Surgical modality for stage IA non-small cell lung cancer among the elderly: analysis of the Surveillance, Epidemiology, and End Results database.

BACKGROUND: The appropriate surgical modality for early-stage non-small cell lung cancer (NSCLC) among the elderly remains controversial; identifying appropriate modalities will be helpful in clinical practice.
METHODS: It's a cohort study and we explored the Surveillance, Epidemiology, and End Results (SEER) database for identifying patients aged ≥70 years with pathologic stage IA NSCLC. Three types of surgeries were compared (lobectomy, segmentectomy, and wedge resection) via survival and stratification analyses.
RESULTS: Overall, 6,197 patients were enrolled. Among patients aged ≥76 years with tumor diameters ≤1 cm, significant differences in survival were noted for segmentectomy vs. lobectomy [hazard ratio (HR) =0.294, P=0.007] and wedge resection vs. lobectomy (HR =0.548, P=0.017) but not in those with tumors diameters >1 cm. Among patients aged 70-75 years with tumor diameters >1-2 cm, significant differences in survival were observed for segmentectomy vs. lobectomy (HR =0.671, P=0.037) and segmentectomy vs. wedge resection (HR =0.556, P=0.003) and for wedge resection vs. lobectomy (HR =1.283, P=0.003) among those with tumor diameters >2-3 cm but not in those with tumor diameters ≤1 cm.
CONCLUSIONS: Both age and tumor size should be considered when selecting the surgical modality. Lobectomy is not recommended for lesions ≤1 cm among patients aged ≥76 years. Segmentectomy was associated with superior prognosis for tumor diameters >1-2 cm and survival favored lobectomy rather than wedge resection for NSCLCs >2-3 cm among patients aged 70-75 years. Surgeons could rely on personal experience to determine the appropriate surgical modality for NSCLCs >1 cm among patients aged ≥76 years and NSCLCs ≤1 cm among patients aged 70-75 years. 2020 Journal of Thoracic Disease. All rights reserved.

Introduction

Non-small cell lung cancer (NSCLC) is the most common cause of cancer-related deaths worldwide, with a median age at diagnosis of 70 years (1); thus, NSCLC is a disease of the elderly. However, age-restrictive exclusion criteria are commonplace in clinical trials. Dedicated studies need to be carefully designed to assess elderly patients due to their unique profile, multiple medical comorbidities, and increased rates of treatment-related morbidity and mortality. Approximately 10–15% of patients with NSCLC are classified into pathological stage IA (2). In the past few decades, the incidence of early-stage NSCLC has increased remarkably, largely owing to the popularization of screening methods, especially those that use low-dose computed tomography (3). Surgical treatment should be recommended as a potential cure for patients with early-stage NSCLC.

The recommended standard treatment for patients with stage IA NSCLC has been lobectomy and mediastinal lymph node dissection, with a 5-year survival rate of approximately 70% (4). However, elderly patients are at risk of being intolerant to this aggressive therapy. Currently, sublobar resection is considered as an alternative surgical modality for elderly patients with early-stage NSCLC, with sublobar resection having the advantage of preserving pulmonary function which is important for these patients, especially for those at high-risk for or with multiple primary lung cancer (5-8). However, compared with lobectomy, sublobar resection is associated with a higher tumor recurrence rate and poorer long-term outcomes (4). Currently, it has not yet been clarified whether sublobar resection is oncologically equivalent to lobectomy in early-stage NSCLC, particularly among the elderly.

Since 1995, a number of retrospective studies and one randomized controlled trial (RCT) have reported results in favor of lobectomy (4,9,10). Recently, some retrospective studies have revealed that the survival in patients with localized stage IA NSCLC after sublobar resection was non-inferior to that in those who underwent lobectomy, particularly among the elderly (2,11-13). However, these controversial findings have mainly focused on assessing outcomes of surgical modalities with regard to tumor size but not the age at diagnosis. Thus, the optimal surgical modality for early-stage NSCLC as a function of age at diagnosis and tumor size remains unclear.

To address this gap in knowledge, we evaluated the population-based Surveillance, Epidemiology, and End Results (SEER) database to assess the survival in patients ≥70 years of age with stage IA NSCLC, who underwent lobectomy, segmentectomy, or wedge resection. We hypothesized that both the age at diagnosis and tumor size could play important roles on the prognosis and should be considered in the selection of a surgical modality for elderly patients with early-stage NSCLC.

We present the following article in accordance with the STROBE reporting checklist (available at http://dx.doi.org/10.21037/jtd-20-2221).

Methods

Study population

The SEER database is a cancer statistics registry in the US and covers almost 30% of the American population. We extracted data that concerned elderly patients (≥70-year-old) who underwent lobectomy or sublobar resection for stage IA NSCLC (from 1998 to 2016). Pathologic stage IA NSCLC was defined as stage T1a/b/c N0 M0, according to the eighth edition of the American Joint Committee on Cancer criteria (14). Data for the study were extracted from 1998, because the SEER database did not differentiate segmentectomy from wedge resection until that year. We excluded patients with more than one primary NSCLC and other malignancies, as well as those who underwent chemotherapy and/or radiation treatment or had an unknown radiation and chemotherapy status. This study was based on a publicly available database; thus, it was exempted from the institutional review board approval. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013).

Study variables

We collected data of baseline patient demographics (sex, age, marital status, and race and/or ethnicity), histopathologic information (grade, histology, and size of the tumor), and surgical modalities. In this study, tumor histology subtypes included squamous cell carcinoma, adenocarcinoma, and other histologic types, such as bronchioalveolar cell carcinoma and large-cell carcinoma. Tumor size was assessed not only as a continuous variable but also as a variable: ≤1 cm (T1a categorical), >1–2 cm (T1b), and >2–3 cm (T1c). Tumors were divided into well-differentiated (grade I), moderately differentiated (grade II), poorly differentiated (grade III), and undifferentiated (grade IV). Pulmonary function was not used as a variable because it was unavailable in the SEER database. Surgical modalities were classified into lobectomy, segmentectomy, and wedge resection. Overall survival (OS) and lung cancer-specific survival (LCSS), provided in the SEER database, were the primary outcomes in this study. OS was defined as the interval from the time of diagnosis to death of any cause, while LCSS was defined as the time from diagnosis to death caused by NSCLC.

Statistical analysis

Continuous variables were analyzed using the two-sample t-test, while categorical variables were compared using Pearson’s chi-squared (χ2) test. Age at diagnosis was a continuous variable with a non-normal distribution; thus, we assessed it as a binary variable using a median age of 76 years (≥76 vs. <76 years). Survival analysis was performed using the Kaplan-Meier method. Predictors were obtained using the Cox proportional hazards model [expressed as hazard ratios (HRs) and 95% confidence intervals (CIs)]. Survival curves among surgical resections stratified by size of the tumor and age at diagnosis were compared using the log-rank test. All statistical analyses were performed using SPSS (version 25; IBM Corporation, Armonk, NY, USA). All tests were two-sided and a P value <0.05 was considered statistically significant.

Results

Baseline characteristics

A total of 6,197 records were included; 3,279 patients were treated using lobectomies and 2918 using sublobar resections (620 segmentectomies and 2,298 wedge resections). The median follow-up time for the lobectomy and sublobar resection groups were 57 and 40.5 months, respectively. The baseline information is presented in , with key information summarized as follows. The median age at diagnosis among patients included in our study was 76 years. The patients were predominantly female (55.1%), married (52.5%), and Caucasian (88.4%). Sublobar resections were tended to be performed in more elderly patients (), females, and those with smaller NSCLC (diameter ≤1 cm). The mean age at diagnosis was higher in the sublobar resection group than in the lobectomy group (76.7 vs. 75.8 years, P<0.001). There was no significant difference in the distribution of ethnicity.

Table 1

Baseline characteristics of patients

Characteristic	Total	No. (%) of patients		P
		Lobectomy (n=3,279)	Sublobar resection (n=2,918)
Age, years (mean ± SD)	76.2±4.6	75.8±4.3	76.7±4.8	<0.001
Age group, years				<0.001
70–75	3,050	1,704 (52.0)	1,346 (46.1)
≥76	3,147	1,575 (48.0)	1,572 (53.9)
Sex				0.046
Male	2,780	1,510 (46.1)	1,270 (43.5)
Female	3,417	1,769 (53.9)	1,248 (56.5)
Marital status				0.100
Married	3,255	1,761 (55.2)	1,494 (53.0)
Not married under common law	2,755	1,432 (44.8)	1,323 (47.0)
Unknown	187
Race				0.100
Caucasian	5,477	2,878 (87.9)	2,599 (89.2)
Non-Caucasian	713	398 (12.1)	315 (10.8)
Unknown	7
Grade				0.072
I	1,184	588 (19.6)	596 (22.0)
II	2,648	1,395 (46.4)	1,253 (46.3)
III	1,773	964 (32.1)	809 (29.9)
IV	106	60 (2.0)	46 (1.7)
Unknown	486
Tumor size, cm				<0.001
≤1	644	194 (5.9)	450 (15.4)
1–2	3,151	1,528 (46.6)	1,623 (55.6)
2–3	2,402	1,557 (47.5)	845 (29.0)
WHO classification				0.24
Squamous cell carcinoma	1,787	918 (28.0)	869 (29.8)
Adenocarcinoma	3,635	1,944 (59.3)	1,691 (57.3)
Others	795	417 (12.7)	378 (13.0)
Median survival, months		73	60	<0.001

No., number; SD, standard deviation.

Figure 1

Stage IA NSCLC patients stratified by age at diagnosis and surgical procedure. NSCLC, non-small cell lung cancer.

Survival analysis of lobectomy vs. sublobar resection

The survival analysis showed that lobectomy resulted in a significantly better OS than did sublobar resection (HR, 0.817; 95% CI, 0.766–0.870; P<0.001; ). There was no significant difference in LCSS between the groups (P=0.677; ). The most common causes of death, other than NSCLC, were heart disease (11.8%) and chronic obstructive pulmonary disease (8.9%). Thus, to exclude the potential confounding causes of death and extract the exact prognostic factors, subsequent analyses were mainly focused on LCSS.

Figure 2

Comparison of OS (A) and LCSS (B) between lobectomy and sublobar resection. LCSS, lung cancer-specific survival; OS, overall survival.

Sublobar resection was further subdivided into segmentectomy and wedge resection. Significant reductions in the OS rate were observed in the wedge resection group () (segmentectomy vs. wedge: HR, 0.843; 95% CI, 0.749–0.949; P=0.005; wedge resection vs. lobectomy: HR, 1.269; 95% CI, 1.186–1.357; P<0.001) and there was no significant difference in OS between lobectomy and segmentectomy groups (P=0.239). However, lobectomy and segmentectomy resulted in different LCSS rates (HR, 1.215; 95% CI, 1.024–1.442; P=0.025), as did wedge resection and segmentectomy (HR, 1.311; 95% CI, 1.099–1.563; P=0.003), with no difference between lobectomy and wedge resection (P=0.15; ). Overall, segmentectomy resulted in superior survival rates than did wedge resection.

Figure 3

OS (A) and LCSS (B) among patients undergoing lobectomy, segmentectomy, or wedge resection; OS (C) and LCSS (D) among patients 70–75 years of age undergoing lobectomy, segmentectomy, or wedge resection; OS (E) and LCSS (F) among patients ≥76 years of age undergoing lobectomy, segmentectomy, or wedge resection. LCSS, lung cancer-specific survival; Lob, lobectomy; NSCLC, non-small cell lung cancer; OS, overall survival; Seg, segmentectomy; Wed, wedge resection.

The survival analyses were also performed according to the age at diagnosis, stratified as 70–75 years and ≥76 years. First, among the NSCLC patients 70–75 years of age, wedge resection resulted in a significant reduction in the OS rate (wedge vs. segmentectomy: HR, 1.296; 95% CI, 1.070–1.569; P=0.008; lobectomy vs. wedge resection: HR, 0.799; 95% CI, 0.722–0.884; P<0.001), with no significant difference in the OS rate between lobectomy and segmentectomy (P=0.720; ). While the segmentectomy group showed superior LCSS rates (lobectomy vs. segmentectomy: HR, 1.346; 95% CI, 1.027–1.763; P=0.031; wedge resection vs. segmentectomy: HR, 1.476; 95% CI, 1.117–1.951; P=0.006), no significant difference in the LCSS rate was found between lobectomy and wedge resection (P=0.214; ). In contrast, among patients ≥76 years of age, lobectomy was associated with superior survival than wedge resection (HR, 0.806; 95% CI, 0.736–0.883; P<0.001), with no significant difference in OS between lobectomy and segmentectomy (P=0.166) or wedge resection and segmentectomy (P=0.135; ). Interestingly, there was no significant difference in the LCSS rate between either surgical modality (wedge vs. segmentectomy: P=0.096; lobectomy vs. wedge resection: P=0.840; lobectomy vs. segmentectomy: P=0.123; ).

Multivariate analysis

Further subgroup analysis was carried out using the Cox regression model to control for potential confounding factors (). The analysis with adjustments for patient and tumor variables indicated that wedge resection was independently associated with lower LCSS after lobectomy and segmentectomy among patients 70–75 years of age (wedge vs. lobectomy: HR: 1.233; 95% CI, 1.055–1.442; P=0.009; segmentectomy vs. wedge resection: HR: 0.699; 95% CI, 0.520–0.939, P=0.017), with no significant difference between lobectomy and segmentectomy. Among patients diagnosed at an age ≥76 years, the OS analysis favored lobectomy over both wedge resection and segmentectomy (P<0.001 and P=0.045, respectively). The HR for LCSS for wedge resection was 1.113 (P=0.16), compared with that for lobectomy; however, this was a trend with the difference not being statistically significant. In addition, higher grade, Caucasian race, larger tumor size, and male sex were identified as independent risk factors for LCSS among patients in the age groups 70–75 years or ≥76 years. It is interesting to note that unmarried patients had worse OS than the married patients in the age groups, 70–75 years and ≥76 years (70–75 years, HR: 1.166, 95% CI, 1.052–1.293, P=0.003; ≥76 years, HR: 1.116, 95% CI, 1.012–1.230, P=0.027), while no statistically significant difference was found in LCSS (P>0.05).

Table 2

Cox regression model for survival among patients with NSCLC diagnosed at ages 70–75 years and ≥76 years

Variables	No. (%) of patients by age at diagnosis and survival type
	70–75 years						≥76 years
	OS			LCSS			OS			LCSS
	Hazard ratio (95% CI)	P		Hazard ratio (95% CI)	P		Hazard ratio (95% CI)	P		Hazard ratio (95% CI)	P
Grade		<0.001			<0.001			<0.001			<0.001
I	1.000 (reference)			1.000 (reference)			1.000 (reference)			1.000 (reference)
II	1.706 (1.459–1.994)	<0.001		1.409 (1.240–1.601)	<0.001		1.409 (1.240–1.601)	<0.001		1.767 (1.440–2.168)	<0.001
III	2.007 (1.704–2.365)	<0.001		1.602 (1.394–1.840)	<0.001		1.602 (1.394–1.840)	<0.001		2.306 (1.848–2.878)	<0.001
IV	2.084 (1.488–2.919)	<0.001		1.983 (1.406–2.797)	<0.001		1.983 (1.406–2.797)	<0.001		4.366 (2.655–7.177)	<0.001
Race
Caucasian	1.000 (reference)			1.000 (reference)			1.000(reference)			1.000 (reference)
Non-Caucasian	0.803 (0.689–0.937)	0.005		0.735 (0.583–0.927)	0.009		0.824 (0.707–0.961)	0.013		0.764 (0.610–0.957)	0.019
Marital status
Married	1.000 (reference)			1.000 (reference)			1.000 (reference)			1.000 (reference)
Not married under common law	1.166 (1.052–1.293)	0.003		1.123 (0.967–1.304)	0.127		1.116 (1.012–1.230)	0.027		1.139 (0.982–1.322)	0.085
Sex
Male	1.000 (reference)			1.000 (reference)			1.000 (reference)			1.000 (reference)
Female	0.722 (0.652–0.800)	<0.001		0.719 (0.619–0.834)	<0.001		0.716 (0.649–0.790)	<0.001		0.673 (0.580–0.781)	<0.001
Tumor size, cm		<0.001			<0.001			<0.001			<0.001
≤1	1.000 (reference)			1.000 (reference)			1.000 (reference)			1.000 (reference)
>1–2	1.435 (1.188–1.734)	<0.001		1.687 (1.281–2.222)	<0.001		1.096 (0.919–1.306)	0.307		1.132 (0.868–1.477)	0.361
>2–3	1.578 (1.296–1.920)	<0.001		1.920 (1.444–2.554)	<0.001		1.344 (1.124–1.609)	0.001		1.580 (1.206–2.069)	0.001
WHO classification		0.006			0.717			0.034			0.974
Squamous cell carcinoma	1.000 (reference)			1.000 (reference)			1.000 (reference)			1.000 (reference)
Adenocarcinoma	0.835 (0.748–0.933)	0.001		0.948 (0.805 to 1.117)	0.526		0.873 (0.788 to 0.969)	0.010		0.982 (0.837 to 1.153)	0.827
Others	0.880 (0.741–1.044)	0.142		0.908 (0.702 to 1.176)	0.465		0.954 (0.813 to 1.119)	0.560		0.997 (0.779 to 1.277)	0.982
Surgery		<0.001			0.009			<0.001			0.327
Lobectomy	1.000 (reference)			1.000 (reference)			1.000 (reference)			1.000 (reference)
Wedge resection	1.318 (1.180–1.473)	<0.001		1.233 (1.055–1.442)	0.009		1.293 (1.170–1.428)	<0.001		1.113 (0.959–1.291)	0.160
Segmentectomy	1.044 (0.861–1.265)	0.662		0.865 (0.651–1.151)	0.321		1.174 (1.003–1.375)	0.045		0.992 (0.784–1.256)	0.950

CI, confidence interval; NSCLC, non-small cell lung cancer; LCSS, lung cancer-specific survival; OS, overall survival; WHO, World Health Organization.

Subgroup analysis

A stratified exploratory analysis was performed to assess the appropriate surgical procedure for patients with early-stage NSCLC. The LCSS analysis for the different surgical procedures, based on the age at diagnosis and tumor size is presented in . On one hand, among the more elderly patients diagnosed at ages ≥76 years, significant reduction in LCSS was observed for tumors with a diameter ≤1 cm after lobectomy (segmentectomy vs. lobectomy: HR, 0.294; 95% CI, 0.122–0.713; P=0.007; wedge resection vs. lobectomy: HR, 0.548; 95% CI, 0.334–0.897; P=0.017), while no significant difference was observed for tumors with a diameter >1–2 or >2–3 cm. In contrast, for patients 70–75 years of age, segmentectomy was associated with better survival in patients with a NSCLC tumor diameter >1–2 cm (segmentectomy vs. lobectomy: HR, 0.671; 95% CI, 0.461–0.976; P = 0.037; segmentectomy vs. wedge resection: HR, 0.556; 95% CI, 0.379–0.814; P=0.003), with lobectomy yielding superior survival rates than wedge resection among patients with a NSCLC tumor diameter >2–3 cm (wedge resection vs. lobectomy: HR, 1.283; 95% CI, 1.001–1.643; P=0.003). However, no significant difference was observed among patients with a tumor diameter ≤1 cm.

Table 3

Subgroup analysis of LCSS according to age at diagnosis and tumor size

Variables	HR (95% CI)	P
70–75 years
≤1 cm		0.749
Wed vs. Lob	1.029 (0.628–1.685)	0.910
Seg vs. Lob	0.738 (0.303–1.801)	0.505
Seg vs. Wed	0.717 (0.304–1.692)	0.448
>1–2 cm		0.008
Wed vs. Lob	1.196 (0.982–1.457)	0.075
Seg vs. Lob	0.671 (0.461–0.976)	0.037
Seg vs. Wed	0.556 (0.379–0.814)	0.003
>2–3 cm		0.143
Wed vs. Lob	1.283 (1.001–1.643)	0.049
Seg vs. Lob	1.087 (0.702–1.685)	0.709
Seg vs. Wed	0.854 (0.535–1.364)	0.509
≥76 years
≤1 cm		0.007
Wed vs. Lob	0.548 (0.334–0.897)	0.017
Seg vs. Lob	0.294 (0.122–0.713)	0.007
Seg vs. Wed	0.538 (0.227–1.272)	0.158
>1–2 cm		0.183
Wed vs. Lob	1.139 (0.932–1.391)	0.203
Seg vs. Lob	0.867 (0.631–1.193)	0.381
Seg vs. Wed	0.764 (0.555–1.052)	0.099
>2–3 cm		0.148
Wed vs. Lob	1.200 (0.975–1.475)	0.085
Seg vs. Lob	1.245 (0.891–1.740)	0.199
Seg vs. Wed	1.042 (0.733–1.480)	0.820

CI, confidence interval; HR, hazard ratio; LCSS, lung cancer-specific survival; Lob, lobectomy; Seg, segmentectomy; Wed, wedge resection

Discussion

Currently, lobectomy is the gold-standard treatment recommended for small size NSCLC (4). However, in clinical practice, lobectomy may not be tolerated by elderly patients due to compromised pulmonary reserve and multiple commodities (15); thus, sublobar resection is an alternative for them due to the reduced morbidity, better preservation of pulmonary function, and shorter operative time (16). Some retrospective studies have suggested that survival after sublobar resection was non-inferior to that after lobectomy among patients with stage IA NSCLC (17-21). Recently, an increasing number of researchers have explored the SEER database and have drawn different conclusions. For example, Moon et al. revealed that lobectomy and segmentectomy yielded equivalent OS and LCSS rates among patients with primary NSCLC with a diameter ≤2 cm without lymph node or distant metastases (22). However, the results obtained by Dai et al. showed that lobectomy was superior to segmentectomy and wedge resection in both patients with NSCLC diameter ≤1 cm and >1–2 cm (23). Thus, there is no consensus regarding individualized treatment for patients with stage IA NSCLC and for the elderly (24). Therefore, we specifically analyzed the survival outcomes in elderly patients with stage IA NSCLC obtained from the SEER database to evaluate the role of surgical modalities in outcome and found that appropriate surgical procedures should be selected based on stratification according to the age at diagnosis and tumor size.

Previous studies have investigated the surgical resection of small size NSCLC and revealed that lobectomy and sublobar resection yielded similar levels of survival among the elderly based on the SEER database. For example, Mery et al. and Wisnivesky et al. assessed patients with early-stage NSCLC and showed that the OS advantage of lobectomy disappeared among the elderly (25,26). Similarly, Moon et al. (22), Smith et al. (27), and Razi et al. (13) demonstrated no superior OS of lobectomy among patients older than 75 years of age with a tumor size ≤2 cm. Consistent with these results, our study indicated no significant difference among surgery groups in patients ≥76 years of age with stage IA NSCLC. Furthermore, our present study revealed that among NSCLC patients 70–75 years of age, the segmentectomy group showed superior LCSS rates, suggesting that for the relatively younger age bracket, segmentectomy is less aggressive and might be used as an alternative to lobectomy and could yield superior long-term outcome than wedge resection.

The existing controversies mainly focus on surgical modalities among elderly patients with early-stage NSCLC based on the tumor size, while few studies have performed subgroup analyses according to the age and tumor size. Our study further confirmed that both tumor size and age at diagnosis should be considered when selecting surgical modalities. In elderly patients ≥76 years of age, a reduction in survival rate after lobectomy was observed for NSCLC tumors with a diameter ≤1 cm, while sublobar resection was non-inferior to lobectomy for tumors with a diameter >1–3 cm. However, in patients 70–75 years of age, segmentectomy yielded superior survival than lobectomy and wedge resection for tumors with a diameter >1–2 cm and lobectomy achieved superior LCSS than wedge resection did for lesions with diameter >2–3 cm, while no significant difference was found for those with a diameter ≤1 cm. To be cautious, high-quality evidence from RCTs is needed to verify our results. Currently, there were two prospective RCTs that have compared lobectomy and sublobar resection in early-stage NSCLC and a non-randomized trial (JCOG1211) that had assessed the efficacy of segmentectomy for lung cancers (28,29). These studies enrolled patients older than 18 years of age and a separate subset analysis for the elderly will be highly anticipated.

Marital status has been previously demonstrated as an independent prognostic factor in many cancer types (30-32). A recent study showed that marital status is an independent prognostic factor for cancer-specific survival in NSCLC patients, and patients who were married had better cancer-specific survival than patients who were unmarried (30). This finding is similar to that observed in our present study where marital status was identified as an independent risk factors for OS among patients in both the age groups of 70–75 years and ≥76 years, while no statistically significant difference was found in LCSS. Furthermore, marital status may play an important role when analyzing quality of life among older adults, suggesting that being married may offer a protective mechanism against depressive symptoms and therefore against mental illnesses during late adulthood (33). Thus, being married may have a positive effect on the OS of the elders with NSCLC while the effect on LCSS may require further investigations.

The SEER database is a robust source of cancer statistics with standardized reporting protocols and annually updated follow-up data. However, our study has some limitations that should be acknowledged. First, these data were retrospectively analyzed; although some advanced statistical methods were applied to balance the covariates among the study groups, some latent biases remained that were not adjusted. A separate subset analysis for elderly patients is anticipated in RCTs to provide prospective evidence (28). Second, the information was not comprehensive. Data regarding patient comorbidities and lung function status, stereotactic body radiotherapy (SBRT), treatment selection criteria, and recurrence rate were not recorded. Data regarding chemotherapy and target therapy were also not provided. However, these therapies were seldom performed in patients with early-stage NSCLC and this limitation could have negligible impact on survival. Another important limitation was that the SEER database did not include the information on thoracotomy and video-assisted thoracoscopic surgery (VATS). Previous studies have found age and thoracotomy as independent predictors of morbidity in patients >70 years old (15). This important factor could not be analyzed in the SEER database and needs further investigation. Finally, this study focused on one primary NSCLC alone. Studies have reported that approximately 8% of patients with NSCLC have multiple lesions (34); thus, further studies are required to assess surgical modalities as a function of age and tumor size among elderly patients with multiple primary NSCLCs. Nonetheless, the results are striking and could affect future treatment planning.

In conclusion, among patients with stage IA NSCLC older than 76 years of age, decreasing survival rates after lobectomy were observed in those with a tumor diameter ≤1 cm; sublobar resection is considered as a viable alternative for these patients. For patients 70–75 years of age, segmentectomy led to better survival rates in those with NSCLC >1–2 cm in diameter, whereas lobectomy achieved superior survival rates than wedge resection did in patients with lesions >2–3 cm in diameter. Surgeons could select the modality of resection based on their own expertise and patient profile for NSCLC with diameter ≤1 cm.

The article’s supplementary files as