Central nervous system (CNS) metastases have become increasingly prevalent in patients with EGFR-mutant lung cancer and continue to be a major driver of morbidity and mortality. A 2025 study reported a cumulative incidence of brain metastases (BrM) and leptomeningeal disease (LMD) of 54% and 12% at 5 years, respectively.1
Despite this high burden, patients with active CNS disease, particularly LMD, are frequently excluded from clinical trials, resulting in a significant evidence gap and limited therapeutic options. Mark Y. Jeng, MD, PhD, and colleagues aimed to address this need in a phase II study evaluating the efficacy of amivantamab in combination with lazertinib in previously treated patients with EGFR-mutant non-small cell lung cancer (NSCLC) and active CNS metastases.2
Dr. Jeng, Assistant Attending in the Thoracic Oncology Service at Memorial Sloan Kettering Cancer Center, New York, recently discussed the trial in an interview with ILCN, highlighting key takeaways and clinical implications.
ILCN: Can you start by introducing the “why” of the study? Specifically, why did the team evaluate amivantamab and lazertinib in this particular patient population?
Dr. Jeng: This trial was motivated by both an unmet clinical need and a scientific rationale. Amivantamab in combination with lazertinib offered a compelling strategy to address both canonical EGFR signaling and key resistance pathways, such as MET amplification.
At the time of study initiation, this combination had not yet been approved, and its CNS activity—particularly in patients with progressive disease—remained largely undefined. We aimed to prospectively evaluate its efficacy in a population that is often underrepresented yet highly clinically relevant.
ILCN: Could you briefly walk readers through the trial design and specific eligibility criteria?
Dr. Jeng: This was a phase II, single-institution study evaluating the combination of amivantamab and lazertinib in patients with EGFR-mutant NSCLC, including both common (exon 19 deletions, L858R) and uncommon EGFR alterations, as well as exon 20 insertions. Eligible patients had CNS involvement, either BrM or LMD, and had experienced disease progression, either intracranially or extracranially.
We enrolled two parallel cohorts: patients with BrM (n = 20) and those with LMD (n = 21). Key exclusion criteria included poor functional status (KPS < 60%) and uncontrolled or worsening neurologic symptoms requiring escalating corticosteroids, ensuring that patients could safely tolerate systemic therapy.
The primary endpoint was the best overall response rate (ORR), defined as a confirmed complete or partial response, using a composite assessment that incorporated both systemic and CNS disease. This was evaluated using RECIST version 1.1 alongside CNS-specific criteria: Response Assessment in Neuro-Oncology Brain Metastases (RANO-BM) for BrM and Response Assessment in Neuro-Oncology Leptomeningeal Metastases (RANO-LM) for LMD.
ILCN: What were some of the most significant takeaways, particularly regarding intracranial response?
Dr. Jeng: We observed encouraging intracranial activity, with CNS response rates of 50% in the BrM cohort and 33% in the LMD cohort. These results are particularly notable given the heavily pretreated population: over half had received two or more prior lines of therapy, and 90% had progressed on osimertinib. Survival outcomes were also clinically meaningful, with a median overall survival (OS) of 17.4 months in patients with BrM and 14.4 months in those with LMD.
A key insight from this study is the limitation of current response assessment tools for LMD. Using the standard RANO-LM criteria, no patients met the formal response criteria. However, when we examined CSF circulating tumor cells (CTCs), 67% of patients demonstrated a reduction in tumor burden. This discordance suggests that conventional metrics may underestimate treatment benefits and highlights the need for more sensitive, quantitative biomarkers in LMD, such as CTCs.
Finally, an important practical takeaway is the feasibility of conducting prospective trials in this population. Despite traditional concerns about enrolling patients with active CNS disease, we completed accrual in under 18 months. This underscores both the unmet need and the willingness of patients and clinicians to participate in trials addressing CNS metastases, including LMD.
ILCN: Given that the enrolled patients had progressed on standard-of-care therapies, do you believe this combination should be considered earlier as a preferred strategy for patients with CNS involvement or LMD, rather than waiting for disease progression?
Dr. Jeng: The treatment landscape for EGFR-mutant lung cancer has evolved significantly since this study was initiated. In the frontline setting, patients now have access to highly effective combination strategies, including amivantamab plus lazertinib and osimertinib combined with chemotherapy.
In general, there is a strong rationale for using our most effective therapies earlier in the disease course, particularly for patients at high risk of CNS progression. Both of these frontline combination regimens have demonstrated robust CNS activity; however, they have not been directly compared and differ in toxicity profiles, administration logistics, and mechanisms of action.
While our study focused on a post-progression population, the observed CNS activity of amivantamab and lazertinib supports their use as part of earlier-line strategies, especially in patients with known CNS involvement or high-risk features.
ILCN: How does this study address the unique challenges associated with LMD, such as increasing prevalence associated with improved survival rates?
Dr. Jeng: Diagnosing and monitoring LMD can be challenging. While standard assessment relies on neurologic examinations, MRI scans of the brain and spine, and CSF cytology, these tools are often insensitive and may not fully capture the disease burden or treatment response.
Our study sought to address these limitations by incorporating correlative analyses, including CSF CTCs and comprehensive genomic profiling of paired pre- and post-treatment samples. CSF CTCs, in particular, provide a quantitative and potentially more sensitive measure of disease burden compared to cytology. In our cohort, 67% of patients experienced a reduction in CSF CTC levels, compared to only 38% achieving cytologic clearance.
Interestingly, genomic analyses identified acquired resistance mechanisms in only a minority of cases (23%), suggesting that additional molecular mechanisms—such as transcriptional, epigenetic, or microenvironmental factors—may play an important role in CNS progression. These findings highlight both the complexity of LMD and the need for more comprehensive approaches to biomarker development.
ILCN: As noted in the study, despite improved CNS disease control with osimertinib, 20% of patients still develop CNS progression. How does this regimen help overcome specific resistance pathways such as this?
Dr. Jeng: One of the most common acquired resistance alterations in patients with EGFR-mutant lung cancer who progress on osimertinib is MET overexpression and/or amplification. In our study, 90% of patients had previously been treated with osimertinib, and one-third were MET-positive, defined as MET IHC 3+ in at least 25% of tumor cells.
Notably, we observed a higher response rate in MET-positive patients compared to MET-negative patients (77% vs. 26%), suggesting that dual targeting of EGFR and MET may be particularly effective in this subgroup. This is biologically plausible, as amivantamab simultaneously inhibits both pathways, potentially overcoming bypass signaling that drives resistance to EGFR-directed therapy alone.
These findings support a more biomarker-driven approach to treatment selection and suggest that MET status could possibly help guide therapeutic decisions following osimertinib progression, particularly in patients with CNS involvement.
ILCN: What are the immediate clinical implications of these findings? Looking ahead, can we anticipate any prospective phase III trials to further validate these results?
Dr. Jeng: Clinically, our findings support the use of amivantamab and lazertinib as a viable treatment option for patients with EGFR-mutant lung cancer and active CNS metastases, including the particularly challenging LMD population. Importantly, they also reinforce the need to interpret treatment responses in LMD with caution, as conventional imaging and cytology may underestimate the clinical benefit.
While, to my knowledge, there are currently no phase III trials specifically focused on this combination in patients with active CNS metastases or LMD, there is growing recognition of the importance of including these patients in prospective studies. At our institution, ongoing trials are beginning to address this gap, including a phase III study of osimertinib with or without bevacizumab (NCT04181060) and a phase II trial evaluating osimertinib with consolidative stereotactic radiosurgery (SRS) versus osimertinib alone (NCT06741085)—both of which permit patients with active or untreated CNS disease.
Looking ahead, I hope that future trials will permit broader inclusion criteria and CNS-specific endpoints to generate more representative and clinically actionable data for this high-risk population.
References:
- 1. Wilcox JA, Jeng MY, Tischfield S, Sui JSY, Nemirovsky D, Heller G, Choudhury NJ, Ross JS, Rudin CM, Riely GJ, Kris MG, Donoghue M, Boire AA, Yu HA. Identifying the genomic landscape of EGFR-mutant lung cancers with central nervous system metastases. Ann Oncol. 2025 Oct;36(10):1142-1153. doi: 10.1016/j.annonc.2025.06.001. Epub 2025 Jun 16. PMID: 40532851; PMCID: PMC12507033.
- 2. Chen MF, Lee JJ, Choudhury NJ, Hui AB, Jeng MY, Zheng J, Aly RG, Sielski N, Ahn L, Pupo A, Nesselbush MC, Jabara I, Wilcox JA, Santomasso BD, Lin AL, Schaff L, Chaft JE, Riely GJ, Kris MG, Arfe A, Yang SR, Young RJ, Diehn M, Boire A, Yu HA. Phase 2 Study of Amivantamab Plus Lazertinib in Previously Treated Patients With EGFR-Mutant Lung Cancers With Brain and Leptomeningeal Metastases. J Thorac Oncol. 2026 Mar;21(3):103505. doi: 10.1016/j.jtho.2025.10.012. Epub 2025 Oct 23. PMID: 41139066.
