Lung cancer is the second most common type of cancer worldwide and in the USA, with its incidence varying depending on geographic and socioeconomic factors.^1–3 According to data from the Global Cancer Observatory: Cancer Incidence, Mortality and Prevalence (GLOBACAN) database, an estimated 2.48 million new cases of lung cancer were reported in 2022.^4,5 Although lung cancer rates in the USA have been decreasing, with a notable decline in age-adjusted rates since the early 1990s for men and since 2006 for women, it continues to be a significant health concern.⁶

Non-small cell lung cancer (NSCLC) accounts for 85% of primary lung neoplasms, with many patients being diagnosed at advanced stages (44%).^4,7 For these patients, palliative systemic treatment stands out as the primary therapeutic option. With advances in molecular diagnostic techniques, targeted therapies that are more effective and less toxic than chemotherapy are available for patients with activating mutations in the epidermal growth factor receptor (EGFR) and anaplastic lymphoma kinase (ALK) translocations, for example.⁸

Brain metastasis is one of the most challenging conditions among oncologists today. Lung cancer is responsible for approximately 50% of all brain metastasis, followed by breast cancer and melanoma. Given the high rates of metastasis associated with these cancers, studies focused on early detection and effective treatments are extremely important.^9–12

The objective of this review is to describe the central nervous system (CNS) as a sanctuary for advanced EGFR-mutant NSCLC and to discuss the therapeutic options available for these patients, focusing on drugs such as osimertinib and amivantamab, which have shown promising results in treating brain metastasis.

Importance of innovative therapies

Despite the high incidence of brain metastases, there were not many treatment options available, besides radiotherapy, for treating these patients, regardless of EGFR mutation status. The reduced intracerebral availability of drugs caused by the efflux pump mechanism in the blood–brain barrier (BBB) contributes to the concept of the CNS as a sanctuary.¹³ However, radiotherapy has limited application due to its association with harmful neurocognitive effects, neuronal damage and long-term demyelination. Therefore, the study of new systemic pharmacotherapeutic options has been the focus of research by many specialists.¹⁴

Recent discoveries have highlighted new therapeutic approaches for brain metastases for NSCLC through immunotherapy, which consist of therapies targeted at oncogene-dependent tumours, such as the advent of targeting EGFR mutations. These therapies exhibit enhanced penetration into the CNS and have shown better outcomes compared with conventional chemotherapy.^13,15,16

Among the innovations in immunotherapy, third-generation EGFR inhibitors exhibit superior ability to penetrate the BBB compared with first- and second-generation inhibitors, which have limited CNS penetration and, consequently, low efficacy in the treatment and progression of brain metastases. As a result, these earlier-generation inhibitors were considered a barrier to effective treatment.¹⁵ However, recent studies highlight the potential of third-generation tyrosine kinase inhibitors (TKIs), advancing the concept of the CNS as an ‘untouchable sanctuary’ for therapeutic targets in the treatment of NSCLC with brain metastasis. What was once a major challenge is now an opportunity for therapeutic success.^13,15

Pathophysiology and therapeutic challenges of the blood–brain barrier

Metastatic invasion of the CNS involves a multistep adaptive process where tumour cells break the BBB. Tumour cells with CNS tropism exhibit molecular markers associated with epithelial–mesenchymal transition (EMT), such as the upregulation of N-cadherin and vimentin, alongside the loss of epithelial markers.^17,18 Overexpression of the EGFR/ERK pathway is strongly correlated with brain metastases in lung cancer, with metastatic lesions demonstrating higher pathway activity compared with primary tumours.¹⁹

The incidence of brain metastasis varies based on the molecular subtype: patients with EGFR mutations, ALK rearrangements or ROS1 fusions exhibit CNS involvement in 50–60% of cases at diagnosis.²⁰ However, ROS1-positive NSCLC appears to have lower rates of brain metastases compared with ALK-rearrangement tumours, though the underlying mechanisms remain unclear.^14,16,21,22 Clinically, the presence of brain metastases in EGFR-mutant NSCLC is associated with shorter survival, even among patients treated with first- and second-generation EGFR TKIs, underscoring the BBB’s role in limiting therapeutic efficacy.^23,24

The BBB restricts drug penetration through both passive mechanisms (e.g. tight junctions between endothelial cells) and active efflux transporters (e.g. P-glycoprotein). Early efforts to enhance drug delivery, such as osmotic disruption or magnetic resonance imaging (MRI)-guided focused ultrasound, aim to transiently open the BBB but face challenges related to inconsistent efficacy and potential neurotoxicity.^25,26 Paradoxically, while the BBB excludes most therapeutics, tumour cells bypass it through mechanisms involving EMT-driven plasticity, interactions with the perivascular niche and non-coding RNA-mediated regulation of metastases pathways.^16,27,28 These adaptions allow tumour cells to colonize the CNS while evading systemic treatments, creating a ‘sanctuary site’ where residual disease persists at subtherapeutic drug concentrations.^21,22

Recent advances in targeted therapies, such as third-generation EGFR inhibitors (osimertinib) and bispecific antibodies (amivantamab), have improved CNS penetration, making this systemic therapy the preferred choice over pre-existing conventional therapies.^21,22

Osimertinib in central nervous system metastases

Osimertinib is a third-generation, irreversible EGFR inhibitor that targets both sensitizing EGFR mutations and the T790M resistance mutation. It acts by covalently binding to the cysteine residue at position 797 within the adenosine triphosphate-binding site of mutant EGFR, thereby inhibiting tumor cell proliferation and promoting cell death.²⁹ Preclinical studies highlight osimertinib’s superior BBB penetration comparted with earlier EGFR TKIs (gefinib, erlotinib and afatinib), attributed to reduced efflux by BBB transporters and a higher brain:plasma concentration ratio.^29–31 This enhanced CNS penetration provides a strong rationale for its efficacy in treating brain metastases.

AURA3 trial

The phase III AURA3 trial (A Phase III, Open Label, Randomized Study of AZD9291 Versus Platinum-Based Doublet Chemotherapy for Patients With Locally Advanced or Metastatic Non-Small Cell Lung Cancer Whose Disease Has Progressed With Previous Epidermal Growth Factor Receptor Tyrosine Kinase Inhibitor Therapy and Whose Tumours Harbour a T790M Mutation Within the Epidermal Growth Factor Receptor Gene [AURA3]; ClinicalTrials.gov identifier: NCT02151981) compared osimertinib with platinum-pemetrexed chemotherapy in 419 patients with T790M-positive NSCLC who progressed on prior EGFR-TKI therapy. Osimertinib demonstrated a median progression-free survival (PFS) of 10.1 versus 4.4 months with chemotherapy (HR 0.30; p<0.0001) and a higher CNS objective response rate (70 versus 31%). Notably, osimertinib showed a superior control of CNS metastases, with longer CNS PFS and a manageable safety profile. Although overall survival (OS) did not differ significantly (26.8 versus 22.5 months; HR 0.87; p<0.277), likely due to the crossover from chemotherapy to osimertinib, these results solidified osimertinib as the standard of care for T790M-positive NSCLC with CNS involvement.^32,33

FLAURA trial

The phase III FLAURA trial (A Phase III, Double-blind, Randomised Study to Assess the Safety and Efficacy of AZD9291 Versus a Standard of Care Epidermal Growth Factor Receptor Tyrosine Kinase Inhibitor as First Line Treatment in Patients With Epidermal Growth Factor Receptor Mutation Positive, Locally Advanced or Metastatic Non Small Cell Lung Cancer; ClinicalTrials.gov identifier: NCT02296125) established osimertinib as a first-line therapy for EGFR-mutant NSCLC. In 556 patients with exon 19 deletions or L858R mutations, osimertinib significantly improved median PFS (18.9 versus 10.2 months; HR 0.46; p<0.001) and OS (38.6 versus 31.8 months; HR 0.80; p=0.046) compared with first-line generation TKIs (gefitinib/erlotinib). Patients with CNS metastases particularly benefited, with a higher CNS objective response rate (ORR) and prolonged CNS PFS. Osimertinib also had a favourable safety profile, with fewer grade ≥3 adverse events (42 versus 45%).^34,35

FLAURA2 trial

The FLAURA2 trial (A Phase III, Open-label, Randomized Study of Osimertinib With or Without Platinum Plus Pemetrexed Chemo, as First-line Treatment in Patients With Epidermal Growth Factor Receptor (EGFR) Mutation Positive, Locally Advanced or Metastatic Non-small Cell Lung Cancer [FLAURA2]; ClinicalTrials.gov identifier: NCT04035486) evaluated osimertinib plus chemotherapy (pemetrexed/platinum) versus osimertinib monotherapy in 557 treatment-naïve patients with EGFR-mutant NSCLC.³⁶ The combination group achieved a median PFS of 25.5 versus 16.7 months with monotherapy (HR 0.62; p=0.001) and a longer duration of response (24 versus 15.3 months). However, the combination group experienced higher rates of grade ≥3 adverse events (e.g. haematologic toxicities), underscoring the need to balance efficacy with tolerability.³⁶

APPLE trial

The phase II APPLE trial (APPLE Trial: Feasibility and Activity of AZD9291 [Osimertinib] Treatment on Positive PLasma T790M in EGFR Mutant NSCLC Patients; ClinicalTrials.gov identifier: NCT02856893) compared upfront osimertinib with sequential therapy (gefitinib followed by osimertinib upon T790M detection or progression) in 156 EGFR-mutant patients with NSCLC. The upfront osimertinib arm showed an 18-month PFS rate of 67.2% (median PFS 22.0 months) versus 20.2 in the sequential arm. Notably, the sequential approach was associated with a higher incidence of CNS metastases (53.8 versus 11.5%), likely due to delayed CNS protection. Despite this, OS was similar between groups, suggesting that both strategies are viable but highlighting the importance of early CNS control with third-generation TKIs.³⁷

Osimertinib’s superior CNS penetration and efficacy in trials such as AURA3, FLAURA and FLAURA2 solidify its role as a central therapy for EGFR-mutant NSCLC with CNS metastases. While combination therapies (e.g. FLAURA2) may enhance efficacy, toxicity remains a concern. The APPLE trial reveals the balance in sequencing strategies: upfront osimertinib reduces CNS progression risk, whereas sequential approaches achieve comparable OS but with higher CNS relapse rates. These findings emphasize personalized treatment based on patient priorities, including CNS protection and tolerability.

Amivantamab in central nervous system metastases

EGFR inhibitors are special drugs when treating advanced EGFR-mutant NSCLC. Most of the EGFR inhibitors act on exon 19 deletion and exon 21 deletion, which accounts for 85% of all EGFR subtype mutations. However, 10–15% of EGFR mutations are due to exon 20 insertion mutation, and these drugs have not been successfully targeted until now.^38–40 In 2021, the FDA approved amivantamab, a drug that acts on the exon 20 insertion mutation.⁴¹

Amivantamab is a human Fc-active immunoglobulin G1 that acts as an antibody against epidermal growth factor (EGF) and mesenchymal-epithelial transition factor (MET) receptors. Two arms are involved: the extracellular arm inhibits the ligation of EGFR to its ligand EGF, and the intracellular arm inhibits the hepatocyte growth factor ligand from binding with the MET receptor. By targeting both receptors, amivantamab provides a novel therapeutic approach for overcoming resistant mechanisms in these cancer types. As a result, there is downstream signalling of pro-growth and pro-survival proteins.^42–48

MARIPOSA-2 trial

The MARIPOSA-2 trial (A Phase 3, Open-Label, Randomized Study of Amivantamab and Lazertinib in Combination With Platinum-Based Chemotherapy Compared With Platinum-Based Chemotherapy in Patients With EGFR-Mutated Locally Advanced or Metastatic Non-Small Cell Lung Cancer After Osimertinib Failure; ClinicalTrials.gov identifier: NCT04988295) evaluated amivantamab combined with chemotherapy, with or without the third-generation EGFR TKI lazertinib, in 657 patients with EGFR-mutant NSCLC (exon 19/L858R) who progressed on osimertinib. Compared with chemotherapy alone, amivantamab–chemotherapy doubled the median PFS from 4.2 to 6.3 months (HR 0.48; p<0.001). The triplet regimen (amivantamab + lazertinib + chemotherapy) extended the median PFS further to 8.3 months (HR 0.44). Notably, amivantamab-containing regimens demonstrated superior control of CNS metastases, with median intracranial PFS reaching 12.5–12.8 months versus 8.3 months for chemotherapy alone. ORR also favoured amivantamab combinations (63–64 versus 36%). However, the addition of lazertinib increased haematologic toxicities, such as neutropenia, leading to protocol adjustments, including delayed lazertinib initiation.⁴⁹

MARIPOSA trial

For untreated EGFR-mutant NSCLC, the MARIPOSA trial (A Phase 3, Randomized Study of Amivantamab and Lazertinib Combination Therapy Versus Osimertinib Versus Lazertinib as First-Line Treatment in Patients With EGFR-Mutated Locally Advanced or Metastatic Non-Small Cell Lung Cancer; ClinicalTrials.gov identifier: NCT04487080) compared amivantamab–lazertinib combination therapy with osimertinib monotherapy in 1,074 patients, including those with CNS involvement. The combination significantly prolonged median PFS to 23.7 versus 16.6 months with osimertinib (HR 0.70), establishing its potential as a first-line option. At the American Society of Clinical Oncology (ASCO) 2024 meeting, Dr Helena Yu presented subgroup data highlighting the regimen’s efficacy in patients with CNS metastases. The combination delayed intracranial progression and showed a trend towards prolonged survival in this high-risk population, reinforcing its role as a future standard for EGFR-mutant NSCLC with brain involvement. These findings were particularly impactful for patients with baseline CNS metastases, where the combination demonstrated robust intracranial activity.^50–52

Despite these advancements, toxicity remains a critical consideration. In MARIPOSA-2, 52% of patients receiving the triplet regimen experienced grade ≥3 adverse events (e.g. cytopenias) compared with 20% with chemotherapy alone. Similarly, the amivantamab–lazertinib combination in MARIPOSA led to higher rates of EGFR- and MET-related toxicities, such as rash and hypoalbuminemia, though these were generally manageable.^49,50 Table 1 summarizes the six studies cited above.^{32–37,49–51}

Adverse effects and safety profiles

Both osimertinib and amivantamab are important options for treating NSCLC with specific EGFR mutations, but they come with different side effect profiles.

For osimertinib, common issues include diarrhoea and skin rash, affecting about half of the patients. Paronychia (inflammation of the nail folds) is less frequent but still occurs in roughly a quarter of cases. Some people also experience dry skin, itching or mouth sores, though these are generally manageable.^{34,36,53–55}

Amivantamab, on the other hand, is more likely to cause infusion-related reactions, which happen in over half of patients, with severe cases being less common. Skin rashes are another frequent complaint, ranging from mild to severe in some cases. Nail infections are also reported more often, and other side effects such as muscle pain, nausea, fatigue and swelling can make the treatment challenging at times.^50,51,56,57

The overlapping adverse effects of these drugs include rash and diarrhoea, although amivantamab is more related to infusion reactions; on the other hand, osimertinib is associated with gastrointestinal issues. While both drugs have their challenges, close monitoring and supportive care can make a big difference in helping patients manage these side effects.

The most common adverse effects of both drugs (compared with placebo) are summarized in Table 2.^{32–37,49–51}

Future directions and research gaps

Amivantamab, a bispecific antibody targeting ECFR and MET, and osimertinib, a TKI, have shown promising evidence in treating CNS metastasis in NSCLC. However, much remains to be understood, particularly regarding resistance mechanisms in the treatment of CNS metastasis. The current strategies focus on combination therapies, such as amivantamab in combination with lazertinib, another EGFR TKI, as demonstrated in the MARIPOSA trial. The combination approach seeks to personalize treatment through biomarker-driven strategies and optimize treatment sequencing, leading to promising results.⁵⁰

While the optimal strategy for treating CNS metastasis remains uncertain, the current guidelines still recommend osimertinib to treat EGFR-mutant NSCLS with CNS involvement. Additionally, combination therapies, including the usage of amivantamab, must be explored further to address existing gaps. One notable clinical trial gap is the comparison between amivantamab plus chemotherapy and osimertinib as the first-line treatment in patients with EGFR-positive NSCLC.

Discussion

The management of brain metastases in patients with advanced EFGR-mutant NSCLC remains a significant challenge due to the inherent limitations of traditional treatment options, particularly chemotherapy and radiotherapy. The BBB significantly restricts the intracranial penetration of many systemic therapies, rendering them ineffective in treating CNS metastases. This phenomenon contributes to the CNS being considered a ‘sanctuary’ for cancer cells, making brain metastasis particularly difficult to treat. However, recent advances in targeted therapies and immunotherapy have provided new hope in overcoming this barrier.

Third-generation EGFR inhibitors, such as osimertinib, have shown superior penetration into the BBB compared with first- and second-generation inhibitors, leading to improved clinical outcomes in patients with brain metastases from EGFR-mutant NSCLC. The ability of osimertinib to cross the BBB effectively has made it a preferred option for patients with CNS involvement. Clinical trials, such as the AUDORA trial, have highlighted the efficacy of osimertinib in improving PFS and OS, particularly in the adjuvant setting. This represents a significant step forward in the management of brain metastases in EGFR-mutant NSCLC.

Similarly, amivantamab, an antibody targeting both EGFR and MET, has emerged as another promising therapeutic option for patients with EGFR-mutant NSCLC, especially those with brain metastases. In combination with other therapies, such as lazertinib, amivantamab has demonstrated the potential to overcome resistance mechanisms and provide valuable insights into the clinical benefit of amivantamab in NSCLC with brain metastases, further emphasizing the importance of personalized, biomarker-driven therapeutic strategies.

Despite these advances, there remain critical gaps in the treatment of brain metastases in NSCLC. Resistance to targeted therapies and the lack of robust clinical data comparing combination treatments highlight the need for further research in this area.

Conclusion

The CNS has long been considered a sanctuary for cancer cells, especially in advanced EGFR-mutant NSCLC. However, recent advancements in targeted therapies, especially with the use of combination treatments such as amivantamab and osimertinib, offer promising potential to improve outcomes for patients with advanced EGFR-mutant NSCLC and brain metastases. Despite these advancements, there is still a critical need for further clinical trials to evaluate the efficacy of these therapies across diverse patient populations and various stages of disease. Ongoing research and the development of new therapeutic regimens will be essential in closing the existing gaps in treatment strategies and enhancing our understanding of how best to manage brain metastases in NSCLC. The rapidly evolving landscape of CNS-targeted therapies holds significant promise, potentially transforming the CNS from a once impenetrable sanctuary into a site of successful therapeutic intervention.