Lung cancer has surpassed colonic and prostate cancers as a leading cause of cancer-related deaths globally, due to improvements in early detection and lowered average ages at diagnosis (1). Authoritative research had shown that the development of non-small cell lung cancer (NSCLC) may be influenced by the epidermal growth factor receptor (EGFR) (2). Among the frequent EGFR gene mutations, the L858R point mutations in exon 21 accounts for 40% and the exon 19 deletion mutations (19del) accounts for 45% (3, 4). Both variants have been named sensitive mutations. Additional EGFR mutations (G719X, S768I, L861Q, etc.) are classified as rare mutations. Since treatment with epidermal growth factor receptor-tyrosine kinase inhibitors (EGFR-TKI) has been demonstrated to dramatically enhance the survival of patients with NSCLC, the identification of EGFR mutations has become a crucial part of NSCLC treatment, particularly for lung adenocarcinoma (LUAD) (5–7).

Currently, EGFR-related studies are focusing on the link between EGFR mutations and NSCLC; nevertheless, this topic is fraught with controversy. The findings of studies by Yotsukura et al. Zhang and et al., which suggest that EGFR mutations are early genomic events in LUAD (8, 9), while Zhu et al. suggest that the frequency of EGFR mutations is not significantly different in minimally invasive adenocarcinoma (MIA) and invasive adenocarcinoma (IAC) (10). Also of great interest is whether or not there is a correlation between the imaging characteristics of LUAD and the frequency with which it undergoes EGFR mutations. Wei et al. found no link between EGFR and ground glass opacity (GGO) development, but Ortiz et al. found an increased probability of EGFR mutations when GGO was present in pulmonary nodules (11, 12). As a result, we aimed to perform a systematic review and meta-analysis of the literature to learn more about the correlation between EGFR mutations and the percentage of ground glass opacity and final histological classification in people with NSCLC.

A growing trend today is the identification of EGFR mutations in patients undergoing early diagnosis and therapy for NSCLC (29, 30). Although the conclusions of these investigations are still up for debate, prior researches have shown a connection between EGFR mutations and the imaging presentation of NSCLC and the ultimate histology of NSCLC. In order to draw more convincing conclusions, we conducted a systematic review and meta-analysis of 9 retrospective studies to examine the relationship between the clinical, imaging and histology characteristics of NSCLC patients and the frequency of EGFR mutations. This meta-analysis study discovered that age, tumor size, female gender, non-smoking individuals, pulmonary nodules presenting as mGGO, pulmonary nodules with histological types of MIA and IAC were significant factors influencing EGFR mutations. And further analysis confirmed that in the L858R point mutation group, there was a higher incidence of pulmonary nodules with histological type IAC or MIA. In contrast, in the exon 19 mutation group, only pulmonary nodules with IAC had a higher incidence, while pulmonary nodules with MIA were not statistically significant.

In this study, we discovered that patients who were substantially older had a higher risk of developing EGFR mutations than patients who were younger (SMD = 0.23; 95% CI: 0.12–0.34; p < 0.001). Multiple studies have shown that the risk of EGFR mutations increases with age in NSCLC patients (31, 32). Despite adjusting for patient histology, smoking status, and pathological staging, one study found an independent statistical difference between EGFR mutations and the age of the patient at diagnosis (33). In response to this conclusion, we suggest that those who are older have an increased risk of developing mutations due to changes in their own hormone levels, diminished nucleotide repair capacity, and increased exposure to carcinogenic stimuli from the environment (34–37). To validate this conclusion, additional large-scale gene sequencing with population representation is required.

The individual’s smoking history is also a significant determinant in EGFR mutations. This meta-analysis indicated that individuals without a history of smoking are more likely to have EGFR mutations (OR = 1.95; 95% CI: 1.42–2.69; p < 0.001). However, this analysis showed relatively high heterogeneity (I² = 51%; p = 0.04). This may be attributable to variations in the statistical criteria for smoking history among the included studies. For example, the study by Wang et al. and Dai et al. defined patients who had quit smoking ≥1 year prior to the surgery were defined as former smokers (21, 27), whereas the remaining included studies did not make this distinction. In addition, we would have liked to use the smoking index, was defined as the number of cigarettes smoked per day multiplied by the number of years smoked, as a statistical measure of the extent of smoking among patients in this analysis to give more credibility. This statistical method was not utilized in any of the included studies, which prevented us from realizing our design. This can be explored further in a subsequent investigation.

Furthermore, the study revealed that the proportion of female patients in the EGFR mutant-positive group was significantly higher than in the EGFR mutant-negative group (OR = 1.73; 95% CI: 1.16–2.57; p = 0.007), meaning that women are more likely to develop EGFR mutations. Based on the findings of this study, we suggest that the estrogen level in females affects the incidence of EGFR mutations. Multiple studies have verified a positive correlation between estrogen receptors and EGFR mutations (38, 39). Additionally, the study by Mazières et al. confirmed that the expression of estrogen receptors was greater in non-smoking women than in smokers (40). This precisely explains why women who do not smoke have a higher lung cancer incidence rate. Chen et al. and Linardou et al. revealed a considerably greater probability of EGFR mutations in women in Asia, notably in East Asia (41, 42). The fact that each of the 9 studies we included was from East Asia further supports the reliability of our analysis.

For determining whether the tumor has an EGFR mutation, it’s also crucial to consider the size of the tumor and the percentage of GGO on the thoracic CT. The results of this meta-analysis revealed that the incidence of EGFR mutations was greater in pulmonary nodules exhibiting mGGO compared to those exhibiting pGGO (OR = 1.43; 95% CI: 1.09–1.88; p = 0.010). And the incidence of EGFR mutation was positively correlated with the imaging size of the tumor (SMD = 0.38; 95% CI: 0.25–0.51; p < 0.001). A study by Cai et al. in 2023 showed that pGGO and mGGO reflect the pathological development and genetic alterations of pulmonary nodules (43). This study suggests that the solid component of pGGO emerged and that the proportion of solid components progressively increased - manifesting as mGGO, which indicates progression of the tumor (i.e., pathologic findings of MIA or IAC) and an increase in the rate of EGFR mutations. Li et al. found that larger diameter tumors did reveal higher frequency and types of mutations in addition to EGFR mutations, such as ALK rearrangements, TP53 mutations, etc., in postoperative genetic testing (44).

EGFR mutations were strongly related with the histological type of the patient’s tumor. The results of the analysis indicate that the ultimate pathological outcomes of the tumors, whether MIA (OR = 1.82; 95% CI: 1.23–2.69; p = 0.003) or IAC (OR = 1.99; 95% CI: 1.14–3.48; p = 0.020), have a higher incidence of EGFR mutations than PGL. However, the meta-analysis of IAC showed a relatively high heterogeneity (I² = 66%; p = 0.007), which might be due to subjective bias in the interpretation of IAC criteria by different institutions and different pathologists in the included studies. Several studies have shown that EGFR amplification is essential for the progression of AIS to MIA and even IAC (9, 45). Moreover, the risk of developing secondary primary lung cancer is marginally increased in MIA patients with EGFR mutations (8).

After conducting additional analysis for various subtypes of mutants, we arrived at contrasting conclusions. For tumors with pathological type IAC, both exon 19 deletion and L858R point mutation have a high incidence of mutations (OR = 2.94, 95% CI: 1.95–4.96, p < 0.001 for exon 19 deletion; OR = 2.79, 95% CI: 1.28–6.04, p = 0.009 for L858R point mutation). In contrast, tumors exhibiting MIA maintained a high mutation incidence only for the L858R point mutation (OR = 1.80; 95% CI: 1.04–3.13; p = 0.04), while losing statistical significance for the exon 19 deletion (OR = 1.81; 95% CI: 0.95–3.44; p = 0.07). It has been shown that L858R point mutations are detected more frequently in MIA with completely different tumor characteristics compared to exon 19 deletions (46). However, as only 3 papers were included in this analysis, the results are perhaps not robust enough, which still needs to be corroborated by subsequent relevant studies.

In 2004, the close link between NSCLC and EGFR mutations was identified for the first time, ushering NSCLC treatment into the period of targeted therapy. More than 80% of EGFR mutations are exon 19 deletion mutations and exon 21 L858R point mutations. According to studies, however, patients with rare EGFR mutations have a worse prognosis than those with sensitive mutations. For instance, the EGFR 20 exon insertion mutations causes a spatial site block in the structure of the EGFR protein, reducing the size of the drug-binding pocket and preventing the EGFR-TKI from binding to its target, thereby allowing the EGFR protein to remain active and the oncogenic signal to persist (47, 48). We had planned to set up further research of rare types of EGFR mutations, but unfortunately, the dearth of studies and the scarcity of data on rare mutations prevented us from implementing our plan, which may be taken into account in a future meta-analysis.

This systematic review and meta-analysis has several major advantages. First, we report for the first time that lung adenocarcinomas presenting as MIA exhibit distinct EGFR mutation features. Secondly, our study included a relevant sample size of 1789 patients whose sources included only articles with a low risk of bias and high quality, which provides a more realistic and convincing reflection of the accuracy of the final results. In addition, the sensitivity analysis performed provides evidence that all the results presented are robust. We also assessed publication bias using Egger’s test to ensure that the results are not biased by publication bias. Most importantly, the search strategy developed by our researchers was thorough to ensure that no valuable literature was missed.

This meta-analysis also has several limitations. First of all, the literature included in our study was all from East Asia, which may compromise the accuracy of the analysis of the incidence of EGFR mutations in non-Asian populations. In addition, all included studies were retrospective studies, and there may be selection bias affecting the final overall results. Furthermore, thoracic CT scan parameters, GGO ratio calculation, and EGFR mutations detection methods differed among the included studies, which may increase the heterogeneity among the studies.

This systematic review and meta-analysis is the most exhaustive and up-to-date examination of the literature concerning the risk factors linked with EGFR mutations. The final results proved that patients with higher age, female gender, and non-smoking individuals would have a relatively higher incidence of EGFR mutations. Also, the risk of EGFR mutations was increased with larger tumor diameter, tumor imaging presentation of mGGO, and tumor pathological findings of MIA or IAC. Importantly, tumors presenting as MIA are more likely to have the incidence of L858R point mutation. This finding is somewhat helpful in predicting and assessing the molecular pathological alterations in stage IA NSCLC. Based on the aforementioned risk factors associated with EGFR mutations, thoracic surgeons can make an initial assessment of the molecular pathology of early-stage lung nodules in order to minimize unnecessary costs to the patient, thereby reducing the patient’s financial burden. It also reduces the rate of underdiagnosis of patients with EGFR mutations, and ensures that patients with EGFR mutations will be able to detect their mutations in a timely manner and receive more rigorous postoperative follow-up.

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding authors.

JQ: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Writing – original draft. ZM: Data curation, Formal analysis, Methodology, Project administration, Supervision, Writing – review & editing. RL: Data curation, Formal analysis, Investigation, Software, Writing – review & editing. CQ: Formal analysis, Investigation, Writing – review & editing. KW: Formal analysis, Software, Writing – review & editing. BL: Data curation, Software, Writing – review & editing. YT: Conceptualization, Project administration, Supervision, Writing – review & editing. HT: Conceptualization, Funding acquisition, Methodology, Project administration, Supervision, Writing – review & editing.