Critical Reviews in Oncology / Hematology 

 

ToXicity profile of epidermal growth factor receptor tyrosine kinase inhibitors for patients with lung cancer: A systematic review and network

meta-analysis
Yi Zhao a, 1, Bo Cheng a, 1, Zisheng Chen a, b, 1, Jianfu Li a, 1, Hengrui Liang a, 1, Ying Chen a, 1, Feng Zhu a, Caichen Li a, Ke Xu a, Shan Xiong a, WeiXiang Lu a, ZhuXing Chen a, Ran Zhong a,
Shen Zhao c, Zhanhong Xie d, Jun Liu a, Wenhua Liang a, e,*, Jianxing He a,*
a Department of Thoracic Surgery and Oncology, The First Affiliated Hospital of Guangzhou Medical University, State Key Laboratory of Respiratory Disease, National Clinical Research Center for Respiratory Disease, Guangzhou 510120, China
b Department of Respiratory Medicine, The Sixth Affliated Hospital of Guangzhou Medical University, Qingyuan People’s Hospital, Qingyuan 511518, China
c Department of Medical Oncology, Sun Yat-sen University Cancer Center, State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Guangzhou 510050, China
d Department of Respiratory Medicine, The First Affiliated Hospital of Guangzhou Medical University, State Key Laboratory of Respiratory Disease, National Clinical
Research Center for Respiratory Disease, Guangzhou 510120, China
e Department of Medical Oncology, The First People’s Hospital of Zhaoqing, Zhaoqing 526020, China

A R T I C L E I N F O

Keywords: ToXicity Adverse event EGFR
Lung cancer
Network meta-analysis

A B S T R A C T

Epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs) are treatments commonly used for lung cancer. The toXicity profile including toXicity incidence, severity, and spectrum (involving various specific adverse events) of each EGFR-TKI are of particular clinical interest and importance. Data from phase II and III randomized controlled trials comparing treatments among EGFR-TKIs (osimertinib, dacomitinib, afatinib, erlo- tinib, gefitinib, and icotinib) and chemotherapy for lung cancer were synthesized with Bayesian network meta-
analysis. The primary outcome was systemic all-grade and grade ≥3 adverse events. The secondary outcome was
specific all-grade adverse events including those of the skin, gastrointestinal tract, lung, etc. 40 trials random- izing 13,352 patients were included. Generally greater toXicity for dacomitinib and afatinib, and safety for icotinib were suggested. Furthermore, we found individual EGFR-TKIs had different toXicity spectrums. These findings provide a compelling safety reference for the individualized use of EGFR-TKIs for patients with lung cancer.

1. Introduction

Lung cancer is the leading cause of cancer-related mortality world- wide (Bray et al., 2018), and approXimately 10–20 % Caucasian and 40
% Asian cases harbor epidermal growth factor receptor (EGFR) muta- tions (Sharma et al., 2007; Kosaka et al., 2004; Rosell et al., 2009). The therapeutic landscape of lung cancer has been revolutionized in recent decades with the molecularly targeted use of EGFR tyrosine kinase in- hibitors (TKIs), including the first-generation erlotinib, gefitinib and icotinib, the second-generation dacomitinib and afatinib, and the third-generation osimertinib (Recondo et al., 2018; Díaz-Serrano et al.,

2018). For the impressive clinical activity, all these EGFR-TKIs (icotinib in China) are recommended for patients with EGFR-mutated lung can- cer, resulting in their broadly common selections and applications for clinical practice (NCCN, 2021; Planchard et al., 2018; Hanna et al., 2021), for example, for these patients, no restrictions have been set for the use of a specific EGFR-TKI in the emerging effective combination, adjuvant, or neoadjuvant treatment strategies (Remon et al., 2019; Cheng et al., 2019; Zhai et al., 2015).
Previous evidence has consistently shown that EGFR-TKIs are well tolerated, however, a few adverse events (AEs) are commonly seen including those of the skin, gastrointestinal tract, and lung (Greenhalgh
* Corresponding authors at: Department of Thoracic Surgery and Oncology, The First Affiliated Hospital of Guangzhou Medical University, State Key Laboratory of Respiratory Disease, National Clinical Research Center for Respiratory Disease, 151 Yanjiang Road, Guangzhou 510120, China.
E-mail addresses: [email protected] (W. Liang), [email protected] (J. He).
1 Authors contributed equally as joint first authors.

https://doi.org/10.1016/j.critrevonc.2021.103305

Received 12 October 2020; Received in revised form 16 March 2021; Accepted 16 March 2021
Available online 20 March 2021
1040-8428/© 2021 Published by Elsevier B.V.

et al., 2016; Shah and Shah, 2019). These side-effects are mild in most cases but can affect the quality of patients’ life and even call for dose reduction, treatment discontinuation, or pharmacotherapeutic inter- vention. Furthermore, different toXicity severities among EGFR-TKIs and special AEs associated with individual EGFR-TKI could be observed (Ding et al., 2017; Takeda et al., 2015). Therefore, for these daily administered EGFR-TKIs in long-term clinical use, a comprehen- sive understanding of the toXicity profile including toXicity incidence, severity and spectrum (involving various specific AEs) is required to aid clinicians in individualized administration of EGFR-TKIs.
However, previous trials could not directly compare some of those EGFR-TKIs, for instance, no head-to-head comparisons among osi- mertinib, dacomitinib, afatinib, and icotinib have been established, making it difficult to prospectively evaluate the relative toXicity of any two EGFR-TKIs based on direct evidence. Thus, we performed this Bayesian network meta-analysis, which enables comparisons between any two treatments by synthesizing both direct and indirect evidence (Cipriani et al., 2013), to systematically investigate the toXicity profile of EGFR-TKIs for patients with lung cancer.
2. Materials and methods
This systematic review and network meta-analysis is conducted following a prespecified protocol registered in the Prospective Register of Systematic Reviews (PROSPERO: CRD42019140710) and the Preferred Reporting Items for Systematic Reviews and Meta-analyses statement extension for network meta-analysis (Supplementary Table S1) (Hutton et al., 2015). All discrepancies during the process were resolved by arbitration by a panel of adjudicators (Y.Z., J.L., S.X., R.Z., W.L., and J.H.).
2.1. Search strategies and study selection
According to the Cochrane Handbook for Systematic Reviews of In- terventions (Higgins et al., 2019), we systematically searched PubMed, Embase, Cochrane Library, Web of Science, and ClinicalTrials.gov for studies published up to June 30, 2020, with no language restrictions. The search strategy is detailed in Supplementary Table S2. Reference lists of relevant studies including reviews and meta-analyses were manually checked for additional records.
We included phase II or III randomized controlled trials among EGFR-TKI monotherapies and chemotherapy for patients with lung cancer and presenting sufficient data on systemic (all-grade and/or grade 3) and/or specific AEs (all-grade). Studies published online ahead of print were eligible, but meeting abstracts were excluded. Studies not adhering to the inclusion criteria were excluded. Other exclusion criteria were: 1) trials in which EGFR-TKIs were used in a combination, maintenance, neoadjuvant, or adjuvant treatment strat- egy; 2) trials comparing EGFR-TKIs with other treatments like mono- clonal antibodies, immunotherapies, and some pathway inhibitors; 3) trials comparing treatments unapproved by any Food and Drug Administration (FDA); and 4) trials whose safety outcomes subsequently published updated data in a mature or longer duration of follow-up. Placebo-controlled trials were excluded in the original analysis as they mainly recruit patients with a mild form of disease to meet ethical and safety requirements for regulatory approval (Kirsch and Moncrieff, 2007); however, these trials were added to a validation analysis.
2.2. Outcome measures and data extraction
The primary outcome was systemic AEs reported as those of all-grade and grade 3, representing incidence and severity of toXicity, respec- tively; the secondary outcome was specific AEs (all-grade) including those from skin, gastrointestinal tract, lung, etc. Two investigators (C.L. and S.Z.) independently extracted information on each study into a predefined spreadsheet mainly including baseline characteristics, the

number of patients evaluable for toXicity, and the number of patients recorded for systemic and specific AEs. AEs specified as treatment- related were preferably used, but if not available in a trial, data of any reported AEs were used instead. Data in supplementary materials were checked and extracted. When necessary, study authors and pharma- ceutical companies were contacted to request complete and updated information. As the control arm in the FLAURA study (Ramalingam et al., 2020) grouped erlotinib and gefitinib, we assumed that they caused the same toXicity when respectively compared with osimertinib.
2.3. Risk of bias assessment
Pairs of investigators (B.C., Z.C., H.L., and F.Z.) independently assessed risk of bias in each study using the Cochrane risk of bias tool (Higgins et al., 2011). The following domains of possible bias were considered: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting, and other bias. Items were scored as low, high, or unclear risk of bias. Review Manager (version 5.3) was used to generate the risk of bias figures.
2.4. Data synthesis and statistical analysis
In STATA (version SE 14.2), a network plot was generated for each primary outcome to illustrate network geometry (Chaimani et al., 2013). Displayed in a forest plot by Review Manager (version 5.3), a random-effects pairwise meta-analysis was conducted for a direct comparison with at least two studies to estimate pooled odds ratio (OR)
and 95 % confidence interval. Between-study heterogeneity was assessed by I2 with values over 50 % indicating substantial heteroge- neity (Higgins et al., 2003).
Network meta-analyses on systemic and specific AEs were conducted within the Bayesian consistency framework using Markov chain Monte Carlo methods in OpenBUGS (version 3.2.3) (Lunn et al., 2009). We applied a random-effects model to allow for apparent between-study heterogeneity in the treatment comparison effects (Lu and Ades, 2004; Jansen et al., 2011). To fit the model, non-informative uniform and normal prior distributions were set (Sutton and Abrams, 2001), gener- ating 200,000 iterations after a burn-in of 100,000 in each of the three different chains. The posterior distributions were obtained as the output of the network meta-analysis estimate, the pooled OR and its corre- sponding 95 % credible interval (CrI). Convergence was ensured using the Brooks-Gelman-Rubin method (Supplementary Fig. S1) (Brooks and Gelman, 1998). For each outcome, we additionally used the surface under the cumulative ranking curve area (SUCRA), which equals one, denoting the most toXic treatment, and zero, denoting the safest treat- ment, to summarize the relative rankings of the treatments (Salanti et al., 2011).
Transitivity (the across-study exchangeability based on the similar
clinical and methodological characteristics) and consistency (the direct and indirect estimates are statistically similar when both are available for a given comparison), are two key assumptions underlying the network meta-analysis that were considered (Cipriani et al., 2013). We set the criteria of only including randomized controlled trials with strict patient allocation and optimized balance to address all treatments for the same condition to ensure transitivity. Transitivity was evaluated using Bayesian meta-regression statistics for study and population baselines, such as publication year, age, and female percentage. Incon- sistency was locally and globally assessed by comparing the pooled ORs from network meta-analysis and pairwise meta-analysis, and by comparing the fit and parsimony of consistency and inconsistency models, respectively (Dias et al., 2014, 2010). The node splitting method was used to calculate the inconsistency in the entire network on particular comparisons (nodes), where direct and indirect evidence were separately contrasted (Dias et al., 2010).
For the primary outcome, we conducted validation analysis by

additionally including the placebo-controlled trials (Supplementary Table S3). To explore the impact of effect modifiers, we also conducted predefined subgroup analyses in EGFR-mutated, untreated, and pre- treated patients as well as EGFR-TKIs approved by the United States FDA, and sensitivity analyses by restricting phase III studies, studies explicitly reporting treatment-related AEs, studies comparing only EGFR-TKIs, and studies in patients with a performance status 0 1, and by excluding the FLAURA study.
3. Results
3.1. Eligible studies and characteristics
Literature search identified 1328 relevant studies. A total of 40 randomized controlled trials (Ramalingam et al., 2020; Wu et al., 2017; Shi et al., 2017; Han et al., 2017; Yang et al., 2017; Paz-Ares et al., 2017; Mok et al., 2017; Kim et al., 2016; Urata et al., 2016; Wu et al., 2015; Cadranel et al., 2015; Soria et al., 2015; Wu et al., 2014; Ramalingam et al., 2014; Kawaguchi et al., 2014; Zhou et al., 2014; Li et al., 2014; Gregorc et al., 2014; Sequist et al., 2013; Garassino et al., 2013; Shi et al., 2013; Karampeazis et al., 2013; Rosell et al., 2012; Han et al., 2012; Ramalingam et al., 2012; Ciuleanu et al., 2012; Kim et al., 2012; Kelly et al., 2012; Sun et al., 2012; Zhou et al., 2011; Maemondo et al., 2010; Mor`ere et al., 2010; Lee et al., 2010; Mitsudomi et al., 2010; Mok et al., 2009; Lilenbaum et al., 2008; Crino` et al., 2008; Kim et al., 2008; Maruyama et al., 2008; Cufer et al., 2006) involving 13,352 patients were finally included after multilayered screening and eligibility assessment (Supplementary Fig. S2). Table 1 summarizes their baseline characteristics. Of all included patients, 6458 (48.3 %) were female and 5429 (40.7 %) were non-Asian. The median duration of follow-up was
18.4 months. The risk of bias assessments are presented in Supplemen- tary Fig. S3. The biggest source of high risk was from the blinding of participants and personnel domain primarily because a high proportion of studies were open-label. The assumption of transitivity was accepted as there was no significant variability identified in the study or patient characteristics possibly associated with treatment effects (Supplemen- tary Table S4).
3.2. Overview
7700 (85.4 %) patients from 24 studies, and 3839 (33.2 %) patients from 31 studies were reported to have at least one systemic all-grade and grade 3 AEs, respectively. Network diagrams are shown in Fig. 1. More than 120 different types of specific AEs were reported, and 18 of them representing the most clinically relevant AEs in practice were con- cerned, including the non-hematologic parameters: rash, diarrhea, anorexia, nausea, vomiting, fatigue, stomatitis, constipation, paro- nychia, pruritus, dry skin, alopecia, and interstitial lung disease (ILD), and the hematologic parameters: leucopenia, neutropenia, anemia, aspartate transaminase (AST) elevation, and alanine transaminase (ALT) elevation. The pooled incidence of systemic AEs based on the frequency of different AEs for each EGFR-TKI (Supplementary Fig. S4) was: osi- mertinib (all-grade 93.1 %; grade 3 29.7 %), dacomitinib (96.9 %; 47.4
%), afatinib (96.6 %; 35.9 %), erlotinib (86.5 %; 23.4 %), gefitinib (83.4
%; 22.8 %), and icotinib (57.7 %; 5.8 %). Compared with chemotherapy,
AEs related to EGFR-TKIs mainly occurred in the skin, gastrointestinal, and pulmonary systems, including rash (P < 0.001), diarrhea (P < 0.001), stomatitis (P 0.03), paronychia (P < 0.001), dry skin (P < 0.001), pruritus (P < 0.001), and ILD (P < 0.001).
3.3. Primary outcome: systemic AEs
In terms of systemic all-grade AEs (Fig. 2A), dacomitinib and afatinib were consistent (OR, 0.99; 95 % CrI, 0.25–4.61) in causing the most toXicity, with significant differences for dacomitinib versus icotinib (OR, 7.19; 95 % CrI, 1.70–31.87), and afatinib versus gefitinib (OR, 3.02; 95

% CrI, 1.01–8.08) and icotinib (OR, 7.26; 95 % CrI, 1.78–25.77). Icotinib (followed by gefitinib) was the safest EGFR-TKI with significant differ- ences when compared with erlotinib (OR, 3.43; 95 % CrI, 1.16–9.48) besides dacomitinib and afatinib. In terms of systemic grade 3 AEs (Fig. 2A), dacomitinib and afatinib were also consistent (OR, 0.97; 95 % CrI, 0.37–2.49) in causing the most toXicity among EGFR-TKIs with significant differences for dacomitinib (OR, 3.18; 95 % CrI, 1.17–8.76) and afatinib (OR, 3.09; 95 % CrI, 1.31–7.33) versus osimertinib, gefiti- nib versus dacomitinib (OR, 0.39; 95 % CrI, 0.17 0.84) and afatinib
(OR, 0.40; 95 % CrI 0.22 0.72), and icotinib versus dacomitinib (OR,
0.26; 95 % CrI, 0.08 0.85) and afatinib (OR, 0.27; 95 % CrI,
0.09 0.77). Icotinib (followed by osimertinib) also appeared to be the safest treatment with significant differences when compared with dacomitinib and afatinib.
Fig. 2B depicts the relative ranking of each treatment based on the SUCRA value. The toXicity ranking of each EGFR-TKI from low to high was: icotinib (SUCRA 0.02), gefitinib (0.26), osimertinib (0.37),
erlotinib (0.45), chemotherapy (0.71), dacomitinib (0.84), and afatinib (0.85) for systemic all-grade AEs; and icotinib (0.10), osimertinib (0.16), gefitinib (0.26), erlotinib (0.51), afatinib (0.74), and dacomitinib (0.75) for systemic grade 3 AEs.
Similar results were noticed in the validation analysis by additionally including the placebo-controlled trials (Supplementary Fig. S5) and all sensitivity analyses by restricting phase III studies (Supplementary Fig. S6), studies explicitly reporting treatment-related AEs (Supple- mentary Fig. S7), studies comparing only EGFR-TKIs (Supplementary Fig. S8), and studies in patients with performance status 0 1 (Supple- mentary Fig. S9), and by excluding the FLAURA study (Supplementary Fig. S10).
3.4. Secondary outcome: specific AEs
Results from the specific AEs (all-grade) analyses (Table 2) revealed osimertinib was significantly safer than afatinib and erlotinib for rash, dacomitinib and afatinib for diarrhea, and erlotinib and gefitinib for ALT elevation; but more toXic than gefitinib for neutropenia. Dacomitinib was significantly safer than afatinib but more toXic than the others for diarrhea. Afatinib was significantly more toXic than gefitinib and icoti- nib, besides osimertinib, for rash, erlotinib and gefitinib for stomatitis, gefitinib for paronychia, and all other EGFR-TKIs for diarrhea. Gefitinib was safer than erlotinib for rash.
Bayesian ranking profile based on the SUCRA results (Table 3) were consistent with the OR estimates. Compared with other EGFR-TKIs, osimertinib had the highest risk of leucopenia (SUCRA 0.79), neu- tropenia (0.81), and ILD (0.82), but the lowest risk of rash (0.25), alo- pecia (0.22), AST elevation (0.23), and ALT elevation (0.16); dacomitinib was associated with a broad toXicity spectrum with the highest risk in anorexia (0.66), anemia (0.60), dry skin (0.93), and al- opecia (0.61), and second highest risk of diarrhea (0.83), nausea (0.55), stomatitis (0.72), constipation (0.54), paronychia (0.76), pruritus (0.64), and ILD (0.62), but the lowest risk of neutropenia (0.16); afatinib had the highest risk of rash (0.94), diarrhea (0.99), stomatitis (0.96), paronychia (0.90), and pruritus (0.77), but the lowest risk of anemia (0.22), constipation (0.18), dry skin (0.42), and ILD (0.33); erlotinib had relatively mild toXicity spectrums with the lowest risk of stomatitis (0.22) and leucopenia (0.23) but predominant in fatigue (0.63); gefitinib had the highest risk of nausea (0.66), vomiting (0.63), constipation (0.63), AST elevation (0.86), and ALT elevation (0.91), but the lowest risk of fatigue (0.16) and paronychia (0.34); icotinib had the lowest risk on diarrhea (0.19), anorexia (0.09), nausea (0.13), vomiting (0.22), and
pruritus (0.35).
3.5. Subgroup analysis
12 studies (n = 3737), 17 studies (n = 5337), 14 studies (n = 6235), and 29 studies (n = 10,888) were involved in the subgroup analyses in

Table 1
Baseline characteristics of studies included in the network meta-analysis.

Study, year (phase) Ethnicity EGFR- mutated

TrAE First-line PS 0-1 (%) No. of patients
(female%)

Median age (y)

Treatment (median follow-up months)

FLAURA, 2020 (III) (Ramalingam et al., 2020)

279 (63.8) 64 Osimertinib 80 mg, QD (35.8)
Multiple Y Y Y 100a 277 (62.1) 64 Gefitinib 250 mg or erlotinib 150 mg, QD
Pemetrexed 500 mg/m2 Cisplatin 75
mg/m2, Q3W, 4 cycles (39.6)
41 (56.1) NG Gefitinib 250 mg, QD (NG)
40 (57.5) NG Pemetrexed 500 mg/m2 + Carboplatin

128 (53.1) NG

AUC = 5, Q4W, 6 cycles (NG)
Erlotinib 150 mg, QD (22.1)
Pemetrexed 500 mg/m2 + Carboplatin

6, Q4W, ≤6 cycles (69.4)

Afatinib 40 mg, QD (18.4)

First-SIGNAL, 2012 (III) (Han et al., 2012) Asian Multiple NG Y 90.9b
113 (68.0) 58 Gemcitabine 1000 mg/m2 + Cisplatin 75
mg/m2, Q3W, ≤6 cycles (16.6)
436 (34.4) 64 Dacomitinib 45 mg, QD (7.1)
436 (36.9) 62 Erlotinib 150 mg, QD (7.1)
150 (28.0) 68 Erlotinib 150 mg, QD (8.9)
150 (29.1) 67 Docetaxel 60 mg/m2, Q3W (8.9)
81 (33.3) 58 Gefitinib 250 mg, QD (10.6)
76 (38.2) 56 Pemetrexed 500 mg/m2, Q3W (10.6)
61 (34.4) 54 Erlotinib 150 mg, QD (14.7)
62 (37.1) 55 Pemetrexed 500 mg/m2, Q3W (14.7)
134 (26.1) 66 Erlotinib 150 mg, QD (32.4)
Pemetrexed 500 mg/m2 or Docetaxel 75
mg/m2, Q3W, 6 cycles (32.4)
229 (63.9) 62 Afatinib 40 mg, QD (16.4)
111 (67.0) 61 Pemetrexed 500 mg/m2 + Cisplatin 75
mg/m2, Q3W, ≤6 cycles (16.4)
107 (29.4) 66 Erlotinib 150 mg, QD (33.0)
104 (33.6) 67 Docetaxel 75 mg/m2, Q3W or 35 mg/m2,
200 (41.2) 57 Icotinib 125 mg, TID (NG)
199 (43.4) 57 Gefitinib 250 mg, QD (NG)
166 (18.7) 65 Erlotinib 150 mg, QD (29.0)
166 (16.9) 66 Pemetrexed 500 mg/m2, Q3W (27.3)
84 (67.4) 65 Erlotinib 150 mg, QD (18.9)
82 (78.2) 65 Docetaxel 75 mg/m2 or gemcitabine 1250
mg/m2 + Cisplatin 75 mg/m2, Q3W (14.4)
159 (88.0) 57 Gefitinib 250 mg, QD (35.0)
150 (89.3) 57 Gemcitabine 1250 mg/m2 + Cisplatin 80
mg/m2, Q3W, ≤9 cycles (35.0)

A7471028, 2012 (II) (Ramalingam et al., 2012)

Multiple Multiple Y N 88.3b

TITAN, 2012 (III) (Ciuleanu et al., 2012) Multiple Multiple Y N 79.9b

Standard docetaxel or pemetrexed dosing schedule (24.8)

Kim et al., 2012 (II) (Kim et al., 2012) Asian Multiple Y N 85.4b 48 (85.4) 56 Erlotinib 150 mg, QD (16.3) Kelly et al., 2012 (IIb) (Kelly et al., 2012) non-Asian NG Y N 100b 101 (32.6) 62 Erlotinib 150 mg, QD (NG)

KCSG-LU08-01, 2012 (III) (Sun et al., 2012)

b 68 (85.3) 58 Gefitinib 250 mg, QD (15.9)
67 (85.1) 64 Pemetrexed 500 mg/m2, Q3W (15.9)

OPTIMAL, 2011 (III) (Zhou et al., 2011) Asian Y Y Y 93.5b 83 (58.5) 57 Erlotinib 150 mg, QD (15.6)
(continued on next page)

Table 1 (continued )
Study, year (phase) Ethnicity EGFR- mutated
TrAE First-line PS 0-1 (%) No. of patients
(female%)
Median age (y)

Treatment (median follow-up months)

72 (59.7) 59 Gemcitabine 1000 mg/m2 + Cisplatin AUC

98.7b114 (63.2) 64c= 5, Q3W, ≤4 cycles (15.6)
Gefitinib 250 mg, QD (17.6)

113 (64.0) 63c Paclitaxel 200 mg/m2 + Carboplatin AUC

0b 43 (11.6) 70= 6, Q3W, ≥3 cycles (17.6) Gefitinib 250 mg, QD (NG)

42 (21.4) 71 Docetaxel 75 mg/m2, Q3W (NG)
a 82 (32.9) 57 Gefitinib 250 mg, QD (NG)
79 (43.0) 58 Docetaxel 75 mg/m2, Q3W (NG)
87 (68.6) 64 Gefitinib 250 mg, QD (2.7)
100b 88 (69.8) 64 Cisplatin 80 mg/m2 + Docetaxel 60 mg/
m2, Q3W, 3—6 cycles (2.7)
607 (79.5) 57 Gefitinib 250 mg, QD (5.6)

IPASS, 2009 (III) (Mok et al., 2009) Asian Multiple NG Y 89.6a

589 (79.1) 57 Carboplatin AUC = 5/6 + Paclitaxel 200
mg/m2, Q3W, ≤6 cycles (5.6)

Lilenbaum et al., 2008 (II) (Lilenbaum et al., 2008)

52 (55.8) NG Erlotinib 150 mg, QD (NG)
non-Asian N Y Y 0b 51 (45.1) NG Carboplatin AUC = 6 + Paclitaxel 200 mg/
m2, Q3W, ≤4 cycles (NG)

INVITE, 2008 (II) (Crino` et al., 2008) Multiple Multiple Y Y 80.0a 94 (22.7) 74 Gefitinib 250 mg, QD (6.4)
INTEREST, 2008 (III) (Kim et al., 2008) Multiple Multiple Y N 99.6a 729 (36.4) 61 Gefitinib 250 mg, QD (7.6)

V-15-32, 2008 (III) (Maruyama et al., 2008)

a 244 (38.4) NG Gefitinib 250 mg, QD (21.0)
239 (38.1) NG Docetaxel 60 mg/m2, Q3W (21.0)

SIGN, 2006 (II) (Cufer et al., 2006) Multiple NG Y N 67.4a 68 (30.8) 63 Gefitinib 250 mg, QD (9.2)
EGFR, epidermal growth factor receptor; TrAE, treatment-related adverse event; CTCAE, Common Terminology Criteria for Adverse Events; PS, performance status; Y, yes; N, no; NG, not given; AUC, area under the concentration-time curve; QD, once daily; TID, three times daily; Q3W, every 3 weeks; Q4W, every 4 weeks.
a World Health Organization performance status.
b Eastern Cooperative Oncology Group performance status.
c Mean age was given instead of median age.

Fig. 1. Network diagrams for comparisons on systemic all- grade (A) and grade ≥3 (B) adverse events.
Circular nodes represent treatments, and the size of each node is proportional to the total number of patients (in parentheses) assigned to the corresponding treatment. Lines represent direct comparisons, and the width of each line is proportional to the number of trials (next to the line) investigating the corre- sponding comparison.

EGFR-mutated, untreated, pretreated patients and in EGFR-TKIs approved by the United States FDA, respectively. In the pretreated subgroup, the front treatment for 93.2 % of patients was chemotherapy. All subgroup analyses (Supplementary Figs. S11–14) yielded commonly consistent results with the original analysis with some new findings: dacomitinib and afatinib were not consistent in causing the most sys- temic all-grade AEs, instead, dacomitinib appeared to cause the most in the EGFR-mutated and untreated subgroups, and afatinib appeared to cause the most in the pretreated subgroup; osimertinib appeared to cause the least systemic grade 3 AEs in the untreated (consistent with icotinib) and pretreated subgroups; and in all United States FDA approved EGFR-TKIs (excluding icotinib which is only approved in China), gefitinib and osimertinib appeared to cause the least systemic

all-grade and grade ≥3 AEs, respectively.

3.6. Heterogeneity and inconsistency assessment
Supplementary Fig. S15 presents the results of the pairwise meta- analysis and heterogeneity estimates for the feasible pairwise compari- sons. Significant heterogeneity was detected in the comparisons of

erlotinib versus chemotherapy for both systemic all-grade and grade 3 AEs (I2 72 % and 83 %, respectively); and afatinib versus chemo- therapy (I2 90 %) and gefitinib versus chemotherapy (I2 63 %) for
systemic grade 3 AEs. The test of global inconsistency showed a similar or better fit of the consistency model than that of the inconsistency model (Supplementary Table S5). There was no evidence of local inconsistency for the outstanding consistent results of pairwise meta- analyses and network meta-analyses (Supplementary Fig. S16). The node splitting analysis showed no evidence of inconsistency (Supple- mentary Table S6).
4. Discussion
By analyzing systemic and specific AEs data from 40 randomized controlled trials, our study suggested a generally greater toXicity for dacomitinib and afatinib, and safety for icotinib followed by osimertinib and gefitinib, with detailed toXicity rankings from low to high: icotinib, gefitinib, osimertinib, erlotinib, dacomitinib, and afatinib for systemic all-grade AEs; and icotinib, osimertinib, gefitinib, erlotinib, dacomiti-
nib, and afatinib for systemic grade ≥3 AEs. In particular, individual

Fig. 2. Pooled estimate (A) and relative rankings of treat- ments (B) regarding systemic all-grade and grade ≥3 adverse events in network meta-analysis.
(A) Numbers in cells are odds ratios, along with 95 % credible intervals in parentheses, which greater (smaller) than 1 in- dicates that the row-defining treatment is more (less) toXic than the column-defining treatment. Bold numbers represent statistically significant results. (B) Numbers in cells are SUCRA values indicating the probability of treatment being ranked highest on toXicity, which is between 0 (certainly the safest treatment) and 1 (certainly the most toXic treatment). Abbre- viation: SUCRA, surface under the cumulative ranking; Osi, osimertinib; Dac, dacomitinib; Afa, afatinib; Erl, erlotinib; Gef, gefitinib; Ico, icotinib; CT, chemotherapy; AE, adverse event.

EGFR-TKIs were found to be associated with different spectrums.
toXicity

similar in design and conduct. EXcellent transitivity and consistency were summarized. The observed heterogeneity in EGFR-TKIs (afatinib,

The mechanisms underpinning the different toXicity profiles of these EGFR-TKIs might be explained by 1) their different affinities for the kinase domain of EGFR and the ability to spare wild-type EGFR. Of note, the third-generation inhibitor osimertinib, different from quinazoline- based reversible (erlotinib, gefitinib, and icotinib) and irreversible EGFR-TKIs (dacomitinib and afatinib) for its distinctive amino- pyrimidine scaffold, is irreversibly selective for EGFR mutations including T790 M while sparing wild type EGFR (Finlay et al., 2014; Russo et al., 2017). 2) Their different pharmacokinetic properties. EXcept for afatinib, EGFR-TKIs are metabolized by various members of the cytochrome P450 (CYP) family, mainly by CYP3A4, CYP2D6 and

erlotinib, and gefitinib) versus chemotherapy could not be explained based on the similar drug dose or key assessable baseline characteristics, except for the unitized various chemotherapy regimens in trials as a single network node in our study. Third, robust results were obtained by the tailored design of validation and sensitivity analyses.
It is essential to understand the toXicity profile of EGFR-TKIs to maximize their advantages, especially at the time when their adminis- tration is moving to earlier-stage lung cancer, as adjuvant strategies in particular (Wu et al., 2020; Zhong et al., 2021; Yue et al., 2018), where drug exposure is usually longer than the metastatic setting and long-term toXicities may increase. Along with the current recommen-

CYP1A2 (Shah and Shah, 2019). The different metabolizing enzyme

dations on prevention and management of toXicities induced by

profiles and sensitivity to those enzymes may consequently introduce the inter-individual variability in toXicity for individual EGFR-TKIs. And
3) other multiple complex mechanisms which have not yet been fully elucidated.
Previous systemic review studies, such as the one by Shah et al. (Shah and Shah, 2019), have indicated the difference in the incidence and severity of EGFR-TKI induced AEs in oncology, but are limited to a su- perficial level. Previous meta-analyses are limited by including fewer studies and treatments (Ding et al., 2017; Takeda et al., 2015), and primarily focusing on certain specific AEs, such as skin toXicity, diar- rhea, and fatigue (Sun and Li, 2019; Li and Gu, 2019; Li and Sun, 2018; Shi et al., 2014). In addition to the methodological power of network meta-analysis identifying relative toXicity between any two treatments, our study has several other key strengths. First, to the best of our knowledge, this is the largest and most comprehensive study on toXicity of EGFR-TKIs in lung cancer. Systemic all-grade and grade 3 AEs were thoroughly investigated as conservative incidence and severity metrics, in addition to each specific AE to know its incidence for each EGFR-TKI. Second, network heterogeneity, transitivity, and inconsistency were completely assessed. To minimize the potential statistical uncertainty and confounding effects in this broad investigation, we included only phase II or III randomized controlled trials which were generally very

EGFR-TKIs, like those focusing on skin reaction (Lacouture et al., 2021, 2011), this network meta-analysis provides clinicians a compelling safety reference for the individualized use of EGFR-TKIs for patients with lung cancer. First, before initiating use of an individual EGFR-TKI, results from this study, 1) could provide a toXicity encyclopedia of EGFR-TKIs for patient counseling. Clinicians could share with patients the toXicity profile of an individual EGFR-TKI including its relative toXicity compared with other EGFR-TKIs and its high incidence on some special AEs, which may improve treatment acceptance and compliance.
2) Guide the most appropriate choice among multiple EGFR-TKIs for some patients with pre-existing conditions. Gefitinib and erlotinib, for example, revealed predominant hepatotoXicity in this present study. Therefore, whenever possible, avoidance of their application for patients with pre-existing hepatic abnormalities or hepatotoXic co-medications should be considered. 3) Enhance the conduction of EGFR-TKI based combination treatments. The combination of EGFR-TKI and chemo- therapy, anti-angiogenic therapy, or other therapies has been vigorously investigated in recent years and has generally shown an impressive ef- ficacy (Hosomi et al., 2020; Noronha et al., 2020; Saito et al., 2019; Nakagawa et al., 2019), and thereinto, erlotinib plus ramucirumab or bevacizumab have become standard treatments for patients with EGFR-mutated lung cancer (NCCN, 2021). However, the combination of

EGFR-TKIs and another treatment may result in more toXicity in general, a reflection of the expected additional toXicity for either combined treatment (Zhao et al., 2019). For example, in patients exposed to osi- mertinib in combination with durvalumab in the TATTON study, up to
64 % of patients experienced ILD (Ahn et al., 2016). Therefore, providing this systematic evaluation of the toXicity of EGFR-TKI mon- otherapy, in this moment of potential shift to combination paradigms, is important for decision making and future trial designs on such combi- nation treatments.
Second, during long-term use, our results facilitate the process of careful monitoring, early recognition, and then timely management of EGFR-TKI related AEs. For instance, osimertinib was surprisingly demonstrated to be associated with the highest risk of ILD, which is an uncommon but potentially life-threatening EGFR-TKI related AE (Inoue et al., 2003). Although it is critical to keep an eye on ILD when admitting any EGFR-TKIs, for patients undergoing treatment with osimertinib versus those with other EGFR-TKIs, it is critical to pay more attention to suspicious respiratory symptoms and radiographic findings indicative of ILD/pneumonitis and its diagnosis by exclusion of other causes at an early phase, resulting in timely management and then longer mainte- nance of osimertinib.
Third, after the uncontrolled AEs (especially some long-term toXic- ities), our results may optimize the solutions by guiding drug switch among EGFR-TKIs. There are many reports of successful and uneventful switching from one EGFR-TKI to another when the first induced serious AEs, such as hepatotoXicity (Ku et al., 2010) and ILD (Chang et al., 2010). Those cases benefited from empirical decisions without known research to base on; our study could provide more evidence and strengthen the recommendations of clinicians when considering which EGFR-TKIs are the most appropriate alternatives. For example, there may be a more ensured success for the switching from gefitinib to afa- tinib than the reported alternative of erlotinib for hepatotoXicity, due to the demonstrated least and second most elevation of aminotransferases for afatinib and erlotinib, respectively. Still, further clinical experience and researches are warranted.
This study has limitations. First, there were a large number of direct
comparisons based on evidence from one trial only, and a low number of trials involving some EGFR-TKIs such as osimertinib (2 trial (Ram- alingam et al., 2020; Mok et al., 2017)) and icotinib (2 trials (Shi et al., 2017, 2013)), which could have weakened the external validity of this analysis and should be considered when interpreting our results. Second, discrepancies existed in the length of follow-up time (ranging from 2.7 to 69.4 months), which meant there were some uncertainties for the reported data in some trials in representing the real frequency of treatment-related AEs. Third, we were not able to investigate some potentially important clinical and demographical modifiers of treatment safety at the individual patient level (e.g., age, sex, race, EGFR mutation subtypes, smoking status), other important safety outcomes that might inform treatment decision making in routine clinical practice (e.g., serious AEs and AEs leading to treatment withdrawal or dose modifi- cation); and some additional AEs associated with individual EGFR-TKI that are not shared widely by the others (e.g., cardiac toXicity has been identified as an emerging safety issue that are more commonly associated with osimertinib (Ewer et al., 2021; Schiefer et al., 2018)). Fourth, the relationship between toXicity and outcome of anticancer treatments has gained attention in the past decade, for incidence, the
predictive value of skin rash for efficacy of EGFR-TKIs in patients with
lung cancer has been identified (Lee et al., 2012; Petrelli et al., 2012). However, no such investigations could be established in the present study and future studies are warranted. Other imprecise estimates were likely: 1) due to the limitation of the FLAURA study (Ramalingam et al., 2020) design, we had to assume the same toXicity outcome for gefitinib and erlotinib in the control arm, but the main results of remaining comparisons did not change based on the sensitivity analysis excluding the FLAURA study (Ramalingam et al., 2020); 2) icotinib was not accessible in some analyses, such as fatigue, stomatitis, constipation, dry

Table 3
Relative rankings of treatments on each specific adverse event (all-grade) in network meta-analysis.
Osimertinib Dacomitinib Afatinib Erlotinib Gefitinib Icotinib Chemotherapy Rash 0.25a 0.75 0.94b 0.76 0.43 0.37 0
Diarrhea 0.53 0.83 0.99b 0.58 0.35 0.19a 0.02
Anorexia 0.32 0.66b 0.38 0.50 0.57 0.09a 0.99
Nausea 0.39 0.55 0.31 0.46 0.66b 0.13a 0.99
Vomiting 0.50 0.43 0.44 0.32 0.63b 0.22a 0.97
Fatigue 0.45 0.31 0.45 0.63b 0.16a NA 0.99
Stomatitis 0.44 0.72 0.96b 0.22a 0.35 NA 0.31
Anemia 0.50 0.60b 0.22a 0.27 0.50 0.43 0.98
Constipation 0.45 0.54 0.18a 0.20 0.63b NA 0.99
Paronychia 0.48 0.76 0.90b 0.52 0.34a NA 0
Pruritus 0.58 0.64 0.77b 0.64 0.47 0.35a 0.05
Dry skin 0.54 0.93b 0.42a 0.63 0.49 NA 0
Alopecia 0.22a 0.61b 0.26 0.55 0.37 NA 0.99
AST elevation 0.23a 0.46 0.45 0.66 0.86b 0.37 0.46
ALT elevation 0.16a 0.42 0.51 0.76 0.91b 0.23 0.52
Leucopenia 0.79b 0.32 0.34 0.23a 0.44 0.42 0.96
Neutropenia 0.81b 0.16a 0.30 0.45 0.30 0.49 0.99
ILD 0.82b 0.62 0.33a 0.48 0.62 0.43 0.19
Numbers in cells are SUCRA values indicating the probability of treatment being ranked highest on toXicity, which is between 0 (certainly the safest treatment) and 1 (certainly the most toXic treatment). SUCRA, surface under the cumulative ranking; EGFR-TKI, epidermal growth factor receptor tyrosine kinase inhibitor; AST, aspartate transaminase; ALT, alanine transaminase; ILD, interstitial lung disease; NA, not applicable.
a Numbers represent the safest treatments among EGFR-TKIs on each specific adverse event.
b Numbers represent the most toXic treatments among EGFR-TKIs on each specific adverse event.
skin, and alopecia, for the data sparseness; 3) validation, subgroup, and sensitivity analyses were not conducted for the secondary outcome to evaluate the influence of potential eff ;ect modifiers to specific AEs.
5. Conclusions
This network meta-analysis depicts the toXicity profile including toXicity incidence, severity and spectrum (involving various specific AEs) of each EGFR-TKI in lung cancer. We found a generally greater toXicity for dacomitinib and afatinib, and less for icotinib followed by osimertinib and gefitinib. Individual EGFR-TKIs were associated with different toXicity spectrums. These findings provide a compelling safety reference for the individualized use of EGFR-TKIs for patients with lung cancer.
Funding
This research received no external funding.

CRediT authorship contribution statement
Yi Zhao: Methodology, Software, Formal analysis, Writing – original draft, Supervision. Bo Cheng: Methodology, Software, Formal analysis, Writing – original draft. Zisheng Chen: Software, Formal analysis, Writing – original draft, Writing – review & editing. Jianfu Li: Valida- tion. Hengrui Liang: Validation, Writing – original draft. Ying Chen: . Feng Zhu: Validation. Caichen Li: Investigation, Data curation. Ke Xu: Data curation. Shan Xiong: Data curation. Weixiang Lu: Methodology, Data curation, Supervision. Zhuxing Chen: . Ran Zhong: Writing – re- view & editing. Shen Zhao: Visualization, Data curation. Zhanhong Xie: Data curation. Jun Liu: Data curation, Writing – original draft. Wenhua Liang: Conceptualization, Project administration. Jianxing He: Conceptualization, Project administration.

Declaration of Competing Interest
The authors report no declarations of interest.

Acknowledgements
None.

Appendix A. Supplementary data
Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.critrevonc.2021.10330 5.
References
Ahn, M.J., Yang, J., Yu, H., Saka, H., Ramalingam, S., Goto, K., et al., 2016. 136O: osimertinib combined with durvalumab in EGFR-mutant non-small cell lung cancer: results from the TATTON phase Ib trial. J. Thorac. Oncol. 11, S115.
Bray, F., Ferlay, J., Soerjomataram, I., Siegel, R.L., Torre, L.A., Jemal, A., 2018. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J. Clin. 68, 394–424.
Brooks, S.P., Gelman, A., 1998. General methods for monitoring convergence of iterative simulations. J. Comput. Graph. Stat. 7, 434–455.
Cadranel, J., Gervais, R., Merle, P., Moro-Sibilot, D., Westeel, V., Bigay-Game, L., et al., 2015. Erlotinib versus carboplatin and paclitaxel in advanced lepidic adenocarcinoma: IFCT-0504. Eur. Respir. J. 46, 1440–1450.
Chaimani, A., Higgins, J.P.T., Mavridis, D., Spyridonos, P., Salanti, G., 2013. Graphical tools for network meta-analysis in STATA. PLoS One 8, e76654.
Chang, S.-C., Chang, C.-Y., Chen, C.-Y., Yu, C.-J., 2010. Successful erlotinib rechallenge after gefitinib-induced acute interstitial pneumonia. J. Thorac. Oncol. 5, 1105–1106.
Cheng, H., Li, X.-J., Wang, X.-J., Chen, Z.-W., Wang, R.-Q., Zhong, H.-C., et al., 2019.
A meta-analysis of adjuvant EGFR-TKIs for patients with resected non-small cell lung cancer. Lung Cancer 137.
Cipriani, A., Higgins, J.P.T., Geddes, J.R., Salanti, G., 2013. Conceptual and technical challenges in network meta-analysis. Ann. Intern. Med. 159, 130–137.
Ciuleanu, T., Stelmakh, L., Cicenas, S., Miliauskas, S., Grigorescu, A.C., Hillenbach, C., et al., 2012. Efficacy and safety of erlotinib versus chemotherapy in second-line treatment of patients with advanced, non-small-cell lung cancer with poor prognosis (TITAN): a randomised multicentre, open-label, phase 3 study. Lancet Oncol. 13, 300–308.
Crino`, L., Cappuzzo, F., Zatloukal, P., Reck, M., Pesek, M., Thompson, J.C., et al., 2008. Gefitinib versus vinorelbine in chemotherapy-naive elderly patients with advanced non-small-cell lung cancer (INVITE): a randomized, phase II study. J. Clin. Oncol. 26, 4253–4260.
Cufer, T., Vrdoljak, E., Gaafar, R., Erensoy, I., Pemberton, K., 2006. Phase II, open-label, randomized study (SIGN) of single-agent gefitinib (IRESSA) or docetaxel as second- line therapy in patients with advanced (stage IIIb or IV) non-small-cell lung cancer. Anticancer Drugs 17, 401–409.
Dias, S., Welton, N.J., Caldwell, D.M., Ades, A.E., 2010. Checking consistency in miXed treatment comparison meta-analysis. Stat. Med. 29, 932–944.
Dias, S., Welton, N.J., Sutton, A.J., Caldwell, D.M., Lu, G., Ades, A.E., 2014. NICE DSU Technical Support Document 4: Inconsistency in Networks of Evidence Based on Randomised Controlled Trials. National Institute for Health and Care EXcellence (NICE), London.
Díaz-Serrano, A., Gella, P., Jim´enez, E., Zugazagoitia, J., Paz-Ares Rodríguez, L., 2018.
Targeting EGFR in lung cancer: current standards and developments. Drugs 78, 893–911.
Ding, P.N., Lord, S.J., Gebski, V., Links, M., Bray, V., Gralla, R.J., et al., 2017. Risk of treatment-related toXicities from EGFR tyrosine kinase inhibitors: a meta-analysis of

clinical trials of gefitinib, erlotinib, and afatinib in advanced EGFR-mutated non- small cell lung cancer. J. Thorac. Oncol. 12, 633–643.
Ewer, M.S., Tekumalla, S.H., Walding, A., Atuah, K.N., 2021. Cardiac safety of osimertinib: a review of data. J. Clin. Oncol. 39, 328–337.
Finlay, M.R.V., Anderton, M., Ashton, S., Ballard, P., Bethel, P.A., BoX, M.R., et al., 2014. Discovery of a potent and selective EGFR inhibitor (AZD9291) of both sensitizing and T790M resistance mutations that spares the wild type form of the receptor.
J. Med. Chem. 57, 8249–8267.
Garassino, M.C., Martelli, O., Broggini, M., Farina, G., Veronese, S., Rulli, E., et al., 2013. Erlotinib versus docetaxel as second-line treatment of patients with advanced non- small-cell lung cancer and wild-type EGFR tumours (TAILOR): a randomised controlled trial. Lancet Oncol. 14, 981–988.
Greenhalgh, J., Dwan, K., Boland, A., Bates, V., Vecchio, F., Dundar, Y., et al., 2016.
First-line treatment of advanced epidermal growth factor receptor (EGFR) mutation positive non-squamous non-small cell lung cancer. Cochrane Database Syst. Rev.
CD010383.
Gregorc, V., Novello, S., Lazzari, C., Barni, S., Aieta, M., Mencoboni, M., et al., 2014. Predictive value of a proteomic signature in patients with non-small-cell lung cancer treated with second-line erlotinib or chemotherapy (PROSE): a biomarker-stratified, randomised phase 3 trial. Lancet Oncol. 15, 713–721.
Han, J.-Y., Park, K., Kim, S.-W., Lee, D.H., Kim, S.-W., Kim, S.-W., et al., 2012. First-
SIGNAL: first-line single-agent iressa versus gemcitabine and cisplatin trial in never- smokers with adenocarcinoma of the lung. J. Clin. Oncol. 30, 1122–1128.
Han, B., Jin, B., Chu, T., Niu, Y., Dong, Y., Xu, J., et al., 2017. Combination of chemotherapy and gefitinib as first-line treatment for patients with advanced lung adenocarcinoma and sensitive EGFR mutations: a randomized controlled trial. Int. J. Cancer 141, 1249–1256.
Hanna, N.H., Robinson, A.G., Temin, S., Baker, S., Brahmer, J.R., Ellis, P.M., et al., 2021. Therapy for stage IV non-small-cell lung cancer with driver alterations: ASCO and OH (CCO) joint guideline update. J. Clin. Oncol. JCO2003570.
Higgins, J.P.T., Thompson, S.G., Deeks, J.J., Altman, D.G., 2003. Measuring inconsistency in meta-analyses. BMJ 327, 557–560.
Higgins, J.P.T., Altman, D.G., Gøtzsche, P.C., Jüni, P., Moher, D., OXman, A.D., et al., 2011. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ 343, d5928.
Higgins, J., Thomas, J., Chandler, J., Cumpston, M., Li, T., Page, M., et al., 2019.
Cochrane Handbook for systematic reviews of interventions version 6.0 (updated July 2019). Cochrane.
Hosomi, Y., Morita, S., Sugawara, S., Kato, T., Fukuhara, T., Gemma, A., et al., 2020. Gefitinib alone versus gefitinib plus chemotherapy for non-small-cell lung cancer with mutated epidermal growth factor receptor: NEJ009 study. J. Clin. Oncol. 38, 115–123.
Hutton, B., Salanti, G., Caldwell, D.M., Chaimani, A., Schmid, C.H., Cameron, C., et al., 2015. The PRISMA extension statement for reporting of systematic reviews incorporating network meta-analyses of health care interventions: checklist and explanations. Ann. Intern. Med. 162, 777–784.
Inoue, A., Saijo, Y., Maemondo, M., Gomi, K., Tokue, Y., Kimura, Y., et al., 2003. Severe acute interstitial pneumonia and gefitinib. Lancet 361, 137–139.
Jansen, J.P., Fleurence, R., Devine, B., Itzler, R., Barrett, A., Hawkins, N., et al., 2011. Interpreting indirect treatment comparisons and network meta-analysis for health- care decision making: report of the ISPOR Task Force on Indirect Treatment Comparisons Good Research Practices: part 1. Value Health 14, 417–428.
Karampeazis, A., Voutsina, A., Souglakos, J., Kentepozidis, N., Giassas, S., Christofillakis, C., et al., 2013. Pemetrexed versus erlotinib in pretreated patients with advanced non-small cell lung cancer: a Hellenic Oncology Research Group (HORG) randomized phase 3 study. Cancer 119, 2754–2764.
Kawaguchi, T., Ando, M., Asami, K., Okano, Y., Fukuda, M., Nakagawa, H., et al., 2014.
Randomized phase III trial of erlotinib versus docetaxel as second- or third-line therapy in patients with advanced non-small-cell lung cancer: Docetaxel and Erlotinib Lung Cancer Trial (DELTA). J. Clin. Oncol. 32, 1902–1908.
Kelly, K., Azzoli, C.G., Zatloukal, P., Albert, I., Jiang, P.Y.Z., Bodkin, D., et al., 2012. Randomized phase 2b study of pralatrexate versus erlotinib in patients with stage IIIB/IV non-small-cell lung cancer (NSCLC) after failure of prior platinum-based therapy. J. Thorac. Oncol. 7, 1041–1048.
Kim, E.S., Hirsh, V., Mok, T., Socinski, M.A., Gervais, R., Wu, Y.-L., et al., 2008. Gefitinib versus docetaxel in previously treated non-small-cell lung cancer (INTEREST): a randomised phase III trial. Lancet 372, 1809–1818.
Kim, S.T., Uhm, J.E., Lee, J., Sun, J.-m, Sohn, I., Kim, S.T., et al., 2012. Randomized phase II study of gefitinib versus erlotinib in patients with advanced non-small cell lung cancer who failed previous chemotherapy. Lung Cancer 75, 82–88.
Kim, Y.S., Cho, E.K., Woo, H.S., Hong, J., Ahn, H.K., Park, I., et al., 2016. Randomized phase II study of pemetrexed versus gefitinib in previously treated patients with advanced non-small cell lung cancer. Cancer Res. Treat. 48, 80–87.
Kirsch, I., Moncrieff, J., 2007. Clinical trials and the response rate illusion. Contemp.
Clin. Trials 28, 348–351.
Kosaka, T., Yatabe, Y., Endoh, H., Kuwano, H., Takahashi, T., Mitsudomi, T., 2004.
Mutations of the epidermal growth factor receptor gene in lung cancer: biological and clinical implications. Cancer Res. 64, 8919–8923.
Ku, G.Y., Chopra, A., GdL, Lopes, 2010. Successful treatment of two lung cancer patients with erlotinib following gefitinib-induced hepatotoXicity. Lung Cancer 70, 223–225. Lacouture, M.E., Anadkat, M.J., Bensadoun, R.-J., Bryce, J., Chan, A., Epstein, J.B., et al.,
2011. Clinical practice guidelines for the prevention and treatment of EGFR inhibitor-associated dermatologic toXicities. Support. Care Cancer 19, 1079–1095.
Lacouture, M.E., Sibaud, V., Gerber, P.A., van den Hurk, C., Fern´andez-Pen˜as, P., Santini, D., et al., 2021. Prevention and management of dermatological toXicities

related to anticancer agents: ESMO Clinical Practice Guidelines. Ann. Oncol. 32, 157–170.
Lee, D.H., Park, K., Kim, J.H., Lee, J.-S., Shin, S.W., Kang, J.-H., et al., 2010. Randomized Phase III trial of gefitinib versus docetaxel in non-small cell lung cancer patients who have previously received platinum-based chemotherapy. Clin. Cancer Res. 16, 1307–1314.
Lee, Y., Shim, H.S., Park, M.S., Kim, J.-H., Ha, S.-J., Kim, S.H., et al., 2012. High EGFR
gene copy number and skin rash as predictive markers for EGFR tyrosine kinase inhibitors in patients with advanced squamous cell lung carcinoma. Clin. Cancer Res. 18, 1760–1768.
Li, J., Gu, J., 2019. Diarrhea with epidermal growth factor receptor tyrosine kinase inhibitors in cancer patients: a meta-analysis of randomized controlled trials. Crit. Rev. Oncol. Hematol. 134, 31–38.
Li, J., Sun, W., 2018. Fatigue with epidermal growth factor receptor tyrosine kinase inhibitors in cancer patients: a meta-analysis of randomized controlled trials.
J. Chemother. 30, 323–331.
Li, N., Ou, W., Yang, H., Liu, Q.-W., Zhang, S.-L., Wang, B.-X., et al., 2014. A randomized phase 2 trial of erlotinib versus pemetrexed as second-line therapy in the treatment of patients with advanced EGFR wild-type and EGFR FISH-positive lung adenocarcinoma. Cancer 120, 1379–1386.
Lilenbaum, R., AXelrod, R., Thomas, S., Dowlati, A., Seigel, L., Albert, D., et al., 2008. Randomized phase II trial of erlotinib or standard chemotherapy in patients with advanced non-small-cell lung cancer and a performance status of 2. J. Clin. Oncol. 26, 863–869.
Lu, G., Ades, A.E., 2004. Combination of direct and indirect evidence in miXed treatment comparisons. Stat. Med. 23, 3105–3124.
Lunn, D., Spiegelhalter, D., Thomas, A., Best, N., 2009. The BUGS project: evolution, critique and future directions. Stat. Med. 28, 3049–3067.
Maemondo, M., Inoue, A., Kobayashi, K., Sugawara, S., Oizumi, S., Isobe, H., et al., 2010.
Gefitinib or chemotherapy for non-small-cell lung cancer with mutated EGFR. N. Engl. J. Med. 362, 2380–2388.
Maruyama, R., Nishiwaki, Y., Tamura, T., Yamamoto, N., Tsuboi, M., Nakagawa, K., et al., 2008. Phase III study, V-15-32, of gefitinib versus docetaxel in previously treated Japanese patients with non-small-cell lung cancer. J. Clin. Oncol. 26, 4244–4252.
Mitsudomi, T., Morita, S., Yatabe, Y., Negoro, S., Okamoto, I., Tsurutani, J., et al., 2010. Gefitinib versus cisplatin plus docetaxel in patients with non-small-cell lung cancer harbouring mutations of the epidermal growth factor receptor (WJTOG3405): an open label, randomised phase 3 trial. Lancet Oncol. 11, 121–128.
Mok, T.S., Wu, Y.-L., Thongprasert, S., Yang, C.-H., Chu, D.-T., Saijo, N., et al., 2009. Gefitinib or carboplatin-paclitaxel in pulmonary adenocarcinoma. N. Engl. J. Med. 361, 947–957.
Mok, T.S., Wu, Y.-L., Ahn, M.-J., Garassino, M.C., Kim, H.R., Ramalingam, S.S., et al., 2017. Osimertinib or platinum-pemetrexed in EGFR T790M-positive lung cancer. N. Engl. J. Med. 376, 629–640.
Mor`ere, J.-F., Br´echot, J.-M., Westeel, V., Gounant, V., Lebeau, B., Vaylet, F., et al., 2010. Randomized phase II trial of gefitinib or gemcitabine or docetaxel chemotherapy in patients with advanced non-small-cell lung cancer and a performance status of 2 or 3 (IFCT-0301 study). Lung Cancer 70, 301–307.
Nakagawa, K., Garon, E.B., Seto, T., Nishio, M., Ponce AiX, S., Paz-Ares, L., et al., 2019. Ramucirumab plus erlotinib in patients with untreated, EGFR-mutated, advanced non-small-cell lung cancer (RELAY): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet Oncol. 20, 1655–1669.
NCCN, 2021. Clinical Practice Guidelines in Oncology for Non-Small Cell Lung Cancer.
Version 3. Available atwww.nccn.org/patientsAccessed March 10, 2021.
Noronha, V., Patil, V.M., Joshi, A., Menon, N., Chougule, A., Mahajan, A., et al., 2020. Gefitinib versus gefitinib plus pemetrexed and carboplatin chemotherapy in EGFR- mutated lung cancer. J. Clin. Oncol. 38, 124–136.
Paz-Ares, L., Tan, E.H., O’Byrne, K., Zhang, L., Hirsh, V., Boyer, M., et al., 2017. Afatinib versus gefitinib in patients with EGFR mutation-positive advanced non-small-cell lung cancer: overall survival data from the phase IIb LUX-Lung 7 trial. Ann. Oncol. 28, 270–277.
Petrelli, F., Borgonovo, K., Cabiddu, M., Lonati, V., Barni, S., 2012. Relationship between skin rash and outcome in non-small-cell lung cancer patients treated with anti-EGFR tyrosine kinase inhibitors: a literature-based meta-analysis of 24 trials. Lung Cancer 78.
Planchard, D., Popat, S., Kerr, K., Novello, S., Smit, E.F., Faivre-Finn, C., et al., 2018.
Metastatic non-small cell lung cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann. Oncol. 29, iv192–iv237.
Ramalingam, S.S., Blackhall, F., Krzakowski, M., Barrios, C.H., Park, K., Bover, I., et al., 2012. Randomized phase II study of dacomitinib (PF-00299804), an irreversible pan- human epidermal growth factor receptor inhibitor, versus erlotinib in patients with advanced non-small-cell lung cancer. J. Clin. Oncol. 30, 3337–3344.
Ramalingam, S.S., J¨anne, P.A., Mok, T., O’Byrne, K., Boyer, M.J., Von Pawel, J., et al., 2014. Dacomitinib versus erlotinib in patients with advanced-stage, previously treated non-small-cell lung cancer (ARCHER 1009): a randomised, double-blind, phase 3 trial. Lancet Oncol. 15, 1369–1378.
Ramalingam, S.S., Vansteenkiste, J., Planchard, D., Cho, B.C., Gray, J.E., Ohe, Y., et al., 2020. Overall survival with osimertinib in untreated, EGFR-Mutated advanced NSCLC. N. Engl. J. Med. 382, 41–50.
Recondo, G., Facchinetti, F., Olaussen, K.A., Besse, B., Friboulet, L., 2018. Making the first move in EGFR-driven or ALK-driven NSCLC: first-generation or next-generation TKI? Nat. Rev. Clin. Oncol. 15, 694–708.
Remon, J., Ahn, M.-J., Girard, N., Johnson, M., Kim, D.-W., Lopes, G., et al., 2019. Advanced-stage non-small cell lung cancer: advances in thoracic oncology 2018. J. Thorac. Oncol. 14, 1134–1155.

Rosell, R., Moran, T., Queralt, C., Porta, R., Cardenal, F., Camps, C., et al., 2009.
Screening for epidermal growth factor receptor mutations in lung cancer. N. Engl. J. Med. 361, 958–967.
Rosell, R., Carcereny, E., Gervais, R., Vergnenegre, A., Massuti, B., Felip, E., et al., 2012. Erlotinib versus standard chemotherapy as first-line treatment for European patients with advanced EGFR mutation-positive non-small-cell lung cancer (EURTAC): a multicentre, open-label, randomised phase 3 trial. Lancet Oncol. 13, 239–246.
Russo, A., Franchina, T., Ricciardi, G.R.R., Smiroldo, V., Picciotto, M., Zanghì, M., et al., 2017. Third generation EGFR TKIs in EGFR-mutated NSCLC: where are we now and where are we going. Crit. Rev. Oncol. Hematol. 117, 38–47.
Saito, H., Fukuhara, T., Furuya, N., Watanabe, K., Sugawara, S., Iwasawa, S., et al., 2019. Erlotinib plus bevacizumab versus erlotinib alone in patients with EGFR-positive advanced non-squamous non-small-cell lung cancer (NEJ026): interim analysis of an open-label, randomised, multicentre, phase 3 trial. Lancet Oncol. 20, 625–635.
Salanti, G., Ades, A.E., Ioannidis, J.P.A., 2011. Graphical methods and numerical summaries for presenting results from multiple-treatment meta-analysis: an overview and tutorial. J. Clin. Epidemiol. 64, 163–171.
Schiefer, M., Hendriks, Lel, Dinh, T., Lalji, U., Dingemans, A.-M.C., 2018. Current perspective: osimertinib-induced QT prolongation: new drugs with new side-effects need careful patient monitoring. Eur. J. Cancer 91, 92–98.
Sequist, L.V., Yang, J.C., Yamamoto, N., O’Byrne, K., Hirsh, V., Mok, T., et al., 2013. Phase III study of afatinib or cisplatin plus pemetrexed in patients with metastatic lung adenocarcinoma with EGFR mutations. J. Clin. Oncol. 31, 3327–3334.
Shah, R.R., Shah, D.R., 2019. Safety and tolerability of epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors in oncology. Drug Saf. 42, 181–198.
Sharma, S.V., Bell, D.W., Settleman, J., Haber, D.A., 2007. Epidermal growth factor receptor mutations in lung cancer. Nat. Rev. Cancer 7, 169–181.
Shi, Y., Zhang, L., Liu, X., Zhou, C., Zhang, L., Zhang, S., et al., 2013. Icotinib versus gefitinib in previously treated advanced non-small-cell lung cancer (ICOGEN): a randomised, double-blind phase 3 non-inferiority trial. Lancet Oncol. 14, 953–961.
Shi, L., Tang, J., Tong, L., Liu, Z., 2014. Risk of interstitial lung disease with gefitinib and erlotinib in advanced non-small cell lung cancer: a systematic review and meta- analysis of clinical trials. Lung Cancer 83, 231–239.
Shi, Y.K., Wang, L., Han, B.H., Li, W., Yu, P., Liu, Y.P., et al., 2017. First-line icotinib versus cisplatin/pemetrexed plus pemetrexed maintenance therapy for patients with advanced EGFR mutation-positive lung adenocarcinoma (CONVINCE): a phase 3, open-label, randomized study. Ann. Oncol. 28, 2443–2450.
Soria, J.-C., Felip, E., Cobo, M., Lu, S., Syrigos, K., Lee, K.H., et al., 2015. Afatinib versus erlotinib as second-line treatment of patients with advanced squamous cell carcinoma of the lung (LUX-Lung 8): an open-label randomised controlled phase 3 trial. Lancet Oncol. 16, 897–907.
Sun, W., Li, J., 2019. Skin toXicities with epidermal growth factor receptor tyrosine kinase inhibitors in cancer patients: a meta-analysis of randomized controlled trials. Cancer Invest. 37, 253–264.
Sun, J.-M., Lee, K.H., Kim, S.-W., Lee, D.H., Min, Y.J., Yun, H.J., et al., 2012. Gefitinib versus pemetrexed as second-line treatment in patients with nonsmall cell lung cancer previously treated with platinum-based chemotherapy (KCSG-LU08-01): an open-label, phase 3 trial. Cancer 118, 6234–6242.
Sutton, A.J., Abrams, K.R., 2001. Bayesian methods in meta-analysis and evidence synthesis. Stat. Methods Med. Res. 10, 277–303.
Takeda, M., Okamoto, I., Nakagawa, K., 2015. Pooled safety analysis of EGFR-TKI treatment for EGFR mutation-positive non-small cell lung cancer. Lung Cancer 88, 74–79.
Urata, Y., Katakami, N., Morita, S., Kaji, R., Yoshioka, H., Seto, T., et al., 2016.
Randomized phase III study comparing gefitinib with erlotinib in patients with previously treated advanced lung adenocarcinoma: WJOG 5108L. J. Clin. Oncol. 34, 3248–3257.
Wu, Y.-L., Zhou, C., Hu, C.-P., Feng, J., Lu, S., Huang, Y., et al., 2014. Afatinib versus cisplatin plus gemcitabine for first-line treatment of Asian patients with advanced non-small-cell lung cancer harbouring EGFR mutations (LUX-Lung 6): an open-label, randomised phase 3 trial. Lancet Oncol. 15, 213–222.
Wu, Y.L., Zhou, C., Liam, C.K., Wu, G., Liu, X., Zhong, Z., et al., 2015. First-line erlotinib versus gemcitabine/cisplatin in patients with advanced EGFR mutation-positive non- small-cell lung cancer: analyses from the phase III, randomized, open-label, ENSURE study. Ann. Oncol. 26, 1883–1889.
Wu, Y.-L., Cheng, Y., Zhou, X., Lee, K.H., Nakagawa, K., Niho, S., et al., 2017.
Dacomitinib versus gefitinib as first-line treatment for patients with EGFR -mutation- positive non-small-cell lung cancer (ARCHER 1050): a randomised, open-label, phase 3 trial. Lancet Oncol. 18, 1454–1466.
Wu, Y.-L., Tsuboi, M., He, J., John, T., Grohe, C., Majem, M., et al., 2020. Osimertinib in resected EGFR-mutated non-small-cell lung cancer. N. Engl. J. Med. 383, 1711–1723.
Yang, J.J., Zhou, Q., Yan, H.H., Zhang, X.C., Chen, H.J., Tu, H.Y., et al., 2017. A phase III randomised controlled trial of erlotinib vs gefitinib in advanced non-small cell lung cancer with EGFR mutations. Br. J. Cancer 116, 568–574.
Yue, D., Xu, S., Wang, Q., Li, X., Shen, Y., Zhao, H., et al., 2018. Erlotinib versus vinorelbine plus cisplatin as adjuvant therapy in Chinese patients with stage IIIA EGFR mutation-positive non-small-cell lung cancer (EVAN): a randomised, open- label, phase 2 trial. Lancet Respir. Med. 6, 863–873.
Zhai, H., Zhong, W., Yang, X., Wu, Y.-L., 2015. Neoadjuvant and adjuvant epidermal growth factor receptor tyrosine kinase inhibitor (EGFR-TKI) therapy for lung cancer. Transl. Lung Cancer Res. 4, 82–93.
Zhao, Y., Liu, J., Cai, X., Pan, Z., Liu, J., Yin, W., et al., 2019. Efficacy and safety of first line treatments for patients with advanced epidermal growth factor receptor mutated, non-small cell lung cancer: systematic review and network meta-analysis. BMJ 367, l5460.

Zhong, W.-Z., Wang, Q., Mao, W.-M., Xu, S.-T., Wu, L., Wei, Y.-C., et al., 2021. Gefitinib versus vinorelbine plus cisplatin as adjuvant treatment for stage II-IIIA (N1-N2) EGFR-mutant NSCLC: final overall survival analysis of CTONG1104 phase III trial. J. Clin. Oncol. 39, 713–722.
Zhou, C., Wu, Y.-L., Chen, G., Feng, J., Liu, X.-Q., Wang, C., et al., 2011. Erlotinib versus chemotherapy as first-line treatment for patients with advanced EGFR mutation- positive non-small-cell lung cancer (OPTIMAL, CTONG-0802): a multicentre, open- label, randomised, phase 3 study. Lancet Oncol. 12, 735–742.
Zhou, Q., Cheng, Y., Yang, J.J., Zhao, M.F., Zhang, L., Zhang, X.C., et al., 2014.
Pemetrexed versus gefitinib as a second-line treatment in advanced nonsquamous nonsmall-cell lung cancer patients harboring wild-type EGFR (CTONG0806): a multicenter randomized trial. Ann. Oncol. 25, 2385–2391.

Yi Zhao is Medical Oncologist at the First Affiliated Hospital of Guangzhou Medical University (Guangzhou, China). He received university diploma in Clinical Medicine from Capital Medical University (Beijing, China), and MD degree in Oncology from Guangzhou Medical University (Guangzhou, China). His current research interests mainly focus on the targeted therapy in lung cancer. He has great experience in data analysis and has published many peer-reviewed papers using the method of systemic review and network meta- analysis.

Bo Cheng is a post-graduate student at Guangzhou Medical University (Guangzhou, China). He focuses on translational research, data mining and clinical trials in lung cancer, especially the characterization of lung cancer stratified by histological and genotypic subtypes, and made major contributions in more than 10 SCI papers..

Zisheng Chen received his MD and PhD degree in Clinical Medicine from Southern Medical University (Guangzhou, China) and currently, he is a postdoc in Oncology at Guangzhou Medical University (Guangzhou, China) with research focusing on the EGFR- mutated non-small cell lung cancer. More than 10 research works were published in peer- reviewed journals. He is a member of the Young Scientific Committee of Guangdong Medical Association (China).

Jianfu Li is a post-graduate student at Guangzhou Medical University (Guangzhou, China). He is also a young researcher focusing on genetic analysis, translational research, early screening, data mining, and clinical trials in thoracic oncology.

Hengrui Liang received MD degree at Guangzhou Medical University (Guangzhou, China) in 2021. He is board certified in thoracic surgery, with clinical interest in minimal invasive thoracic surgery and research interest in lung cancer as well as artificial intelligence related medical application. He has published over 35 first-author scientific papers indexed by SCIE. He won 2017 WCLC Developing Award, 2018 ESMO Asia Merit Award, and 2019 JLCS Travel Award.

Ying Chen is a nurse-in-charge and researcher in the Department of Thoracic Surgery and Oncology at the First Affiliated Hospital of Guangzhou Medical University (China), and in recent 5 years, her research mainly focuses on the EGFR-mutated lung cancer.

Feng Zhu received university diploma in Clinical Medicine from Sun Yat-sen University (Guangzhou, China). She is currently pursuing internal medicine training in the US. Her main research interest is in lung cancer therapy. With extensive experience in statistical analysis, she has published peer-reviewed papers in different journals.

Caichen Li received MD degree in Oncology from Guangzhou Medical University (Guangzhou, China) and now is a researcher at the First Affiliated Hospital of Guangzhou Medical University (Guangzhou, China). His research mainly focuses on disease screening, data mining and clinical trials in thoracic oncology.

Ke Xu had graduated in Clinical Medicine from Guangzhou Medical University (Guangzhou, China). At present, he is receiving his MD education in Oncology and being trained in the minimally invasive thoracoscopic surgery in the Department of Thoracic Surgery and Oncology of the First Affiliated Hospital of Guangzhou Medical University (Guangzhou, China).

Shan Xiong is a post-graduate student at Guangzhou Medical University (Guangzhou, China). She has been receiving sub-specialized training in medical oncology since July 2019. She co-authored 17 peer-reviewed papers. She won the National Scholarship and Nan-Shan Scholarship in 2020.

WeiXiang Lu graduated from Hubei University of Medicine in 2017 (Shiyan, China). He is currently taking the residency training at the First Affiliated Hospital of Guangzhou Medical University (Guangzhou, China). His research interests are in lung cancer and artificial intelligence.

Zhuxing Chen received MBBS degree in Clinical Medicine at the University of South China in 2013. He worked as a thoracic surgeon in Peoples’ Hospital of Sanshui District (Foshan, China) from 2014 to 2019. He is taking the master’s education in Oncology at Guangzhou Medical University (Guangzhou, China). Currently, his research interesting focuses on the anti-tumor treatment and microbiota in lung cancer.

Ran Zhong graduated in Clinical Medicine from Guangzhou Medical University (Guangzhou, China). Currently, she is a postgraduate student in Oncology at Guangzhou Medical University with research focusing on the lung cancer. She co-authored several peer-reviewed papers.

Shen Zhao received MD degree in Oncology from Cancer Center of Sun Yat-sen University. She has researching skills and clinical expertise in drug therapy for lung cancer. She is good at network meta-analysis with many relevant publications.

Zhanhong Xie is a doctor and researcher in respiratory diseases at the First Affiliated Hospital of Guangzhou Medical University (Guangzhou, China).

Jun Liu is Professor at the First Affiliated Hospital of Guangzhou Medical University (Guangzhou, China). He obtained MD and PhD degrees at Guangzhou Medical University (Guangzhou, China). He has engaged in clinical research of individualized treatment for lung cancer. So far, he has published more than 30 SCI papers as first/corresponding author.

Wenhua Liang is Professor of medical oncology at the First Affiliated Hospital of Guangzhou Medical University. He received MD and PhD degrees from Sun Yat-sen Uni- versity (Guangzhou, China). He is the Associated Chief Physician and Associated Professor of Thoracic Oncology at The First Affiliated Hospital of Guangzhou Medical University (Guangzhou, China) since 2014. He dedicates in management of lung cancer and relevant clinical/translational research, and his recent researching works can be found at http s://www.researchgate.net/profile/Wenhua_Liang2/publications. He is the Chief Asso- ciate Editor of Translational Lung Cancer Research.

Jianxing He MD, Clinical medicine, 1985, Guangzhou medical university. PhD, Cardio- vascular Surgery, 2000, Guangdong Institute of Cardiovascular Disease. Director of the Guangzhou Institute of Respiratory Disease (China). Director of the Department of Thoracic Surgery and Oncology at the First Affiliated Hospital of Guangzhou Medical University (China). Chairman of Guangdong Thoracic PF-00299804 Surgeon Association (China). EX- ecutive Editor-in-Chief of Journal of Thoracic Disease; Editor-in-Chief of Annals of Trans- lational Medicine. Research/Clinical Interest: Thoracic oncology and transplantation. Major contribution in more than 330 papers about thoracic diseases in SCI indexed Journals.