If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Corresponding author. Address for correspondence: Dr. Saverio Caini, Cancer Risk Factors and Lifestyle Epidemiology Unit, Institute for Cancer Research, Prevention and Clinical Network (ISPRO), Via Cosimo il Vecchio 2, 50139, Florence, Italy.
Lung cancer (LC) remains a disease with poor prognosis despite recent advances in treatments. Here, we aimed at summarizing the current scientific evidence on whether quitting smoking at or around diagnosis has a beneficial effect on the survival of LC patients.
Methods
We searched MEDLINE and EMBASE for articles published until 31st October, 2021, that quantified the impact on LC patients’ survival of quitting smoking at or around diagnosis or during treatment. Study-specific data were pooled into summary relative risk (SRR) and corresponding 95% confidence intervals (CI) using random effect meta-analysis models.
Results
Twenty-one articles published between 1980 and 2021 were included, which encompassed a total of over 10,000 LC patients. There was substantial variability across studies in terms of design, patients’ characteristics, treatments received, criteria used to define smoking status (quitters or continued), and duration of follow-up. Quitting smoking at or around diagnosis was significantly associated with improved overall survival (SRR 0.71, 95% CI 0.64–0.80), consistently among patients with non-small cell LC (SRR 0.77, 95% CI 0.66–0.90, n studies = 8), small cell LC (SRR 0.75, 95% CI 0.57–0.99, n studies = 4), or LC of both or unspecified histological type (SRR 0.81, 95% CI 0.68–0.96, n studies = 6).
Conclusions
Quitting smoking at or around diagnosis is associated with a beneficial effect on the survival of LC patients. Treating physicians should educate LC patients about the benefits of quitting smoking even after diagnosis and provide them with the necessary smoking cessation support.
Lung cancer (LC) is among the most common and deadliest malignancies worldwide. According to Global Cancer Incidence, Mortality, and Prevalence, LC ranks second globally in terms of incidence, with over 2.2 million new cases estimated to have occurred in 2020, and first in terms of mortality, with nearly 1.8 million caused deaths.
Despite the continuous improvement of treatment modalities (i.e. immunotherapy, new targeted therapies, radiotherapy, etc.), the prognosis of patients with LC remains dismal in comparison with tumors at other body sites, as the 5-year relative survival rate is 25% (all stages combined) for the NSCLC type and 7% for SCLC according to data from the U.S. Surveillance, Epidemiology, and End Results program.
Cigarette smoking is, by far, the most important modifiable risk factor for LC occurrence, accounting for nearly two-thirds of all LC cases globally according to the latest report from the Global Burden of Disease initiative.
GBD 2019. Respiratory tract cancers collaborators. Global, regional, and national burden of respiratory tract cancers and associated risk factors from 1990 to 2019: a systematic analysis for the Global Burden of Disease Study 2019.
The fight against tobacco use (which should include the issuing of strict tobacco control laws and active bans on the advertising, promotion, and sponsorship of tobacco products, and the adoption of other measures as detailed in the WHO Framework Convention on Tobacco Control)
Centers for Disease Control and Prevention National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health. Best practices for comprehensive tobacco control programs.
It is also key that smokers are actively encouraged to quit smoking and given adequate support when making an attempt to quit or exhibiting a willingness to try because maintained smoking cessation substantially reduces LC mortality (and all other major cause-specific mortality) over time.
It is comparatively much less clear to date whether smokers diagnosed with LC have survival advantages if they quit smoking at or around diagnosis. In a systematic literature review and meta-analysis published in 2010 (which encompassed articles published up to December 2008), Parsons et al.
found suggestive evidence that quitting smoking at or around diagnosis could lead to an improvement in the recurrence-free survival and overall survival (OS) of patients with LC. However, the strength of the conclusions that could be drawn at the time was considerably curbed by the small number of studies available then regarding the topic (10, of which only six reported on the effect of smoking cessation on OS); all the more so when attempting to separately evaluate the two main types of LC (NSCLC and SCLC).
In recent years, the number of studies on the subject has been increasing, so that updating the previous meta-analysis seems appropriate and timely. Therefore, we conducted a systematic literature review and meta-analysis of the studies that evaluated the prognostic effect of quitting smoking at or around diagnosis among patients with LC.
Materials and Methods
Literature Search, Articles Selection, and Data Extraction
The protocol of this literature review and meta-analysis was registered in the International prospective register of systematic reviews database (registration number CRD42021245560).
The objectives of the review were defined according to the following PECO criteria: population (P), which included smokers diagnosed with LC; exposure (E), which was quitting smoking at or around diagnosis; comparison (C) of quitters versus continued smokers; and outcome (O), which was survival. The literature search was conducted on MEDLINE and EMBASE on October 31, 2021, using the following search string: “(smok∗) AND (cease OR cessation OR quit∗ OR stop∗) AND (cancer OR carcinoma OR tumor OR malignancy) AND (survival OR prognos∗ OR outcome OR mortality).” No time, geographic, or language restriction was applied as long as an English abstract was available to decide on eligibility. After removing duplicates, the articles were first screened on the basis of their title and abstract, and those deemed as potentially eligible for inclusion were read in full copy. The literature search and articles selection was conducted independently by three researchers (SC, MDR, and VV), and any disagreement was resolved by consensus or by asking for advice from a fourth senior researcher (SG). We finally included in the review and meta-analysis all original full papers that evaluated the effect of quitting smoking at diagnosis, at some point thereafter, or up to 12 months before diagnosis (compared with those who continued smoking) on the survival of patients with LC (any OS; disease-specific survival; disease-, recurrence-, or progression-free survival (PFS); and local or locoregional control). To be retained, an article also had to report a measure of relative risk (RR) obtained through fitting survival analysis models (e.g. a hazard ratio (HR) calculated by means of Cox’s proportional hazards regression or Fine and Gray subdistribution hazard models), and a corresponding measure of statistical uncertainty (e.g. 95% confidence intervals (CI), SE, or exact p value), or provide a Kaplan-Meier (KM) survival curve allowing to calculate an unadjusted HR using the method by Parmar et al.
Editorials, commentaries, and letters not presenting original data were excluded and conferences abstract (the latter, because they generally lack most of the information that is needed for a correct interpretation of the results and an assessment of the study quality). The reference lists of all eligible papers and previously published literature reviews and meta-analyses were inspected by means of backward citation chaining in an attempt to find other additional articles covering the same topic.
Data extraction was done using an internally piloted spreadsheet. The following information was extracted from each eligible article: (1) country and year in which the study was conducted; (2) study design; (3) the number of smoking patients with LC who were included and their breakdown into those who continued smoking and those who quit; (4) the exact definition of those categorized as quitters (i.e. the timing at which they ceased smoking in relation to LC diagnosis); (5) the distribution of patients with LC in terms of sex, age, tumor type (NSCLC and SCLC), and stage at diagnosis; (7) the duration of follow-up (median or mean and maximum, when available); (8) any treatments received; and (9) details on statistical analysis methods and variables used for estimates adjustment. The RR and 95% CI for the association between smoking cessation at or around diagnosis and the survival of patients with LC were inverted when necessary to ensure that the reference group was always that of continued smokers and then transformed into log(RR) and corresponding variance using Greenland’s formula.
In case multiple articles included fully or partially overlapping study populations, we extracted and entered in the analysis the RR given in the article with the largest number of patients with LC included or, in the case of equal study size, the most adjusted RR.
Statistical Analysis and Quality Assessment
For the patients’ OS, study-specific log(RR) were pooled into a summary relative risk (SRR) through random effects models with maximum likelihood estimation, and the corresponding 95% CI was calculated by assuming an underlying t distribution.
In all analyses, quitters were compared with continued smokers, the latter being the reference group; therefore, an SRR of less than 1.00 means there is a gain in survival associated with quitting smoking at or around diagnosis, and vice versa for SRR greater than 1.00. Fewer than five eligible independent articles were available for survival types other than OS; their results were presented in a dedicated table but no SRR was calculated because of the small numbers. The heterogeneity of RRs across studies was quantified using the I
; when this occurred, we used meta-regression, subgroup analysis, and leave-one-out sensitivity analysis to detect possible sources of the observed heterogeneity. Study characteristics considered for meta-regression and subgroup analyses were the country and year of publication, patient’s demographics (age and sex), the tumor stage at diagnosis, and whether the estimates were adjusted or not. Because of its clinical relevance, subgroup analysis according to LC type was planned and conducted a priori—that is, regardless of the heterogeneity as quantified by the I
statistics. In addition, stratified analyses were also conducted after splitting the studies into those in which patients with LC labeled as quitters had quit strictly at or after diagnosis and those in which the category of quitters included patients with LC who had stopped smoking up to 12 months before diagnosis. Moreover, we used the Macaskill test to evaluate whether the summary results were affected by publication bias.
Statistical analyses were conducted using the Statistical Analysis System software, version 9.4 (SAS Institute Inc., Cary, NC) and R software, version 4.1.1 (R Core Team, Vienna, Austria). All tests were two-sided and statistical significance was set at p values below 0.05.
Results
The literature search in MEDLINE and EMBASE produced a total of 11,621 nonduplicate entries, and further 296 articles were found in reference lists (Fig. 1). After removing articles on the basis of title (n = 10,783) and abstract (n = 568), a total of 566 were read in full text. Of these, 545 were removed for not matching the inclusion criteria; the main reasons for exclusion were failure to consider smoking cessation at or around diagnosis as exposure of interest (n = 243), and a focus on outcomes other than patient survival (n = 134). Finally, a total of 21 studies were included in the systematic review
Effect of smoking during radiotherapy, respiratory insufficiency, and hemoglobin levels on outcome in patients irradiated for non-small-cell lung cancer.
Continued cigarette smoking by patients receiving concurrent chemoradiotherapy for limited-stage small-cell lung cancer is associated with decreased survival.
were not independent of each other but were both retained as they provided data on different patient survival analyses (disease-free survival in the former and OS in the latter).
Figure 1Flowchart of the literature search and article selection for the systematic review and meta-analysis on the effect of quitting smoking at or around diagnosis and the survival of patients with LC. LC, lung cancer.
The 21 articles that were included in the systematic review were published between 1980 and 2019 and included NSCLC (n = 10, encompassing a total of 5315 patients), SCLC (n = 5, 1133 patients overall), or patients with LC of both or unspecified subtypes or whose subtype was not specified (n = 6, total of 4490 patients) (Table 1). The median or mean age at LC diagnosis was between 60 and 70 years in most studies and the proportion of men ranged between 40.2% and 91.8%. The studies differed markedly in the length of follow-up (from up to 12 mo in Gemine et al.
), in the proportion of patients that underwent the different types of available treatments, and in the proportion of patients that quitted smoking at or around diagnosis. Substantial variability among studies also existed as regards the criteria used to classify patients as quitters or continued smokers (Table 1). In most studies, quitters were those who stopped smoking at diagnosis or at some point thereafter (e.g. within 3 mo of diagnosis in Gemine et al.,
was the only one in which the survival of quitters and continued smokers was compared not directly, but using a third group of patients (never-smokers) as the reference group. The RR from each study was reported in Table 2 along with details on variables that were used for adjustment. The study quality was generally fair (Supplementary Data 1); the main limitations possibly affecting the validity of some studies were the selection of patients on the basis of clinical characteristics or early disease course (e.g. tumor stage, presence of distant metastasis at diagnosis, and perioperative mortality), the fact that smoking status was self-reported, the inability to adjust for potential confounders, and the susceptibility to immortal time bias (when smoking status was first measured only sometime after LC diagnosis, or when patients were only included if they had survived for a certain period from diagnosis).
Table 1Main Characteristics of the Articles Included in the Systematic Review and Meta-Analysis on the Effect of Quitting Smoking At or Around Diagnosis and the Survival of Patients with LC
Effect of smoking during radiotherapy, respiratory insufficiency, and hemoglobin levels on outcome in patients irradiated for non-small-cell lung cancer.
The study by Saito-Nakaya et al.23 was the only one in which the survival of quitters and continued smokers was not compared directly, but instead used a third group of patients (never-smokers) as the reference group.
Continued cigarette smoking by patients receiving concurrent chemoradiotherapy for limited-stage small-cell lung cancer is associated with decreased survival.
was the only one in which the survival of quitters and continued smokers was not compared directly, but instead used a third group of patients (never-smokers) as the reference group.
Table 2RR and Corresponding 95% CI for the Association Between Quitting Smoking At or Around Diagnosis (vs. Continued Smoking, Taken as the Reference Group) and OS of Patients with LC
The RR for disease-specific survival as outcome were close to those for OS (0.751, 95% CI 0.58–0.9817, 95% CI 0.77–1.0520: men 0.78, 95% CI 0.59–1.01, women 1.15, 95% CI 0.85–1.56) and their use in meta-analysis models did not substantially affect the summary results.
0.67
(0.53–0.85)
age, sex, BMI, stage, histologic diagnosis, treatment, PY, other
Continued cigarette smoking by patients receiving concurrent chemoradiotherapy for limited-stage small-cell lung cancer is associated with decreased survival.
The RR for disease-specific survival as outcome were close to those for OS (0.751, 95% CI 0.58–0.9817, 95% CI 0.77–1.0520: men 0.78, 95% CI 0.59–1.01, women 1.15, 95% CI 0.85–1.56) and their use in meta-analysis models did not substantially affect the summary results.
The RR for disease-specific survival as outcome were close to those for OS (0.751, 95% CI 0.58–0.9817, 95% CI 0.77–1.0520: men 0.78, 95% CI 0.59–1.01, women 1.15, 95% CI 0.85–1.56) and their use in meta-analysis models did not substantially affect the summary results.
The RR and 95% CI were inverted as they were calculated using the quitters (instead of continued smokers) as the reference group in the article.
(0.56–0.89)
None
BMI, body mass index; CI, confidence interval; KM, Kaplan-Meier; LC, lung cancer; OS, overall survival; PY, pack-years; RR, relative risk.
a The RR and 95% CI were inverted as they were calculated using the quitters (instead of continued smokers) as the reference group in the article.
b The RR for disease-specific survival as outcome were close to those for OS (0.751, 95% CI 0.58–0.9817, 95% CI 0.77–1.0520: men 0.78, 95% CI 0.59–1.01, women 1.15, 95% CI 0.85–1.56) and their use in meta-analysis models did not substantially affect the summary results.
Quitting smoking at or around diagnosis was significantly associated with a better OS for patients with LC regardless of the histologic subtype (p value for the difference = 0.812). In particular, the SRR comparing the OS between quitters and continued smokers (taken as the reference group) was 0.77 (95% CI 0.66–0.90) for patients with NSCLC (Table 3 and Fig. 2) and 0.75 (95% CI 0.57–0.99) for patients with SCLC (Table 3 and Fig. 3), on the basis of data from eight and four independent studies, respectively (Table 2). The SRR calculated from the six articles that included patients with LC of both subtypes (or in which the histologic subtype was not specified (Table 2) was similar: 0.81 (95% CI 0.68–0.96) (Table 3 and Fig. 4). Because the strength of the association did not practically differ among histologic subtypes, we worked out an overall SRR and found that smokers who quit smoking at or around LC diagnosis have a 29% improvement in their OS compared with those who continue smoking (SRR 0.71, 95% CI 0.64–0.80) (Table 3). Of note, the association with improved OS tend to be stronger when restricting the analysis to studies in which the category of quitters only included those patients with LC who had quit smoking strictly at or after diagnosis (the only minor exception was the study by Dobson Amato et al.,
which extended the timing of cessation to up to 30 days before LC diagnosis) instead of also including those patients with LC who had quit up to 12 months before diagnosis (Table 3). In particular, the difference between the two subsets of studies was nearly significant (SRR 0.65 vs. 0.82, p value for difference 0.052) when pooling all studies regardless of the LC subtype (Table 3).
Table 3SRR and Corresponding 95% CI for the Association Between Quitting Smoking At or Around Diagnosis
Studies in which all patients categorized as quitters had quit at or after diagnosis, and studies in which quitters included patients that had quit smoking up to 12 months before LC diagnosis).
According to the eligibility criteria (see Methods), papers were included when they evaluated the effect of quitting smoking at diagnosis, at some point thereafter, or up to 12 months before diagnosis (compared with those who continued smoking) on the survival of patients with LC.
According to the eligibility criteria (see Methods), papers were included when they evaluated the effect of quitting smoking at diagnosis, at some point thereafter, or up to 12 months before diagnosis (compared with those who continued smoking) on the survival of patients with LC.
According to the eligibility criteria (see Methods), papers were included when they evaluated the effect of quitting smoking at diagnosis, at some point thereafter, or up to 12 months before diagnosis (compared with those who continued smoking) on the survival of patients with LC.
According to the eligibility criteria (see Methods), papers were included when they evaluated the effect of quitting smoking at diagnosis, at some point thereafter, or up to 12 months before diagnosis (compared with those who continued smoking) on the survival of patients with LC.
a Versus continued smoking, which was taken as the reference group.
b NSCLC, SCLC, LC of both subtypes or unspecified subtype.
c Studies in which all patients categorized as quitters had quit at or after diagnosis, and studies in which quitters included patients that had quit smoking up to 12 months before LC diagnosis).
d According to the eligibility criteria (see Methods), papers were included when they evaluated the effect of quitting smoking at diagnosis, at some point thereafter, or up to 12 months before diagnosis (compared with those who continued smoking) on the survival of patients with LC.
Figure 2Forest plot for the association between quitting smoking at or around diagnosis and overall survival of patients with NSCLC. CI, confidence intervals; RR, relative risk; SCC, squamous cell carcinoma.
Figure 3Forest plot for the association between quitting smoking at or around diagnosis and overall survival of patients with SCLC. CI, confidence intervals; RR, relative risk.
Figure 4Forest plot for the association between quitting smoking at or around diagnosis and overall survival of patients with LC (both subtypes or subtype not specified). CI, confidence intervals; RR, relative risk.
was 59.5%, 50.7%, and 59.6% for studies including patients with NSCLC, SCLC, and LC of both or unspecified subtypes, respectively, and 54.8% for the overall SRR calculated using all available data. We did not find any significant factor explaining between-study heterogeneity: the p value was 0.37 for country (Asia vs. Europe vs. the United States), 0.71 for publication year, and 0.10 for whether the estimates were adjusted by age (for other factors there were not enough studies to investigate their influence on OS estimates). In addition, there was no indication of publication bias (p = 0.87). In the study by Saito-Nakaya et al.,
patients with LC who smoked had a worse prognosis compared with never-smokers; but, consistently with the other articles, the OS was longer among those who quit smoking 1 to 12 months before diagnosis (HR 2.9, 95% CI 1.4–6.3) than among those who continued smoking (HR 6.4, 95% CI 1.7–23.9).
Quitting smoking at or around diagnosis seemed to be beneficial for the prognosis of patients with LC even when evaluating survival types other than OS (Supplementary Table 1). In particular, when compared with continued smokers (taken as reference), quitters were reported to experience a significantly longer PFS in the study by Sheikh et al.
Continued cigarette smoking by patients receiving concurrent chemoradiotherapy for limited-stage small-cell lung cancer is associated with decreased survival.
(HR 0.79, 95% CI 0.67–0.94), which were both based on patients with SCLC. Furthermore, the locoregional control of the disease was found to be significantly longer among quitters than continued smokers in the study by Rades et al.
Effect of smoking during radiotherapy, respiratory insufficiency, and hemoglobin levels on outcome in patients irradiated for non-small-cell lung cancer.
(HR 0.54, 95% CI 0.35–0.94). On the contrary, there seemed not to be an improvement in PFS and local disease control among quitters (vs. continued smokers) in the NSCLC study by Roach et al.,
reported a significantly more-than-halved HR for OS (0.496, p = 0.0004) among quitters versus persistent smokers in a sample of 1274 patients with NSCLC from the United States. This study could not be included because there was no explanation of when smoking cessation had to occur for a patient to be classified as a quitter (which, instead, was one of the predefined eligibility criteria). Karlsson et al.
reported that recent quitters experienced a better overall and cancer-specific survival compared with persistent smokers, more evident among patients with NSCLC (HR 0.83 and 0.84, respectively, p < 0.05) than SCLC (HR 0.89 and 0.91, respectively, p > 0.05). However, the article was excluded because smoking cessation status was entered by means of deep learning algorithms instead of patient self-report or directly ascertained by the investigators.
We conducted a systematic review and meta-analysis of articles published up to October 31, 2021 that investigated whether quitting smoking at or around diagnosis improved the survival of smokers diagnosed with LC. A total of 21 articles published in a time range of nearly three decades (1980–2021) were included, which encompassed a total of over 10,000 patients with LC. The studies differed under several aspects, including study design, patient demographics, duration of follow-up, treatments received, and criteria used to define quitters. With regard to the latter point, we chose to include studies in which quitters were defined as those who ceased smoking either at LC diagnosis, during treatment, or shortly before (up to 12 mo) LC diagnosis. These multiple sources of variability resulted in moderate heterogeneity of HR across studies, yet the results of the meta-analysis strongly pointed toward the existence of a beneficial effect of smoking cessation at or around diagnosis on patients on LC survival, which emerged with comparable magnitude (and always achieving statistical significance at the 0.05 level) in the studies restricted to either patients with NSCLC (SRR 0.77) or SCLC (SRR 0.75) and in studies that included patients of both LC types (SRR 0.81). Our findings were strengthened when the analysis was restricted to those studies in which patients with LC labeled as quitters were those who had stopped smoking strictly at or after LC diagnosis because the observed improvement in patients’ OS associated with smoking cessation was maintained and enhanced. Finally, quitting smoking at or around diagnosis also seemed to positively affect the patients with LC’s disease-free survival, although the limited number of available studies advised against calculating a pooled estimate.
To the best of our knowledge, only a single meta-analysis has been published to date that evaluated the relationship between around or postdiagnosis smoking cessation and LC survival. Parsons et al.
found suggestive evidence of a harmful effect of smoking continuation on LC survival; however, their conclusions for OS were on the basis of only four (for NSCLC) and two (for SCLC) observational studies published up to December 2008. Here, we updated the work of Parsons et al.
by substantially increasing the number of reports that were included, and we confirmed and reinforced their early claims.
There are several plausible biologic mechanisms able to explain the observed effect of smoking cessation on patients with LC’s survival. Tobacco smoke promotes tumor growth, progression, and dissemination; decreases the efficacy of, and tolerance to, radiation and systemic therapy; and increases the risk of postoperative complications and second primary cancers.
The underlying mechanisms are mostly unknown and very little data are available, but an important role may be played by the epigenetic changes induced by tobacco smoke and its cessation, with, for instance, interesting data that suggested an impact on smoking cessation survival-benefit according to DNA methylation level in patients with lung adenocarcinoma.
It is important, however, to acknowledge that several sources of bias may also be at play, particularly confounding, arising from systematic differences between quitters and continued smokers that are insufficiently (or not at all) adjusted for in the analyses. For instance, quitters might differ from continued smokers in terms of demographics (e.g. age) or tumor characteristics (e.g. stage) that affect survival. In addition, patients with LC who manage to quit and remain abstinent after diagnosis might be more likely to receive resection with curative intent and, inversely, it could be hypothesized that patients with LC with lower-stage cancer might be more motivated to quit smoking to enhance their likelihood of recovery. However, studies that looked at predictors of smoking cessation after LC or cancers at other sites have provided highly conflicting results,
and a higher odds of quitting among those with more advanced (instead of localized) stage cancers was observed in the United States in a large investigation on the basis of data from Surveillance, Epidemiology, and End Results.
Because of this uncertainty, it would have been of the utmost importance that all possible confounders (factors that affect both the likelihood of quitting smoking and survival) were adjusted for in the analyses, although this was not the case in all of the studies included in the present meta-analysis. Whereas our findings strongly suggest a beneficial prognostic effect of smoking cessation, it remains uncertain how much of this effect is because of residual confounding possibly affecting some of the studies.
A further point to consider is that the included papers gave scarce attention to the effect of smoking cessation on the efficacy of subsequent or ongoing therapy. In fact, there is not much updated data available to evaluate the effect of continued smoking with regard to the different therapeutic options currently available to treat patients with LC, and further studies are urgently needed, especially considering the introduction of immunotherapy in the extended-stage SCLC setting and the fact that OS and PFS can differ consistently when considering oncogene versus non–oncogene-addicted NSCLC. Continued smoking during treatment surely has an impact on comorbidities, but it also seems to be true for treatment efficacy. For instance, O’Malley et al.
pointed out the increased clearance of medications such as erlotinib and irinotecan (with relative lower systemic exposure) in smokers compared with never- and former smokers, which may play a role in the resulting lower adverse events rates and outcome in such patients. More attention has been given to the impact of smoking on the efficacy of immune checkpoint inhibitors. Limited, yet suggestive data found greater benefit from immunotherapy in smokers, which could be explained by the association of smoking history with increased tumor mutational burden and programmed death-ligand 1 expression levels.
The effect of smoking status on efficacy of immune checkpoint inhibitors in metastatic non-small cell lung cancer: A systematic review and meta-analysis.
Smoking status during first-line immunotherapy and chemotherapy in NSCPatients with LC: A case-control matched analysis from a large multicenter study.
Finally, it is worth emphasizing how former smokers achieved better survival outcomes than current smokers in the subgroup analyses of the KEYNOTE-024 trial, thus, also emphasizing in this specific clinical setting the importance to stop smoking among patients with LC as soon as possible at diagnosis.
The implications of our findings extend beyond the clinical setting. There is intense ongoing research on the feasibility and cost-effectiveness of LC screening programs.
Because heavy smokers would be primarily targeted, LC screening could serve as a “teachable moment” to help participants quit smoking by integrating a structured cessation program into the screening activities.
Our findings greatly support this view by illustrating that success in quitting smoking would benefit not only those testing negative from the screening tests but also those eventually diagnosed with LC.
The main strength of this systematic review and meta-analysis lies in the fair number of studies included, which allowed for obtaining summary estimates stratified by LC subtype and by the exact timing of smoking cessation. The observed moderate heterogeneity in effect estimates across studies was, to some extent, unavoidable, given the multiple differences in study design, patient characteristics included, the definition of exposure, and duration of follow-up. As a rule, the limitations affecting the studies that are pooled in meta-analyses can, in turn, affect the validity of the summary estimates that are calculated, and our meta-analysis is not an exception. In addition to the aforementioned possibility of residual confounding, immortal time bias is a possible cause for concern because it can exaggerate the strength of an association or even generate a spurious association. Immortal time bias can occur when the assessment of smoking status (quitters or continued) is done after a certain time (e.g. months) has elapsed since diagnosis, or when only patients with LC who have survived up to a given time after diagnosis are included in the analysis. Moreover, smoking status was mostly self-reported and confirmed by more reliable methods (e.g. monitoring of exhaled carbon monoxide) only in a few studies. Finally, all studies included were conducted in high-income countries, whereas greater diversity under this regard would be desirable.
LC remains a disease with a generally poor prognosis, despite the advances that have occurred over the past decade, including the introduction of immune checkpoint inhibitors (alone or in combination) effective for both NSCLC and SCLC.
Single or combined immune checkpoint inhibitors compared to first-line platinum-based chemotherapy with or without bevacizumab for people with advanced non-small cell lung cancer.
The articles that we reviewed and summarized here are still limited in number and have some limitations, particularly in study design and methods. Studies carefully designed to avoid confounding and other sources of bias, and investigating any interaction of changes in smoking status with the existing treatment options, are urgently needed. Despite these limitations, the data currently available are, in our opinion, robust enough to conclude with the recommendation that inviting patients to quit smoking at diagnosis or during treatment or follow-up encounters, and giving them all the support necessary to succeed in their attempts to quit and remain abstinent during and after therapy, should arguably become a nonoptional part of the management of these patients given the potentially large gain in survival that can ensue.
CRediT Authorship Contribution Statement
Saverio Caini: Conceptualization, Data curation, Investigation, Methodology, Project administration, Resources, Validation, Visualization, Writing – original draft.
Marco Del Riccio: Conceptualization, Data curation, Investigation, Methodology, Project administration, Writing – review and editing.
Virginia Vettori: Conceptualization, Data curation, Investigation, Methodology, Writing – review and editing.
Vieri Scotti: Investigation, Supervision, Validation, Writing – original draft.
Chiara Martinoli: Data curation, Formal analysis, Methodology, Software, Visualization, Writing – review and editing.
Sara Raimondi: Data curation, Formal analysis, Funding acquisition, Methodology, Software, Visualization, Writing – review and editing.
Giulio Cammarata: Data curation, Formal analysis, Software, Visualization, Writing – review and editing.
Marco Banini: Investigation, Writing – original draft.
Giovanna Masala: Conceptualization, Resources, Supervision, Writing – review and editing.
Sara Gandini: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Resources, Software, Validation, Visualization, Writing – review and editing.
Acknowledgments
This work was supported by the Italian Ministry of Health (Ricerca Corrente and 5 x 1000 funds, no grant number applicable). The funder had no role in the study design, the collection, analysis, and interpretation of data, the writing of the report, and the decision to submit the article for publication.
GBD 2019. Respiratory tract cancers collaborators. Global, regional, and national burden of respiratory tract cancers and associated risk factors from 1990 to 2019: a systematic analysis for the Global Burden of Disease Study 2019.
National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health. Best practices for comprehensive tobacco control programs.
Effect of smoking during radiotherapy, respiratory insufficiency, and hemoglobin levels on outcome in patients irradiated for non-small-cell lung cancer.
Continued cigarette smoking by patients receiving concurrent chemoradiotherapy for limited-stage small-cell lung cancer is associated with decreased survival.
The effect of smoking status on efficacy of immune checkpoint inhibitors in metastatic non-small cell lung cancer: A systematic review and meta-analysis.
Smoking status during first-line immunotherapy and chemotherapy in NSCPatients with LC: A case-control matched analysis from a large multicenter study.
Single or combined immune checkpoint inhibitors compared to first-line platinum-based chemotherapy with or without bevacizumab for people with advanced non-small cell lung cancer.
Lung cancer is currently the second most common cancer in the United States and the leading cause of cancer mortality worldwide. A large number of previous epidemiologic studies have confirmed that smoking is a risk factor for various malignancies, including lung cancer, increasing the risk of tumor-related mortality and all-cause mortality.1 On the basis of the theory of planned behavior, timely feedback of tumor screening and abnormal screening results can effectively improve the smoking cessation rate.
Thoracic oncologists have many reasons to feel upbeat about recent advances in the management of lung cancer. Numerous biologic targets have been identified that have led to effective therapeutic strategies and improved outcomes.1 Targeting the EGFR or ALK has significantly improved long-term outcomes, particularly in those patients with light or never-smoking history. Advances in the understanding of tumor immunology and the development of checkpoint inhibitors have improved outcomes among patients with more complex tumors frequently related to smoking and a high mutational burden.
03404-3&id=gr1.jpg){kind=link}
03404-3&id=gr2.jpg){kind=link}
03404-3&id=gr3.jpg){kind=link}
03404-3&id=gr4.jpg){kind=link}