Phase III Study to Determine the Efficacy of Durvalumab in Combination With Chemotherapy in Completely Resected Stage II-III Non-small Cell Lung Cancer (NSCLC): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Phase III Study to Determine the Efficacy of Durvalumab in Combination With Chemotherapy in Completely Resected Stage II-III Non-small Cell Lung Cancer (NSCLC) (NCT04385368), a randomized phase 3 study. The primary outcome is Disease-free Survival (DFS) in FAS (Using Investigator Assessments According to Response Evaluation Criteria in Solid Tumors 1.1 [RECIST 1.1]).

2. Clinical Question

3. Trial Design

MERMAID-1 is described as a Phase III, randomized, multicenter, double-blind, placebo-controlled, parallel-arm study. It compared durvalumab plus standard-of-care chemotherapy with placebo plus standard-of-care chemotherapy in participants with completely resected stage II-III NSCLC who were MRD-positive after surgery. AstraZeneca is listed as the sponsor. The study status is completed.

The registry record lists enrollment of 89. The primary endpoint analysis population, the FAS, included all randomized participants. The arm-level records list 45 participants in the durvalumab group and 44 in the placebo group. The study began on 2020-07-17, and primary completion is listed as 2023-08-31. The primary endpoint timeframe specifies assessments every 12 weeks, with a window of ± 1 week, until disease recurrence or the primary DFS analysis, up to 33.28 months.

The listed primary outcome was median disease-free survival using investigator assessments according to RECIST 1.1. Secondary outcomes included DFS in the MRD-positive analysis set and overall survival in both the FAS and the MRD-positive analysis set. The registry record lists three secondary outcomes and four outcome measures, but does not provide posted analysis entries. It does not report the randomization procedure, allocation ratio as a protocol specification, planned sample-size justification, detailed statistical analysis plan, interim analyses, or further operational details of multicenter conduct.

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.

Before assignment: A total of 89 participants were randomized in a 1:1 ratio to durvalumab given concurrently with standard of care (SoC) chemotherapy followed by durvalumab monotherapy, or to a matched placebo given concurrently with SoC chemotherapy followed by placebo.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The FAS included all randomized participants.

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 Full analysis set (FAS) included all randomized participants.

6. Primary Endpoint

The registry lists the primary outcome as Disease-free Survival (DFS) in FAS (Using Investigator Assessments According to Response Evaluation Criteria in Solid Tumors 1.1 [RECIST 1.1]), measured over Every 12 weeks (q12w) ± 1 week until appearance of RECIST 1.1-defined disease recurrence or until primary DFS analysis, up to 33.28 months.

Disease-free survival (DFS) measures the time from randomization to a defined event. In this trial, the first qualifying event was disease recurrence identified using investigator RECIST 1.1 assessments, or death from any cause. Recurrence included local or regional recurrence, distant recurrence, or a second primary NSCLC. The endpoint was assessed in the FAS, which included all randomized participants.

General statistical context: a time-to-event endpoint uses both the occurrence of an event and the time at which it occurs. It can therefore describe not only whether an event was recorded, but also how long participants remained without a recorded event during observation. A median is the time by which half of the relevant group has experienced the event, under the survival analysis definition. A confidence interval conveys uncertainty around an estimate; a wider interval generally indicates less precision.

The posted primary outcome is identified as median DFS with a 95% confidence interval, but the median and the complete confidence interval are not reported in the registry record. Consequently, the endpoint definition is available, while the primary endpoint estimates needed to compare the groups are not.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the median represent?

General statistical context: for a time-to-event outcome, the median is the point in time at which the estimated event-free proportion has reached the midpoint under the survival analysis method. It is not an average duration and does not imply that every participant was followed until an event. Participants without a recorded event may contribute information up to their last known observation through censoring. The registry record does not report the number of DFS events, censoring details, or the method used to estimate the median.

The primary endpoint is labeled as median DFS in months and is associated with a 95% confidence interval. However, neither arm has a posted median estimate or a complete confidence interval. A reported lower confidence limit without its corresponding estimate and upper limit cannot be read as the median itself. It also does not convey the full uncertainty around the estimate.

How should a confidence interval be read?

General statistical context: a confidence interval represents uncertainty in an estimated quantity under a statistical procedure. It is not a guarantee that the true value lies within a particular interval, and it does not by itself indicate whether a treatment difference is important. Its width reflects precision in the context of the method and data. For a comparison, the interval for the contrast is often more directly useful than separate intervals for each group, but the registry record does not post a complete contrast estimate or interval for DFS.

In this record, the listed values of 14.06 and 21.91 months are lower limits, not complete arm-level results. No upper limits or median values are reported. It would therefore be inappropriate to compare the two lower limits as though they were medians, calculate a difference from them, or use them to infer the direction or strength of the treatment comparison.

What does a hazard ratio add?

General statistical context: a hazard ratio summarizes the relative instantaneous event rate between groups over follow-up, conditional on the model and its assumptions. It is not the ratio of median survival times and is not an absolute probability of an event. Its confidence interval is important because it indicates the uncertainty of the estimated relative effect. The registry record does not report a hazard ratio or confidence interval for the primary endpoint. The stated kind of result indicates that only sponsor-reported hazard ratios are used for time-to-event material, but no such primary DFS estimate is posted in the record.

Interpreting a hazard ratio also depends on the analysis specification, including the handling of censoring and the form of the model. The registry record does not report those details. General explanation of the statistical measure should not be mistaken for a claim that a particular model, assumption, or analysis was used in MERMAID-1.

Why do analysis populations matter?

The primary outcome is specified for the FAS, defined in the registry record as all randomized participants. The record also identifies an MRD-positive analysis set for secondary outcomes. Analysis populations matter because they identify which participants contribute to each estimate. Results from different populations may not be interchangeable, especially when the population definitions differ. Further details on the MRD-positive analysis set and its membership are not reported in the registry record.

The randomized allocation is relevant to the intended comparison, but a complete interpretation also depends on the observed endpoint data and the analysis approach. The record lists arm counts of 45 and 44, but does not post DFS event counts, withdrawals, missing endpoint data, or analysis output. Those omissions prevent an independent assessment of how much information supported the primary estimates.

What can planning illustrations tell readers?

General statistical context: a sample-size calculation is a planning exercise based on assumptions about the effect to detect, error rates, allocation, and a statistical model. It is not evidence that a trial had a particular probability of finding a result, and it is not an observed treatment effect. The registry record's planning assumptions describe a two-sided significance level of 0.05, equal allocation, a normal approximation, and effects expressed as a fraction of the observed effect. For time-to-event calculations, the stated planning quantity is the total number of events.

Those assumptions are described as planning illustrations rather than the trial's own design justification. The registry record does not report the trial's sample-size rationale or a prespecified target effect. The observed enrollment of 89 should therefore not be treated as confirmation of a design target, nor should planning illustrations be interpreted as observed power. Since the primary estimates are incomplete, they also cannot support a grounded assessment of how the observed treatment effect compares with an assumed planning effect.

10. Multiplicity, Missing Data and Analysis Populations

The registry record lists one primary outcome and three secondary outcomes. The secondary outcomes concern DFS in the MRD-positive analysis set and overall survival in the FAS and MRD-positive analysis set. However, posted analysis entries are not reported in the registry record, and numerical results for the secondary outcomes are not included. The record does not describe a multiplicity-control strategy, a testing hierarchy, or which secondary analyses, if any, were intended to be confirmatory. These details are not reported in the registry record.

General statistical context: when several outcomes or comparisons are tested, the chance of at least one apparently notable result can increase unless the analysis plan accounts for multiple testing. Methods such as hierarchical testing or adjusted significance thresholds can address that issue, but the appropriate interpretation depends on the prespecified plan. Because the record does not state such a plan, no claim about multiplicity adjustment can be made.

The FAS is defined as all randomized participants, and it is the population named for the primary DFS outcome. The record does not report how missing endpoint observations were handled, how censoring was defined in the analysis, whether participants were excluded from any analysis, or whether sensitivity analyses were conducted. Further details about the MRD-positive analysis set are also not reported in the registry record.

General statistical context: missing outcome information can affect precision and, depending on its pattern and handling, can also affect interpretation. Censoring is a normal feature of time-to-event analysis, but the assumptions and procedures used to handle it matter. Without the analysis plan and outcome-level accounting, readers cannot assess the impact of missing or censored observations. The incomplete posted primary estimates make those limitations especially relevant.

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: pneumonia (6.7%), colitis (6.7%), acute kidney injury (4.5%), covid-19 (4.5%), anaemia (4.4%).

The safety record lists serious adverse events affecting 14 of 45 participants in the durvalumab plus standard-of-care group and 10 of 44 participants in the placebo plus standard-of-care group. Other adverse events are listed for 43 of 45 and 44 of 44 participants, respectively. The record also lists 6 deaths in the durvalumab group and 5 in the placebo group. These figures describe the recorded safety counts; they do not by themselves establish that an event or death was caused by study treatment.

The serious adverse event terms listed include Pneumonia and Colitis, each at 6.7%; Acute kidney injury and COVID-19, each at 4.5%; Anaemia at 4.4%; and Infusion related reaction at 2.3%. The registry record does not specify the arm attribution for these listed percentages, so they should not be assigned to either group. It also does not provide detailed definitions, event-level timing, duration, severity grades, or attribution assessments for these terms.

General statistical context: safety counts are usually interpreted alongside exposure time, event definitions, severity, and causal assessment. These details are not reported in the registry record. The listed counts are descriptive and should not be treated as a randomized comparison of the primary DFS endpoint.

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

The registry record establishes that MERMAID-1 was designed as a randomized comparison of durvalumab plus standard-of-care chemotherapy with placebo plus standard-of-care chemotherapy in MRD-positive participants after complete resection of stage II-III NSCLC. It defines the primary endpoint as DFS in the FAS, using investigator assessments according to RECIST 1.1, with death from any cause also counted as an event. These details clarify the clinical question and the outcome the study intended to assess.

The posted primary results are not complete enough to quantify the comparison. Median DFS values and complete 95% confidence intervals are not reported in the registry record. Nor are a hazard ratio, its confidence interval, or a test result. As a result, the posted information does not show the size or precision of a relative or absolute difference in primary DFS between groups.

This limitation is distinct from saying that the trial found no difference. An unavailable or incomplete estimate is not evidence of equivalence, and it does not establish that the groups had the same outcomes. It also does not support a conclusion that one group had longer DFS. The two lower confidence limits alone cannot resolve those questions.

The safety counts provide descriptive information about participants recorded with serious and other adverse events, as well as deaths. They do not replace a complete efficacy analysis and do not establish event causation. The listed secondary outcomes also lack numerical results in the registry record. Thus, the record allows readers to understand the study design, the endpoint definition, and some safety counts, but it does not allow a quantitative interpretation of the primary treatment comparison or the secondary outcome results.

General statistical context: the strength of a trial's evidence depends on the estimates, their uncertainty, and the analysis methods, not only on the study label or design. Randomization describes how the comparison was organized; it cannot supply missing outcome estimates. Conclusions should remain limited to the information actually posted in the registry record.

13. Why This Trial Matters Statistically

This trial raises an important statistical question about how to evaluate a time-to-event endpoint after surgery and chemotherapy in a defined population. The population is restricted to participants with completely resected stage II-III NSCLC who were MRD-positive after surgery, and the primary outcome is DFS in all randomized participants. The endpoint definition includes several forms of recurrence as well as death from any cause, making the event definition central to interpretation.

Randomized, double-blind, placebo-controlled design features are relevant to how a treatment comparison is structured. At the same time, readers need outcome estimates and their uncertainty to understand what the observed data indicate. In this record, the primary DFS medians and complete confidence intervals are missing, and no hazard ratio or comparison test result is posted. That contrast between a clearly described study question and incomplete outcome reporting is statistically consequential: the design can be described, but the primary estimate cannot be assessed.

The distinction between the FAS and the MRD-positive analysis set also matters. The primary outcome is assigned to the FAS, while some secondary outcomes concern the MRD-positive set. Analysis populations define the data included in a result, and different populations should not be silently treated as interchangeable. The record does not supply further detail on membership or analysis procedures for the latter set.

Finally, the safety record illustrates why descriptive counts need context. Counts of participants with adverse events can be reported, but interpretation also depends on event definitions, exposure, timing, severity, and attribution. Those details are not reported in the registry record. Statistical reading therefore requires attention to what is defined, what is estimated, and what remains unavailable, rather than relying on trial phase or design labels alone.

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 principal limitation for interpreting this trial's primary statistical result is that the registry record does not report median DFS estimates or complete 95% confidence intervals for either arm. Although lower confidence limits are listed, they do not constitute complete estimates. The record also does not report a hazard ratio, a confidence interval for a treatment contrast, or a comparison test result. The magnitude, direction, and precision of the primary comparison therefore cannot be assessed from the posted results.

DFS event counts, censoring details, missing endpoint accounting, and the analysis procedures used for the primary endpoint are not reported in the registry record. Without these details, readers cannot evaluate how much endpoint information was available or how incomplete observation was addressed. The record specifies investigator RECIST 1.1 assessments and an assessment schedule, but not further operational or analytical details.

Secondary outcomes are identified but their numerical results are not reported. Multiplicity handling, the trial's sample-size justification, any interim analyses, and the detailed statistical analysis plan are also not reported in the registry record. The described planning assumptions are illustrations, not documentation of the trial's design rationale.

Safety information is limited to participant counts, deaths, and a list of serious adverse event terms with percentages. The record does not provide arm attribution for the listed percentages or detailed event-level information, severity grading, timing, or causality assessment. These gaps constrain interpretation of safety patterns as well as efficacy results. The record's limitations should not be mistaken for evidence that an unreported result was null or that an unreported analysis was not performed.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Phase III Study to Determine the Efficacy of Durvalumab in Combination With Chemotherapy in Completely Resected Stage II-III Non-small Cell Lung Cancer (NSCLC) — NCT04385368.
• Registry record last updated 2024-08-30; status last verified 2024-08.
• Results first posted 2024-08-30.
• 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

MERMAID-1 is a completed Phase III randomized, multicenter, double-blind, placebo-controlled study sponsored by AstraZeneca. It enrolled 89 participants with completely resected stage II-III NSCLC who were MRD-positive after surgery. The comparison was durvalumab plus standard-of-care chemotherapy versus placebo plus standard-of-care chemotherapy.

The primary outcome was median DFS in the FAS, defined as time from randomization to recurrence by investigator RECIST 1.1 assessment or death from any cause. The arm records list 45 and 44 participants, but complete median estimates and confidence intervals are not reported in the registry record. No primary hazard ratio or test result is posted. Safety counts and several serious adverse event terms are listed, while numerical secondary outcome results and further analysis details are not reported in the registry record.