Personalizing upfront treatment options for advanced EGFRm NSCLC

This article will summarize the latest developments in the first-line management of advanced EGFRm NSCLC and highlight key consideration factors.

Introduction

2024 marks the 20th anniversary of the discovery of activating epidermal growth factor receptor mutations (EGFRm) in non-small cell lung cancer (NSCLC), which paved the way for precision oncology in the management of lung cancer. Tyrosine kinase inhibitors (TKI) now form the backbone of treatment for patients with advanced NSCLC harbouring sensitizing EGFRm, which has significantly improved survival and quality of life compared to conventional chemotherapy. Yet, treatment resistance to TKI invariably develops. Recent studies have explored upfront combination strategies to improve survival outcomes. However, this occurs at the expense of increased toxicities, and the availability of multiple treatment options poses a dilemma for patients and physicians.

A brief history of EGFR TKI

The role of EGFRm as a predictive biomarker was first reported in the landmark IPASS trial, which compared gefitinib, a first-generation (1G) EGFR TKI, against platinum-doublet chemotherapy (PDC) in treatment-naïve East Asian patients with advanced lung adenocarcinoma without a significant smoking history. [N Engl J Med 2009;361:947-957] Thereafter, two other confirmatory phase III trials – WJTOG3405 and NEJ002 – were published, demonstrating that gefitinib conferred a significant progression-free survival (PFS) benefit over PDC in treatment-naïve EGFRm NSCLC and further established gefitinib as the standard of care. [Lancet Oncol 2010;11:121-128; N Engl J Med 2010;362:2380-2388] The second-generation (2G) EGFR TKIs afatinib and dacomitinib were subsequently developed, which demonstrated superiority over chemotherapy and gefitinib, respectively. [J Clin Oncol 2013;31:3327-3334; Lancet Oncol 2014;15:213-222; Lancet Oncol 2017;18:1454-1466]

FLAURA was an important practice-changing trial that established osimertinib, a third-generation (3G) EGFR TKI, as the new standard of care in the first-line setting. Osimertinib demonstrated superior intracranial efficacy, improved toxicity profile, and significantly longer PFS and overall survival (OS) benefit over 1G EGFR TKI. [N Engl J Med 2018;378:113-125; N Engl J Med 2020;382:41-50] Following the success of osimertinib, multiple other 3G TKIs (lazertinib, aumolertinib, furmonertinib, befotertinib) have been developed, but primarily in mainland China, hence limiting their use in other populations.

Upfront combinatorial strategies

Upfront treatment intensification by combining gefitinib with platinum-pemetrexed (PP) chemotherapy was evaluated in two phase III trials, which both demonstrated improved PFS with the addition of chemotherapy to gefitinib but inconsistent improvements in OS at the expense of increased toxicities. [J Clin Oncol 2020;38:115-123; J Clin Oncol 2020;38:124-136] Subsequently, the addition of anti-VEGF to first-line osimertinib was attempted in two trials, with conflicting results in PFS improvements. [J Thorac Oncol 2022;17:1098-1108; Clin Lung Cancer 2018;19:213-220.e4] In light of these findings, osimertinib was widely accepted as the standard first-line treatment for advanced EGFRm NSCLC for many years.

FLAURA2 investigated the combination of PP chemotherapy plus osimertinib compared with osimertinib alone in 557 patients with advanced nonsquamous NSCLC harbouring exon 19 deletion (Ex19del) or L858R EGFR mutations. In the combination arm, pemetrexed plus either cisplatin or carboplatin was administered for four cycles, followed by pemetrexed maintenance Q3W in combination with osimertinib until disease progression.

The median PFS (mPFS) per investigator assessment was extended by 8.8 months to 25.5 months with the addition of chemotherapy vs 16.7 months with osimertinib alone (hazard ratio [HR], 0.62, 95 percent confidence interval [CI], 0.49–0.79; p<0.001). [N Engl J Med 2023;389:1935-1948] In addition to PFS, objective response rate (ORR) was also superior in the combination arm at 83 percent vs 76 percent. At the second interim analysis of FLAURA2 with 41-percent maturity of OS events (median follow-up ≥2.5 years), a trend toward an OS benefit was observed for the osimertinib plus chemotherapy combination (HR, 0.75, 95 percent CI, 0.57–0.97; p=0.0280), but this has not reached statistical significance. [ELCC 2024, abstract 4O]

MARIPOSA was the next phase III trial that demonstrated a significant PFS benefit of upfront combinatorial therapies over osimertinib monotherapy. The combination of amivantamab, an EGFR-MET bispecific antibody, and lazertinib, another 3G EGFR TKI, was compared with osimertinib alone in 858 patients with advanced NSCLC harbouring Ex19del or L858R EGFR mutations. Additionally, the study design included a third arm of 216 patients randomized to receive lazertinib monotherapy, allowing for the assessment of the contribution of components in the amivantamab-lazertinib combination.

The primary results at a median follow-up of 22 months showed that amivantamab plus lazertinib significantly improved PFS per blinded independent central review vs osimertinib alone (median 23.7 vs 16.6 months; HR, 0.70, 95 percent CI, 0.58–0.85; p<0.001). [N Engl J Med 2024;391:1486-1498]. Importantly, a significant OS benefit of amivantamab plus lazertinib over osimertinib was reported at ELCC 2025, with median OS not reached in the combination arm vs 36.7 months in the osimertinib arm (HR, 0.75, 95 percent CI, 0.61–0.92; p<0.005). [ELCC 2025, abstract 4O] Notably, crossover was not allowed in MARIPOSA, whereas 81 percent of patients in the osimertinib monotherapy arm in FLAURA2 received platinum-based chemotherapy upon progression, which could have confounded OS results.

Weighing in on toxicities and the risk-benefit ratio

The survival gains in both FLAURA2 and MARIPOSA were at the expense of increased toxicities, resulting in higher frequencies of grade ≥3 adverse events (AEs) vs osimertinib alone. In FLAURA2, AEs were mostly chemotherapy-related (eg, myelosuppression, gastrointestinal toxicities), whereas in MARIPOSA, AEs were attributed to MET and EGFR inhibition, such as cutaneous toxicities and peripheral oedema. Other key AEs identified in MARIPOSA are venous thromboembolism, which required 4 months of anticoagulation prophylaxis, and infusion-related reactions (IRRs).

To mitigate amivantamab-related toxicities, strategies such as oral dexamethasone prophylaxis to reduce IRRs (SKIPPirr trial) and prophylactic oral antibiotics and emollients (COCOON trial) to reduce cutaneous toxicities have been proposed. [J Thorac Oncol 2025:S1556-0864(25)00051-6; ELCC 2025, abstract 10MO] Yet, these are at the expense of increased burden on patients and healthcare costs. In addition, both PP chemotherapy and amivantamab are intravenous medications that increase hospital dwell time for patients, whereas osimertinib as an oral-only therapy is logistically much easier to administer. While a subcutaneous formulation of amivantamab has been developed, it is not yet available in Singapore. [J Clin Oncol 2024;42:3593-3605]

Therefore, it is crucial to recognize that while upfront combination strategies undoubtedly improve survival outcomes, this comes at the price of increased treatment-related toxicities and healthcare costs. Patients need to be counselled appropriately on these consideration factors so they can make an informed decision with regard to a treatment choice that is in line with their goals of care.

Biomarkers to identify patients who may benefit more from upfront combinatorial approaches

Patients with central nervous system (CNS) metastases at baseline, as well as those with tumours harbouring EGFR L858R, are traditionally regarded as poorer prognostic groups, which may argue for an upfront treatment intensification approach.

In the subgroup analyses of FLAURA2, PFS was improved with osimertinib plus PP vs osimertinib alone in patients with and without CNS metastases at baseline [Table 1]. However, patients with baseline CNS metastases derived a greater magnitude of PFS benefit from the combination (HR, 0.47 vs 0.75). In MARIPOSA, the magnitude of PFS benefit from the amivantamab and lazertinib combination vs osimertinib alone was identical between patients with and without a history of brain metastases (HR, 0.69 vs 0.69).

Tumours with the L858R mutation generally do not derive as much benefit from osimertinib monotherapy as compared to Ex19del. [N Engl J Med 2018;378:113-125] Strikingly, first-line osimertinib plus PP achieved comparable PFS efficacy between tumours harbouring the Ex19del or L858R mutation (HR, 0.60 vs 0.63), whereas the efficacy of amivantamab-lazertinib vs osimertinib was more pronounced in Ex19del vs L858R.

ctDNA as a prognostic vs predictive biomarker

A prognostic biomarker relates to the natural history of the disease, irrespective of treatment. In contrast, a predictive biomarker informs the benefit of a specific treatment. Hence, the latter should be used to make treatment decisions rather than the former.

Circulating tumour DNA (ctDNA) is an easily obtainable blood test that can be used to confirm the presence of EGFRm. Clearance of plasma EGFRm after TKI initiation has also been shown to be a good prognostic biomarker, whereas patients with persistently positive ctDNA generally have inferior survival outcomes. [J Thorac Oncol 2024;19:1525-1538] Whether ctDNA (or detectable EGFRm in plasma) can be used to select patients for upfront combination therapy was evaluated as an exploratory endpoint in both FLAURA2 and MARIPOSA. [Janne, P, AACR 2024; Ann Oncol 2024;35:805-816]

Patients who were ctDNA-positive at baseline in both FLAURA2 and MARIPOSA derived a greater magnitude of benefit from combinatorial treatment as compared to patients who had undetectable ctDNA at baseline [Table 2]. On the other hand, patients with undetectable ctDNA at baseline did not derive a significant PFS benefit from either osimertinib plus chemotherapy or amivantamab plus lazertinib over osimertinib monotherapy. [AACR 2024, abstract CT017; Ann Oncol 2024;35:805-816] Taken together, this suggests baseline ctDNA positivity is a predictive biomarker that may be used to select patients for upfront combinatorial therapies.

In both FLAURA2 and MARIPOSA, the mPFS was shorter in patients who did not achieve early ctDNA EGFRm clearance, regardless of the treatment received. Meanwhile, the magnitude of PFS improvement with the combination regimens vs osimertinib alone generally remained consistent between patients with ctDNA EGFRm clearance and those without (mPFS 12.4- vs 8.5-month extension [FLAURA2] and 7.5- vs 7.4-month extension [MARIPOSA]). Therefore, ctDNA clearance is prognostic but not predictive of treatment benefit for combinatorial strategies in the first-line setting.

Shared decision-making for complexities with more than one ‘right’ answer

Treatment decisions in the first-line management of advanced EGFRm NSCLC have become increasingly complex due to a progressively crowded landscape with the availability of multiple treatment options. Improving risk stratification, such as through adaptive trial designs, is an important research priority. While extending survival is typically highly pursued, we should also consider each patient’s needs and preferences in the decision-making process, rather than adopting a “one-size-fits-all” approach.

Figure 1 summarizes key considerations in choosing the treatment approach in the first-line setting. With osimertinib being introduced in the earlier stage setting, it is foreseeable that combinatorial strategies will play a bigger role in advanced-stage disease. [N Engl J Med 2020;383:1711-1723; N Engl J Med 2024;391:585-597; Future Oncol 2021;17:4045-4055; Tsuboi, M, et al, WCLC 2021]

To that end, physicians should familiarize themselves with the management of AEs to help patients cope with treatment-related toxicities, such that quality of life can be maintained along with survival improvements.