Complete surgical excision remains the greatest potential

ORIGINAL ARTICLE Wedge Resection for Non-small Cell Lung Cancer in Patients with Pulmonary Insufficiency: Prospective Ten-Year Survival John P. Griffin, MD,* Charles E. Eastridge, MD, Elizabeth A. Tolley, PhD, and James W. Pate, MD Background: Possibility of curative resection by lobectomy for non small cell lung cancer is often denied patients with compromised pulmonary reserve. Analysis of survival of such patients treated by wedge resection was compared with that of patients treated by standard resection, with both groups followed for 10 years. Design: A prospective 5-year cohort study. Methods: From 1988 to 1992, an observational cohort of 127 consecutive resected patients at Memphis VA Medical Center was divided into those receiving lobectomy in 81 cases and pneumonectomy in 15 cases (group I) versus 31 patients with compromised pulmonary reserve (group II), who had complete tumor excision by wedge resection. Preoperative clinical staging was corrected to surgical pathological staging after demonstration of its superiority. Survival estimates were obtained by Kaplan Meier method with curves compared by log rank tests, with all-cause mortality calculated from date of surgery. Results: Extent of disease in group I was 58% stage I, 19% stage II, and 23% stage III. In group II, extent of disease was 84% stage I, 3% stage II, and 14% stage III. Group I median survival was 26 months with 30% 5-year survival; for group II, median survival was 30 months and 32%. Kaplan Meier survival plots showed similar curves in groups I and II. Realizing less extent of disease in group II, another Kaplan Meier plot restricted to stage I and II patients showed overlapping survival curves for groups I and II. Conclusion: Survival during 10-year observation was similar for patients with pulmonary insufficiency treated by wedge resection to that of patients receiving standard resection in this single-institution consecutive cohort. Divisions of *Pulmonary and Critical Care Medicine, Preventive Medicine and Medicine Division of Biostatistics and Epidemiology, and Department of Surgery, University of Tennessee Health Science Center, Memphis, TN; and Thoracic Surgery Section Surgery Service, VA Medical Center, Memphis, TN. Address for correspondence: John P. Griffin, M.D., 956 Court Avenue, Room H314, Memphis, TN 38163. E-mail: jpgriffin@utmem.edu Deceased. Presented at 11th World Conference on Lung Cancer, July 3 6, 2005, Barcelona, Spain. No financial or potential conflicts of interest exist among any of the authors. Copyright 2006 by the International Association for the Study of Lung Cancer ISSN: 1556-0864/06/0109-0960 Key Words: Lung cancer, Surgery, Diagnosis and staging, Statistics, Survival analysis, Lung physiology, Lung pathology. (J Thorac Oncol. 2006;1: 960 964) Complete surgical excision remains the greatest potential curative therapy for patients with non-small cell lung cancer (NSCLC). Of the American Cancer Society s 2004 estimate of 173,700 cases of newly diagnosed bronchogenic carcinoma, complete resection could only be performed in approximately 20% of patients, which presumably influenced the 160,440 deaths from this disease expected that year 1 and the overall 5-year survival rate of about 15%. A high incidence of poor surgical-risk patients (with poor surgical risk based on comorbidity factors such as COPD and cardiac disease) contributes to these unfortunate results. Although many thoracic surgeons had previously favored limited resection in such patients by segmentectomy 2,3 or wedge excision, 4 the randomized study conducted by the Lung Cancer Study Group (LCSG) 5 was considered by many to have established lobectomy as the standard minimal operation for this disease, based on reported mortality and recurrence rates. As Shields 6,7 pointed out in 1989, adequate local resection is not analogous to incomplete resection, in which gross or microscopic tumor is left in situ. More aggressive attempts to perform lobectomy in individuals with marginal pulmonary physiologic reserve led to evaluation of cardiopulmonary exercise studies as well as simpler dynamic testing, such as stair climbing, 8 to discriminate between high- and low-risk patients for surgical resection. An alternative proposal in the present investigation examined the feasibility of grossly complete wedge resection for patients who would otherwise be denied the opportunity for surgical treatment. 9 In 1988, our VA hospital s decentralized computer program permitted single-institution evaluation of survival in a 5-year cohort of patients with NSCLC, 22% of whom received thoracotomy for a similar number of adenocarcinomas and squamous cell tumors of stage I and II clinical extent. 10 The goal of this study was to determine whether equivalent crude survival rates could be achieved using wedge resection for patients with poor pulmonary reserve compared with those achieved using standard lobectomy/pneumonectomy in patients without physiologic compromise. 960 Journal of Thoracic Oncology Volume 1, Number 9, November 2006

Journal of Thoracic Oncology Volume 1, Number 9, November 2006 Wedge Resection for Non-small Cell Lung Cancer METHODS A prospective observational cohort of 127 consecutive patients, all male, had resection of NSCLC by the same surgeon (C.E.E.) at Memphis VA Medical Center from 1988 through 1992. By design, this cohort was divided into two subcohorts according to type of resection and respiratory reserve status, which were completely confounded. When cardiac abnormalities were identified by usual clinical findings, cardiology consultation was obtained for clearance for general anesthesia and thoracotomy prior to entry into this prospective investigation. Group I patients received standard lobectomy in 81 instances and pneumonectomy in 15 instances. Group II consisted of 31 other patients who were considered at increased risk for standard resection because of poor physiologic respiratory reserve, based on either static pulmonary function testing demonstrating FEV 1 2 liters or dynamic measurement of lack of three-flight stair-climbing ability 11,12 to qualify for lobectomy. At surgery, the tumor was identified by palpation and completely resected with several millimeters of adjacent normal tissue, usually with staplers, and sometimes with multiple clamp and suture technique; frozen sections were taken of any marginal areas that appeared suspicious, and further local excisions were done until all margins were clear. Anatomic segmental resection was rarely used, and no thoracoscopic resections were done. Formal lymphadenectomy was not performed, but lymph nodes in the hilum, subcarina, and paratracheal mediastinum were evaluated by visualization and palpation, and grossly abnormal nodes were removed for pathological evaluation. When no abnormal nodes were found, random sampling of nodes in each of these areas was done for staging purposes. Prospective patient allocation during the 1988 to 1992 study period required use of the 1986 international clinical staging system of Mountain, 13 but because of the increased accuracy of pathological staging demonstrated in the 1997 revision of the international system, 14,15 review and modification of staging in both patient groups was subsequently performed. All patients were followed by the authors until their demise, at 3-month intervals the first year, at 6-month intervals the second year, and annually thereafter for the 10-year period. Concurrent usual thoracic surgical and pulmonary medical care were provided patients in both treatment groups, including palliative radiotherapy, chemotherapy, and supportive measures as indicated. The primary end point of this study was a comparison of all-cause mortality in the two subcohorts, and no secondary end points, such as tumor recurrence rates or comorbidities, were proposed. Survival estimates of both groups were obtained by the Kaplan Meier 16 method, and the survival curves were compared with log rank tests, which included those patients experiencing 30-day surgical mortality. Allcause mortality was calculated from the date of surgical treatment until death. To determine whether survival rates of the subcohorts were affected by confounding with staging, data were further analyzed for only patients with stage I or II extent of disease. RESULTS Table 1 portrays the distribution of cell types by technique of resection, showing a minimal predominance of squamous cell carcinomas in group I and adenocarcinomas in group II patients. Table 2 shows the distribution of pathological staging by technique of resection, demonstrating after postsurgical staging a predominance of more advanced disease in group I compared with group II. Surgical mortality included five patients who died within 30 days of their resection, all occurring in group I (three lobectomy, two pneumonectomy). Patients treated by wedge resection (group II) were an average age of 69 years (52 87). Representative pulmonary function studies in these men showed a mean FEV 1 of 1.4 liters (1.2 1.8), a mean FEV 1/FVC of 48% (42 60), and an average MVV of 42% (20 55) of predicted. All patients who were treated by the more limited resection had failed to complete the required three-flight stair-climb test to qualify for lobectomy. Mean diameter of resected adenocarcinomas was 2.6 cm (1.5 7.0), and mean diameter of squamous cell tumors was 2.8 cm (0.8 8.2). Radiation therapy was administered in an adjuvant mode in eight patients (26%) who received wedge resection; only four patients (13%) received chemotherapy, all for palliation. Although the primary end point in this study was all-cause mortality, nine of these patients (29%) were clinically free of lung cancer at the time of their death from other causes. Interestingly, 5- and 10-year crude survival cure rates of the group I and II subcohorts were almost identical. TABLE 1. Distribution of Cell Types by Technique of Resection for a Consecutive Surgical Series of Patients with Non-small Cell Lung Cancer From 1988 to 1992 at Memphis VA Medical Center Resection Cell Type Group I Group II n % n % Squamous 46 48 13 42 Adenocarcinoma 40 42 16 52 Large cell 6 6 1 3 Adenosquamous 3 3 1 3 Bronchioalveolar 1 1 0 0 Total 96 31 TABLE 2. Distribution of Pathological Staging by Technique of Resection for a Consecutive Surgical Series of Patients with Non-small Cell Lung Cancer From 1988 to 1992 at Memphis VA Medical Center Resection Stage Group I Group II n % n % I 56 58 26 84 II 18 19 1 3 IIIA 16 17 2 7 IIIB 6 6 2 7 Total 96 31 Copyright 2006 by the International Association for the Study of Lung Cancer 961

Griffin et al. Journal of Thoracic Oncology Volume 1, Number 9, November 2006 Twenty-nine patients (30%) who received lobectomy or pneumonectomy lived for five more years, and 10 patients (32%) who were treated by wedge resection reached this end point. After 10 years, only 11% of group I patients, all of whom had lobectomy, were alive, compared with 10% of group II patients. Table 3 lists a median survival rate of 23 months for patients undergoing lobectomy or pneumonectomy; for those who received the more limited resection, median survival was 31 months. Figure 1 shows a Kaplan Meier plot demonstrating similar survival of patients in groups I and II. Because of an excess of patients with more advanced disease in the standard resection group, Table 4 depicts median survival rates only for patients with stage I or stage II disease, showing 29 months for group I and 35 months for group II. Furthermore, Figure 2 compares a Kaplan Meier plot of patients in these two groups to include only those with pathological stages I or II in extent; again, the TABLE 3. Survival Estimates (months) for a Consecutive Surgically Resected Series of Patients with Non-small Cell Lung Cancer From 1988 to 1992 at Memphis VA Medical Center, Based on All-Cause Mortality Median 25% 75% Mean Standard Error Group I 23 9.5 66 45.1 5.1 Group II 31 14.0 76 47.0 7.5 FIGURE 1. Kaplan Meier plot for a consecutive surgically resected series of patients with non-small cell lung cancer from 1988 to 1992 at Memphis VA Medical Center, based on all-cause mortality. TABLE 4. Survival Estimates (months) for a Consecutive Surgically Resected Series of Patients with Pathological Stage I or II Non-small Cell Lung Cancer From 1988 to 1992 at Memphis VA Medical Center, Based on All-Cause Mortality Median 25% 75% Mean Standard Error Group I 29 14 74 53.4 6.2 Group II 35 22 77 52.0 8.2 FIGURE 2. Kaplan Meier plot for a consecutive surgically resected series of patients with pathological stage I or II nonsmall cell lung cancer from 1988 to 1992 at Memphis VA Medical Center, based on all-cause mortality. survival curves of both groups are remarkably similar. This similarity of survival curves does not appear to be influenced by surgical pathological staging. DISCUSSION The LCSG 5 collected data from a prospective multicenter clinical trial with 276 patients, comparing lobectomy with limited procedures such as segmentectomy or wedge resection. After a 3-year follow-up, there was a higher local recurrence rate with limited resection, and the LCSG 5 stated in 1995 that because of a lower death rate and fewer local regional recurrences, lobectomy was the minimal acceptable operation for patients with stage I NSCLC. Using a permissive alpha error of p 0.10, they described a significant increase in total mortality and cancer-specific mortality in patients receiving limited forms of resection. However, based on absence of differences in selected variables between treatment groups, only 247 of the 276 randomized patients were analyzed, and when the entire population of patients was compared in Kaplan Meier survival plots, the difference between treatment groups lost statistical significance. In an attached editorial, Benfield 17 stated that data are insufficient to support dogma about the superiority of lobectomy because its main proven advantage over smaller resections is fewer postoperative local recurrences...if the goal of resection is to prevent death due to systemic cancer, the LCSG results could be interpreted as having failed to show that lobectomy is better than limited resection. Prior to the LCSG 5 randomized study, others had recommended local resection rather than lobectomy in poor-risk patients. In 1972, Le Roux 2 reported 17 patients whose small peripheral bronchial carcinomas were managed by segmentectomy, with a survival rate no worse than that after more extensive resections. Two years later, Shields and Higgins, 18 in a report for the VA Surgical Group on more than 3000 patients undergoing thoracotomy for treatment of lung can- 962 Copyright 2006 by the International Association for the Study of Lung Cancer

Journal of Thoracic Oncology Volume 1, Number 9, November 2006 Wedge Resection for Non-small Cell Lung Cancer cer, found that limited resection in 40 high-risk patients showed a 5-year survival rate of 28.5%. Jensik and colleagues 19 published a continued favorable experience with segmental resection for peripheral stage I lung cancer with a 53% 5-year survival rate in 168 patients. Jacobson et al. 4 evaluated the safety of types of limited excision and favored wedge resection, with 7% major complications versus 19% with segmentectomy. In 1982 Peters 20 stated that in patients with insufficient pulmonary reserve, lobectomy cannot be the procedure of choice for early-stage lung cancer, and that a lesser resection is indicated. Errett and colleagues 21 reported 100 patients who had intentional wedge resections for peripheral clinical stage I NSCLC, with actuarial survival rates at 2 years virtually identical between wedge and lobectomy groups. Miller and Hatcher 22 in 1987 reported on limited resection of bronchogenic carcinoma for patients with marked impairment of pulmonary function. Pastorino et al., 23 as well as Read and associates, 24 studying survival after limited resection for T1N0M0 NSCLC, stated that wedge resection was the definitive treatment for very small peripheral primary lung tumors. Warren and Faber 25 found similar long-term survival for stage I NSCLC patients treated with lobectomy or limited resection, but local recurrence was greater after segmentectomy. Since the report of the LCSG 5 randomized study, continued interest in the role of limited resection for early stages of NSCLC has been evident in the medical literature. 26 Cerfolio and associates 27 reported the Mayo Clinic experience in 1996 of 35 patients with compromised pulmonary function who had local resection but yielded a 5-year survival of 44%. Wedge resection versus lobectomy for stage I NSCLC was also evaluated by Landreneau et al., 28 with 117 and 102 patients, respectively; the Kaplan Meier survival curves were nearly identical 1 year after surgery, and after 5 years, patients having wedge resection had a significantly greater non cancer-related death rate. Kodama and associates 29 evaluated segmentectomy with lymph node dissection for 46 patients with T1 N0 M0 NSCLC compared with 77 concurrent patients who received lobectomy; multivariate analysis showed that limited resection was not associated with decreased survival. Okada et al. 30 prospectively enrolled 70 patients with NSCLC of 2 cm or smaller who were able to tolerate lobectomy; the 5-year survival of patients with segmentectomy was 87%, whereas a concurrent group of similar patients receiving lobectomy experienced an identical 5-year survival rate. The usual coexistence of COPD and major compromise of pulmonary function indicates a need for conserving lung parenchyma when feasible. Many patients with lung cancer were so physiologically impaired that insistence on lobectomy as a minimal acceptable resection would deny them the opportunity of receiving the only possibly curative treatment. Multiple synchronous and metachronous occurrences of primary NSCLC give further support to the need for conservative excision techniques so that multiple potentially curative surgical resections might be possible. 7 This prospective observational study of physiologically compromised patients with NSCLC resected by the same surgeon (C.E.E.), in a single institution and with follow-up of all patients for 10 years or until their demise, adds support to this concept. CONCLUSION This group of pulmonary physiologically compromised patients who received complete wedge resection for NSCLC had survival results during 10 years of follow-up and analysis that were not statistically different from those acceptablerisk patients who concurrently received lobectomy or pneumonectomy for resection of their lung cancer. REFERENCES 1. American Cancer Society. Cancer statistics, 2004. CA Cancer J Clin 2004;54:8 29. 2. Le Roux BT. Management of bronchial carcinoma by segmental resection. Thorax 1972;27:70 74. 3. Jensik RJ, Faber LP, Milloy FJ, et al. Segmental resection for lung cancer. J Thorac Cardiovasc Surg 1973;66:563 572. 4. Jacobson MJ, Zand L, Fox RT, et al. A comparison of wedge and segmental resection of the lung. Thorax 1976;31:365 368. 5. Lung Cancer Study Group, Ginsberg RJ, Rubinstein LV. Randomized trial of lobectomy versus limited resection for T1 N0 non-small cell lung cancer. Ann Thorac Surg 1995;60:615 623. 6. Shields TW. The incomplete resection. Ann Thorac Surg 1989;46: 487 488. 7. Shields TW. Surgical therapy for carcinoma of the lung. Clin Chest Med 1982;3:369 387. 8. Pate P, Tenholder MF, Griffin JP, et al. Preoperative assessment of the high-risk patient for lung resection. Ann Thorac Surg 1996;61:1494 1500. 9. Griffin JP, Eastridge CE. Lung cancer registry: value of limited resection. Chest 1999;116:354S. 10. Griffin JP, Eastridge CE, Burch MA, et al. Lung cancer registry: Efficacy of surgical therapy. Amer Rev Respir Crit Care Med 1995;151: A847. 11. Bolton J, Weiman D, Haynes J, et al. Stair climbing as an indicator of pulmonary function. Chest 1987;92:783 788. 12. Olsen G, Bolton J, Weiman DS, et al. Stair climbing as an exercise test to predict the postoperative complications of lung resection. Chest 1991;99:587 590. 13. Mountain CF. A new international staging system for lung cancer. Chest 1986;89:225S 233S. 14. Mountain CF. Revisions in the international system for staging lung cancer. Chest 1997;111:1710 1717. 15. Mountain CF, Dressler C. Regional lymph node classification for lung cancer staging. Chest 1997;111:1718 1723. 16. Kaplan E, Meier P. Non-parametric estimation from incomplete observations. J Am Stat Assoc 1959;53:457 481. 17. Benfield JR. Lobectomy versus limited resection in T1 N0 non-small cell lung cancer. Ann Thorac Surg 1995;60:623. 18. Shields TW, Higgins GA Jr. Minimal pulmonary resection in treatment of carcinoma of the lung. Archiv Surg 1974;108:420 422. 19. Jensik RJ, Faber LP, Kittle F. Segmental resection for bronchogenic carcinoma. Ann Thorac Surg 1979;28:475 483. 20. Peters RM. The role of limited resection in carcinoma of the lung. Am J Surg 1982;143:706 710. 21. Errett LE, Wilson J, Chiu RC, et al. Wedge resection as an alternative procedure for peripheral bronchogenic carcinoma in poor-risk patients. J Thorac Cardiovasc Surg 1985;90:656 661. 22. Miller JI, Hatcher CR Jr. Limited resection of bronchogenic carcinoma in the patient with marked impairment of pulmonary function. Ann Thorac Surg 1987;44:340 343. 23. Pastorino U, Valente M, Bedini V, et al. Limited resection for Stage I lung cancer. Eur J Surg Oncol 1991;17:42 46. 24. Read RC, Yoder G, Schaeffer RC. Survival after conservative resection for T1N0M0 non-small cell lung cancer. Ann Thorac Surg 1990;49: 391 400. Copyright 2006 by the International Association for the Study of Lung Cancer 963

Griffin et al. Journal of Thoracic Oncology Volume 1, Number 9, November 2006 25. Warren WH, Faber LP. Segmentectomy versus lobectomy in patients with stage I pulmonary carcinoma. J Thorac Cardiovasc Surg 1994;107: 1087 1094. 26. Sugarbaker DJ. The case for limited surgical resection in non-small cell lung cancer. Thorax 2003;58:639 641. 27. Cerfolio RJ, Allen MS, Trastek VF, et al. Lung resection in patients with compromised pulmonary function. Ann Thorac Surg 1996;62:348 351. 28. 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;113:691 698. 29. Kodama K, Doi O, Higashiyama M, et al. Intentional limited resection for selected patients with T1 N0 M0 non-small-cell lung cancer: A single institution study. J Thorac Cardiovasc Surg 1997;114:347 353. 30. Okada M, Yoshikawa K, Hatta T, et al. 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;71:956 961. 964 Copyright 2006 by the International Association for the Study of Lung Cancer