Impact of smoking on resected lung cancer depends on epidermal growth factor receptor mutation

Abstract

OBJECTIVES

Smokers comprise the majority of surgical patients with primary lung cancer. Epidermal growth factor receptor (EGFR) mutation-negative status impacts the treatment of recurrence. However, the prognostic impact of cigarette smoking stratified by EGFR mutation status has not been reported. Therefore, we assessed its impact on patients with resected lung cancer.

METHODS

We retrospectively analysed 362 consecutive patients who underwent complete resection for stage 1 primary lung cancer at our institution between 2012 and 2021. The EGFR mutation status was evaluated using the real-time polymerase chain reaction. We compared the overall survival (OS) and disease-free survival (DFS) between patients with and without a history of smoking.

RESULTS

The EGFR mutation-negative group included 194 patients, of whom 160 (83%) had a history of smoking. Male sex (P < 0.01), forced expiratory volume in 1 s (P < 0.01) and adenocarcinoma (P < 0.01) showed significant differences between the groups. In the EGFR mutation-positive group, the 5-year OS and DFS were similar regardless of smoking status (OS: 86% vs 75%; DFS: 73% vs 73%). In the EGFR mutation-negative group, the 5-year OS and DFS were significantly poorer in the smoking group (OS: 87% vs 65%, P = 0.05; DFS: 84% vs 54%, P = 0.01). Deaths from other diseases were relatively high (n = 19, 53%).

CONCLUSIONS

Cigarette smoking may be associated with a poor prognosis in EGFR mutation-negative lung cancer but had no impact on the prognosis of the EGFR mutation-positive group. This finding underscores the potential influence of smoking on the treatment of lung cancer recurrence but also highlights its significance in contributing to death from other diseases.

Complete surgical resection provides the best possibility of cure in patients with early-stage primary lung cancer [1].

Graphical abstract

INTRODUCTION

Complete surgical resection provides the best possibility of cure in patients with early-stage primary lung cancer [1]. Cigarette smoking has been shown to affect the incidence of primary lung cancer; a previous study demonstrated that the incidence is 4.2–4.5 times higher in smokers than in non-smokers [2]. Therefore, smokers account for the majority of surgical patients with primary lung cancer. According to a national database of surgical cases in Japan, the proportion of smokers reached 64% [3]. The prognosis of smokers is poorer than that of non-smokers because of low respiratory function, preoperative comorbidity, high risk of postoperative complications and perioperative mortality [3–6].

The postoperative prognosis is not only affected by perioperative outcomes; treatment for postoperative recurrence also has a significant influence [7, 8]. The discovery of epidermal growth factor receptor (EGFR) mutations can be regarded as the pioneer of driver mutations [9, 10]. The EGFR-tyrosine kinase inhibitors (EGFR-TKIs) have had a great impact not only on lung cancer treatment but also on cancer treatment as a whole [11]. The prognosis of patients with EGFR mutation-positive lung cancer has greatly improved with the elucidation of the mechanism of EGFR mutation resistance [12, 13]. Up to the third generation of EGFR-TKIs, such as osimertinib, are proven first-line treatments for recurrent and advanced lung cancer [14], and the results of the ADAURA trial supported their use as postoperative adjuvant therapy [15]. Further improvement in treatment outcomes is expected. However, EGFR mutations are more common in non-smokers, women and Asians; therefore, patients with a smoking history do not benefit from molecular-targeted agents due to a lack of EGFR mutations [16]. Hence, cigarette smoking and EGFR mutation-negative status are independent prognostic factors in primary lung cancer. Nonetheless, the prognostic impact of cigarette smoking stratified by EGFR mutation status has not been reported.

Our hypothesis was that EGFR mutation-positive patients would have controlled lung cancer and fewer lung cancer-related deaths. In contrast, EGFR mutation-negative patients would have more lung cancer-related deaths. Accordingly, the prognostic impact of cigarette smoking in resected lung cancer would depend on the EGFR mutation status. Therefore, the goal of this study was to compare the postoperative survival of patients with resected lung cancer according to their smoking and EGFR mutation status to explore prognostic factors and provide basic data to improve the prognosis of smokers. The results of this study may provide an opportunity to derive new therapeutic strategies for this population with a poor prognosis.

MATERIALS AND METHODS

Ethical statement

This study was approved by the ethical review board of Hamamatsu University School of Medicine on 26 April 2019 (approval number: 19-041). The requirement for written informed consent was waived owing to the retrospective design of the study.

Study design and population

We retrospectively reviewed the medical records of patients who underwent complete resection for clinical stage 1 lung cancer at our institution between January 2012 and December 2021. Patients’ clinicopathological data were extracted and analysed according to their smoking and EGFR mutation status. We evaluated the postoperative survival, overall survival (OS) and disease-free survival (DFS). Multivariate analyses for OS and DFS were also performed due to confounding by various clinical factors.

Smoking history was defined as habitual smoking for any interval that was carefully evaluated in the preoperative outpatient department and on the ward at the time of admission. Chronic obstructive pulmonary disease was diagnosed in patients with a history of cigarette smoking who met one or more of the following 3 criteria: (i) obstructive changes defined by a forced expiratory volume of 1 s % of less than 70%; (ii) emphysematous changes in high-resolution computed tomography (CT) images as evaluated by surgeons (KS, AK, YM, YT and KF) who referred to radiologic reports written in the radiology department of our institution; or (iii) a previously established diagnosis of chronic obstructive pulmonary disease with treatment. Interstitial pneumonia was evaluated based on high-resolution CT images by surgeons (KS, AK, YT, and KF) who referred to radiologic reports described by the radiology department, according to the criteria of the Japanese Respiratory Society, which are consistent with the 2011 Guidelines of the American Thoracic Society [17]. The preoperative risk was evaluated using the simplified comorbidity index score [18]. Tumours were staged according to the 7th edition of the TNM Classification by the International Union for Cancer Control [19]. The EGFR mutation status was evaluated in resected specimens using real-time polymerase chain reaction (Roche Cobas EGFR mutation test v2).

Postoperative follow-up

Patients were examined at our outpatient clinic at 3- to 6-month intervals for a minimum of 5 years. Follow-up evaluations included a physical examination, chest radiography, blood tests including pertinent tumour markers and a periodic chest CT scan. In patients in whom symptoms or signs of recurrence were observed, further examinations were also performed, including chest and abdominal CT scans, brain magnetic resonance imaging, bone scintigraphy and integrated positron emission tomography. Recurrence was diagnosed based on a compatible physical examination and diagnostic imaging findings and confirmed histologically when clinically feasible.

Statistical analyses

Differences between groups were analysed using Fisher’s exact test for categorical variables or the t-test for continuous variables. All cumulative survival curves were estimated using the Kaplan–Meier method, and differences between groups were evaluated using the log-rank test. The OS was defined as the time from the date of the initial operation to the date of death of any cause or the date of the last follow-up. Disease-free survival was defined as the time from the date of the initial operation to the first recurrence of lung cancer, the date of death of any cause or the date of the last follow-up examination. Multivariate analysis was performed using Cox's proportional hazards model. All P-values were reported using two-sided analyses, and the statistical significance level was set at P < 0.05. All statistical analyses were performed using R software for Windows GUI front-end ver. 4.0.2 (R Development Core Team 2013, R Foundation for Statistical Computing, Vienna, Austria; http://www.r-project.org).

RESULTS

Patient clinicopathological characteristics

A total of 362 consecutive patients underwent complete resection for lung cancer during the study period. Among them, 259 (72%) had a history of smoking.

Compared to the non-smoking group, the smoking group had a higher proportion of men (84% vs 11%, P < 0.01), significantly lower forced expiratory volume in 1 s (FEV1) values (73% vs 78%; P < 0.01), a higher simplified comorbidity index score (8 vs 1; P < 0.01) and higher proportions of non-adenocarcinoma histologic (32% vs 6%, P < 0.01) analyses. No significant differences were observed in the distribution of the pathological stages (84% vs 88%, P = 0.32) or the median age at the time of the operation (70 vs 72 years, P = 0.59). In the smoking group, 139 (43%) patients started smoking in their teens and had a median Brinkman index of 1020; a total of 170 (54%) patients started smoking in their 20s, with a median Brinkman index of 870. No significant difference was observed in the amount of smoking between those who started smoking in their teens versus in their 20s (P = 0.14).

The EGFR mutation status was evaluated in 300 (83%) patients (213 in the smoking group and 87 in the non-smoking group). The clinicopathological characteristics of the EGFR mutation-positive and negative groups are shown in Table 1. The EGFR mutation-positive group included 106 patients, 53 (50%) of whom had a history of smoking, whereas the EGFR mutation-negative group included 194 patients, 160 (82%) of whom had a history of smoking.

Table 1:

Comparison of patient characteristics according to smoking history and epidermal growth factor receptor mutation status.

		EGFR mutation-positive			EGFR mutation-negative
		Smoking group (n = 53)	Non-smoking group (n = 53)	P-value	Smoking group (n = 160)	Non-smoking group (n = 34)	P-value
Age		68	73	<0.01	72	71	0.20
Sex (%)	Male	39 (74)	3 (6)	<0.01	137 (86)	4 (12)	<0.01
	Female	14 (26)	50 (94)		23 (14)	30 (88)
Smoking history (%)	Current	42 (79)	–	–	35 (22)	–	–
	Past	11 (21)	–	–	125 (78)	–	–
Brinkman index		600	–	–	980	–	–
Respiratory function	%VC	98	99	0.28	97	107	0.19
	FEV1%	77	78	0.48	72	78	<0.01
	%DLCO	105	94	0.20	95	94	0.43
Pulmonary	COPD	11 (21)	0 (0)	<0.01	81 (50)	0 (0)	<0.01
disease (%)	IP	1 (2)	0 (0)	1.0	17 (11)	0 (0)	0.05
Simplified comorbidity index score		8	0	<0.01	8	1	<0.01
Surgical procedure (%)	Wedge	7 (13)	3 (6)	0.32	25 (16)	1 (3)	0.05
	Segmentectomy	4 (7)	3 (6)		5 (3)	5 (15)
	Lobectomy	42 (80)	46 (86)	0.29	130 (81)	27 (79)	1.0
	Bilobectomy	0 (0)	1 (2)		0 (0)	1 (3)
pStage	I	46 (87)	49 (92)	0.53	131 (82)	29 (85)	0.81
	II	4 (7)	2 (4)		21 (13)	4 (12)
	III	2 (4)	2 (4)		7 (4)	1 (3)
	IV	1 (2)	0 (0)		1 (1)	0 (0)
Histologic diagnosis (%)	Adenocarcinoma	53 (100)	53 (100)	1.0	109 (68)	33 (97)	<0.01
	Squamous cell carcinoma	0 (0)	0 (0)		38 (24)	1 (3)
	LCNEC	0 (0)	0 (0)		7 (4)	0 (0)
	Small	0 (0)	0 (0)		1 (1)	0 (0)
	Large	0 (0)	0 (0)		2 (1)	0 (0)
	Other	0 (0)	0 (0)		3 (2)	0 (0)
Lymphatic invasion (%)		9 (17)	10 (19)	1.0	78 (49)	10 (29)	0.06
Vascular invasion (%)		12 (23)	15 (28)	0.66	104 (65)	11 (32)	<0.01

Data are presented as number (percentage) and median.

COPD: chronic obstructive pulmonary disease; DLCO: diffusion lung capacity for carbon monoxide; EGFR: epidermal growth factor receptor; FEV1: forced expiratory volume in 1 s; IP: interstitial pneumonia; LCNEC: large cell neuroendocrine carcinoma; VC: vital capacity.

In the EGFR mutation-positive group, compared with the non-smoking subgroup (53 patients), the smoking subgroup (53 patients) had a higher proportion of men (74% vs 6%; P < 0.01) and younger age when operated on (68 vs 73; P < 0.01). No significant differences were observed in the pathological stage or histologic analysis; only adenocarcinoma was included in both subgroups.

The EGFR mutation-negative group, compared with the non-smoking subgroup, included 34 patients; the smoking subgroup, which included 160 patients, had a higher proportion of men (86% vs 12%; P < 0.01), lower FEV1 values (72% vs 78%; P < 0.01), higher simplified comorbidity index scores (8 vs 1; P < 0.01) and a higher proportion of non-adenocarcinoma histologic diagnoses (32% vs 3%, P < 0.01). No significant differences were observed in the pathological stage. Anaplastic lymphoma kinase (ALK) gene translocation was observed in 1 patient in the non-smoking group.

The EGFR mutation-negative smoking group included non-adenocarcinoma patients (n = 51, 32%), and we performed a subgroup analysis regarding adenocarcinoma histologic diagnosis (n = 109). The clinicopathological characteristics of the EGFR mutation-negative adenocarcinoma in the smoking group are shown in Table 2. Compared with the non-smoking subgroup, which included 33 patients, the smoking subgroup included 109 patients and had a higher proportion of men (84% vs 12%; P < 0.01), lower FEV1 values (72% vs 78%; P < 0.01) and higher simplified comorbidity index scores (8 vs 1; P < 0.01). No significant differences were observed in the pathological stage between the 2 subgroups.

Table 2:

Comparison of patient characteristics according to smoking history and EGFR mutation-negative adenocarcinoma histologic diagnosis.

		EGFR mutation-negative
		Smoking group (n = 109)	Non-smoking group (n = 33)	P-value
Age		70	71	0.95
Sex (%)	Male	92 (84)	4 (12)	<0.01
	Female	17 (16)	29 (88)
Smoking history (%)	Current	23 (21)	–	–
	Past	86 (79)	–	–
Brinkman index		900	–	–
Respiratory function	%VC	106	108	0.43
	FEV1%	72	78	<0.01
	%DLCO	135	94	0.22
Pulmonary disease (%)	COPD	50 (46)	0 (0)	<0.01
	IP	6 (6)	0 (0)	0.34
Simplified comorbidity index score		8	1	<0.01
Surgical procedure (%)	Wedge	14 (13)	1 (3)	0.19
	Segmentectomy	3 (3)	5 (15)
	Lobectomy	92 (84)	26 (79)	0.43
	Bilobectomy	0 (0)	1 (3)
pStage	I	89 (82)	28 (85)	0.80
	II	14 (13)	4 (12)
	III	5 (4)	1 (3)
	IV	1 (1)	0 (0)
Lymphatic invasion (%)		50 (46)	9 (27)	0.07
Vascular invasion (%)		61 (56)	10 (30)	0.02

Data are presented as number (percentage) or median.

COPD: chronic obstructive pulmonary disease; DLCO: diffusion lung capacity for carbon monoxide; EGFR: epidermal growth factor receptor; FEV1: forced expiratory volume in 1 s; IP: interstitial pneumonia; VC: vital capacity.

Postoperative survival

The median follow-up duration for all patients was 46 months. Compared with the non-smoking group, the smoking group had significantly poorer 5-year OS (83% vs 65%, P = 0.02; Supplementary Material, Fig. 1a) and DFS (79% vs 58%, P = 0.01; Supplementary Material, Fig. 1b). When stratified based on the amount of cigarette smoking, the median Brinkmann index was 920. We subdivided the smoking group into light smokers (Brinkmann index <900; n = 128) and heavy smokers (≥900; n = 131). Compared to light smokers, heavy smokers showed significantly poorer prognosis, both in OS (P < 0.01) and DFS (P < 0.01); the 5-year OS rates were 82% versus 51%, respectively (light smokers vs heavy smokers); the 5-year DFS rates were 74% versus 45%, respectively.

Figure 1:

Comparison of survival rates in the epidermal growth factor receptor mutation-positive group. (a) Overall survival; (b) disease-free survival.

When stratified based on the EGFR mutation status, there were no significant differences in postoperative survival between the smoking and non-smoking subgroups in the EGFR mutation-positive group (OS: 86% vs 75%, P = 0.50; DFS: 73% vs 73%, P = 0.60; Fig. 1a, b). Recurrence was observed in 17 patients: 6 patients in the smoking group and 9 patients in the non-smoking group. Sixteen patients (94%) received ERGF-TKIs. One patient in the non-smoking group did not receive EGFR-TKIs for age-related reasons (> 80 years at recurrence). However, in the EGFR mutation-negative group, there was a clear tendency for poorer prognosis in the smoking group, although the level of significance was not met for the OS (OS: 87% vs 65%, P = 0.05; DFS: 84% vs 54%, P = 0.01; Fig. 2a, b). Recurrence was observed in 31 patients: 26 patients in the smoking group and 5 patients in the non-smoking group. Twenty-eight patients (90%) received systemic treatment for the recurrence. The ALK-TKIs were administered to a patient with ALK gene translocation. Twenty-four patients in the smoking group and 3 patients in the non-smoking group received systemic chemotherapy. The subgroup analysis of postoperative survival focused on adenocarcinoma in the EGFR mutation-negative smoking group (n = 109) as in the OS patients. There was a clear tendency for a poorer prognosis in the smoking group, although the level of significance was not met for OS (OS: 87% vs 66%, P = 0.06; DFS: 84% vs 54%, P = 0.02; Fig. 3a, b).

Figure 2:

Comparison of survival rates in the epidermal growth factor receptor mutation-negative group. (a) Overall survival; (b) disease-free survival.

Figure 3:

Comparison of survival rates in those with epidermal growth factor receptor mutation-negative adenocarcinoma. (a) Overall survival; (b) disease-free survival.

The causes of death are presented in Table 3. In the EGFR mutation-positive group, 10 patients died during the follow-up period, 5 (50%) of lung cancer and 5 (50%) of other diseases, without significant differences between the 2 subgroups. In the EGFR mutation-negative group, 39 patients died during the follow-up period, and the most frequent cause of death was lung cancer. However, death due to other diseases was relatively high in the smoking subgroup (n = 19, 53%).

Table 3:

Causes of death in each group.

	EGFR mutation-positive			EGFR mutation-negative
	Smoking group (n = 53)	Non-smoking group (n = 53)	P-value	Smoking group (n = 160)	Non-smoking group (n = 34)	P-value
Lung cancer	2	3	1.0	17	2	0.74
Other disease	3	2	1.0	19	1	0.21
Other cancer	2	0		4	0
Cardiovascular disease	0	1		5	0
Pulmonary disease	1	0		3	1
Cerebrovascular disease	0	0		1	0
Other	0	0		5	0
Unknown	0	1		1	0

EGFR: epidermal growth factor receptor.

In the EGFR mutation-positive group, 22 patients reported a recurrence (13 patients in the non-smoking group and 9 in the smoking group). Nineteen patients (86%) received EGFR-TKIs (11 patients in the non-smoking group and 8 in the smoking group).

Table 4 shows the results of the multivariate analyses of prognostic factors for OS and DFS in the EGFR cohort (n = 300). Older age [hazard ratio (HR), 1.08; P < 0.01] was a significantly unfavourable prognostic factor in OS. Older age (HR, 1.03; P < 0.01) and advanced pathological stage (HR, 2.40; P < 0.1) were significantly unfavourable prognostic factors in DFS. Cigarette smoking (HR, 1.30; P = 0.61 in OS, 2.18; P = 0.05 in DFS) and an EGFR mutation-negative status (HR, 1.41; P = 0.27 in OS, 1.32; P = 0.28 in DFS) were unfavourable prognostic factors but without statistical significance.

Table 4:

Multivariate analyses of risk factors for overall survival and disease-free survival (n = 300).

		Number of patients	OS		DFS
			HR (95% CI)	P-value	HR (95% CI)	P-value
Age		–	1.08 (1.04–1.12)	<0.01	1.06 (1.03–1.09)	<0.01
Sex	Female	117	Ref		Ref
	Male	183	1.84 (0.71–4.71)	0.21	1.00 (0.51–4.78)	0.98
Smoking history	Absent	87	Ref
	Present	213	1.30 (0.47–3.64)	0.61	2.18 (1.00–2.10)	0.05
pStage	I	255	Ref
	II or III or IV	45	1.70 (0.98–2.98)	0.06	2.41 (1.56–3.72)	<0.01
Histologic diagnosis	Ad	248	Ref
	Non-ad	52	1.33 (0.73–2.42)	0.21	1.24 (0.74–2.10)	0.41
EGFR mutation	Present	106	Ref
	Absent	194	1.41 (0.39–1.31)	0.27	1.32 (0.46–1.26)	0.28

Ad: adenocarcinoma; DFS: disease free survival; EGFR: epidermal growth factor receptor; HR: hazard ratio; Non-ad: non-adenocarcinoma; OS: overall survival.

DISCUSSION

Patients with a smoking history had a negative trend for OS that did not reach statistical significance but significantly poorer DFS in the EGFR mutation-negative group. All patients in the EGFR mutation-positive group had only adenocarcinoma, whereas other histologic types and vascular and lymphatic invasions were frequently observed in the EGFR mutation-negative group. However, an analysis focused on EGFR mutation-negative adenocarcinoma showed that smoking history also had a negative trend for OS and a significantly poorer DFS. A frequent cause of death in the EGFR mutation-negative group was not only lung cancer but also death of other diseases. Multivariate analyses indicated that smoking history (HR; 1.30 in OS, 2.18 in DFS) and an EGFR mutation-negative status (HR; 1.41 in OS, 1.32 in DFS) were unfavourable prognostic factors, but the findings were not statistically significant.

Cigarette smoking leads to a poor prognosis after curative resection. Our study showed a significant difference in the 5-year OS and DFS between the smoking and non-smoking groups. Previous studies have also demonstrated poorer survival in patients with a history of smoking [5, 6, 20]. Maeda et al. found a statistically significant difference in the OS between smokers and non-smokers (92% vs 76%, P < 0.001) and suggested that the biological aggressiveness of primary lung cancer was one of the main causes of the poorer prognosis of smokers.

Although multiple factors influence survival after complete resection, we focused on EGFR mutation status. The excellent therapeutic effect of EGFR-TKIs has been described extensively, and the efficacy of perioperative adjuvant therapy has recently been demonstrated [15, 21]. In the FLAURA trial, the hazard ratio of progression-free survival reached only 0.46 with osimertinib and was lower than that with other EGFR-TKIs (gefitinib or erlotinib) [1, 14]. The majority of patients with a smoking history have an EGFR mutation-negative status, and only a few have an EGFR mutation-positive status (20% in this study).

In this study, the prognosis of patients who were EGFR mutation-positive did not differ based on smoking status; the 5-year OS and DFS were similar between the 2 subgroups. In the EGFR-positive patients, the age of the patients in the non-smoking group was significantly older than that of the smoking group (P < 0.01). The EGFR positive non-smoking group also showed significant differences in female sex (P < 0.01) and simplified comorbidity index score (P < 0.01), not only older age. Although older age was a significant prognostic factor, the HR was only 1.08 (OS) and 1.06 (DFS). Therefore, other favourable prognostic factors led to the result. A past study demonstrated that a simplified comorbidity index score had a prognostic impact [22]. The prognosis of the non-smoking group would have been better than that of the smoking group if the ages of the 2 groups had been adjusted in the analysis. Furthermore, deaths of lung cancer were observed in only 5 (5%) of the 106 patients in this group. A previous study reported long-term survival after recurrence with EGFR-TKIs [23]. After complete resection, monitoring for recurrence is key to improving survival [23]. The therapeutic effects of EGFR-TKIs may also have contributed to the results of our study.

Our findings only partially confirmed our hypothesis. Patients with a smoking history had a poorer prognosis, particularly those in the EGFR mutation-negative group. Notably, the proportion of deaths from other diseases was as high as 53% in the EGFR mutation-negative patients with a history of smoking. This result suggested that the prognosis of smokers is not influenced only by the outcome of lung cancer treatment. Smoking history is a risk factor for cardiovascular disease due to the promotion of arteriosclerosis [24]. Iso et al. showed that current smokers have a relative risk of cardiovascular disease of 1.60 for males and 2.06 for females compared with non-smokers [25]. In addition, cigarette smoking increases the risk for other cancers; oesophageal and pharyngeal cancers have widely been documented to be associated with a history of smoking [26, 27]. The incidence of gastric and colon cancer was 2.0 and 1.27 times higher in smokers than in non-smokers, respectively [28, 29]. These diseases affected the postoperative survival in our study.

EGFR gene mutations are more common in non-smokers and women [16]. We chose the prognostic factors based on past reports about postoperative recurrence and these confounding factors [6, 16, 23]. In the multivariable analyses, cigarette smoking (HR, 1.30 in OS, 2.18 in DFS) and EGFR mutation-negative status (HR, 1.41 in OS, 1.32 in DFS) were unfavourable prognostic factors. Older age and advanced pathological stage were extracted as significant factors; hence our results were reproducible when compared with those of past studies. Because these factors did not reach statistical significance, we need to reconsider the results with more cases.

When patients who died of other diseases were excluded from the analysis, no significant differences in lung cancer-specific survival rates were observed between ever- and non-smokers [4]; however, the reason remains unclear. In this study, the number of smokers dying of other diseases was lower in EGFR mutation-positive patients compared to EGFR mutation-negative patients. The EGFR mutation-positive group included light smokers (the median Brinkman index was 600). Conversely, the EGFR mutation-negative group included heavy smokers (the median Brinkman index was 980). This difference was statistically significant (P < 0.01). Thus, the amount of cigarette smoking also affected the prognosis. In a subgroup analysis of EGFR mutation-negative adenocarcinoma, the results were the same as those of the population including other histologic diagnoses. The influence of death of other diseases related to cigarette smoking was greater than that of lung cancer-specific death.

Our study has some limitations. First, it was a retrospective study that included only Japanese patients from 1 institution. Our database is extensive and has been consistently updated with uniform follow-up. The median follow-up term was 46 months, so it may be inadequate to evaluate long-term outcomes. However, although a general rule was applied to all patients, postoperative surveillance was inconsistent, and sample size (n = 362) was small for drawing definitive conclusions. Second, although there was no significant difference in the expression of programmed death-ligand 1 (PD-L1), 12 patients in the EGFR mutation-negative smoking group were positive for PD-L1. Because PD-L1 expression was evaluated in only 15 patients, a more detailed examination of the factors that significantly affect the treatment strategy is needed. Third, several values are missing, e.g. diffusion lung capacity for carbon monoxide, tumour spread through air spaces and PD-L1; thus, we should improve the precision of data collection. Fourth, Asians are more likely to have EGFR mutations than other races. This study included only Asians; therefore, we could not evaluate the prognostic impact in other populations.

The latest regimen that combines systemic chemotherapy and multiple immune checkpoint inhibitors improves the survival of patients without an EGFR mutation [30]. The POSEIDON study included patients with a younger median age and provided no description of comorbidities, but the proportion of patients with a history of smoking was more than 80%. Because a history of smoking is a risk factor for various diseases, and death of other cancers and other diseases contributes to the prognosis, we should carefully consider the treatment strategy for patients with lung cancer with a history of smoking. Although new regimens are expected to improve the survival of EGFR mutation-negative patients in the future, comprehensive management, including that for comorbidities, is essential.

CONCLUSIONS

Cigarette smoking might be associated with a poor prognosis in EGFR mutation-negative lung cancer but had no impact on the prognosis of EGFR mutation-positive lung cancer. These findings have implications not only for the treatment of lung cancer recurrence but also for death of other diseases.

Glossary

ABBREVIATIONS

ALK

Anaplastic lymphoma kinase

ALK-TKIs

Anaplastic lymphoma kinase tyrosine kinase inhibitors

COPD

Chronic obstructive pulmonary disease

CT

Computed tomography

DFS

Disease-free survival

EGFR

Epidermal growth factor receptor

EGFR-TKIs

Epidermal growth factor receptor–tyrosine kinase inhibitors

OS

Overall survival

SUPPLEMENTARY DATA

Supplementary material is available at Interactive CardioVascular and Thoracic Surgery online.

FUNDING

This work was supported in part by the Smoking Research Foundation [Grant Number 108].

Conflict of interest: none declared.

DATA AVAILABILITY

The data underlying this article will be shared on reasonable request to the corresponding author.

Author contributions

Keigo Sekihara: Data curation; Investigation; Methodology; Writing—original draft; Writing—review and editing. Akikazu Kawase: Methodology; Writing—original draft; Writing—review and editing. Yuta Matsubayashi: Data curation; Investigation; Writing—original draft; Writing—review and editing. Tomoya Tajiri: Data curation; Methodology; Validation; Writing—original draft; Writing—review and editing. Motohisa Shibata: Data curation; Investigation; Writing—original draft; Writing—review and editing. Takamitsu Hayakawa: Data curation; Writing—original draft; Writing—review and editing. Norihiko Shiiya: Data curation; Project administration; Writing—original draft; Writing—review and editing. Kazuhito Funai: Data curation; Formal analysis; Funding acquisition; Investigation; Methodology; Project administration; Resources; Writing—original draft; Writing—review and editing.

Reviewer information

Interactive CardioVascular and Thoracic Surgery thanks Toru Bando and the other anonymous reviewers for their contributions to the peer review process of this article.

 

Presented at the 40th Annual Meeting of the Japanese Association for Chest Surgery, 13 July 2023, Niigat city, Niigata prefecture, Japan.