The epidermal growth factor receptor (EGFR) is a member of the ErbB family. It is involved in biological processes such as cell proliferation, angiogenesis, inhibition of apoptosis, adhesion, metastasis, and is associated with the progression of lung cancer [1-3]. These EGFR aberrations will promote cell proliferation and resists apoptosis by excessive active of downstream signaling pathways, and thus the EGFR is also recognized as a proto-oncogene [2]. The clinical significance of EGFR mutations has been widely explored in non-small cell lung cancer (NSCLC). Galvez et al. [4] enrolled 282 radically treated patients with lung adenocarcinoma, of which 142 were EGFR mutated and 140 were EGFR wild. It was found that there was no difference in median progression-free survival (PFS) between the two groups, but patients with EGFR-mutated lung adenocarcinoma were more prone to recurrence and metastasis (97% vs 68%, P=0.007). It has been demonstrated in clinical practice that EGFR mutations are correlated with high response rates to EGFR tyrosine kinase inhibitors (EGFR-TKI).

NSCLC patients are prone to brain metastasis, the incidence of brain metastasis in NSCLC is about 30%, and the cases in which the brain is the site of first recurrence are about 15%-40% [5]. The mechanism is unknown, and hematogenous metastasis is currently considered the most likely route of metastasis. The entry of tumor cells into the brain parenchyma through the blood-brain barrier requires a series of cascading steps [6], which mainly include the invasion of cancer cells into the surrounding stroma by breaking through the basement membrane, blood circulation, invasion as well as disruption of the blood-brain barrier, central nervous system invasion, and colonization of the brain. The advent of EGFR-TKI marked a milestone in NSCLC treatment, providing a new therapeutic approach for patients with brain metastases. This meta-analysis included 26 studies [7], including 4,007 EGFR-mutated NSCLC patients and 10,022 EGFR-wild NSCLC patients, and the pooled results indicated that EGFR-mutated patients were more likely to suffer from brain metastases (OR=1.58, 95%CI: 1.36-1.84, P<0.05). This evidence-based finding carries significant real-world implications, suggesting that the EGFR mutation status should be considered by clinicians when formulating treatment plans.

For patients with NSCLC brain metastases harboring EGFR mutations, systemic treatment primarily involves targeted therapies. The BRAIN study [8] from China provides clinical evidence for the use of first-generation EGFR-TKI, which is also the first phase III clinical trial comparing EGFR-TKI and whole-brain radiotherapy plus chemotherapy for the treatment of EGFR-mutant NSCLC brain metastases. The median follow-up period of the study was 16.5 months. The intracranial PFS was 4.8 months (95%CI: 2.4-7.2) in the combined group and 10.0 months in the icotinib group (95% CI: 5.6-14.4), which was statistically significant (HR=0.56, 95% CI :0.36-0.90, P=0.014, equivalent to a 44% lower risk of intracranial lesion progression or death events with icotinib). The safety profile of icotinib was also superior to that of the combined group, with rates of grade 3 or higher adverse events of 8% (7/85) and 38% (28/73), respectively.

There is fewer data for the second-generation EGFR-TKI studies. LUX-Lung 3, a phase III clinical study compared afatinib versus cisplatin in combination with pemetrexed for the treatment of patients with EGFR-mutated lung adenocarcinoma, showed that the PFS for patients with brain metastases was 11.1 months vs. 5.4 months (HR=0.54, P=0.1378). LUX-Lung 6, a phase III clinical study compared afatinib with cisplatin in combination with gemcitabine for the treatment of patients with EGFR-mutated lung adenocarcinoma and the results showed that for patients with brain metastases, the PFS was 8.2 months vs. 4.7 months (HR=0.47, P=0.1060). When further summarizing survival data for patients with brain metastases in both trials [9], there was a significant improvement in PFS with afatinib compared to chemotherapy (8.2 vs. 5.4 months, HR=0.50, P=0.0297).

The third-generation EGFR-TKI prolongs patient survival with the best clinical data on efficacy against brain metastases. It is well known that several large clinical trials have demonstrated superior PFS as well as higher quality of life (toxicities) with EGFR-TKI compared to chemotherapy in EGFR mutation-positive advanced NSCLC [10]. The survival data from the FLAURA study are even more encouraging [11,12]. Compared to the first-generation EGFR-TKI (gefitinib/erlotinib), the survival benefit was significantly prolonged in the third-generation EGFR-TKI (osimertinib), with a median PFS (18.9 months vs. 10.2 months, HR=0.46, 95% CI: 0.37-0.57, P<0.001), and a median overall survival (38.6 months vs. 31.8 months, HR=0.80, 95% CI: 0.64-1.00, P=0.046). In the subgroup analysis of brain metastases, the median PFS of osimertinib was also superior to that of gefitinib/erlotinib (15.2 months vs. 9.6 months, HR=0.47, 95% CI: 0.30-0.74, P<0.001). The first-generation EGFR-TKI and second-generation EGFR-TKI resistance mechanism is mainly EGFR-T790M mutation. Osimertinib has a strong inhibitory ability on tumor cells with EGFR-T790M (especially L858R mutation). It shows better efficacy in clinic by inducing the degradation of mutant EGFR as well as blocking the homodimerization of mutant EGFR to inhibit receptor activation, blocking downstream signaling pathways, and inhibiting the proliferation of tumor cells [13].

EGFR-TKI targeted therapy has a high response rate for intracranial lesions in patients with EGFR-mutant NSCLC brain metastases, but the control is not persistent. Radiotherapy is an effective treatment for brain metastases. EGFR-TKI combined with radiation therapy may have some advantages over other ways [14], but the optimal way of using it in combination with EGFR-TKI is still unclear. In a multicenter retrospective study [15], 95 patients with EGFR-mutated NSCLC brain metastases were enrolled to compare the survival prognosis of patients who received either EGFR-TKI (92.3% received osimertinib) or EGFR-TKI (97.7% received osimertinib) in combination with brain radiotherapy, and the results showed that PFS, intracranial PFS, and time to treatment failure were similar in both groups. A meta-analysis [16] included 12 clinical studies to investigate the difference in efficacy of EGFR-TKI combination radiotherapy or EGFR-TKI alone in patients with brain metastases from first-diagnosed EGFR mutation-positive NSCLC. The results indicated that for patients with symptomatic brain metastases, elderly patients, and patients with 19Del mutation, EGFR-TKI combined with radiotherapy improved overall survival and intracranial PFS compared with EGFR-TKI treatment alone. Some scholars suggest that EGFR-TKI has great potential in the treatment of NSCLC brain metastases. However, due to a variety of factors, such as the widespread brain metastases in EGFR-mutant NSCLC, EGFR-TKI possibly serve as radiosensitizers in combination and the possible sudden worsening of disease (including brain metastases) due to abrupt withdrawal of EGFR-TKI, the optimal treatment and sequence for brain metastases requires individual consideration [17].

The prognosis of patients with brain metastases from NSCLC is poor. First, the appearance of brain metastasis suggests that the cancer cells have spread, which means that NSCLC patients enter the advanced stage and have a short survival time; second, the growth of brain metastases in the skull is prone to cause cognitive dysfunction, which seriously affects the quality of life of patients; third, the existence of the blood-brain barrier makes it difficult for chemotherapeutic drugs to enter the skull to exert their therapeutic effects. New generations of EGFR-TKI have extended patient survival times, but their use is contingent on the EGFR mutation status. This meta-analysis carries significant clinical practice implications. However, the included studies are all retrospective analyses, and the conclusions only describe a biological characteristic. It is well known that NSCLC has a high rate of EGFR mutations, and EGFR is related to tumor development [1-3]. The reason why patients with EGFR mutations have a higher risk of brain metastasis is unknown. Wang et al. [18] performed next-generation sequencing technologies on paired lung primary foci and brain metastatic foci in 61 cases of NSCLC, and found that mutations of the EGFR, KRAS, TP53, and ALK genes in the lung primary and metastatic foci were highly concordant (>80%). Brain lesions had higher tumor mutational load and genomic instability than lung lesions, in which variants in genes encoding the PI3K signaling pathway were enriched in brain metastases. The PI3K signaling pathway was significantly correlated with an increased risk of NSCLC brain metastasis, and patients with the activated PI3K signaling pathway had a shorter brain metastasis-free survival time. EGFR mutations can activate the downstream signaling pathway AKT-PI3K-mTOR [2], and future studies can further explore the correlation between the PI3K signaling pathway and NSCLC brain metastasis.

This Meta-analysis reminds us that cancer is heterogeneous at the molecular level from a clinical perspective. The treatment of cancer should be individualized, and the appropriate treatment method and treatment sequence should be selected according to the clinical symptoms, tumor stage, and different mutation status of the patient. Currently, the search for individualized biomarkers at the genetic level is a hot topic, including genetic testing for EGFR-TKI resistance, exploration of new gene targets, detection of minimal residual disease in liquid biopsy, and biomarkers for predicting immunotherapy efficacy, etc., aiming to provide more evidence for the individualized management of NSCLC patients [19]. As the Chinese proverb goes, "a thousand people have a thousand faces," the treatment of cancer should be individualized.