FLAME Study Shows Osimertinib Plus Chemotherapy Improves PFS in ctDNA-Persistent EGFR-Mutant NSCLC

June 1, 2026 | Krista Rosa

The addition of carboplatin and pemetrexed (Alimta) to osimertinib (Tagrisso) improved progression-free survival (PFS) compared with osimertinib alone in patients with advanced EGFR-mutant non–small cell lung cancer (NSCLC) who had persistent circulating tumor DNA (ctDNA) EGFR mutation 3 weeks after initiating first-line osimertinib, according to data from the randomized FLAME study (NCT04769388) presented at the 2026 ASCO Annual Meeting.¹

At a median follow-up of 17.9 months (95% CI, 12.5-23.1) in the combination arm (n = 40) and 12.7 months (95% CI, 7.1-16.0) in the monotherapy arm (n = 40), the median PFS by investigator assessment was 23.1 months (95% CI, 12.65-29.24) with osimertinib plus chemotherapy vs 12.7 months (95% CI, 12.45-18.07) with osimertinib alone (HR, 0.54; 90% CI, 0.34-0.86; P = .028). The primary analysis was conducted at 68.8% data maturity, with 24 PFS events in the combination arm and 31 in the monotherapy arm.

Moreover, the PFS benefit favoring the combination was consistent across all prespecified subgroups. The effect was most pronounced among patients with brain metastases (HR, 0.24; 95% CI, 0.08-0.69) and was also observed in patients younger than 65 years (HR, 0.44; 95% CI, 0.23-0.85) and those with an ECOG performance status of 1 (HR, 0.51; 95% CI, 0.29-0.89).

FLAME was designed to test whether dynamic, on-treatment ctDNA could identify patients who derive less benefit from osimertinib monotherapy and stand to gain from early treatment intensification, according to presenting author Jia Zhong, MD, of the Department of Medical Oncology at the Cancer Hospital, Chinese Academy of Medical Sciences, in Beijing, China. “These findings add evidence to the dynamic ctDNA-guided individualized escalation treatment strategy,” Zhong said.

How was the FLAME trial of osimertinib plus chemotherapy designed?

FLAME was a prospective, multicenter, randomized study that enrolled patients at least 18 years of age with locally advanced or metastatic EGFR-mutant NSCLC harboring an exon 19 deletion or L858R mutation.^1,2 All patients had received 3 weeks of first-line osimertinib and had persistent plasma ctDNA EGFR mutation at 3 weeks, as determined by Super ARMS-PCR, without disease progression per RECIST 1.1. Patients had an ECOG performance status of 0 or 1. Those with asymptomatic, stable brain metastases were eligible.

A total of 80 patients were randomly assigned 1:1 to osimertinib at 80 mg once daily plus pemetrexed at 500 mg/m² and carboplatin at area under the curve 5 every 3 weeks for 4 to 6 cycles, followed by maintenance osimertinib plus pemetrexed, or to osimertinib monotherapy at 80 mg once daily. Randomization was stratified by brain metastasis status (yes vs no) and EGFR mutation type (exon 19 deletion vs L858R mutation). PFS by investigator assessment per RECIST 1.1 served as the primary end point; secondary end points included 18-month overall survival (OS) rate, objective response rate (ORR), disease control rate (DCR), duration of response (DOR), safety, and resistance mechanisms.

Of 448 patients assessed for eligibility, 310 had cleared plasma ctDNA EGFR mutation and 135 had persistent ctDNA EGFR mutation; 80 were ultimately randomly assigned. Baseline characteristics were balanced between arms, including in terms of brain metastases, which were present in 35% of patients in each group. The data cutoff date was January 26, 2026.

What was the safety profile of osimertinib plus chemotherapy in the FLAME trial?

In the combination arm, patients received a median of 4 cycles of carboplatin (range, 1-6), with 81% completing at least 4 cycles, and a median of 9 cycles of pemetrexed (range, 1-43). The median duration of osimertinib exposure was 18.6 months (range, 1.4-38.8) in the combination arm vs 12.7 months (range, 1.4-33.7) in the monotherapy arm.

Grade 3 or higher treatment-emergent adverse effects (TEAEs) occurred in 73% of patients in the combination arm (n = 37) vs 17% in the monotherapy arm (n = 41), and serious TEAEs occurred in 49% vs 15%, respectively. Dose interruption of any study drug was reported in 73% vs 7% of patients, dose reduction in 32% vs 0%, and treatment discontinuation of any study drug in 16% vs 5%. One death occurred in each arm.

Zhou attributed higher grade 3 or higher event rate in the combination arm primarily to chemotherapy-related myelosuppression. She added that the safety profile was consistent with the established profiles of osimertinib and platinum-pemetrexed chemotherapy and was broadly comparable with that of FLAURA2 (NCT04035486), with no new safety signals.

What did the ctDNA biomarker analysis show in EGFR-mutant NSCLC?

In a parallel observational analysis comparing patients with 3-week ctDNA EGFR persistence (n = 59) and those with 3-week clearance (n = 59), persistence was associated with higher intratumor heterogeneity, more frequent co-alterations, and a numerically higher plasma ctDNA concentration after initial exposure to osimertinib. Zhou framed these molecular features as supporting the rationale for early combination therapy in patients with persistent ctDNA at 3 weeks.

She concluded that FLAME represents the first prospective proof-of-concept study to show that ctDNA-guided early intensification can improve outcomes in this population and that the data support further investigation of treatment escalation strategies based on molecular features.

Disclosures: Zhong had no relationships to disclose.

References

Wang Z, Zhuo M, Zhong J, et al. Osimertinib monotherapy versus osimertinib plus chemotherapy in advanced EGFR-mutant NSCLC with persistent ctDNA EGFR mutation at three weeks after first line osimertinib monotherapy initiation: a multicenter randomized trial (FLAME study). Presented at: 2026 ASCO Annual Meeting; May 29-June 2, 2026; Chicago, IL. Abstract LBA101.

Osimertinib plus chemotherapy vs osimertinib in EGFRm NSCLC with persistence week-3 ctDNA EGFRm after 1L osimertinib (FLAME). ClinicalTrials.gov. Updated January 12, 2026. Accessed May 31, 2026. https://clinicaltrials.gov/study/NCT04769388