Is limited pulmonary resection equivalent to lobectomy for surgical management of stage I non-small-cell lung cancer?

Abstract

A best evidence topic in thoracic surgery was written according to a structured protocol. The question addressed was: is limited pulmonary resection equivalent to lobectomy in terms of morbidity, long-term survival and locoregional recurrence in patients with stage I non-small-cell lung cancer (NSCLC)? A total of 166 papers were found using the reported search; of which, 16 papers, including one meta-analysis and one randomized control trial (RCT), represented the best evidence to answer the clinical question. The authors, journal, date and country of publication, patient group studied, study type, relevant outcomes and results of these papers are tabulated. With regards to 5-year survival rates, the evidence is conflicting: a 2005 meta-analysis and six other retrospective or prospective nonrandomized analyses did not find any statistically significant difference when comparing lobectomy with limited resection. However, three studies found evidence of a decreased overall survival with limited resection, including the only randomized control trial, which showed a 50% increase in the cancer-related death rate (P = 0.09), and a 30% increase in the overall death rate in patients undergoing limited resection (P = 0.08). Age, tumour size and specific type of limited resection were also factors influencing the survival rates. Four studies, including the RCT, found increased locoregional recurrence rates with limited resection. There is also evidence that wedge resections, compared with segmentectomies, lead to lower survival and higher recurrence rates. In conclusion, lobectomy is still recommended for younger patients with adequate cardiopulmonary function. Although limited resection carries a decreased rate of complications and shorter hospital stays, it may also carry a higher rate of loco-regional recurrences. However, limited resection may be comparable for patients >71 years of age, and those with small peripheral tumours.

INTRODUCTION

A best evidence topic was constructed according to a structured protocol. This is fully described in the ICVTS [1].

THREE-PART QUESTION

In [patients with stage I non-small-cell lung cancer] is [limited resection] equivalent to [lobectomy] in terms of [morbidity, long-term survival and locoregional recurrence].

CLINICAL SCENARIO

A 74-year old male patient presents to your thoracic clinic complaining of shortness of breath and weight loss. Preoperative staging confirm an early-stage NSCLC in the right upper lobe with no mediastinal nodal involvement. He is a lifelong smoker. The lung function test returns an FEV1 of 1.1 l and diffusing capacity of 40%. Although a lobectomy would represent the standard treatment, you are concerned about the fitness for surgery of the patient and wonder if a limited pulmonary resection, as a less invasive surgery than lobectomy, would carry similar survival and cancer recurrence rates. You resolve to check the literature yourself.

SEARCH STRATEGY

MedLine 1950 to December 2011 using OVID interface [Non-small cell lung cancer.mp. OR Carcinoma, Non-Small-Cell Lung/] AND [lobectomy.mp] AND [limited pulmonary resection.mp OR wedge resection.mp OR segmentectomy.mp OR sublobar.mp]

The search was limited to English language articles and Human studies only.

In addition, the reference lists of each publication were searched.

SEARCH OUTCOME

A total of 166 papers were found using the reported search. From these, 16 papers provided the best evidence to answer the question. These are presented in Table 1.

Table 1:

Best evidence papers

Author, date, country, level of evidence	Patient group	Outcomes		Key results	Comments
		5-year (or specified) survival rate	Locoregional recurrence
Nakamura et al. (2005), Japan Meta-analysis (level 1a)	Lobectomy: 1887 (Wedge resection + Segmentectomy): 903	Survival difference (5 years) = 3.6%	N/A	No significant difference in survival	Considerable heterogeneity, particularly after 5 years (Q = 33.6 P = 0.0004)
Ginsberg et al. (1995), USA Randomized control trial (level 1b)	Lobectomy: 125 (Wedge resection + Segmentectomy): 122	Overall death rate (persons/year): Lobectomy: 0.089 Limited resection: 0.117	(persons/year) Lobectomy: 0.020 Limited resection: 0.060	30% increase in the death rate with limited resection (P = 0.08) 300% increase in locoregional recurrence (P = 0.008)	Inconsistent mediastinal lymph node sampling
Kraev et al. (2007) USA Retrospective cohort study (level 2b)	Lobectomy: 215 Wedge resection: 74	Survival times: Lobectomy: 5.8 ± 0.3 years Wedge resection: 4.1 ± 0.3 years	N/A	No significant difference in the overall survival. Increased survival with lobectomy for tumours <3 cm in size (P = 0.029)	Small-cell lung cancers were also included
Landreneau et al. (1997), USA Retrospective cohort study (level 2b)	Lobectomy: 117 Wedge resection (open): 42 Wedge resection (VATS): 60	Lobectomy: 70% Wedge resection (Open): 58% Wedge resection (VATS): 65%	Wedge resection (Open): 30% Wedge resection (VATS): 26% Lobectomy-19%	Significant increase in survival with lobectomy (P = 0.02) No significant difference in locoregional recurrence	The 5-year survival differences were mainly due to non-cancer-related deaths
El-Sherif et al. (2006) USA Retrospective cohort study (level 2b)	Lobectomy: 577 Segmentectomy: 85 Wedge resection: 122	Lobectomy: 54% (Wedge resection + Segmentectomy)–40%	Lobectomy: 28% (Wedge resection + Segmentectomy): 29%	Significant increase in survival with lobectomy (P = 0.0038) No difference in locoregional recurrence Decrease in 5-year disease-free survival from 58% (lobectomy) to 50% (sublobar) in stage 1B patients (P = 0.0093)	Surgical approach unknown
Koike et al. (2002), Japan Prospective non-randomized analysis (level 2b)	Lobectomy: 159 Segmentectomy: 60 Wedge resection: 14	Lobectomy: 90% (Wedge resection + Segmentectomy)–89%	Lobectomy: 1.3% (Wedge resection + Segmentectomy): 2.7%	No significant difference in survival No significant difference in locoregional recurrence	Inconsistent mediastinal lymph node sampling
Keenan et al. (2004), USA Retrospective analysis (level 2b)	Lobectomy: 146 Segmentectomy: 54	(Kaplan-Meier 4-year survival) Lobectomy: 67% Segmentectomy: 62%	Lobectomy: 7.5% Segmentectomy: 11.1%	No significant difference in survival No significant difference in locoregional recurrence	Inconsistent mediastinal lymph node sampling
Read et al. (1990), USA Retrospective study (level 2b)	Lobectomy: 113 Segmentectomy: 107 Wedge resection: 6	Lobectomy: 74% Segmentectomy: 84%	Lobectomy: 11.5% Segmentectomy: 4.4%	No significant difference in survival
Warren et al. (1994), USA Retrospective analysis (level 2b)	Lobectomy: 103 Segmentectomy: 66	Lobectomy: 65% Segmentectomy: 45% Calculated from survival graphs	Lobectomy: 4.9% Segmentectomy: 22.7%	Significant increase in overall survival with lobectomy. (P = 0.035) No significant survival difference for tumours <3 cm Significant increase in locoregional recurrence with segmentectomy	Surgical approach unknown
Okada et al. (2006), Japan Prospective nonrandomized study (level 2b)	Lobectomy: 305 (Wedge resection + Segmentectomy): 262	Lobectomy: 89.6% (Wedge resection + Segmentectomy): 89.1%	N/A	No significant difference in survival	Significant crossover between groups
Okada et al. (2001), Japan Prospective nonrandomized study (level 2b)	Lobectomy: 139 Segmentectomy: 70 Stage 1 NSCLC <2 cm	Lobectomy: 87.3% Segmentectomy: 77.7%	N/A	No significant difference in survival	Significant crossover between groups
Harpole et al. (1995), USA Retrospective analysis (level 2b)	Lobectomy: 193 Wedge resection: 75 Pneumonectomy: 21	Lobectomy: 62% Wedge resection: 61% Pneumonectomy: 270%	Lobectomy: 9.3% Wedge resection: 16% Pneumonectomy: 214.3%	No significant difference in survival No significant difference in locoregional recurrence	Trend towards decreased survival in wedge resection over lobectomy with tumours >3 cm (P = 0.067)
Mery et al. (2005), USA Retrospective cohort study (level 2b)	Lobectomy: 9875 (Wedge resection + Segmentectomy)–1403 Pneumonectomy: 292	Median survival time: <65 years–71 months 65–74 years–47 months ≥75 years–28 months	N/A	Significant increase in survival with lobectomy in patients <75 years old	Considerable heterogeneity of data. Stage II NSCLC also included in study
Rami-Porta et al. Spain (2009) Evidence-based clinical practice guidelines		Sublobar resections should be avoided in patients who can tolerate lobectomy (Grade of recommendation: A) In patients unable to undergo lobectomy, sublobar resection is an alternative that will confer similar prognosis (Degree of recommendation: B) Wedge resection should be reserved for patients aged >71 years (Degree of recommendation: B) If the intra-operative diagnosis of bronchioloalveolar carcinoma is certain, because of its non-invasive nature and lack of nodal involvement, sublobar resection with no systematic nodal dissection may be sufficient treatment for this particular type of tumour (Degree of recommendation: B)
Okumura et al. (2007), Japan Retrospective study (level 2b)	Lobectomy: 1241 Segmentectomy: 144	Tumours ≤2 cm: Lobectomy: 18% Segmentectomy: 83% Tumours >2 cm: Lobectomy: 78% Segmentectomy: 58%	N/A	Significant increase in survival with lobectomy in tumours >2 cm (P = 0.057)
Nakamura et al. (2011), Japan Prospective non-randomized study (level 2b)	Lobectomy: 289 Segmentectomy: 38 Wedge resection: 84	Lobectomy: 82.1% Segmentectomy: 87.2% Wedge resection: 55.4%	Lobectomy: 18.0% Segmentectomy: 7.9% Wedge resection: 15.5%	Significant decrease in survival with wedge resection (vs. lobectomy and segementectomy) (P = 0.0003)

RESULTS

Is there a better survival with lobectomy or limited resection for stage I NSCLC? Out of the 16 papers included in this literature review, seven papers, including the only meta-analysis conducted on the topic, found no significant difference in the overall 5-year survival rate when comparing lobectomy and limited resection [2–8].

Three studies showed a decreased survival in the sublobar resection patient population. Ginsberg and Rubinstein [9] conducted the only published randomized control trial comparing lobectomy and limited resection, showing a 30% increase in the overall death rate, and a 50% increase in the cancer-related death rate (P = 0.08 and 0.09, respectively). Landreneau et al. [10] discovered a significant survival difference at 5 years—58% survival for open wedge resection and 70% for lobectomy (P = 0.02). El-Sherif et al. [11] also reported a decreased survival with limited resection (lobectomy, 54%; limited resection, 40%; P = 0.0038). However, further analysis of the data showed that limited resection patients tended to be older and have limited node sampling. After adjusting for those variables, there was no significant difference in the survival.

Some studies showed differences in the survival, but with certain limitations. Mery et al. [12] showed a decreased survival with limited resection with patients younger than 71 years of age, and Nakamura et al. [13] reported a decreased survival specifically with wedge resections, whereas segmentectomy was comparable to lobectomy. The tumour size was also a significant factor: Okada et al. [6, 7], Okumura et al. [14], Warren and Faber [15], El-Sherif et al. [11] and Koike et al. [3], all showed no significant difference in the survival for tumours ≤2 cm in size. Above this landmark, lobectomy has been associated with an improved survival compared with limited resection: Harpole et al. [8] reported a non-significant trend towards a decreased survival with wedge resection over lobectomy in patients with tumours >3 cm (P = 0.067), and Okumura et al. [14] showed a decrease in the survival from 78% (lobectomy) to 58% (segmentectomy) for patients with tumours >2 cm (P = 0.057). The one potential exception is the study by Kraev et al., which demonstrated an improved survival with lobectomy for tumours <3 cm (P = 0.029) [16].

On the question of increased local recurrence with limited resection compared with lobectomy, Ginsberg et al. [9] reported a 300% increase in the locoregional recurrence rate in the limited resection population (P = 0.008), and similarly Warren et al. [15] also showed a significant increase in the recurrence rate (lobectomy, 4.9%; segmentectomy, 22.7%). Harpole et al. [8] determined a non-significant trend towards an increased locoregional recurrence in wedge resection patients. With regards to disease-free survival, El-Sharif et al. reported a decrease in the survival in sublobar resection patients, but this was only limited to stage 1B tumours (lobectomy, 58%; sublobar, 50%, P = 0.0093) [11]. However, as described by Rami-Porta and Tsuboi, this is generally a reasonable trade for sparing lung parenchyma [17].

Does sublobar resection translate into a reduction in morbidity? Some studies did support this: Harpole et al. [8] indicated a lower complication rate, Landreneau et al. [10] reported shorter hospital stays and Keenan et al. [4] showed a preservation of pulmonary function in patients who underwent limited resection compared with lobectomy. However, Ginsberg et al. [9] showed no significant difference in perioperative morbidity or postoperative pulmonary function between lobectomy and limited resection patients.

Some of the conflicting evidence provided by these studies may be attributed to various combinations of wedge resections and segmentectomies, which often create a single studied limited resection group [2, 7, 9, 11, 12]. Wedge resections and segmentectomies are not technically or oncologically equivalent, and there is increasing evidence that there are differences in terms of survival and recurrent rates [17, 18]. An example of this would be the recent study by Nakamura et al. [13], which showed a significantly lower survival rate wedge resection patients (55.4%) compared with segmentectomy or lobectomy (87.2 and 82.1%, respectively).

Evidence-based clinical guidelines published in 2009 state that sublobar resections should be avoided in patients who can tolerate lobectomy, with the possible exception of pure bronchioloalveolar carcinoma, where limited resection may be sufficient. However, in patients who unable to undergo lobectomy, sublobar resection is an alternative that will confer similar prognosis [17]. The 2011 NICE lung cancer guidelines also offer a similar recommendation [19].

CLINICAL BOTTOM LINE

Lobectomy remains the most reliable surgical treatment for early-stage NSCLC in young patients with adequate pulmonary reserve and cardiac function. Although limited resection carries a decreased rate of complications and shorter hospital stays, it may also carry a higher rate of loco-regional recurrences. As for older patients with cardio-pulmonary impairment or small peripheral tumours, limited resection may achieve similar survival rates when compared with lobectomy.

Conflict of interest: none declared.