Key words metabolic syndrome - lung cancer - incidence - mortality - meta-analysis
Introduction
Lung cancer is a widely prevalent malignancy that has a significant impact on the
global population [1 ]
[2 ]. Global cancer statistics from 2020 reveal
that lung cancer comprised 11.4% of all cancer cases and contributed to
18.0% of cancer-related deaths worldwide [3 ]. Histologically, lung cancer can be classified into non-small cell
lung cancer (NSCLC) and small cell lung cancer (SCLC), with treatment options
primarily consisting of surgery, radiation therapy, chemotherapy, and targeted drug
therapy [4 ]
[5 ]. However, despite of the above comprehensive treatment strategies, the
prognosis of patients with lung cancer remains poor, highlighting the importance of
primary prevention. Traditional risk factors for lung cancer, such as male gender,
advancing age, and smoking, have been widely acknowledged [6 ]. However, investigating additional factors
associated with an elevated risk of lung cancer in the general populace is
imperative for enhancing screening and prevention measures [7 ]. Furthermore, there is evidence suggesting
that metabolic disorders may have a detrimental impact on the occurrence and
prognosis of lung cancer, such as obesity [8 ],
hyperglycemia [9 ], and dyslipidemia [10 ].
Collectively, metabolic syndrome (MetS) is a conglomeration of metabolic disorders
distinguished by the pathophysiological manifestation of central obesity, insulin
resistance, hypertension, and dyslipidemia [11 ]
[12 ]. As the global population
ages, MetS has emerged as a prevalent health concern in both developed and
developing nations, affecting approximately 10–30% of adult
populations [13 ]. Given that both MetS and
cancer share a common underlying mechanism of low-grade chronic inflammation [14 ]
[15 ],
it is plausible to hypothesize that MetS may be associated with an increased risk
of
cancer. An initial meta-analysis revealed a potential association between MetS and
an elevated risk of developing cancer overall, although the findings varied
depending on the specific site of the cancer [16 ]. A subsequent meta-analysis indicated that MetS may not be a
contributing factor to the development of lung cancer [17 ]. However, it is important to note that this
latter analysis only included five cohort studies, and a few relevant studies have
been published subsequently [18 ]
[19 ]
[20 ]
[21 ]
[22 ]
[23 ]
[24 ]. Given the conflicting
results from previous researches, we conducted a comprehensive meta-analysis to
ascertain the connection between MetS and the incidence and mortality rates of lung
cancer in the adult population.
Materials and Methods
The Meta-analysis of Observational Studies in Epidemiology (MOOSE) guideline [25 ] and Cochrane Handbook [26 ] were followed in this systematic review and
meta-analysis.
Database search
In order to identify studies that met the meta-analysis’ objectives, the
following terms were combined: (1) “metabolic syndrome” OR
“insulin resistance syndrome” OR “syndrome X”;
(2) “lung” OR “pulmonary”; and (3)
“carcinoma” OR “cancer” OR
“tumor” OR “malignancy” OR
“malignant” OR “neoplasm”. Electronic databases
including PubMed, Embase, Cochrane Library, and Web of Science were searched
with the combined terms from inception of the databases to June 5, 2023. Our
selection criteria were limited to studies conducted on humans and published in
English as full-length papers. Additionally, we manually checked the references
of the related original and review articles to identify the original studies
that were not included.
Study identification
The PICOS criteria were followed in determining study selection criteria:
P (Participants): Adult population without a known diagnosis of cancer at
baseline.
I (Intervention): People with MetS at baseline. The diagnosis of MetS was
in accordance with the criteria used in the original studies.
C (Control): People without MetS at baseline.
(Outcome): At least one of the following outcomes was reported: the
incidence of lung cancer and/or the incidence of lung-cancer
specific mortality during follow-up durations.
S (Study design): Observational studies with longitudinal follow-up,
including cohort studies, post-hoc analyses of clinical trials, and
nested case-control studies (NCC).
Reviews, meta-analyses, editorials, studies enrolling patients with known cancer
at baseline, studies without longitudinal follow-up, studies did not investigate
MetS as exposure, or studies with no relevant outcomes were excluded.
Study quality assessment and data extraction
For the purpose of assessing the study quality, the Newcastle-Ottawa Scale (NOS)
[27 ] was used, which was composite of
three domains involving defining groups of the study, comparing groups between
them, and validating outcomes. The NOS incorporates nine criteria, and each
study receives one point if it meets a specific criterion. As detailed above,
two authors conducted electronic database searches, extracted study data
independently, and assessed study quality independently. Disagreements between
the two authors should be discussed in order to resolve them. The data collected
were: (1) study information (authors, countries, publication year, and study
design); (2) sources and sample sizes of the included population and number of
adults included in each study; (3) diagnostic criteria for MetS; (4) mean
follow-up durations, outcomes reported, and methods for validating the outcomes;
and (5) variables included in the multivariate regression analysis for the
association between MetS and risks of lung cancer incidence and mortality.
Statistical methods
Risk ratios (RRs) and 95% confidence intervals (CIs) were used to assess
the association between MetS and lung cancer related outcomes. For variance
stabilization and normalization, we performed a logarithmical transformation
followed by a calculation of the RRs and standard errors (SE) [26 ]. An evaluation of heterogeneity was
conducted using the Cochrane Q test and an I2 statistic [28 ]. If I2 >50%,
heterogeneity was considered significant. In order to synthesize data, we used a
randomized-effects model, which incorporates between-study heterogeneity and
provides a more generalized result [26 ].
Subgroup analysis was carried out to evaluate whether the association between
MetS and lung cancer related outcomes were significantly affected by study
characteristics such as design, country, sex of the participants, definition of
MetS, follow-up duration, adjustment of smoking, and different quality scores.
In order to reflect publication bias, funnel plots were constructed, and
symmetry was examined visually. In addition, publication bias was simultaneously
evaluated using Egger’s regression asymmetry test [29 ]. The RevMan (Version 5.1; Cochrane
Collaboration, Oxford, UK) and Stata (version 12.0; Stata Corporation, College
Station, TX) software were employed for the statistical analyses.
Results
Database search results
An overview of the database search process is shown in [Fig. 1 ]. As a result of the initial
literature search, 619 articles were found; after excluding duplications, 495
articles remained. As a result of screening the titles and abstracts, an
additional 468 studies were excluded from the meta-analysis. A full-text review
was conducted on the remaining 27 studies, of which 13 were further excluded for
the reasons listed in [Fig. 1 ]. As a
final step, fourteen observational studies [18 ]
[19 ]
[20 ]
[21 ]
[22 ]
[23 ]
[24 ]
[30 ]
[31 ]
[32 ]
[33 ]
[34 ]
[35 ]
[36 ] were eligible for this
meta-analysis.
Fig. 1 Flowchart of database search and study inclusion.
Characteristics of the included studies
Characteristics of the included studies are displayed in [Table 1 ]. Overall, nine prospective
cohorts [18 ]
[19 ]
[23 ]
[24 ]
[30 ]
[31 ]
[32 ]
[34 ]
[36 ], three retrospective cohort studies [20 ]
[33 ]
[35 ], and two NCC [21 ]
[22 ] were included in the meta-analysis. These studies were published
between 2008 and 2023, and performed in Italy, the United States, Japan, the
Netherlands, Korea, China, the United Kingdom, and Spain. Most of the included
studies enrolled community-derived adult population, while two of them included
patients with vascular diseases [34 ] and
hepatitis B virus infection [20 ]. In
total, 12 562 361 adults who were not with a known diagnosis of cancer at
baseline were included in this meta-analysis. By definition, MetS was diagnosed
via the National Cholesterol Education Program Adult Treatment Panel-III
(NCEP-ATP III) criteria in 11 studies [19 ]
[20 ]
[21 ]
[22 ]
[23 ]
[24 ]
[30 ]
[32 ]
[34 ]
[35 ]
[36 ], via the International
Diabetes Foundation (IDF) criteria in one study [18 ], and via the both above criteria in two studies [31 ]
[33 ]. The mean follow-up duration was 2.7 to 18.5 years. The outcome
of lung cancer incidence was reported in 11 studies [19 ]
[20 ]
[21 ]
[22 ]
[23 ]
[24 ]
[30 ]
[31 ]
[33 ]
[34 ]
[35 ], and the outcome of lung cancer mortality was also reported in
four studies [18 ]
[24 ]
[32 ]
[36 ]. The outcomes were
validated with databases such as national cancer registries, national death
index, medical records, or death certificates, etc. Multivariate analyses were
used to estimate the association between MetS and lung cancer related outcomes
in all of the included studies. Variables such as age and sex were adjusted in
all the studies, while other variables such as smoking etc. were adjusted in
some of the studies [18 ]
[19 ]
[20 ]
[21 ]
[22 ]
[23 ]
[24 ]
[31 ]
[32 ]
[33 ]
[35 ]
[36 ]. A good quality study was indicated by a NOS range of seven to
nine stars ([Table 2 ]).
Table 1 Characteristics of the included
studies.
Author year [Ref]
Location
Design
General status of the population
Number of participants
Definition of MetS
Follow-up duration
LC outcome reported
Validation of LC outcome
Variables adjusted
Russo 2008 [30 ]
Italy
P
Community-derived population
16677
NCEP-ATP III
2.7
LC incidence
Local cancer registry
Age and sex
Jaggers 2009 [32 ]
USA
P
Community-derived men
33230
NCEP-ATP III
14
LC mortality
National death index
Age, examination year, height, smoking, alcohol intake,
physically inactive, hypercholesterolemia, cardiovascular
disease, family history of cancer and cardiorespiratory
fitness
Inoue 2009 [31 ]
Japan
P
Community-derived population aged 40 years or older
27724
NCEP-ATP III and IDF
10.2
LC incidence
National cancer registry
Age, sex, study area, smoking, weekly alcohol intake, and
TC
Osaki 2012 [33 ]
Japan
R
Community-derived population
23625
NCEP-ATP III and IDF
9.1
LC incidence
Local cancer registry
Age, sex, smoking, and heavy drinking
van Kruijsdijk 2013 [34 ]
the Netherlands
P
Patients with vascular diseases
6172
NCEP-ATP III
5.5
LC incidence
National cancer registry
Age, sex, and calendar year
Ko 2016 [35 ]
Korea
R
Community-derived population
61758
NCEP-ATP III
10.4
LC incidence
National cancer registry
Age, smoking status, alcohol intake, and exercise
Gathirua 2017 [36 ]
USA
P
Community-derived population
19106
NCEP-ATP III
13.1
LC mortality
National death index
Age, sex, race, education, smoking, alcohol intake, and
concurrent medications
Watanabe 2019 [18 ]
Japan
P
Community-derived population
10379
IDF
18.5
LC mortality
Death certificates
Age, sex, smoking, alcohol drinking, education, physical
activity, and occupation category
Sin 2020 [19 ]
Korea
P
Community-derived population
9586753
NCEP-ATP III
6.3
LC incidence
National cancer registry
Age, sex, physical activity, and smoking status
Choe 2021 [20 ]
Korea
R
HBV-infected people over 40 years
1504880
NCEP-ATP III
4.9
LC incidence
National cancer registry
Age, sex, BMI, smoking, alcohol consumption, physical
activity
Li 2022 [21 ]
China
NCC
Community-derived population aged 45 years or older
17708
NCEP-ATP III
4.5
LC incidence
Medical chart
Age, sex, education, smoking status, drinking status, and
depression score
Shao 2022 [23 ]
UK
P
Community-derived population
450482
NCEP-ATP III
6.3
LC incidence
National cancer registry
Age, sex, physical activity, and smoking status
Lopez 2022 [22 ]
Spain
NCC
Community based population over 40 years
732992
NCEP-ATP III
5.5
LC incidence
National cancer registry
Age, sex socioeconomic status, smoking status, and
nationality
Van Hoang 2023 [24 ]
Japan
P
Community-derived population
70875
NCEP-ATP III
6.1
LC incidence and mortality
National cancer registry and death certificates
Age, sex, smoking, and worksite
LC: Lung cancer; MetS: Metabolic syndrome; P: Prospective; R:
Retrospective; NCC: Nested case-control; NCEP-ATP III: National
Cholesterol Education Program Adult Treatment Panel-III; IDF:
International Diabetes Foundation; HBV: Hepatitis B virus; TC: Total
cholesterol; BMI: Body mass index.
Table 2 Study quality evaluation via the Newcastle-Ottawa
Scale.
Cohort Study [Ref]
Representativeness of the exposed cohort
Selection of the non-exposed cohort
Ascertainment of exposure
Outcome not present at baseline
Control for age and sex
Control for other confounding factors
Assessment of outcome
Enough long follow-up duration
Adequacy of follow-up of cohorts
Total
Russo 2008 [30 ]
1
1
1
1
1
0
1
0
1
7
Jaggers 2009 [32 ]
1
1
1
1
1
1
1
1
1
9
Inoue 2009 [31 ]
1
1
1
1
1
1
1
1
1
9
Osaki 2012 [33 ]
0
1
1
1
1
1
1
1
1
8
van Kruijsdijk 2013 [34 ]
1
1
1
1
1
0
1
1
1
8
Ko 2016 [35 ]
0
1
1
1
1
1
1
1
1
8
Gathirua 2017 [36 ]
1
1
1
1
1
1
1
1
1
9
Watanabe 2019 [18 ]
1
1
1
1
1
1
1
1
1
9
Sin 2020 [19 ]
1
1
1
1
1
1
1
1
1
9
Choe 2021 [20 ]
0
1
1
1
1
1
1
0
1
7
Li 2022 [21 ]
0
1
1
1
1
1
1
0
1
7
Shao 2022 [23 ]
1
1
1
1
1
1
1
1
1
9
Lopez 2022 [22 ]
0
1
1
1
1
1
1
1
1
8
Van Hoang 2023 [24 ]
1
1
1
1
1
1
0
1
1
8
Association between Mets and lung cancer risk
Eight of the included studies [18 ]
[19 ]
[22 ]
[30 ]
[31 ]
[33 ]
[34 ]
[35 ] reported the results according to the
sex of the participants. Accordingly, these datasets were included independently
into the meta-analysis. Overall, 18 datasets from 11 studies [19 ]
[20 ]
[21 ]
[22 ]
[23 ]
[24 ]
[30 ]
[31 ]
[33 ]
[34 ]
[35 ] reported the association between MetS and lung cancer incidence.
Pooled results showed that MetS was associated with a higher risk of lung cancer
incidence (RR: 1.15, 95% CI: 1.05 to 1.26, p=0.002;
I2 =89%; [Fig.
2a ]). Subgroup analysis suggested that the association was not
significantly affected by study country, design, sex of the participants,
adjustment of smoking, or different study quality scores (p for subgroup
difference all>0.05; [Table 3 ]).
The association was predominantly contributed by studies with MetS defined by
the NCEP-ATP III criteria rather than those with MetS defined by the IDF
criteria, and the association seemed to be stronger in studies with follow-up
within 6 years than those over 6 years (p for subgroup difference=0.03
and 0.04, respectively; [Table 3 ]). In
addition, pooled results of five datasets from four studies [18 ]
[24 ]
[32 ]
[36 ] showed that MetS was associated with a
higher risk of lung cancer mortality (RR: 1.46, 95% CI: 1.19 to 1.79,
p<0.001; I2 =0%; [Fig. 2b ]).
Fig. 2 Forest plots for the meta-analysis of the association
between MetS and lung cancer risk. a : Forest plots for the
meta-analysis of the association between MetS and lung cancer incidence;
and b : Forest plots for the subgroup analysis of the association
between MetS and lung cancer mortality.
Table 3 Subgroup analyses for the association between MetS
and the incidence of lung cancer.
Study characteristics
Datasets number
RR (95% CI)
I2
p for subgroup effect
p for subgroup difference
Country
Asian
11
1.10 [0.97, 1.24]
92%
0.14
Western
7
1.25 [1.06, 1.48]
84%
0.01
0.22
Design
PC
10
1.14 [1.04, 1.24]
82%
<0.001
RC or NCC
8
1.13 [0.91, 1.39]
84%
<0.001
0.93
Sex
Men
7
1.15 [1.01, 1.31]
57%
0.04
Women
7
1.12 [0.94, 1.33]
81%
0.20
0.84
Definition of MetS
NCEP-ATP III
18
1.15 [1.05, 1.26]
89%
0.002
IDF
4
0.82 [0.61, 1.11]
0%
0.20
0.03
Follow-up duration
Within 6 years
8
1.31 [1.09, 1.57]
89%
0.004
Over 6 years
10
1.08 [1.01, 1.16]
60%
0.03
0.04
Adjustment of smoking status
Yes
14
1.11 [1.00, 1.23]
90%
0.04
No
4
1.38 [0.98, 1.95]
90%
0.07
0.23
Quality score
NOS=7
4
1.15 [0.90, 1.48]
86%
0.27
NOS=8
9
1.24 [0.97, 1.58]
81%
0.08
NOS=9
5
1.09 [1.01, 1.17]
76%
0.03
0.56
PC: Prospective cohort; RC: Retrospective cohort; NCC: Nested
case-control; RR: Risk ratio; CI: Confidence interval; NCEP-ATP III:
National Cholesterol Education Program Adult Treatment Panel-III; IDF:
International Diabetes Foundation.
Publication bias
[Fig. 3 ] shows the funnel plots regarding
the association between MetS the risk of lung cancer incidence in adult
populations. According to visual inspection, the plots are symmetrical, which
suggested risk of publication bias is low. Additionally, Egger’s
regression test also indicated a low risk of publication bias (p=0.49).
The publication bias for the meta-analysis of MetS and lung cancer mortality was
unable to be determined because only five datasets were available.
Fig. 3 Funnel plots for the publication bias underlying the
meta-analysis of the association between MetS and lung cancer
incidence.
Discussion
According to the findings of this meta-analysis, there is evidence to suggest that
the adult population diagnosed with MetS may have a 15% higher likelihood of
developing lung cancer compared to adults without MetS. Further analysis revealed
that this association was primarily observed in studies that utilized the NCEP-ATP
III criteria for diagnosing MetS. Additionally, it was observed that the strength
of
the association between MetS and increased lung cancer risk appeared to be more
pronounced in studies with a mean follow-up duration of less than 6 years, as
opposed to those with a follow-up duration exceeding 6 years. Furthermore, the
aforementioned correlation appeared to exhibit consistency among both males and
females and was not significantly influenced by factors such as the country of
study, research design, adjustment for smoking, and variations in study quality
scores. Additionally, our findings indicate that MetS is associated with a
46% elevated likelihood of mortality due to lung cancer in the adult
population. Taken together, these outcomes imply that MetS could potentially serve
as a risk factor for the occurrence of lung cancer events in adults.
As far as we know, few meta-analyses few meta-analyses have evaluated the association
between MetS and lung cancer risk. In an early meta-analysis evaluating the risk of
overall cancer in people with MetS, a subgroup analysis including four studies
suggested that MetS was not associated with an increased risk of lung cancer [16 ]. However, limited data were included and
studies reporting lung cancer incidence and lung cancer mortality were combined in
the meta-analysis, which may confound the results [16 ]. Although both the lung cancer incidence and lung cancer mortality
indicate lung cancer related events, these two outcomes are not always consistent
because lung cancer mortality outcome is also affected by therapeutic factors. A
subsequent meta-analysis in 2020 included five cohort studies and showed that MetS
was not related to a higher risk of lung cancer incidence [17 ]. However, the number of available data
remained limited, and the results need to be updated in view of the fact that a few
relevant studies were published after the meta-analysis. In the present systematic
review and meta-analysis, we conducted separate analyses to examine the correlation
between MetS and the incidence and mortality of lung cancer. A thorough search of
four commonly utilized electronic databases yielded 14 observational studies that
fulfilled the objective of the meta-analysis. The quantity of included studies and
the overall sample size of the participants were considerably greater than those of
previous meta-analyses. Furthermore, it is noteworthy that all the studies
incorporated in this analysis employed a longitudinal follow-up design, suggesting
a
potential longitudinal correlation between MetS and the incidence and mortality of
lung cancer. Moreover, it is pertinent to mention that all the included studies
utilized multivariate analysis to ascertain the risk of lung cancer events
associated with MetS. Consequently, these findings lend support to the notion that
the relationship between MetS and the risk of developing lung cancer may be
independent of confounding variables, such as age, gender, and smoking status. In
aggregate, the results of this meta-analysis underscore the significance of lung
cancer screening and prophylaxis in people with MetS.
The subgroup analysis conducted in our study revealed that the relationship between
MetS and the risk of developing lung cancer was not significantly influenced by the
country in which the study was conducted or the sex of the participants. This
suggests that the aforementioned association may not be susceptible to variations
in
ethnicity and sex within the population. Furthermore, our findings indicated that
the association between MetS and lung cancer risk appeared to be more pronounced in
studies that utilized the NCEP-ATP III criteria for diagnosing MetS, compared to
those employing the IDF criteria. However, it is important to exercise caution when
interpreting these results due to the limited availability of only four datasets for
the subgroup of IDF studies, as well as the presence of significant heterogeneity
within the subgroup of NCEP-ATP III studies. In addition, it was suggested that the
association between MetS and lung cancer risk may be stronger in studies with
follow-up duration within 6 years (RR: 1.31) compared to those over 6 years (RR:
1.08). Since MetS has become more prevalent in middle-aged population [37 ], these findings may suggest the importance
of lung cancer screening in middle-aged people with MetS.
The association between Metabolic Syndrome (MetS) and lung cancer is likely
influenced by a multifactorial process. Pathophysiologically, the presence of
low-grade systemic inflammation may serve as an intermediate mechanism in the
development of MetS and the pathogenesis of lung cancer [38 ]. Furthermore, emerging evidence suggests
that insulin resistance, a fundamental mechanism in MetS [39 ], is also implicated in the pathogenesis of
lung cancer [40 ]. Moreover, several components
of MetS, including central obesity [41 ],
hyperglycemia [9 ], and dyslipidemia [42 ], have been increasingly linked to an
elevated risk of developing lung cancer. These findings may also be the mechanisms
underlying the association between MetS and lung cancer risk.
This study possesses certain limitations. The meta-analysis conducted on lung cancer
mortality is constrained by the inclusion of a limited number of studies,
necessitating the need for additional prospective cohort studies to authenticate the
findings. Furthermore, caution must be exercised when interpreting the outcomes of
certain subgroup analyses due to the scarcity of available datasets pertaining to
these subgroups, such as those involving studies utilizing the IDF diagnostic
criteria for MetS. Additionally, since lung cancers of different histopathological
type may have different biological features, the association between MetS and
different histopathological type of lung cancer should be analyzed. However, since
data according to the histopathological type of lung cancer were not reported in
either of the included cohort studies, we were unable to evaluate the outcomes
according to the histopathological type of lung cancer. Future studies are warranted
in this regard. Furthermore, it should be noted that although all the studies chosen
for analysis employed multivariate regression analysis, the presence of residual
confounding factors, including the potential impact of dietary [43 ] and other lifestyle factors [44 ] associated with lung cancer risk, could not
be entirely eliminated. Moreover, the confounding factors adjusted in the original
studies were different among the included studies, which may also affect the results
of the meta-analysis. Lastly, due to the inclusion of observational studies in this
meta-analysis, it is not feasible to establish a causal relationship between MetS
and lung cancer solely based on these findings.
Conclusion
According to the results of the meta-analysis, there appears to be a potential
correlation between the presence of MetS in adults and an elevated incidence and
mortality rate of lung cancer. These findings indicate that MetS could potentially
serve as a risk factor for the occurrence of lung cancer in the adult population,
thereby emphasizing the significance of implementing lung cancer screening and
prevention measures among individuals with MetS.
Notice
This article was changed according to the erratum on
November 17, 2023.
Erratum
In the above-mentioned article, the authors Zhao Zhang and
Qinxiang Liu contributed equally. The affiliations of the coauthors
were corrected in the online version
