Limited Resection for Non-small Cell Lung Cancer

Limited Resection for Non-small Cell Lung Cancer

The standard treatment for stage I non-small cell lung cancer (NSCLC) is lobectomy with mediastinal lymph node sampling or dissection. This standard of care was effectively established by the landmark publication from the Lung Cancer Study Group in 1995 demonstrating decreased local recurrence rates and a trend toward improved survival after lobectomy compared with sublobar resections, including anatomic segmentectomies requiring individual pulmonary arterial and bronchial division, as well as nonanatomic pulmonary wedge resections (Ginsberg and Rubinstein. Ann Thorac Surg. 1995;60[3]: 615).

Limited or sublobar resections have been restricted in North America and Europe to patients with diminished pulmonary reserve. Renewed enthusiasm for limited resection for NSCLC has been driven by the growing number of smaller tumors detected from the recent widespread use of CT scanning. In addition, recent retrospective observational data from the Western and Japanese literature, comparing limited resection to lobectomy, have questioned the lobectomy standard, especially for patients with peripheral tumors less than 2 cm in size (T1a in Edge et al, eds. AJCC Cancer Staging Manual. Philadelphia, PA: Springer; 2010 [AJCC 7th Edition]). In addition, clinicians are increasingly confronted with an aging population with significant comorbidities who might benefit from lesser resections (Table 1). In this review, we will examine the literature supporting the use of limited resection for stage I non-small cell lung cancer and focus on what groups of patients should be considered for such resections.

Table 1. Selected Studies of Limited Resection for Stage I NSCLC

Limited Resection Studies
North American and European studies of sublobar resection are mainly limited to analysis of patients with impaired pulmonary function. Warren and Faber compared local recurrence and survival in 68 patients who underwent segmentectomy to 105 patients who underwent lobectomy for stage I NSCLC (T1N0, T2N) (AJCC 6th Edition). Wedge resection or segmentectomy was performed at the discretion of the surgeon. A segmentectomy was performed in those patients with small peripheral lesions who would have otherwise tolerated a lobectomy. In this analysis, the rate of local recurrence was higher for the segmentectomy group than for the lobectomy group (22.7% vs 4.9%, respectively). Although there was a survival benefit for patients undergoing lobectomy for tumors greater than 3 cm, no significant survival difference was seen in patients with tumors less than 3 cm (Warren and Faber. J Thorac Cardiovasc Surg. 1994;107[4]:1087). Similarly, in an analysis of 219 patients with stage I NSCLC who underwent wedge resection or lobectomy, Landreneau and colleagues demonstrated no statistical difference in 5-year survival between the wedge (both open and thoracoscopic) and lobectomy groups (5-year survival: 58% open wedge, 65% thoracoscopic wedge, and 70% lobectomy, P=.056). In this study, the patients undergoing wedge resections had significantly worse pulmonary function (Landreneau et al. J Thorac Cardiovasc Surg 1997;113[4]:691). To control for this and other potential confounding factors, a Cox proportional hazards model was created that, in turn, demonstrated lobectomy to be associated with superior survival compared with wedge (hazard ratio 0.80, P=.042). In a comparison of 784 patients with stage I NSCLC, 207 patients treated by sublobar resection and the remaining 577 patients by lobectomy, El-Sherif and colleagues demonstrated in 2006 identical, disease-free survival of 65% at 7 years (P=.308) for patients with stage IA disease (Ann Thorac Surg. 2006; 82[2]:408).

Whereas the aforementioned Western studies describe the use of limited resection in patients with compromised pulmonary function, the Japanese literature reflects a longer standing tradition of sublobar resection in medically fit patients with early lung cancer, who would otherwise tolerate a lobectomy. In an analysis by Okada and colleagues of 139 patients with cT1N0 NSCLC, 2 cm or less in size, 70 patients underwent “extended” segmentectomy (segmentectomy with parenchymal division slightly outside segmental boundary) if they had no intraoperative evidence of nodal disease. There was no difference in 5-year survival between the two groups (87.1% in the extended segmentectomy group vs 87.7% in the lobectomy group (P=.8)) (Okada et al. Ann Thorac Surg. 2001;71[3]: 956). A similar survival was reported by Yoshikawa and coworkers in a multi-institutional Japanese trial of 55 patients with peripheral NSCLC lesions, 2 cm or less, who underwent extended segmentectomy (81.8% 5-year overall survival) (Yoshikawa et al. Ann Thorac Surg. 2002;73[4]:1055). In another analysis of 233 patients with peripheral T1N0 NSCLC, 2 cm or less (60 segmentectomies, 14 wedge, 159 lobectomies), overall 5-year survival was no different between the limited resection and lobectomy groups (89.1% vs 90.1%, respectively) (Koike et al. J Thorac Cardiovasc Surg. 2003;125[4]:924). Similarly, Watanabe and colleagues reported a study of 91 patients with stage I NSCLC with lesions less than 2 cm, who underwent either a limited resection or lobectomy (57 lobectomy, 14 wedge, 20 extended segmentectomy). There was no survival difference at 5 years between the two groups. Those who underwent an extended segmentectomy had a 5-year survival of 93%, while the lobectomy group had a 5-year survival of 84% (Watanabe et al. Jpn J Thorac Cardiovasc Surg. 2005;53[1]:29-35).

Special Populations
Small Tumors
Tumor size has been established by several studies as a prognostic factor for survival. The 7th edition of the AJCC staging system recognizes this and has further subdivided T1 tumors into T1a (= 2 cm), T1b (2-3 cm), and has placed the largest lesions more than 7 cm) into a T3 designation. In a review of 244 patients with stage IA NSCLC, we previously reported that the overall 5-year survival for patients with tumors = 2 cm to be superior to those with tumors more than 2 cm (77.2% vs 60.3%, P=.03, respectively) (Port et al. Chest. 2003;124[5]:1828). Similarly, in a study of 83 patients with subcentimeter tumors by Lee and colleagues, 5-year survival was 94%, with a diseasespecific survival of 100% (Lee et al. J Thorac Cardiovasc Surg. 2006;132[6]:1382). These findings are most likely related to the link between nodal metastases and increasing tumor size (Lee et al. Ann Thorac Surg 2007;84[1]:177). Because nodal disease would preclude a limited resection for curative intent, it would be reasonable to suggest that a limited resection be considered for T1a tumors (= 2 cm) that would be associated with a 4.8% rate of occult mediastinal nodal disease.

Elderly Population
The elderly represent a specific subset of patients in whom a limited resection may be advantageous. Not only has age been shown to be an independent predictor of mortality for patients undergoing resection, but it is also this group of patients who harbor the most significant comorbidities, with limited physiologic reserve. Whether the survival benefit from lobectomy in early stage lung cancer persists in the elderly must be questioned, as well. Analysis of the Surveillance, Epidemiology, and End Results (SEER) database, a large multiinstitutional cancer registry, by Mery and coworkers, demonstrated no survival benefit for lobectomy when compared with limited resection in patients 71 years or older (Mery et al. Chest. 2005;128[1]:237). Similarly, in an analysis of the SEER-Medicare database of patients 65 years or older, with stage IA NSCLC with lesions less than 2 cm, Wisnivesky and colleagues showed no survival benefit when lobectomy was compared with limited resection (Wisnivesky et al. Ann Surg. 2010;251[3]:550).

Second Primary Tumors
Another subset of patients where limited resection may be useful is in those patients with a second primary who have previously undergone resection. A limited resection might allow for improved postoperative function, improved pulmonary reserve, and greater likelihood that adjuvant therapies might be delivered. To support this theory, a Veteran’s Administration study has shown that both morbidity and 30-day mortality for re-resectional lung surgery is higher, with the mortality increasing with the extent of resection (Linden et al. Ann Thorac Surg. 2007;83[2]:425).

Bronchoalveolar Carcinoma
Bronchoalveolar carcinoma (BAC) is a subtype of adenocarcinoma defined as a lung tumor without invasion. BACs often present in women and nonsmokers and can be multifocal. Therefore, multiple resections may be offered to these patients. Because these lesions are often discovered as small ground glass opacities by incidental CT scan, their characterization, diagnosis, and treatment are controversial. Many clinicians have suggested more limited resection for these favorable, slow-growing lesions.

Recent reviews of the treatment of BAC have reported excellent long-term survival for limited resection, especially for small lesions. Disease-free survival has approached 95% or greater at 5 years in highly select patients (Barlesi et al. Eur J Cardiothorac Surg. 2003;24[1]:159; Fukui et al. J Thorac Oncol. 2007;2[8]:546; Nakayama et al. Ann Thorac Surg. 2007; 84[5]:1675). Subset analysis of larger ongoing trials, such as the CALGB 140503 discussed below, should help clarify the role of limited resection for BACs.

Technical Considerations
The location of a tumor must also be considered when a limited resection is planned. Wedge resections are most practical for tumors that are close to the pleural surface, preferably within the outer one-third of the lung periphery. While no standard has been accepted for the size of the wedge resection, most surgeons will attempt to obtain a margin at least equivalent to size of the lesion that is resected. For more central tumors, anatomic segmentectomies are a more appropriate option, as long as the lesion is clearly contained within the anatomic boundaries of the segment.

Limited resection can be performed either via open thoracotomy or by video-assisted thoracoscopic (VATS) techniques. Traditionally, a segmentectomy was performed by open thoracotomy because of its increased technical complexity, requiring isolation of segmental pulmonary artery and bronchi.

However, recently there has been an increased interest in performing VATS segmentectomies. To date, several reports have shown the technique to be safe and, at least, equivalent to open techniques (Schuchert et al. J Thorac Cardiovasc Surg. 2009;138[6]:1318).

Future Directions
Brachytherapy
To address the concern for a possible increase in local recurrence rates following sublobar resection, several investigators have offered adjuvant radiation therapy. Radiation can be delivered in the form of external beam, endobronchial or intraoperative brachytherapy with iodine 125 seeds placement (d’Amato TA et al. Chest. 1998;114[4]:1112). In a retrospective analysis of 124 patients undergoing sublobar resection for pT1N0 disease, 60 patients had intraoperative iodine 125 seed placement (Fernando et al. J Thorac Cardiovasc Surg 2005;129[2]:261). In this report, the local recurrence rate decreased from 17.7% to 3.3% with seed implantation. Other single institutional results report similar findings. Confirmation of these promising results may come from the American College of Surgeons Oncology Group (Z04042) trial, which is a randomized trial of sublobar resection (segmentectomy or wedge) with and without intraoperative iodine 125.

CALGB 140503
The question of whether it is appropriate to offer medically fit patients a limited resection for small peripheral lesions is under investigation in a randomized prospective controlled study. Patients with stage IA NSCLC with lesions = 2 cm are currently being randomized to limited resection vs lobectomy in the Cancer and Leukemia Group B phase III trial (CALGB 140503). An N0 nodal status is confirmed intraoperatively prior to randomization. Disease-free survival is the primary endpoint, with overall survival and rates of local and systemic recurrences as secondary endpoints. This trial, along with a similar Japanese cooperative trial ( JCOG0802), will yield a better understanding of the role of limited resection for early stage NSCLC.

Conclusions
Currently, a lobectomy remains the standard of care for patients with resectable NSCLC. Limited or sublobar resections, including wedge and segmentectomy, have been reserved, for the most part, for patients with poor pulmonary function. Recent institutional series, as well as analyses of large multi-institutional registries, suggest that lobectomy and limited resection may lead to similar disease-free and, possibly, overall survival in appropriately selected patients with early disease. It appears reasonable to offer a limited resection to medically compromised patients, the elderly, to those with multiple primary lesions, and to patients with BAC tumors. We await the results from the ongoing CALGB 140503 trial to further elucidate the role for expanding these indications to a broader patient population.

Subroto Paul, MD
Assistant Professor of Cardiothoracic Surgery; and
Jeffrey L. Port, MD
Associate Professor ofCardiothoracic Surgery
New York Presbyterian Hospital-
Weill Cornell Medical Center
New York, NY

Editor’s Insights

This short summary of the concepts and science of limited pulmonary resection for early non-small cell lung cancer represents information from a large and complex international body of clinical work published over the last 20 to 30 years. The Lung Cancer Study Group data from 1995, despite several statistical and design flaws, is the only prospective, randomized, multicenter trial on this topic published to date and is still felt by many to be the definitive reference and opinion on this issue.

Since this study, however, there has been a large accumulation of data on patients treated with limited resections for selected tumors, as referenced here, which implies that, in the appropriately selected patient, these limited resections may, in fact, lead to comparable disease-free and long-term survival. At present, the standard treatment for a stage I non-small cell lung cancer remains formal lobectomy with mediastinal lymph node sampling or dissection. From these data, however, several issues concerning limited resection in patients who would otherwise tolerate lobectomy are suggested:

1. Limited resection may be an appropriate definitive treatment for older patients (more than 70 years), for patients with either synchronous or second primary tumors, and for patients with small bronchoalveolar cancers (BAC).

2. Limited resection procedures, at present, should only be considered for patients with tumors = 2 cm in diameter, should include tumor-free margins at least equal in length to the diameter of the tumor (ie, M/T 1), and should only be offered to patients proven to be free of lymph node metastases.

At present, there are several large prospective randomized trials underway that are designed to answer some of the key questions regarding this approach to early lung tumors, such as whether or not limited resection provides comparable survival to lobectomy and whether or not adjuvant radiation therapy can improve local recurrence rates following limited resection.

It would not be surprising to find that, like patients with breast cancer, small lung cancers can be treated with limited resections, including lymph node sampling or dissection along with intraoperative or postoperative radiation therapy and achieve similar long-term outcomes to the larger operations that are presently the standard of care.

Dr. Loren J. Harris, FCCP
Deputy Editor,
Pulmonary Perspectives