Osimertinib (Tagrisso) 8 Years On
- Eight-year data from the ADAURA trial confirm that the survival advantage of osimertinib (Tagrisso) for patients with resected, EGFR-mutated non-small cell lung cancer (NSCLC) persists long after the 3‑year adjuvant course, with 8‑year OS of 79% vs. 64% overall and 74% vs. 58% in stage II–IIIA.
- Hazard ratios (~0.52–0.53) mirror the 5‑year results, though the L858R subgroup shows a less certain effect (HR 0.72; CI crosses 1.0).
- The update reinforces long‑term benefit and the necessity of EGFR testing, while leaving open questions about optimal duration, resistance patterns, and next‑line strategies.
Adjuvant osimertinib (Tagrisso) keeps patients with resected, EGFR-mutated non-small cell lung cancer (NSCLC) alive longer, and a new landmark analysis indicates that advantage is still measurable eight years out, well after the three-year treatment course has ended.
The exploratory 8-year overall survival (OS) update from the phase III ADAURA trial (NCT02511106) extends the survival data that reshaped adjuvant practice for this population.
“These unprecedented data represent the longest-ever survival data reported in a global phase III trial in this setting and reinforce Tagrisso as the standard of care and backbone therapy for people with EGFRm lung cancer,” Eder Martins, vice president and U.S. franchise head for lung cancer at AstraZeneca, tells SurvivorNet Connect.
In the overall trial population (stage IB–IIIA), the hazard ratio for death was 0.52 (95% CI, 0.39–0.71), a 48% relative reduction in the risk of death. Eight-year OS reached 79% with osimertinib versus 64% with placebo.
According to the investigators, this is the longest OS follow-up reported from a global phase III adjuvant trial in EGFR-mutated early-stage NSCLC.
For clinicians who already prescribe osimertinib after resection, the update works as a durability check rather than a practice change. It shows that the relative benefit has held.
What The 8-Year Analysis Shows
Two populations anchor the report.
- In the overall population (stage IB–IIIA), the OS HR was 0.52 (95% CI, 0.39–0.71), with 8-year OS of 79% versus 64% and median follow-up of 93.3 months on osimertinib versus 79.6 months on placebo.
- In the primary DFS population (stage II–IIIA), the OS HR was 0.53 (95% CI, 0.38–0.75), with 8-year OS of 74% versus 58% and median follow-up of 92.0 versus 68.5 months.
ADAURA randomized 682 patients 1:1 to osimertinib 80 mg once daily or placebo for up to three years. Because this is a long-term landmark update, the survival curves now describe outcomes roughly five years beyond the end of protocol treatment for most patients. The early separation between the arms has not closed.
“These results reinforce the importance of people with EGFRm NSCLC working closely together with their doctors to manage side effects early and maintain good communication, to ensure they have the best chance of completing the full three years of treatment and achieving the best possible outcomes,” Martins explains.
Building on Earlier Data
ADAURA has reported in stages, and each release answered a different question.
The 2020 primary analysis reported a disease-free survival (DFS) HR of 0.17 (99.06% CI, 0.11–0.26) in stage II–IIIA and 0.20 (99.12% CI, 0.14–0.30) in the overall population, an effect large enough that the independent data monitoring committee recommended unblinding two years ahead of schedule.
That result established DFS benefit but left a familiar question in adjuvant oncology open. Would delaying recurrence translate into longer life, or only postpone the reckoning? The 2023 final analysis answered this.
Both populations showed an OS HR of 0.49 (95% CI, 0.33–0.73 in stage II–IIIA; 95.03% CI, 0.34–0.70 in stage IB–IIIA), a 51% reduction in the risk of death, with 5-year OS of 88% versus 78% overall and 85% versus 73% in stage II–IIIA.
Against those figures, the 8-year update reads as confirmation of durability. The hazard ratios of 0.52 and 0.53 sit close to the 0.49 seen at five years, so the relative survival benefit has been essentially maintained across an additional three years of follow-up. Absolute OS has declined in both arms over time, as expected, while the gap between them has stayed wide.
The Subgroup That Complicates the Picture
The OS benefit was reported across subgroups, but the mutation-type breakdown deserves a closer look before it gets summarized as uniform.
By EGFR mutation, the OS HR was 0.45 (95% CI, 0.29–0.68) for exon 19 deletions and 0.72 (95% CI, 0.46–1.11) for L858R. The L858R confidence interval crosses 1.0, so the OS benefit in that subgroup did not reach statistical significance in this exploratory analysis.
The point estimate still favors osimertinib, and this is a subgroup with fewer events rather than a signal of no effect, but the distinction matters when you are quantifying expected benefit for an L858R patient. A smaller or less certain effect in L858R disease is consistent with patterns seen across EGFR-mutated NSCLC.
Reading the Exploratory Design Honestly
The strength of this update is the length of follow-up. Its main limitation is how that follow-up was assembled.
This was a post hoc, exploratory analysis. Survival data were collected from 431 of the 558 patients alive at the January 2023 final OS analysis, tracked through yearly follow-up to the May 2026 data cutoff or through available records from their last contact date.
The 127 patients without additional follow-up were censored at the 2023 cutoff, their survival time unchanged. That approach is reasonable for a long-term landmark look, but it means the 8-year estimates rest on continued follow-up of a subset of the original cohort rather than complete ascertainment. Read the hazard ratios and OS rates as supportive long-term evidence, not as a new confirmatory endpoint.
What it Means for Practice
Adjuvant osimertinib is already guideline-recommended and widely used for resected EGFR-mutated stage IB–IIIA NSCLC, and nothing here changes that positioning. What the update adds is a longer-term answer to the question patients ask in the room after surgery: does the benefit last? Eight-year OS of 79% overall and 74% in stage II–IIIA puts a concrete number behind that conversation.
The data also reinforce the upstream dependency that makes any of it possible. None of this reaches the patient without EGFR testing at diagnosis.
Open questions still remain. The optimal duration of adjuvant osimertinib is still undefined. Three years was a protocol choice, not a biologically derived stopping point, and whether a longer or shorter course would do as well or better is unknown.
Resistance patterns after adjuvant exposure, and the best next line for patients who recur despite it, are under active study. For now, the 8-year data give clinicians evidence that the survival advantage is real, and lasting.
