A Study of Osimertinib With or Without Chemotherapy as 1st Line Treatment in Patients With Mutated Epidermal Growth Factor Receptor Non-Small Cell Lung Cancer (FLAURA2): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Osimertinib With or Without Chemotherapy as 1st Line Treatment in Patients With Mutated Epidermal Growth Factor Receptor Non-Small Cell Lung Cancer (FLAURA2) (NCT04035486), a randomized phase 3 study. The primary outcome is Adverse Events Graded by Common Terminology Criteria for Adverse Event v5 (Safety Run-In Treatment Arms Only).

2. Clinical Question

FLAURA2 is a Phase III randomized study in non-small cell lung cancer with EGFR mutations. Its rationale in the registry record is that resistance to osimertinib may develop over time, and adding chemotherapy is being studied as a way to delay resistance and cancer worsening. That rationale describes why the comparison was undertaken; it is not itself evidence that either outcome occurred.

The randomized comparison is between osimertinib plus chemotherapy and osimertinib alone. The study also includes a safety run-in with chemotherapy combinations, and the primary outcome title specifically limits its population to those safety run-in treatment arms. The randomized and run-in results therefore address different groups and should not be blended into one treatment comparison.

Among the secondary outcomes listed for the randomized component are overall survival, landmark overall survival at 1, 2, and 3 years, and objective response rate. The registry record also lists outcomes for the safety run-in. The numerical results for these efficacy outcomes are not reported in the registry record details, so their comparative effects cannot be described.

3. Trial Design

FLAURA2 is a Phase III, randomized, open-label, parallel-group study sponsored by AstraZeneca. It evaluates first-line treatment in people with EGFR mutation-positive, locally advanced or metastatic non-small cell lung cancer. The record lists the study start date as 2019-07-02, primary completion date as 2023-04-03, and status as active, not recruiting.

The randomized component compares osimertinib alone with osimertinib plus platinum-based chemotherapy. In the combination arm, pemetrexed is given with cisplatin or carboplatin for 4 cycles, followed by daily osimertinib with pemetrexed maintenance every 3 weeks. The osimertinib-only arm is described as osimertinib 80 mg once daily. Both arm descriptions allow dose reduction to manage treatment-related toxicity.

A separate safety run-in includes chemotherapy combinations with carboplatin or cisplatin. The registry’s primary outcome is adverse events among participants in these run-in treatment arms. The posted safety results distinguish the two run-in groups from the randomized groups; the run-in data are not a randomized comparison of combination treatment against osimertinib alone.

The registry lists enrollment as 587. Its brief summary describes a plan to enroll approximately 586 participants, divided between an approximate safety run-in group and an approximate main-trial group. These are differently stated enrollment figures and should not be treated as interchangeable exact counts.

Open-label means treatment assignment was not masked according to the registry. The record does not report the randomization procedure, allocation concealment, or other operational details. The parallel design indicates that participants were assigned to a treatment group rather than crossing between treatments; further details of follow-up and adherence are not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Participants from the Safety Analysis Set, which includes all participants in the Safety Run-In Treatment Arms who received at least 1 dose of study treatment (osimertinib, cisplatin, carboplatin, or pemetrexed).

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: The Safety Run-In treatment arms are based on the Safety set, which includes all participants allocated to the safety run-in group of this study and who receive at least 1 dose of study treatment (Osimertinib, Cisplatin, Carboplatin, or Pemetrexed). The Randomized treatment arms are based on the Full Analysis set, which includes all participants allocated to the randomized group of this study and who were randomized to a treatment arm.

6. Primary Endpoint

The registry lists the primary outcome as Adverse Events Graded by Common Terminology Criteria for Adverse Event v5 (Safety Run-In Treatment Arms Only), measured over From first dose date to 28 days following last dose, up to 45 months.

The listed primary outcome is adverse events graded by the Common Terminology Criteria for Adverse Events, version 5.0, in the safety run-in treatment arms. The measure is the count of participants, and events are summarized by each participant’s maximum reported grade. The timeframe runs from the first dose to 28 days after the last dose, up to 45 months. The definition includes events beginning on or after the first dose and up to and including 28 days after treatment discontinuation, provided they occur before a new anti-cancer therapy starts.

The grade descriptions distinguish increasing clinical severity, from mild symptoms or observations without intervention through severe events, life-threatening consequences, and death related to an adverse event. Such grading helps describe the intensity of observed toxicity, while a participant count indicates how many people experienced an event in a category. These are distinct pieces of information: an event’s grade does not by itself describe how often it occurred.

Safety outcomes are used to characterize harms during treatment and the specified follow-up period. They do not establish comparative efficacy. The registry record’s primary-outcome information covers the safety run-in, and the listed outcome has multiple categories or time points rather than a single comparison reduced to one estimate. The registry record does not report a single primary endpoint result that can be used to quantify a treatment difference between the randomized arms.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

General statistical explanation: a randomized comparison is intended to assess differences between assigned treatment groups while reducing the influence of baseline differences on average. Randomization does not make every group identical, and it does not remove the need to report how outcomes were analyzed. The registry identifies the study as randomized and parallel, but the randomization method and detailed analysis plan are not reported in the registry record.

What do participant counts describe?

The safety results are counts of participants with events. A participant count is a binary outcome at the person level: each person is classified as having or not having the event under the stated definition. These counts differ from the number of individual events, because one participant can experience more than one event. The record labels the primary measure as participants and describes adverse events by maximum reported grade.

The posted safety figures give a group size at risk and a participant count affected. Reading both values together keeps the denominator visible. The counts should not be converted into a treatment effect without a defined comparison and analysis. No new percentages, risk differences, or ratios are necessary to describe what the record reports.

What does grading add?

Severity grading describes the most severe reported event for a participant under the primary outcome’s definition. The listed categories range from mild through life-threatening, with a separate grade for death related to an adverse event. Summarizing by maximum grade can make the most severe toxicity experienced by each participant visible, but it does not show the timing, duration, recurrence, or full set of events for that participant. Those details are not reported in the registry record details.

Why keep the run-in separate?

The primary outcome is explicitly limited to safety run-in treatment arms. Those groups are not the randomized comparison of chemotherapy plus osimertinib against osimertinib alone. Combining their counts with the randomized groups would mix different parts of the study and would not create a valid randomized treatment contrast. The registry presents the two run-in combinations separately, and the randomized treatment groups separately; interpretation should preserve that distinction.

What can a confidence interval say?

General statistical explanation: a confidence interval describes the range of effect estimates compatible with the data under a specified statistical method. A narrow interval indicates greater precision than a wide interval, while an interval that includes the null value may leave considerable uncertainty about the direction or size of an effect. The relevant interval depends on the effect measure and model. No confidence intervals for the safety comparisons or randomized efficacy outcomes are reported in the registry record details, so precision of a treatment contrast cannot be assessed from them.

Why can event counts be hard to compare?

General statistical explanation: counts depend on how many people were observed and for how long, as well as on the chance of an event. When follow-up differs, a simple comparison of counts may not represent a comparison of event rates over time. Likewise, baseline risk, treatment exposure, and discontinuation can influence observed safety counts. The record specifies the primary safety outcome’s event window, but does not provide the follow-up distribution or exposure-adjusted analysis needed to evaluate these factors.

General statistical explanation: over-dispersion matters when the variability in event counts is greater than a chosen model expects. It can make model-based uncertainty too small if ignored. This issue is most directly relevant to counts of repeated events; the primary outcome is described as the number of participants in adverse-event categories, not the total number of events per participant. The registry record does not report a model or an over-dispersion assessment.

What does the sample-size illustration mean?

The registry record describes planning illustrations based on the observed effect, using a two-sided significance level of 0.05, equal allocation, and a normal approximation. It says that the effect to detect is a stated fraction of the observed effect, and distinguishes participant-per-arm illustrations from total-event illustrations for time-to-event outcomes. Such calculations are planning illustrations, not observed power and not proof that a study can detect a particular effect. No illustration values are included in the registry record. The trial’s own sample-size justification and design assumptions are not reported in the registry record.

The registry lists 32 outcome measures and 12 analyses. These counts indicate that multiple outcomes and analyses are recorded, but do not reveal which analyses were prespecified, which were primary or exploratory, or how decisions were made across them. Those methodological details are not reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 32 outcome measures and 12 analyses, including safety and efficacy outcomes and outcomes tied to different study components. When many outcomes or analyses are considered, the chance of at least one apparently notable result can rise if no adjustment or interpretation strategy is specified. That is a general statistical concern; the number of measures alone does not show that a particular analysis is invalid.

The multiplicity approach—such as whether outcomes were organized in a testing hierarchy or adjusted for multiple comparisons—is not reported in the registry record. The registry entries do not identify which secondary outcomes were confirmatory, whether interim analyses were planned, or whether any decision rules were used. These should not be inferred from the outcome list or analysis count.

The primary outcome has several categories or time points and is not reduced to one comparison. This limits a direct summary to a single effect estimate. The registry record also does not report a single randomized primary-outcome comparison, confidence interval, or significance test in the registry record.

The safety population is described as participants in the safety run-in treatment arms who received at least one dose of study treatment. Other analysis populations, including the exact population definitions for the randomized efficacy outcomes, are not reported in the registry record details. Missing-data methods, handling of treatment discontinuation, and rules for censoring time-to-event outcomes are likewise not reported in the registry record. Without those details, readers cannot assess how incomplete follow-up or missing outcome information may have affected the analyses.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: retinal tear (6.7%), nausea (6.7%), small intestinal obstruction (6.7%), allergy to vaccine (6.7%), covid-19 (6.7%).

The safety results distinguish serious adverse events from other adverse events and give both the group size at risk and the number of participants affected. For the randomized combination group, the record lists 276 at risk, 104 affected by serious adverse events, and 273 affected by other adverse events. For the osimertinib-alone group, it lists 275 at risk, 53 affected by serious adverse events, and 258 affected by other adverse events.

In the safety run-in, the carboplatin and pemetrexed group has 15 at risk, with 5 affected by serious adverse events and 15 by other adverse events. The cisplatin and pemetrexed group has 15 at risk, with 6 affected by serious adverse events and 14 by other adverse events. The run-in groups should not be treated as randomized comparators for the main treatment question.

The registry lists Retinal tear, Nausea, Small intestinal obstruction, Allergy to vaccine, COVID-19, and Parainfluenzae virus infection among the top serious terms, each at 6.7%. The record also lists deaths by group, but the listed death counts do not establish that deaths were caused by an adverse event or by treatment. Seriousness, severity grade, and causal attribution are not interchangeable concepts; the counts alone do not establish causation.

12. What the Result Does, and Does Not, Mean

The registry record establishes that FLAURA2 is a randomized Phase III study comparing osimertinib plus platinum-based chemotherapy with osimertinib alone in first-line treatment, alongside a separate safety run-in. It also provides descriptive safety counts for the run-in and randomized groups. Those data can be read as counts of participants recorded with serious and other adverse events, within the definitions used in the record.

They do not by themselves establish that one treatment caused more events than another. A causal interpretation requires attention to the study population, outcome definitions, follow-up, missing data, and analysis method. The record does not provide enough detail in the registry record to quantify a comparative effect with its precision or to assess all these considerations.

The safety outcome is not a substitute for the study’s efficacy outcomes. The registry lists overall survival, landmark overall survival, and objective response rate among the randomized-component secondary outcomes, but numerical results for them are not reported in the registry record details. Thus the material does not establish whether adding chemotherapy changes survival, response, resistance, or cancer worsening.

Likewise, an absence of a reported efficacy estimate in the registry record is not evidence that there was no effect. It means that the relevant result cannot be evaluated from the information described. The primary outcome is a safety measure restricted to the run-in arms, and the record indicates that it contains several categories or time points rather than one reduced comparison.

The appropriate statistical conclusion is therefore bounded: the record reports participant counts for adverse-event categories and deaths in the named groups, but does not supply a complete comparative primary-outcome analysis or the numerical randomized efficacy results. The counts are descriptive, not a stand-alone judgment of benefit, harm, or treatment choice.

13. Why This Trial Matters Statistically

FLAURA2 presents a clinically relevant statistical structure: a randomized comparison of a targeted therapy alone against the same therapy combined with chemotherapy, alongside an independent safety run-in. Keeping those components distinct is essential. The run-in contributes safety information, while the randomized component is the part designed to compare assigned treatment strategies.

The study also illustrates why endpoint labels and populations matter. Although the stated research aim concerns the combination versus osimertinib alone, the registry’s primary outcome is adverse events in the safety run-in arms. A reader should not assume that a study’s broad rationale and its listed primary endpoint are the same statistical question. Nor should a safety count from a run-in be read as a randomized treatment-effect estimate.

Many outcomes are listed, spanning safety, response, and survival. This breadth makes transparent reporting of outcome definitions, analysis populations, multiplicity strategy, missing-data handling, and uncertainty especially valuable. The registry record details do not report several of these details or the numerical efficacy analyses, which constrains what can be concluded from the posted material.

Finally, presenting group sizes alongside affected-participant counts allows readers to understand the scope of the safety observations without turning descriptive counts into causal claims. The trial is statistically important as a setting for evaluating how design, endpoint selection, and reporting determine which questions a set of results can answer.

14. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The registry’s primary outcome is limited to the safety run-in treatment arms, while the central treatment question concerns the randomized comparison. The available primary-outcome description specifies grading and a time window, but the registry entries do not give the full grade-specific results or a single treatment comparison. This makes it inappropriate to present the posted safety counts as the primary randomized result.

The record lists safety counts and deaths by group, but the registry entries do not describe comparative models, uncertainty intervals, adjustment for follow-up, or causal attribution. The listed deaths should not be interpreted as treatment-related. The serious-event terms and their reported frequencies also do not provide a complete account of event timing, duration, recurrence, or individual participant histories.

The randomized component’s secondary outcomes include survival and response measures, but their numerical results are not reported in the registry record details. Nor are the statistical plan, multiplicity procedures, interim analyses, missing-data methods, or all analysis-population definitions reported. These omissions limit assessment of precision, robustness, and the role of incomplete follow-up.

The registry lists enrollment as 587, while its brief summary describes approximately 586 planned participants. The summary’s approximate run-in and main-trial targets should not be treated as exact realized group sizes. The posted arm counts are separately identified for the safety groups, and should be kept distinct from approximate planning descriptions.

Finally, the open-label design means treatment assignment was not masked according to the registry. The record does not describe additional safeguards for outcome assessment. No inference about the practical impact of this feature is possible in the registry record alone, but it is relevant context when interpreting outcomes that may involve judgment in their assessment.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Study of Osimertinib With or Without Chemotherapy as 1st Line Treatment in Patients With Mutated Epidermal Growth Factor Receptor Non-Small Cell Lung Cancer (FLAURA2) — NCT04035486.
• Registry record last updated 2026-09-03; status last verified 2026-08.
• Results first posted 2024-08-06.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

18. Record Summary

FLAURA2 (NCT04035486) is a Phase III, open-label, randomized, parallel-group study sponsored by AstraZeneca. It compares first-line osimertinib alone with osimertinib plus platinum-based chemotherapy in EGFR mutation-positive, locally advanced or metastatic non-small cell lung cancer. The record lists 587 as enrollment, with status active, not recruiting.

The listed primary outcome is adverse events graded by CTCAE version 5.0 among safety run-in treatment arms. The record also lists randomized-component secondary outcomes including overall survival, landmark overall survival, and objective response rate. Safety counts are reported for the run-in and randomized groups, but numerical efficacy comparisons and a single reduced primary-outcome comparison are not reported in the registry record details.

The registry record lists 32 outcome measures and 12 analyses. It does not report the multiplicity strategy, missing-data methods, or sample-size justification in the registry record. The study began on 2019-07-02, and its primary completion date is listed as 2023-04-03.