Prognostic Factors in Resected Satellite Nodule T4 Non-Small Cell Lung Cancer

Prognostic Factors in Resected Satellite Nodule T4 Non-Small Cell Lung Cancer Jagan Rao, FRCS(C-Th), Rana A. Sayeed, FRCS(C-Th), Sandra Tomaszek, Stefan Fischer, MD, Shaf Keshavjee, MD, FRCSC, and Gail E. Darling, MD, FRCSC Division of Thoracic Surgery, Toronto General and Princess Margaret Hospitals, University Health Network, University of Toronto, Toronto, Ontario, Canada Background. The 1997 non-small cell lung cancer staging revisions assigned a T4 descriptor to satellite nodules in the primary tumor lobe. We reviewed our experience of satellite nodule T4 non-small cell lung cancer following these revisions and evaluated prognostic factors for this group. Methods. All patients who underwent resection of non-small cell lung cancer between April 1997 and June 2005 with satellite nodule(s) confirmed at pathologic examination were identified from our institutional Lung Tumor Registry. Case notes and pathology reports were reviewed and data collected on possible prognostic factors. Survival was modeled using the Kaplan Meier method, and survival differences between groups were analyzed using the log-rank test. Results. From 1,276 non-small cell lung cancer patients who underwent resection, 137 were staged pt4, and 35 were T4-satellite nodules. Median follow-up was 25 months (range, 1 to 102 months). Median main tumor size was 3.0 cm (range, 1 to 9.8 cm). Adenocarcinoma or bronchioloalveolar carcinoma was the predominant histologic diagnosis (n 28; 80%). One-, 3- and 5-year survival was 86%, 69%, and 57%, respectively; median survival was 68 months. During the same period, 137 patients undergoing resection for all T4 lesions had a 1-, 3-, and 5-year survival of 68%, 53%, and 18%, respectively. Adenocarcinoma or bronchioloalveolar carcinoma histologic diagnosis (adenocarcinoma or bronchioloalveolar carcinoma versus squamous, 75% versus 67% 3-year survival; p 0.0026), female gender (66% versus 49% for males, 5-year survival; p 0.041), and absence of vascular invasion (no invasion versus vascular invasion, 74% versus 20% 5-year survival; p 0.0101) were significant predictors of better survival. Conclusions. Survival for resected T4 non-small cell lung cancer with satellite nodule(s) in the primary lobe is better than for other T4 lesions, and the T4 descriptor may unduly upstage these cases. The current T4 descriptor represents a heterogeneous population. (Ann Thorac Surg 2007;84:934 9) 2007 by The Society of Thoracic Surgeons Satellite nodules are defined as well-circumscribed accessory malignant foci clearly separated from the main tumor with identical histologic characteristics [1]; the presence of satellite nodules is predictive of poor prognosis in certain malignancies such as breast cancer [2] and melanoma [3]. Satellite nodules in non-small cell lung cancer (NSCLC) were not specifically considered in early staging classifications: intrapulmonary malignant nodules were designated metastatic (M1) disease irrespective of location [4]. However, several groups reported intermediate survival for tumors with satellite nodules between those without satellite nodules and those with extrapulmonary metastatic disease [1, 5]. The 1992 American Joint Committee on Cancer and 1993 International Union against Cancer staging revisions therefore differentiated between contralateral pulmonary metastases (M1), nodules within an ipsilateral nonprimary lobe (T4), Accepted for publication April 23, 2007. Presented at the Forty-third Annual Meeting of The Society of Thoracic Surgeons, San Diego, CA, Jan 29 31, 2007. Address correspondence to Dr Darling, Division of Thoracic Surgery, Toronto General Hospital, 9N955, 200 Elizabeth Street, Toronto, Ontario, M5G 2C4, Canada; e-mail: gail.darling@uhn.on.ca. and primary lobe satellite nodules (T descriptor of the primary increased by one level) [6]. In the most recent revision of the TNM staging system for NSCLC in 1997, primary lobe satellite nodules were designated T4 and ipsilateral nonprimary lobe nodules were grouped with contralateral metastases as M1 disease [7]; 4-year survival was 8% for all T4 tumors and 2% for M1 disease. However, the T4 descriptor in this revision encompasses a diverse group of tumors, including tumors with mediastinal invasion or with an associated malignant pleural effusion, in the same category as tumors with satellite nodules. A review by Urschel and colleagues [8] reported a 20% 5-year survival for patients with satellite nodule T4 disease, higher than expected for the overall clinical T4 group. The prognostic implications of satellite nodules in NSCLC are unclear: we therefore reviewed our institutional experience with resected satellite nodule T4 NSCLC following the 1997 staging revision. Patients and Methods We performed a retrospective survey of our institutional Lung Cancer Registry. One thousand two hundred seventysix patients underwent surgical resection for NSCLC from 2007 by The Society of Thoracic Surgeons 0003-4975/07/$32.00 Published by Elsevier Inc doi:10.1016/j.athoracsur.2007.04.097

Ann Thorac Surg RAO ET AL 2007;84:934 9 SATELLITE NODULE T4 LUNG CANCER 935 Fig 1. Kaplan Meier survival plot for resected satellite nodule pt4 non small cell lung cancer. January 1997 to June 2005; 137 patients had pathologic stage T4 according to the 1997 revision of the International System for Staging Lung Cancer [7]. The surgical pathology reports of these pt4 cases were carefully reviewed to identify patients with confirmed satellite nodule(s). To be defined as a satellite nodule, the nodule was required to have identical histologic characteristics to the main tumor, or of the bronchioloalveolar carcinoma (BAC) subtype for an invasive adenocarcinoma main tumor. Bronchioloalveolar carcinoma tumors were classified according to the 2004 World Health Organization definition [9]. Multifocal pure BAC without an invasive component was excluded. Nodules with different histologic characteristics were regarded as additional primary tumors and were not included in this analysis. Patients who had been previously treated for lung cancer and those with carcinoid tumors were also excluded. Tumor characteristics, type of resection, tumor recurrence, and patient survival were determined for each patient. Follow-up information was acquired from clinic follow-up notes and from the patient s primary care physician. Survival from the date of initial pulmonary resection was modeled using the Kaplan Meier method, and survival differences between groups were analyzed using the logrank test; a probability value less than 0.05 was considered to be significant. The Princess Margaret Hospital Research Ethics Board, University Health Network, granted approval for this study, with no requirement for patient consent as this was a retrospective study. Table 1. Univariate Analysis of Prognostic Factors for Resected Satellite-Nodule Non Small Cell Lung Cancer Survival Factor n (%) 1-year (%) 3-year (%) 5-year (%) median (months) a p Value Main tumor size 3 cm 18 (51.4) 86 69 69 0.55 3 cm 17 (48.6) 82 68 46 43 Lymph nodes N0 18 (51.4) 89 64 64 0.70 N1 N3 17 (48.6) 82 76 47 43 Histologic type Adenocarcinoma or BAC 28 (80.0) 93 76 62 0.0026 Squamous cell 3 (8.6) 67 67 Other 4 (11.4) 50 6 Vascular invasion Absent 26 (74.3) 92 74 74 0.010 Present 9 (25.7) 70 50 20 37 Resection Complete 30 (85.7) 90 74 67 0.010 Incomplete 5 (14.3) 60 40 0 19 Satellite nodules Solitary 32 (91.4) 84 69 62 0.18 Multiple 3 (8.6) 100 67 33 37 Gender Male 17 (48.6) 71 58 49 37 0.041 Female 18 (51.4) 100 77 66 Smoking history Nonsmoker 6 (17.1) 100 100 75 0.099 Ex-smoker 21 (60.0) 85 69 69 Current 8 (22.9) 75 41 21 25 a Asterisk indicates median survival not reached. BAC bronchioloalveolar carcinoma.

936 RAO ET AL Ann Thorac Surg SATELLITE NODULE T4 LUNG CANCER 2007;84:934 9 Fig 2. Kaplan Meier survival plots for resected satellite nodule pt4 non small cell lung cancer grouped by patient gender (A), tumor histology (B), presence of vascular invasion (C), and completeness of resection (D). (BAC bronchioloalveolar carcinoma.) Results Between January 1997 and June 2005, 35 patients underwent resection for NSCLC that was ultimately staged pt4 based on the presence of satellite nodule(s). The presence of satellite nodule(s) was established preoperatively in 18 cases (51%) by computed tomography. In these cases all satellites were solid rather than ground-glass opacities. The presence of satellite nodule(s) was confirmed in all cases on definitive histologic examination, and results for the whole pt4 group are presented. Eighteen patients were female (51%), and median age was 62.9 years (range, 45 to 80 years). No cases were lost to follow-up, and median follow-up duration was 25 months (range, 1 to 102 months). During the same study period, 10 patients had tumor nodules in an ipsilateral nonprimary lobe (M1 disease), and 6 patients had nodules with histologic features different from the main tumor (separate primaries): these 16 patients were not included in subsequent analysis. Lobectomy was undertaken in 66% (23 patients), pneumonectomy in 11% (4 patients), and a sublobar resection, either segmentectomy or wedge resection, in the remaining 23% (8 patients). There was no in-hospital mortality. Median main tumor size was 3.0 cm (range, 1.0 to 9.8 cm). There was a solitary satellite nodule in 91% (32 patients) and two nodules in the remaining 9% (3 patients). Tumor histologic diagnosis was adenocarcinoma in 54% (19 patients), adenocarcinoma with BAC features in 26% (9 patients), large-cell carcinoma in 9% (3 patients), squamous cell carcinoma in 9% (3 patients), and mixed adenosquamous carcinoma in 2% (1 patient). In the pt4 (satellite nodule) group, 1-, 3-, and 5-year actuarial survival was 86%, 69%, and 57%, respectively; median survival was 68 months (Fig 1). By comparison, for the 137 patients undergoing surgical resection for any T4 lesion, the 1-, 3-, and 5-year actuarial survival was 68%, 53%, and 18%, respectively. Prognostic factors evaluated were main tumor size, nodal status, histologic type, presence of vascular invasion, completeness of resection, number of satellite nodules, gender, and smoking history (Table 1). Female sex, adenocarcinoma or BAC histology, the absence of vascu-

Ann Thorac Surg RAO ET AL 2007;84:934 9 SATELLITE NODULE T4 LUNG CANCER 937 lar invasion, and complete resection were significant predictors of better survival by univariate analysis. Female patients had a 66% 5-year survival versus 49% for males (p 0.041; Fig 2A). Patients with adenocarcinoma or BAC histology had a 3-year survival of 75% versus 67% for patients with squamous carcinoma and 0% for other histologic type (p 0.0026; Fig 2B). Vascular invasion predicted a poor outcome with a 5-year survival of 20% versus 74% for tumors without evidence of vascular invasion (p 0.0101; Fig 2C). Patients with a complete (R0) resection had a 5-year survival of 67% compared with 0% for an incomplete (R1 or R2) resection (p 0.0104; Fig 2D). There were trends toward better survival for nonsmokers or ex-smokers versus current smokers and for solitary versus multiple satellite nodules, but these did not reach significance; there was no difference between nodenegative N0 versus node-positive N1 through N3 tumors or between main tumors equal to or less than 3 cm in diameter or those greater than 3 cm diameter (Table 1). Comment The prognostic significance of satellite nodules in NSCLC has been uncertain as reflected in the changes in the T descriptor for a tumor associated with nodules within the primary lobe with each of the previous revisions of the TNM classification [4, 6, 7]. This study reports a 57% 5-year survival for resected satellite nodule pt4 NSCLC, which is much higher than the 18% 5-year survival for all T4 NSCLC at our institution and the 8% 4-year survival reported with the 1997 staging revision [7]. These findings suggest that satellite nodule T4 tumors represent a group of tumors with better prognosis than other T4 disease and the prognosis for this group is significantly better than for reported series of resected T4 tumors. A review of 11 earlier series after the 1997 staging revisions demonstrates a 20% 5-year survival for patients with nodules in the primary lobe or in ipsilateral nonprimary lobes, but primary lobe satellite nodule tumors had a better prognosis than those with nodules in ipsilateral nonprimary lobes [5]. The review concluded that the revised TNM classification appeared to upstage patients with satellite nodules in the primary tumor lobe. Other authors have confirmed that patients with primary lobe satellite nodules have a better prognosis than patients with extrapulmonary metastasis, but there has been disagreement whether satellite nodule NSCLC has a prognosis similar to T1 through T3 NSCLC without satellite nodules (as suggested by our data), or to other resectable T4 disease. A better prognosis for satellite nodule T4 tumors has been reported by Osaki and associates [10], who found a 27% 5-year survival for satellite nodule T4 tumors, 18% for mediastinal invasion, and 0% for malignant pleural involvement. Yano and colleagues [11] reported a 33% 5-year survival for patients with satellite nodules in the primary lobe compared with 0% for patients after complete resection of other T4 tumors or tumors with nodules in an ipsilateral nonprimary lobe. Yano and coworkers also found that their survival was similar between patients with no satellite nodule and those with primary lobe satellite nodules. These findings agree with our results, demonstrating superior survival for satellite nodule T4 tumors compared with other T4 disease. In contrast, Okumura and colleagues [12] reported a significant difference in 5-year survival in patients with no satellite nodules (60%) compared with those with primary lobe satellite nodules (34%), but there were no survival differences between patients with primary lobe satellite nodules and those with other resected T4 disease (34% 5-year survival). Nor was there any survival difference between patients with ipsilateral nonprimary lobe nodules (11%) and those with extrapulmonary metastatic disease (6%). Nakagawa and associates [13] reported similar findings with a 5-year survival of 39% for primary lobe satellite nodules compared with 31% for other resected pt4 tumors and 19% for patients with ipsilateral nonprimary lobe metastases versus 28% for those with extrapulmonary metastatic disease. Earlier series have reported a 23% to 31% 5-year survival after complete (R0) resection of any T4 NSCLC [13 15]. The 57% 5-year survival for resected satellite nodule T4 tumors reported here is strikingly better, and this difference may be related to the high percentage of adenocarcinoma or BAC in this study, perhaps reflecting a different biology of tumors that develop satellite nodules compared with those demonstrating aggressive local invasion or nonprimary lobe metastasis. These results support the need for the current ongoing review of the TNM staging system. The current T4 designation defines a heterogeneous group, and as such, although primary lobe satellite nodules may have an adverse effect on prognosis, the magnitude of that effect appears to be overstated with a T4 designation. The T4 designation as a result of primary lobe satellite nodules clearly does not portend as poor a prognosis as T4 attributable to malignant effusions, nor is it similar to nodules in the nonprimary lobe, which appear to be appropriately staged as M1. Whether satellite nodules in the primary lobe have a similar prognosis to other resected T4 tumors is difficult to determine as the complex resections required for other T4 tumors may have a significant impact on the outcome of these patients irrespective of their cancer staging. The prognosis for satellite nodule T4 NSCLC appears to be different from other resectable tumors designated T4 by virtue of mediastinal or vertebral invasion, and we believe that the current TNM staging upstages this particular subset of tumors. Several theories have been proposed regarding the origin of satellite nodules, including (1) spread through tumor thrombus in the pulmonary artery, (2) lymphogenous metastasis, (3) spread through bronchial artery invasion, (4) airborne metastasis, and (5) hematogenous spread. Shimizu and coworkers [16] postulated that most intrapulmonary metastases occurred through pulmonary arterial or retrograde lymphogenous spread because of the low incidence of local or mediastinal nodal involvement as found in our series, with 18 of 35 patients (51.4%) having pn0 disease. Nodal status has been found to be an important prognostic factor in T4 NSCLC [7, 12, 16,

938 RAO ET AL Ann Thorac Surg SATELLITE NODULE T4 LUNG CANCER 2007;84:934 9 17], and we observed a trend toward poorer survival with pn1 and pn2 disease (Table 1). Other factors that also contribute to the good survival reported here are the low incidence of vascular invasion (26%), a high frequency of adenocarcinoma or BAC histologic diagnosis (adenocarcinoma 54%, adenocarcinoma with BAC features 26%), and the high rate of complete resection (86%; Table 1). Vascular invasion was a significant negative prognostic factor in this series, with a 74% 5-year survival in patients without vascular invasion compared with 20% when vascular invasion was present, in agreement with earlier reports. Fujisawa and associates [18] studied the importance of vascular invasion with satellite nodules and found a significantly better outcome in patients without vascular invasion (34% 5-year survival) compared with patients with malignant vascular involvement (15%). The correlation of vascular invasion with prognosis is not just confined to tumors with satellite nodules: Brechot and colleagues [19] found that vascular invasion correlated with T descriptor and ptnm stage, with a higher prevalence of vascular and lymphatic invasion in advanced ptnm stages. Histologic type was also found to be a significant predictor of survival, with better survival for adenocarcinoma or BAC histologic diagnosis. The relatively high proportion of BAC histologic diagnosis in the satellite nodule(s) (9 of 35 patients, 26%) may explain the good survival in this series; indeed, 64% 5-year survival after resection of multifocal BAC has been reported [20]. Completeness of resection was the other significant prognostic factor in this series, and this has also been demonstrated before [10, 13]. Nakagawa and coworkers [13] have reported that tumor size is an important prognostic factor for primary lobe satellite nodule tumors, but it was not a significant factor in this study (Table 1). The good survival for resected satellite nodule T4 NSCLC reported in this study agrees with the findings of Battafarano and colleagues [21] who reported a 66.5% 3-year survival after resection of node-negative ipsilateral multifocal disease. That study did not differentiate between synchronous primary lung cancers and satellite nodules, but there was no survival difference between patients with multiple tumors of the same histologic type and patients with tumors of different histologic type. These findings add weight to the recommendation to continue offering surgical resection for satellite nodule T4 NSCLC. In conclusion, satellite nodules within the primary tumor lobe should not be considered a contraindication to surgical resection and should not be equated with metastatic disease; specifically, they are not equivalent in prognostic implication to nodules within the nonprimary lobes. Completeness of resection, absence of vascular invasion, and adenocarcinoma or BAC histologic type are predictors of better survival within this subset. Nonsmall cell lung cancer with satellite nodules represents a subset of T4 tumors with a better prognosis than suggested by the current TNM classification, and this should be addressed in the next revision of the TNM staging system for lung cancer. References 1. 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. 2. Andea AA, Bouwman D, Wallis T, Visscher DW. Correlation of tumor volume and surface area with lymph node status in patients with multifocal/multicentric breast carcinoma. Cancer 2004;100:20 7. 3. Balch CM, Soong SJ, Gershenwald JE, et al. Prognostic factors analysis of 17,600 melanoma patients: validation of the American Joint Committee on Cancer melanoma staging system. J Clin Oncol 2001;19:3622 34. 4. Hermanek P, Sobin LH, eds. TNM classification of malignant tumors, 4th ed. Berlin: Springer, 1987. 5. Naruke T, Goya T, Tsuchiya R, Suemasu K. Prognosis and survival in resected lung carcinoma based on the new international staging system. J Thorac Cardiovasc Surg 1988; 96:440 7. 6. Hermanek P. TNM supplement 1993: a commentary on uniform use. Berlin: Springer-Verlag, 1993. 7. Mountain CF. Revisions in the International System for Staging Lung Cancer. Chest 1997;111:1710 7. 8. Urschel JD, Urschel DM, Anderson TM, Antkowiak JG, Takita H. Prognostic implications of pulmonary satellite nodules: are the 1997 staging revisions appropriate? Lung Cancer 1998;21:83 7. 9. Travis WD. Pathology and genetics of tumours of the lung, pleura, thymus and heart. Lyon: IARC Press, 2004. 10. 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. 11. Yano M, Arai T, Inagaki K, Morita T, Nomura T, Ito H. Intrapulmonary satellite nodule of lung cancer as a T factor. Chest 1998;114:1305 8. 12. Okumura T, Asamura H, Suzuki K, Kondo H, Tsuchiya R. Intrapulmonary metastasis of non-small cell lung cancer: a prognostic assessment. J Thorac Cardiovasc Surg 2001;122:24 8. 13. Nakagawa T, Okumura N, Miyoshi K, Matsuoka T, Kameyama K. Prognostic factors in patients with ipsilateral pulmonary metastasis from non-small cell lung cancer. Eur J Cardiothorac Surg 2005;28:635 9. 14. Macchiarini P, Silvano G, Janni A, Mussi A, Chella A, Angeletti CA. Results of treatment and lessons learned from pathologically staged T4 non-small cell lung cancer. J Surg Oncol 1991;47:209 14. 15. Hsu CP, Hsu NY, Chen CY. Surgical experience in treating T4 lung cancer: its resectability, morbidity, mortality and prognosis. Eur J Surg Oncol 1996;22:171 6. 16. Shimizu N, Ando A, Date H, Teramoto S. Prognosis of undetected intrapulmonary metastases in resected lung cancer. Cancer 1993;71:3868 72. 17. Okada M, Tsubota N, Yoshimura M, Miyamoto Y, Nakai R. Evaluation of TMN classification for lung carcinoma with ipsilateral intrapulmonary metastasis. Ann Thorac Surg 1999;68:326 30. 18. Fujisawa T, Yamaguchi Y, Saitoh Y, Hiroshima K, Ohwada H. Blood and lymphatic vessel invasion as prognostic factors for patients with primary resected nonsmall cell carcinoma of the lung with intrapulmonary metastases. Cancer 1995;76: 2464 70. 19. Brechot 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;78:2111 8. 20. Roberts PF, Straznicka M, Lara PN, et al. Resection of multifocal non-small cell lung cancer when the bronchioloalveolar subtype is involved. J Thorac Cardiovasc Surg 2003;126:1597 602. 21. Battafarano RJ, Meyers BF, Guthrie TJ, Cooper JD, Patterson GA. Surgical resection of multifocal non-small cell lung cancer is associated with prolonged survival. Ann Thorac Surg 2002;74:988 94.

Ann Thorac Surg RAO ET AL 2007;84:934 9 SATELLITE NODULE T4 LUNG CANCER 939 DISCUSSION DR DAVID C. RICE (Houston, TX): I was just wondering if you felt that your excellent results may be influenced by the high proportion of pure BAC (bronchioloalveolar carcinoma) in your series of 26%. Did you perform a survival analysis excluding these patients to see if it changed? DR SAYEED: I agree that the high proportion of BAC is likely to be one factor in the good survival of our group, but we did not perform a separate survival analysis excluding these tumors. DR JOHN F. DEROSIMO (Charleston, SC): Could you please tell us what percentage of patients underwent postoperative chemotherapy in the cohort that you had? DR SAYEED: The routine use of postoperative chemotherapy only became established practice at our institution in 2004. So in fact, just about 20% of this cohort, toward the end of the study, underwent adjuvant chemotherapy. DR FRANK C. DETTERBECK (New Haven, CT): I have a couple of comments and a question as well. You have shown what many other people have shown, that a satellite lesion in the same lobe is a different beast. It does not really affect the prognosis that much, it should not be lumped in with other T4s, and I echo that. I think that you should have excluded the pure bronchioloalveolar carcinomas. A pure BAC, defined strictly, is a tumor that has a different behavior. So I think you should take those out. I think it is unfair to include them. Furthermore, as I reflect on papers from Japan and other Asian countries, I get the sense that we may be dealing with a bit of a different beast there as well. These patients have a higher incidence of satellite lesions as well as other subtle differences. I get the feeling that they are seeing a different type of patient population than what I am seeing here. I do not know exactly how to get at that. My specific question for you is regarding an observation that I have seen on a number of papers about satellite lesions. The observation is that most of the satellite nodules are peripheral to the primary tumor, suggesting that this is, perhaps, some form of local spread within that lobe, separate from lymphatic spread or something like that. I wonder if you have made that observation as well. DR SAYEED: I agree with your comments that BAC tumors behave differently. It would be possible for us to look at our data again and repeat the analyses excluding the BAC tumors, and it would be possible to look at the pattern of satellite nodules on the scans, but I am afraid I do not have those results to hand. DR ARA A. VAPORCIYAN (Houston, TX): I wanted to reecho what Dr Detterbeck said about the bronchioloalveolar diagnosis, but the other question I had is since these T4 tumors were identified pathologically, how many of these were incidental findings of satellite nodules on the final-pass specimen? Conversely, how many of them were actually clinically identified? Because your conclusion is suggesting that they are the same thing. DR SAYEED: In 23 of the 35 cases was the diagnosis established preoperatively, so that is about two thirds. And I have repeated these analyses, just looking at cases where the diagnosis was established preoperatively: the 5-year survival is 60%, about the same, and the factors that predict a better survival are the same. DR VAPORCIYAN: Mostly BAC again, is that again one of the factors that predict survival? DR SAYEED: Yes, adenocarcinoma or BAC histology predicted the best survival. DR GIUSEPPE CARDILLO (Rome, Italy): I have two questions. The percent of sublobar resections, 23%, seems to be too high. Could you please comment on this? And the second question is: The nodal status did not affect the prognostic survival. I wonder if it depends on the small number of patients? DR SAYEED: Answering the second part of your question first, there was a trend toward better survival without mediastinal nodal involvement. This is a small retrospective series, so it may be that with a larger series one might see that nodal status is a significant factor. I think that is what one would expect.