I started a clinic database to complement my operative database in about 2002 and have it for 2003 onward. Now everybody we see in the clinic gets entered by me later that night in a prospective database. The OR (operating room) I do in between cases on my computer in my operating room office. The people we see in the clinic we do later that night. So we don t have good data for the first year or two, but we do now. But, again, it s based on the amount of lung that s going to be resected, the risk to the patient, and then really the nodal status. If they are N2 disease, we won t resect them until they get neoadjuvant or restaged.