Optimizing upfront combination therapy in advanced epidermal growth factor receptor mutation (EGFR)-mutated non-small cell lung cancer

Introduction

Epidermal growth factor receptor mutation (EGFR) tyrosine kinase inhibitors (TKIs) have revolutionized the treatment of EGFR-mutated advanced non-small cell lung cancer (NSCLC), becoming the first-line standard of care. However, acquired resistance inevitably develops, even with third-generation TKIs such as osimertinib, prompting extensive exploration of upfront combined strategies to improve long-term survival. Recent phase III studies of these upfront combination regimens have yielded conflicting survival outcomes, creating uncertainty about the optimal first-line paradigm for this patient population. In this editorial, we focus on a head-to-head comparison of four landmark clinical trials that define the current first-line treatment landscape for EGFR-mutated advanced NSCLC: Kanda et al. (JCOG1404/WJOG8214L), FLAURA2 (NCT04035486), MARIPOSA (NCT04487080), and NorthStar (NCT05870319) (1-4).

Kanda et al.’s study, published in Clinical Cancer Research in 2025, enrolled 501 treatment-naive patients with advanced non-squamous NSCLC harboring EGFR exon 19 deletion or exon 21 L858R mutation. Patients were randomized to EGFR-TKI monotherapy or TKI plus intercalated cisplatin-pemetrexed chemotherapy, with a key design feature being the switch from gefitinib to osimertinib in October 2018, resulting in a mixed-cohort population. The study reported a significant progression-free survival (PFS) benefit with intercalated chemotherapy but no improvement in overall survival (OS). Notably, patients with stable central nervous system (CNS) metastases derived no significant benefit from intercalated chemotherapy, a finding in stark contrast to FLAURA2 (1).

In contrast, the FLAURA2 study, published in the New England Journal of Medicine in 2026, enrolled 557 patients with EGFR-mutant advanced NSCLC and designated PFS as its primary endpoint, randomly assigning patients to osimertinib monotherapy or osimertinib plus concurrent platinum-pemetrexed chemotherapy with pemetrexed maintenance. This study achieved a breakthrough in its primary PFS endpoint and also saw notable improvement in OS with mature follow-up data. A preplanned CNS-specific analysis further demonstrated that the combination therapy reduced the risk of CNS progression and improved the depth of CNS response. Notably, the FLAURA2 trial did not include pre-specified mechanistic analyses to explain the biological basis of this CNS benefit; the observed intracranial efficacy may be partially explained by the well-documented superior blood-brain barrier (BBB) penetration of osimertinib from independent preclinical and clinical studies, as well as the potential for upfront concurrent chemotherapy to eradicate intracranial micrometastases before the emergence of treatment-resistant clones. As a distinct, fully chemotherapy-free intensified first-line paradigm, the MARIPOSA study, another key trial evaluating intensified first-line therapy, also designated PFS as its primary endpoint, comparing amivantamab plus lazertinib a CNS-penetrant third-generation TKI to osimertinib or lazertinib alone, reporting a prolonged median PFS in the dual-targeting arm versus osimertinib monotherapy, with consistent benefits across patients with and without brain metastases though CNS PFS data remain unreported (1,3).

The divergent outcomes among these studies stem from four core differences: TKI generational selection, chemotherapy timing, study endpoint design, and efficacy in patients with CNS metastases. In this editorial, we aim to dissect the key design and mechanistic drivers of the conflicting survival results across these four trials, define the optimal upfront combination paradigm for EGFR-mutated advanced NSCLC, and provide evidence-based guidance for clinical decision-making and future clinical trial design in this field.

Comparison between studies

Kanda et al.’s study is distinguished by its mixed TKI cohort design, with the backbone TKI switching from gefitinib to osimertinib midway through enrollment. The chemotherapy regimen was administered in an intercalated fashion: 8 weeks of TKI treatment, followed by a 2-week drug-free washout period, then 3 cycles of cisplatin-pemetrexed, with no subsequent maintenance therapy. The study’s primary endpoint was PFS, which was significantly prolonged in the combination arm, but this benefit did not translate to OS. Notably, subgroup analyses stratified by the backbone TKI revealed a disparity in survival outcomes between the gefitinib and osimertinib cohorts. For the gefitinib cohort (n=389), combination with intercalated chemotherapy failed to yield any statistically significant improvement in OS compared with gefitinib monotherapy [median OS: 35.2 vs. 34.8 months; hazard ratio (HR) =0.97, 95% confidence interval (CI): 0.79–1.19, P=0.78]. This null OS benefit was consistent across all prespecified subgroups, including patients with EGFR 19del or L858R mutation, and those with or without stable CNS metastases. In contrast, the osimertinib cohort (n=112) exhibited a trend toward improved PFS (median PFS: 18.9 vs. 15.4 months; HR =0.67, P=0.046) and a numerical OS benefit that did not reach statistical significance due to limited sample size (median OS: 48.6 vs. 42.3 months; HR =0.72, P=0.13), with no meaningful benefit in patients with stable CNS metastases. The potential survival benefit of the osimertinib cohort was diluted by the large gefitinib cohort, resulting in an OS-neutral outcome for the entire study population. Safety-wise, the combination arm had a grade ≥3 adverse event (AE) rate of 65.6–70.2%, primarily driven by neutropenia and nausea, with no fatal AEs attributed to the combination (2).

FLAURA2 adopted a fixed third-generation TKI design, with all 557 enrolled patients receiving osimertinib and the study setting PFS as its primary endpoint. Chemotherapy was administered concurrently from treatment initiation: 4 cycles of platinum-pemetrexed followed by pemetrexed maintenance therapy. The study achieved a significant improvement in its primary PFS endpoint in the combination arm, along with a notable improvement in OS with mature follow-up data. A preplanned CNS-specific analysis highlighted notable benefits in patients with baseline CNS metastases: the combination reduced the 24-month CNS progression rate and improved the depth of CNS response. However, this significant survival benefit came with a clinically meaningful increase in toxicity burden that requires critical consideration in routine clinical practice. The combination arm reported a substantially higher rate of grade ≥3 AEs (70% vs. 34% with osimertinib monotherapy), driven predominantly by chemotherapy-related myelosuppression, including neutropenia, thrombocytopenia, and anemia. Most critically, the combination arm had a 6% rate of treatment-related fatal AEs, with the primary causes being infectious complications (pneumonia and sepsis) secondary to myelosuppression, thromboembolic events, respiratory failure, and cardiac failure. This fatal toxicity risk directly explains the early crossing of the survival curves observed in the trial, which was driven by early treatment-related deaths in vulnerable, frail patients with poor performance status or significant baseline cardiopulmonary comorbidities. Notably, this fatal AE rate is higher than that observed in the intercalated chemotherapy regimen in Kanda et al.’s study, which reported no treatment-related fatal events, highlighting the core trade-off between upfront concurrent chemotherapy’s superior long-term efficacy and its higher acute toxicity risk.

MARIPOSA evaluated a dual-targeting strategy without chemotherapy, enrolling patients with advanced EGFR-mutated NSCLC and designating PFS as its primary endpoint, which was significantly prolonged in the dual-targeting arm versus osimertinib monotherapy. Benefit was consistent across subgroups, including patients with and without brain metastases, though CNS PFS data have not yet been reported. Notably, amivantamab, a large bispecific antibody with limited BBB penetration, nevertheless showed indirect CNS activity in a phase II study of pretreated patients with active CNS disease, with intracranial responses observed in 35% of those with parenchymal brain metastases. Safety-wise, the dual-targeting arm had a grade ≥3 AE rate, with infusion-related reactions being the most common treatment-related AEs (3).

Kanda’s study’s “PFS-OS disconnect” is largely attributed to several factors, such as the mixed TKI cohort, first-generation TKI-chemotherapy combinations being less efficacious, delayed chemotherapy allowing resistant clone expansion, and a lack of maintenance chemotherapy. In contrast, FLAURA2’s success in its primary PFS endpoint and subsequent OS benefit stems from its "fixed third-generation TKI + concurrent chemotherapy + maintenance" model, which minimizes TKI-chemotherapy antagonism, eradicates early resistant clones including CNS micro-metastases, and sustains tumor suppression. MARIPOSA’s dual-targeting approach also achieved its primary PFS endpoint, offering an alternative for patients ineligible for chemotherapy, leveraging complementary mechanisms to delay resistance without chemotherapy-related toxicity (2) (Table 1).

TKI generation difference as a watershed for combination therapy efficacy

NEJ009 and Kanda’s gefitinib cohort consistently demonstrate that first-generation TKI-chemotherapy combinations fail to improve OS (5). Kanda’s study further confirms that the negative outcomes of the gefitinib cohort potentially diluted the potential benefit of the smaller osimertinib cohort, resulting in an overall OS-neutral result even with a PFS benefit. In contrast, FLAURA2’s exclusive use of osimertinib avoided generational mixing, allowing the true efficacy of combined therapy to emerge in both its primary PFS endpoint and subsequent OS data. Osimertinib also has superior BBB penetration compared to first-generation TKIs, laying the foundation for benefits in CNS metastasis patients (1).

A notable recent addition to third-generation TKI-based combination strategies is the NorthStar study, a phase II randomized trial presented as an abstract at the 2025 European Society for Medical Oncology (ESMO) Congress (NCT05870319) (4). This study evaluated osimertinib with or without local consolidative therapy (LCT) in EGFR-mutated metastatic NSCLC patients, including TKI-naive individuals and those with acquired T790M without prior third-generation TKI exposure. After 6–12 weeks of osimertinib induction therapy, non-progressing patients were randomized to continue osimertinib alone or receive LCT (radiation in 59%, surgery in 29%, or combined modalities in 12%) plus osimertinib. Despite most patients having polymetastatic disease (>3 lesions in 57–59% of both arms), the osimertinib + LCT arm achieved a significantly prolonged median PFS (25.4 vs. 17.0 months; HR =0.60, 95% CI: 0.40–0.92, P=0.02). The combination was feasible and safe, with grade ≥3 AEs occurring in 29% of the LCT arm (vs. 16% in osimertinib alone). While mature OS data and CNS-specific efficacy remain pending, the NorthStar study highlights that third-generation TKI can serve as a robust backbone for non-chemotherapy intensification strategies (4).

Dual-targeting strategies represent another effective approach, leveraging complementary mechanisms to target EGFR-driven tumors without chemotherapy and achieving significant PFS improvement as seen in MARIPOSA. While amivantamab has limited BBB penetration, its combination with a CNS-penetrant TKI has shown indirect CNS activity, offering a chemotherapy-sparing option for vulnerable patients. Clinically, third-generation TKI may be a prerequisite for effective combined therapy. Consequently, first-generation TKI-chemotherapy combinations should be avoided, even in resource-limited settings, as they offer no OS benefit but with increased toxicity. For patients with limited access to osimertinib, participation in clinical trials of third-generation TKIs may be preferred over first-generation TKI-chemotherapy combinations.

Notably, this treatment paradigm shift toward third-generation EGFR-TKI-based first-line therapy is not only supported by landmark randomized controlled trial (RCT) data, but also mirrored in real-world clinical practice. A recent nationwide analysis of real-world treatment patterns in the United States (6) demonstrated a rapid, widespread shift toward third-generation EGFR-TKIs as the first-line standard of care for EGFR-mutated advanced NSCLC in routine clinical practice, with further insights into the real-world uptake of intensified upfront combination regimens, treatment outcomes, and associated economic impact. This real-world evidence confirms the generalizability of the RCT findings highlighted in this editorial, and underscores the urgent clinical need to define the optimal upfront intensified combination paradigm based on third-generation EGFR-TKIs.

Study endpoint design

PFS has become a widely accepted primary endpoint in advanced NSCLC clinical trials, as it directly reflects disease control and early therapeutic efficacy, and is less susceptible to confounding factors such as subsequent therapies compared to OS. Kanda’s study set OS as its primary endpoint and achieved a significant PFS improvement, but this benefit failed to translate to OS due to the high proportion of patients receiving potent subsequent therapies that may have compensated for the first-line treatment difference. Furthermore, intercalated chemotherapy may only “delays” rather than “reverses” resistance, with PFS benefits representing temporary disease control rather than altered disease natural history. This indicates that PFS benefits must be contextualized with OS outcomes and patient-reported outcomes (PROs) to determine long-term clinical value.

FLAURA2 and MARIPOSA both designated PFS as their primary endpoint, a rational choice for evaluating intensified first-line therapies in EGFR-mutated NSCLC, as it allows for the early assessment of therapeutic efficacy and the rapid translation of positive results into clinical practice. FLAURA2’s mature follow-up data further demonstrated a significant OS improvement, confirming that the PFS benefit from combination therapy translated into a meaningful survival advantage and fundamentally altered the natural history of EGFR-mutated NSCLC by eradicating early resistant clones. This result has led to the inclusion of osimertinib plus concurrent platinum-pemetrexed chemotherapy as a preferred first-line option in the 2025 National Comprehensive Cancer Network (NCCN) Guidelines for NSCLC. This recommendation is fully aligned with the 2025 ESMO Clinical Practice Guidelines, which also list this combination regimen as a preferred first-line option for medically fit patients; in contrast, the 2024 American Society of Clinical Oncology (ASCO) Living Guideline provides a conditional recommendation for this combination, with additional emphasis on patient fitness stratification and shared decision-making to balance the survival benefit against the increased risk of treatment-related toxicity. This finding highlights that a PFS benefit supported by subsequent mature OS improvement could be considered a gold standard for evaluating definitive first-line treatment strategies for EGFR-mutated advanced NSCLC (7,8). MARIPOSA’s focus on PFS was also appropriate for evaluating a chemotherapy-sparing dual-targeting strategy, and mature OS data are needed to confirm its long-term clinical value (3). For patients and clinicians, the choice of PFS as a primary endpoint underscores the importance of early disease control, while OS remains the definitive measure of long-term therapeutic success.

Chemotherapy timing and optimization for individualized administration

Kanda’s delayed intercalation strategy has inherent drawbacks that undermine the sustainability of its PFS benefit and prevent translation to OS. Osimertinib’s maximum tumor regression may require more than 8 weeks, and the 2-week washout period may “release” EGFR pathway-dependent tumors, leading to potential proliferation and reduced chemotherapy efficacy. Kanda’s study showed no PFS benefit in the osimertinib + 19del subgroup, suggesting this risk. For CNS metastases, delayed chemotherapy may allow micro-metastases to develop resistance, reducing the efficacy of subsequent chemotherapy due to limited BBB penetration, another possible reason for the lack of CNS benefit in Kanda’s study.

FLAURA2’s initial concurrent chemotherapy addresses these limitations by eradicating TKI-insensitive clones early, before resistance can develop, which may be a core reason for its significant and sustained PFS benefit and subsequent OS improvement. Preclinical studies suggest that concurrent administration of osimertinib and chemotherapy targets both proliferating and quiescent tumor cells, reducing drug-tolerant persister (DTP) cell expansion and preventing the early emergence of resistance (9). The “induction + maintenance” mode further sustains efficacy, especially benefiting high-risk patients and maintaining long-term disease control. However, concurrent chemotherapy increases toxicity, and may also cause early treatment interruption in vulnerable patients and leading to initial survival curve crossing, indicating that “one-size-fits-all” concurrent therapy is not suitable for all patients.

Circulating tumor DNA (ctDNA) dynamic monitoring may resolve this dilemma (10). For patients with rapid ctDNA clearance, chemotherapy can be delayed avoiding early toxicity; for those with persistent ctDNA positivity or slow decline, concurrent chemotherapy should be initiated promptly to block resistant clones. This strategy, supported by recent MARIPOSA data showing reduced resistance alterations with early dual-targeting, balances efficacy and safety, representing a key future research direction for optimizing chemotherapy timing and maximizing the clinical benefits (11). Notably, this ctDNA-guided strategy remains hypothesis-generating and has not yet been validated in prospective phase III trials, thus it is not yet practice-changing in routine clinical care

Future research directions

Future research could focus on ctDNA-guided individualized timing trials, conducting phase III studies comparing different chemotherapy timing strategies to validate the optimal timing for different ctDNA dynamic subgroups, particularly in CNS metastasis patients, with the goal of maximizing PFS benefits and translating them into OS improvement while minimizing toxicity. CNS-targeted combination strategies warrant exploration, such as osimertinib + chemotherapy + BBB-penetrating agents or intrathecal therapy for leptomeningeal disease patients, building on FLAURA2’s CNS success to further improve intracranial disease control and contribute to long-term survival. Mechanism-driven combination optimization is needed, investigating combinations targeting DTP cells to prevent resistance expansion during TKI washout, which is a key factor in the failure of delayed intercalation strategies to translate PFS benefits to OS. Long-term follow-up of trials is necessary to assess late toxicities and their impact on quality of life, and to further confirm the sustainability of OS benefits from intensified first-line therapy. Additionally, more studies are needed to explore the optimal duration of maintenance chemotherapy in the FLAURA2 regimen, balancing efficacy and the practical inconvenience of long-term infusions for patients.

Conclusions

The contrasting results of Kanda’s study and FLAURA2, contextualized with recent CNS-specific and mechanistic evidence, clarify the core principles of combined therapy for EGFR-mutated NSCLC. Notably, this includes that third-generation TKI is the foundation for efficacy, PFS as a primary endpoint with concurrent chemotherapy initiated early from treatment onset achieves significant and sustained disease control and can translate to meaningful OS improvement, and chemotherapy timing and patient stratification directly impact both PFS and OS outcomes especially in CNS metastasis patients. Future research should focus on ctDNA-guided individualized timing, CNS-targeted strategies, and mechanism-driven combinations to advance precision medicine, maximize PFS benefits, and ensure their translation to long-term OS improvement for patients with EGFR-mutated NSCLC.

Acknowledgments

None.

Footnote

Provenance and Peer Review: This article was commissioned by the editorial office, Shanghai Chest. The article has undergone external peer review.

Peer Review File: Available at https://shc.amegroups.com/article/view/10.21037/shc-2026-1-0012/prf

Funding: None.

Conflicts of Interest: Both authors have completed the ICMJE uniform disclosure form (available at https://shc.amegroups.com/article/view/10.21037/shc-2026-1-0012/coif). A.C.T. declares that he has served in the consulting or advisory roles for Amgen, AstraZeneca, Bayer, Pfizer; all outside the submitted work. The other author has no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Jänne PA, Planchard D, Kobayashi K, et al. Survival with Osimertinib plus Chemotherapy in EGFR-Mutated Advanced NSCLC. N Engl J Med 2026;394:27-38. [Crossref] [PubMed]
2. Kanda S, Niho S, Kurata T, et al. Randomized Phase III Study of EGFR Tyrosine Kinase Inhibitor and Intercalated Platinum-Doublet Chemotherapy for Non-Small Cell Lung Cancer Harboring EGFR Mutation. Clin Cancer Res 2025;31:2317-26. [Crossref] [PubMed]
3. Yang JC, Lu S, Hayashi H, et al. Overall Survival with Amivantamab-Lazertinib in EGFR-Mutated Advanced NSCLC. N Engl J Med 2025;393:1681-93. [Crossref] [PubMed]
4. Elamin YY, Gandhi S, Antonoff M, et al. LBA72 NorthStar: A phase II randomized study of osimertinib (OSI) with or without local consolidative therapy (LCT) for metastatic EGFR-mutant non-small cell lung cancer (NSCLC). Ann Oncol 2025;36:S1614.
5. Miyauchi E, Morita S, Nakamura A, et al. Updated Analysis of NEJ009: Gefitinib-Alone Versus Gefitinib Plus Chemotherapy for Non-Small-Cell Lung Cancer With Mutated EGFR. J Clin Oncol 2022;40:3587-92. [Crossref] [PubMed]
6. Gourzoulidis G, Kastanioti C, Mavridoglou G, et al. Real-World Prevalence, Treatment Patterns, and Economic Impact of EGFR- and ALK-Targeted Therapies in Non-Small Cell Lung Cancer: A Nationwide Analysis from Greece. Curr Oncol 2025;32:542. [Crossref] [PubMed]
7. Owen DH, Halmos B, Puri S, et al. Therapy for Stage IV Non-Small Cell Lung Cancer Without Driver Alterations: ASCO Living Guideline, Version 2025.1. J Clin Oncol 2025;43:e45-e58. [Crossref] [PubMed]
8. Hendriks LE, Kerr KM, Menis J, et al. Oncogene-addicted metastatic non-small-cell lung cancer: ESMO Clinical Practice Guideline for diagnosis, treatment and follow-up. Ann Oncol 2023;34:339-57. [Crossref] [PubMed]
9. Criscione SW, Martin MJ, Oien DB, et al. The landscape of therapeutic vulnerabilities in EGFR inhibitor osimertinib drug tolerant persister cells. NPJ Precis Oncol 2022;6:95. [Crossref] [PubMed]
10. Bordi P, Del Re M, Minari R, et al. From the beginning to resistance: Study of plasma monitoring and resistance mechanisms in a cohort of patients treated with osimertinib for advanced T790M-positive NSCLC. Lung Cancer 2019;131:78-85. [Crossref] [PubMed]
11. Felip E, Cho BC, Gutiérrez V, et al. Amivantamab plus lazertinib versus osimertinib in first-line EGFR-mutant advanced non-small-cell lung cancer with biomarkers of high-risk disease: a secondary analysis from MARIPOSA. Ann Oncol 2024;35:805-16. [Crossref] [PubMed]