Trials · for clinicians

Flaura 2 trial

10/31/2025 · 5 min read · updated 27 Oct 2026

Bottom line

FLAURA2 adds ~10 months of median OS (HR 0.77) at the cost of more grade ≥3 toxicity (70% vs 34%). Strongest case: CNS metastases, L858R, high disease burden. Osimertinib alone remains reasonable for frail patients or by preference.

Design

Phase 3, open-label; n=557; untreated advanced EGFR ex19del/L858R NSCLC. Osimertinib 80 mg + pemetrexed + cisplatin/carboplatin ×4, then osimertinib + pemetrexed maintenance vs osimertinib alone.

Results

PFS (primary, investigator-assessed, NEJM 2023): 25.5 vs 16.7 mo; HR 0.62 (95% CI 0.49–0.79; p<0.001); BICR HR 0.62 (29.4 vs 19.9 mo). 24-mo PFS 57% vs 41%. ORR 83% vs 76%; median DoR 24.0 vs 15.3 mo. [NEJM 2023]
OS (final, ESMO 2025/NEJM): 47.5 vs 37.6 mo; HR 0.77 (95% CI 0.61–0.96; p=0.02); 57% maturity; median OS follow-up ~51 mo. [NEJM 2025]
OS at 24 mo: 79.7% vs 71.5%; 36 mo: 63.1% vs 50.9%; 48 mo: 49.1% vs 40.8% (exploratory). [AZ]
Baseline CNS metastases (exploratory): median OS 40.9 vs 29.7 mo; HR 0.72 (0.52–0.99). [LBA77]
Grade ≥3 AEs (all causes): 70% vs 34% — mainly haematological, chemotherapy-related. AEs leading to osimertinib discontinuation: 12% vs 7%. [NEJM 2025]

FLAURA2: survival outcomes

Median PFS: osi + chemo25.5 mo

Median PFS: osi16.7 mo

Median OS: osi + chemo47.5 mo

Median OS: osi37.6 mo

FLAURA2: overall survival rate at 3 and 4 years

36-mo: osi + chemo63.1%

36-mo: osi50.9%

48-mo: osi + chemo49.1%

48-mo: osi40.8%

AstraZeneca/NEJM final OS data (cut-off 12 June 2025).

Where it fits (Oct 2026)

NCCN/ESMO list three preferred first-line options for common EGFR mutations: osimertinib, osimertinib + chemotherapy (FLAURA2) and amivantamab + lazertinib (MARIPOSA, also with OS benefit vs osimertinib). Choice depends on toxicity tolerance, logistics (IV vs oral), and disease features.

I favour FLAURA2 for fit patients with brain metastases or high tumour burden; osimertinib monotherapy remains entirely reasonable for elderly/frail patients or those prioritising fewer hospital visits.

Subgroups worth knowing

Exploratory final-OS subgroups (ESMO 2025 LBA77); HRs all consistent with the overall 0.77, CIs mostly cross 1 — hypothesis-generating. [LBA77]

Subgroup	n (osi+CT / osi)	mOS osi+CT	mOS osi	HR (95% CI)
CNS mets: yes	116 / 110	40.9	29.7	0.72 (0.52–0.99)
CNS mets: no	163 / 168	NR	43.9	0.77 (0.57–1.05)
L858R	106 / 107	38.1	32.4	0.76 (0.55–1.07)
Ex19del	172 / 169	NR	43.0	0.76 (0.56–1.02)
Plasma EGFRm detected	148 / 161	38.4	32.5	0.79 (0.60–1.03)
Tissue TP53 altered	46 / 40	51.1	43.1	0.71 (0.40–1.27)

CNS metastases: the only subgroup whose CI excludes 1, with an ~11-month median OS gain — the main reason many oncologists favour the combination here.
L858R vs exon 19 deletion: identical HRs (0.76); L858R does worse in absolute terms on either arm, a reasonable argument for intensification.
ctDNA / co-mutations (TP53): poorer-prognosis biology still benefits, but small numbers.

Toxicity in practice

Most extra toxicity came from chemotherapy (anaemia, neutropenia, thrombocytopenia, nausea) and was concentrated in the first 4 cycles. Pemetrexed maintenance requires folic acid and B12 premedication and regular renal checks. At final analysis, AEs led to osimertinib discontinuation in 12% vs 7%; the excess grade ≥3 AEs were described as reversible. [NEJM 2025]

The competing option: MARIPOSA

Amivantamab + lazertinib improved PFS over osimertinib (23.7 vs 16.6 months, HR 0.70) [NEJM 2024] and, at final analysis (median follow-up 37.8 mo), overall survival: HR 0.75 (95% CI 0.61–0.92; p=0.005); 3-year OS 60% vs 51%; grade ≥3 AEs 80% vs 52%. [NEJM 2025] It requires IV infusions and carries high rates of rash, paronychia, infusion reactions and venous thromboembolism (prophylactic anticoagulation is recommended). No head-to-head trial compares it with FLAURA2.

Practical sequencing

Get EGFR mutation subtype right first; uncommon mutations (G719X, L861Q, S768I) and exon 20 insertions were excluded from FLAURA2.
At progression on either strategy, re-biopsy (tissue or plasma) to look for MET amplification, histological transformation (small cell) or other resistance mechanisms.
The protocol allowed cisplatin 75 mg/m² or carboplatin (AUC 5) for 4 cycles; carboplatin is the practical choice for older patients or those with renal/hearing concerns.

The original oncosphere.in article was image-based and could not be recovered; this is a rewritten, referenced version under the same title.

Sources

Educational summary, not medical advice. Opinions are labelled as such and are the author's own. Approvals and availability in India (CDSCO) may differ from the US/EU.

← Back to blog