Brain metastasis is a common complication in patients with cancer, with lung cancer being the most frequent origin of brain metastases (1). Lung cancer has an ~40% chance of metastasizing to the brain during the course of disease progression (1). Current standard treatments for lung cancer brain metastasis patients include surgery, stereotactic radiosurgery and whole-brain radiation therapy (1). It has long been considered that the blood-brain barrier (BBB) hampers the delivery of chemotherapeutic agents to the metastatic brain lesion, and systemic antineoplastic therapy has been reported to demonstrate limited or no efficacy in brain metastases (2,3).

In addition to chemotherapeutic agents, molecular targeting agents, including erlotinib, have begun to serve an important role in the treatment of lung cancer. Erlotinib is an orally active epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor (TKI), which is effective against advanced non-small cell lung cancer (NSCLC) with EGFR mutations. Potentially due to their low molecular mass, EGFR-TKIs are able to cross the BBB and have been reported to demonstrate anticancer activity against NSCLC brain metastases by penetrating the BBB (1,2,4). However, the underlying mode of action or molecular mechanisms of metastatic brain lesion resistance to EGFR-TKIs remain unclear. Although immunohistological features of resected lung cancer following EGFR-TKI treatment have been reported in an attempt to elucidate the mechanisms of resistance, no immunohistological study of EGFR status has been performed on brain metastasis while still responding to EGFR-TKI. The present study reports, to the best of our knowledge, the first case of EGFR-mutated NSCLC brain metastasis that was surgically resected while the lesion was responding to erlotinib, and aims to elucidate the mode of response of metastatic brain lesions from EGFR-mutated NSCLC to EGFR-TKIs, and investigate the possible underlying mechanisms involved in drug resistance.

Previous case reports have advocated the efficacy of EGFR-TKI for patients with brain metastasis arising from EGFR-mutated lung cancer (10,11). Berger et al (11) reported that the response rates using EGFR-TKI in brain metastasis with EGFR mutation were <80%, which is markedly high when the expected response rate for conventional treatments such as whole-brain radiation therapy is taken into account. Although the concentration of erlotinib in the cerebrospinal fluid is reported to be decreased compared with that in the plasma (3), a disrupted BBB around the metastatic lesion is considered to lead to an increase in the local drug concentration (2). Consistent with previous studies (1,12,13), pre-operative MRI in the present study demonstrated a slight but notable regression of brain metastasis following initiation of EGFR-TKI therapy (Figs. 1 and 2).

Despite a marked amount of data associated with the clinical response of EGFR-mutated lung cancers to EGFR-TKIs (11,12), there is a lack of evidence of the histopathological changes in cancer tissues following EGFR-TKI treatment. A previous study by Lara-Guerra et al (14) demonstrated the histopathological features of lung cancer tissues treated with gefitinib. In the previous study, EGFR-mutated tumors following EGFR-TKI therapy demonstrated decreased tumor cellularity and a decreased Ki-67 labeling index within the fibrous stroma compared with wild-type EGFR adenocarcinoma tumors. Furthermore, the study described that histological features were associated with the clinical response of the tumor to EGFR-TKI (14). In the present study, viable metastatic cells existed as foci within a large necrotic area. It was also confirmed that those viable cells presented decreased proliferative activity evaluated using Ki-67 labeling similar to the study by Lara-Guerra et al (14). These results confirm that EGFR-TKI demonstrates cytotoxic action against brain metastatic lesions, potentially beyond the BBB.

Furthermore, EGFR mutation status during treatment with EGFR-TKI was evaluated using IHC and F-PHFA gene analysis. IHC demonstrated that viable cells did not express L858R mutant EGFR, whereas the use of F-PHFA demonstrated that these cells did have the L858R EGFR mutation. Although viable cells that are resistant to EGFR-TKI treatment possessed the L858R EGFR mutation, EGFR expression was potentially markedly downregulated in those cells. This model is also supported by the fact that these cells demonstrated negative staining when anti-pan-EGFR antibodies were used for immunostaining. The secondary EGFR mutation T790M, which is observed in ~50% of EGFR-TKI-resistant patients (8), is one of the primary mechanisms that is considered important in the development of EGFR-TKI resistance. The T790M EGFR mutation was also evaluated using F-PHFA, demonstrating that viable cells were negative for the T790M EGFR mutation. It is hypothesized that the T790M mutation occurs during EGFR-TKI therapy and may exist prior to treatment (11,15). EGFR-TKI-resistant brain metastatic lesions have been reported to have the T790M EGFR mutation (16), similar to primary cancer (8). The present study is markedly distinct from previously reported cases in that the tumor was resected and evaluated prior to the lesion demonstrating clinical resistance to EGFR-TKI. The results of the present study demonstrate that viable cells, although possessing reduced proliferative capability, are present while the entire lesion responds to EGFR-TKI. Furthermore, to the best of our knowledge, the present study is the first to provide evidence that EGFR-mutated NSCLC cells are able to evade the cytotoxic effect of EGFR-TKI by downregulating EGFR expression, without exhibiting the T790M EGFR mutation.