Impact of Smoking on Response to the First-Line Treatment of Advanced ALK-Positive Non-Small Cell Lung Cancer: A Bayesian Network Meta-Analysis

Abstract

Background: The impact of smoking on the efficacy of anaplastic lymphoma kinase (ALK)-positive non-small cell lung cancer (NSCLC) treatment is controversial and has not been systematically explored in the first-line setting. We performed a systematic review based on a pairwise meta-analysis and a Bayesian network meta-analysis (NMA) to address this issue.Methods: PubMed, Embase, Web of Science, Cochrane Library, Clinical-Trials.gov, and other resources were searched until 5 January 2022. Progression-free survival (PFS) was considered the main outcome of interest. Randomized controlled trials with smoking status analysis were included. Cochrane Risk of Bias Tool was performed to assess the risk of bias. Random effects models were adopted conservatively in meta-analysis. The NMA was performed in a Bayesian framework using the “gemtc” version 1.0–1 package of R-4.1.2 software.Results: A total of 2,484 patients from nine studies were eligible for this study, with 1,547 never-smokers (62.3%) and 937 smokers (37.7%). In a pairwise meta-analysis, in the overall population, no significant difference was found between never-smokers and smokers. However, in the subgroup analyses based on crizotinib-controlled studies, anaplastic lymphoma kinase tyrosine kinase inhibitors (ALK-TKIs) derived better PFS in the smoking group over the never-smoking group in the Asian population (HR = 0.17, 95%CI = 0.09–0.31 in the smoking group, HR = 0.39, 95%CI = 0.24–0.65 in the never-smoking group, p = 0.04, low quality of evidence). In NMA, among never-smokers, lorlatinib ranked the highest for PFS (SUCRA = 96.2%), but no significant superiority was found among the new-generation ALK-TKIs except for ceritinib. In smokers, low-dose alectinib performed best (SUCRA = 95.5%) and also demonstrated a significant superiority over ensartinib (HR = 0.23, 95%CI = 0.08–0.68, very low quality of evidence), brigatinib (HR = 0.38, 95%CI = 0.14–0.99, low quality of evidence), ceritinib (HR = 0.24, 95%CI = 0.09–0.66, low quality of evidence), crizotinib (HR = 0.18, 95%CI = 0.08–0.41, moderate quality of evidence), and chemotherapy (HR = 0.11, 95%CI = 0.05–0.28, low quality of evidence).Conclusion: In general, smoking may not affect the treatment efficacy of advanced ALK-positive NSCLC in the first-line setting. However, alectinib may perform better in the smoking Asian population. Moreover, lorlatinib in never-smokers and low-dose alectinib in smokers could be considered optimal first-line therapy for advanced ALK-positive NSCLC. Acceptable limitations of evidence, such as study risk of bias, inconsistency, and imprecision, were present in this NMA.

Integration of artificial intelligence in lung cancer: Rise of the machine

“… 73 The Bayesian network approach helps address this issue by producing novel outputs such as treatment rankings, odds ratios, and probability distributions. 74 Several such studies have been performed in lung cancer, including evaluating safety and efficacy of bevacizumab biosimilars, 75 determining optimal platinum-based chemotherapy for early-stage resected NSCLC, 76 how smoking status influences effect of targeted therapy, 77 and choice of first-line treatment for patients on the basis of PD-L1 expression. 78 , 79 , 80 …”