Osimertinib (AZD9291) in First-line Locally Advanced or Metastatic NSCLC Patients With EGFR and EGFR T790M: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Osimertinib (AZD9291) in First-line Locally Advanced or Metastatic NSCLC Patients With EGFR and EGFR T790M (NCT02841579), a na phase 2 study. The primary outcome is Objective Response Rate (ORR).

2. Clinical Question

3. Trial Design

This is a phase IIa clinical trial of osimertinib in patients with first-line, locally advanced or metastatic non-squamous non-small cell lung cancer with EGFR mutations and a concomitant EGFR T790M mutation at diagnosis. The registry lists MedSIR as the sponsor and records the study as completed. The recorded start date is 2016-09-02, and primary completion is listed as 2018-12. Enrollment is recorded as 22; that figure is not, by itself, a statement about how many participants were included in the primary outcome analysis.

The design is single-group, with no allocation to comparison groups and no masking. The intervention description specifies osimertinib (AZD9291), 1 tablet of 80 mg by mouth daily, until disease progression or unacceptable side effects, up to 78 weeks from the first administered dose. There is no control treatment described, so observed outcomes are descriptive for the treated study group rather than estimates of a randomized difference between treatments.

Objective response rate is the primary outcome. The registry defines it using complete and partial response under revised RECIST guidelines, version 1.1, and gives a timeframe through study completion, from baseline up to approximately 28 months. Secondary outcomes listed include overall survival, duration of response, disease control rate, tumor shrinkage, plasma EGFR mutation status, acquired resistance, adverse events, and progression-free survival. The record lists 14 outcome measures, but does not report the analysis population or a statistical analysis plan.

The design can describe response and safety observations in the enrolled study group, subject to the availability and completeness of outcome reporting. Because there is no concurrent control arm, differences from other populations or treatments cannot be attributed to osimertinib using this design alone. The registry record does not report a sample-size justification, a planned interim analysis, or details of how participants with incomplete follow-up were to be handled.

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: Fifty-one patients were screened of whom 22 patients with advanced NSCLC harbouring pre-treatment sensitising EGFR and Thr790Met mutations were enrolled and started to receive first-line osimertinib.

Overall Study

Reasons for not completing this period:

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.

6. Primary Endpoint

The registry lists the primary outcome as Objective Response Rate (ORR), measured over Through study completion. From baseline up to approximately 28 months..

Objective response rate is defined as the percentage of patients who achieve complete response or partial response under the revised RECIST guidelines, version 1.1. For target lesions, complete response means disappearance of all target lesions. Partial response means at least 30% decrease in the sum of target-lesion diameters, using the baseline sum of diameters as the reference. The endpoint timeframe is through study completion, from baseline up to approximately 28 months.

This endpoint reduces a patient's tumor assessment to a binary classification for the response calculation: the patient either meets the stated complete- or partial-response definition, or does not. The percentage is a summary of how commonly response was observed in the assessed study group. A confidence interval describes the statistical uncertainty around that percentage; it is important because a percentage from a limited study group may be imprecise.

As a general statistical point, objective response is often used in clinical research because it provides a defined measure of tumor change over an assessment period. It is not the same as overall survival, duration of response, or progression-free survival, which are separately listed outcomes in this record. The registry record does not report the observed objective response rate or its confidence interval, so the endpoint's definition is available but its result is not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Objective response rate is a binary-outcome summary. Each assessed participant is classified according to whether the specified complete- or partial-response criteria are met. The group percentage is calculated from the response classifications and the participants included in the endpoint analysis. The registry record does not report those analysis counts or the observed percentage, so no response estimate can be described from the record.

What does the confidence interval represent?

The primary outcome lists a 95% confidence interval as its dispersion measure. In general, a confidence interval communicates the uncertainty associated with an estimate. For a response percentage, it helps readers see whether the estimate is precise or whether the data remain compatible with a wider range of response rates. The interval depends on the observed response classifications and the analysis denominator, as well as the method used to calculate it. The registry record does not report the interval, its calculation method, or the numerator and denominator used for the primary endpoint. Therefore, neither the estimate nor its precision can be assessed.

A confidence interval is not a guarantee that the true value lies inside the particular interval. It is a statistical procedure with a long-run coverage interpretation under its assumptions. The interval also does not address design limitations such as the lack of a comparator, or potential biases in who entered the study and who had outcome assessment. Precision and validity are related but distinct questions: a narrow interval would concern random uncertainty, while design and measurement issues may affect how well the estimate applies beyond the study group.

Why does the single-group design matter?

The registry describes a single-group study, with allocation not applicable and masking none. There is no randomized comparison group. As a general principle, a treatment outcome observed in a single group cannot by itself distinguish the treatment's contribution from other explanations, including the natural course of disease, participant selection, concurrent care, or how outcomes are assessed. A comparison with a separate study or historical group would not have the protections of a concurrent randomized comparison unless the relevant differences and assumptions were addressed; those methods are not reported in the registry record.

Masking is listed as none. In general, lack of masking can matter when outcome assessment involves judgment. The response endpoint is tied to RECIST criteria, which provide defined lesion-change rules, but the record does not describe who assessed scans, whether review was independent, or how disagreements were handled. It would be inappropriate to infer those operational details.

How should the endpoint definition be read?

The stated complete-response rule is disappearance of all target lesions. The partial-response rule is at least 30% decrease in the sum of target-lesion diameters relative to the baseline sum. These are classification thresholds, not a reported finding that any participant achieved them. The endpoint combines complete and partial responses into a percentage, so it does not by itself show the distribution between those two response categories. The registry record does not report a separate count for either category.

The timeframe is through study completion, from baseline up to approximately 28 months. A timeframe defines the period of observation for the endpoint, but it does not establish that all participants were observed for the same duration or had an evaluable assessment. The registry record does not report the assessment schedule, the handling of missed assessments, or the definition of evaluability.

What is not described about the analysis?

The record does not report a statistical analysis plan, a sample-size justification, or interim analyses. It also does not state how missing outcome assessments were handled, what population was used for analysis, or whether any sensitivity analyses were conducted. These omissions matter because analysis choices can affect both the response percentage and its uncertainty. A denominator based on all enrolled participants can differ from one based only on participants with evaluable assessments; the record does not identify which approach was used.

As a general concept, over-dispersion concerns variation greater than a model's assumptions allow. It can lead to uncertainty estimates that are too small if ignored. The primary endpoint is reported as a percentage with a confidence interval, but the record does not identify a model or report enough information to assess whether over-dispersion was relevant or considered. No model-based result should be inferred from the endpoint description alone.

The record lists 14 outcome measures, while no analysis records are listed. An outcome list indicates what was intended for measurement, not that results were reported or analyzed in a particular way. The record does not specify how the analysis addressed multiple outcomes, whether outcome priorities were prespecified beyond identifying the primary endpoint, or whether any adjustment for multiple testing was planned.

Planning calculations, when used, answer a different question from observed trial results. A sample-size illustration based on an observed effect is a hypothetical design exercise and is not observed power. The registry record does not report the trial's own sample-size justification. It would therefore be misleading to use planning assumptions as evidence about the strength or certainty of the trial's findings.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Osimertinib")
n      <- c(22)
events <- c(17)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Osimertinib  22  17  77.3  54.6  92.2  56.6  89.9  56.3  89.8

11. Multiplicity, Missing Data and Analysis Populations

The registry record identifies objective response rate as the primary outcome and lists 14 outcome measures, including secondary outcomes such as overall survival, duration of response, disease control rate, tumor shrinkage, progression-free survival, and safety measures. It does not report a statistical analysis plan describing how multiple outcomes or multiple tests were to be handled. No adjustment strategy, hierarchy of secondary outcomes, or plan for controlling false-positive findings is reported. Without those details and without outcome analyses, the strength of any secondary finding cannot be assessed from the record.

Missing-data methods are also not reported in the registry record. In general, missing assessments can affect a response percentage if participants without an evaluable assessment are excluded or classified in a particular way. The record does not state whether missing assessments occurred, how they were handled, or whether sensitivity analyses were performed. It also does not define the population used for the primary analysis. The enrollment entry of 22 should not be assumed to be the analysis denominator.

The record lists no analysis records. It therefore does not provide results that would clarify the analysis population, missing-data handling, multiplicity, or the uncertainty around the primary estimate. These are not merely reporting details: they help readers understand which observations contribute to each result and how much confidence to place in inferences drawn from several outcomes.

12. 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: pulmonary embolism (9.1%), respiratory failure (9.1%), atrioventricular block (4.5%), elevated creatine phosphokinase levels in blood (4.5%), pleural effusion (4.5%).

The registry's adverse-event information records 10 affected among 22 at risk for serious adverse events. For other adverse events, it records 22 affected among 22 at risk. It also records 1 death. These are descriptive counts in the study group, not comparative risk estimates. The record does not report a control group against which to compare them, and the counts alone do not establish the cause of an event or death.

The serious-event terms listed with percentages are pulmonary embolism, 9.1%; respiratory failure, 9.1%; atrioventricular block, 4.5%; elevated creatine phosphokinase levels in blood, 4.5%; pleural effusion, 4.5%; and diarrhoea, 4.5%. These percentages describe the registry entries for the listed terms; the record does not give additional detail about event timing, duration, recurrence, or attribution. They should not be read as treatment-versus-control differences.

Safety interpretation also depends on exposure and follow-up, and the registry record does not provide enough detail to characterize those for each event. The regimen description allows treatment until progression or unacceptable side effects, up to 78 weeks from the first dose, but does not report individual exposure durations. The figures are therefore best read as reported safety observations, not as complete evidence about comparative safety.

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

The registry record establishes that the study examined osimertinib in a defined first-line population with locally advanced or metastatic non-squamous non-small cell lung cancer and both EGFR mutation and EGFR T790M mutation at diagnosis. It specifies a daily 80 mg regimen and defines objective response using revised RECIST guidelines, version 1.1. It also lists secondary efficacy and safety outcomes. These details describe the study question and measurement framework.

The record does not report the observed objective response rate or its 95% confidence interval. As a result, it does not show how many participants achieved complete or partial response, how precise an estimated percentage was, or whether participants had evaluable primary outcome assessments. A specified endpoint is not the same as a reported result. The available record therefore cannot establish the magnitude or precision of response in the study group.

Even if a response estimate were available, a single-group design would describe outcomes among participants receiving the study treatment rather than establish a causal difference versus another treatment. The lack of a comparator means that the record cannot provide a randomized estimate of relative efficacy or safety. General comparisons with other populations would require careful consideration of differences in eligibility, assessment, follow-up, and analysis.

The safety entries provide descriptive affected counts and serious-event terms, including a recorded death. They do not establish that the treatment caused each event, nor do they quantify excess risk relative to a control group. The registry record does not report the context needed for such a comparison.

Accordingly, the result that can be stated from this record is limited: the trial's primary endpoint and treatment regimen are defined, but the primary endpoint result is not reported in the registry record. The record does not support a conclusion about treatment benefit or lack of benefit. It also does not support a comparative safety conclusion. Readers should distinguish the design and endpoint information from numerical outcome evidence.

14. Why This Trial Matters Statistically

This trial is statistically notable because its registry entry clearly identifies a response endpoint but does not report the primary response estimate. That gap illustrates why a registry record's endpoint definition and its results must be read separately. Knowing how response was defined does not reveal the observed percentage, the analysis denominator, or the uncertainty around the estimate.

The single-group design also makes the estimand—what the study can describe—important. The study can characterize outcomes in the treated group if results are reported, but it cannot, on its own, estimate a randomized difference between osimertinib and a control. This distinction is central to interpreting single-arm response studies: an outcome proportion is not automatically a treatment effect relative to an alternative.

Precision matters as well as the point estimate. The primary outcome specifies a 95% confidence interval, but the interval is not reported. Without it, readers cannot assess the statistical uncertainty associated with the response estimate. The enrollment entry does not substitute for either an analysis denominator or an interval, and should not be used to infer the precision of an unreported result.

The record also lists a range of secondary outcomes and safety observations. Multiple outcomes raise questions about prespecification, analysis choices, and interpretation across endpoints. Because the analysis plan and analyses are not reported, those questions remain unresolved. Safety counts are informative as descriptive entries, but absence of a control arm limits comparative interpretation.

These features make the record a useful example of how design, endpoint definition, analysis reporting, and outcome availability jointly determine what can be concluded. Its most important statistical lesson is not a numerical effect estimate, because that estimate is absent; it is the need to separate a clearly stated research question from evidence that would answer it.

15. 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 main trial-specific limitation in the registry record is the absence of a reported result for the primary objective response rate. Although the endpoint definition and its 95% confidence-interval measure are stated, the record does not give the estimate, interval, response counts, or analysis denominator. The enrollment entry of 22 cannot be assumed to represent the participants assessed for response.

The study is single-group and unmasked, with no comparator described. Consequently, the registry record cannot support a randomized comparison of efficacy or safety. The primary outcome is based on response classifications under RECIST, but the record does not describe who assessed imaging, whether review was independent, or how assessments were timed and managed when incomplete.

Several analysis details are not reported: the analysis population, missing-data handling, statistical analysis plan, sample-size justification, interim analyses, and any sensitivity analyses. These omissions prevent assessment of how the reported endpoint was intended to be analyzed and how robust it would be to incomplete observations or alternative analysis choices. The record lists 14 outcome measures but no analysis records, leaving the status and interpretation of secondary outcomes unclear.

The safety data are also limited in scope. The registry includes affected and at-risk counts, a death count, and selected serious-event terms with percentages, but does not provide a comparator, individual exposure durations, event timing, or attribution details. The treatment description gives a maximum treatment period of up to 78 weeks, but does not establish actual exposure for each participant. These limitations constrain descriptive as well as comparative interpretation of safety.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Osimertinib (AZD9291) in First-line Locally Advanced or Metastatic NSCLC Patients With EGFR and EGFR T790M — NCT02841579.
• Registry record last updated 2025-01-28; status last verified 2025-01.
• Results first posted 2025-01-28.
• 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.

19. Record Summary

The registry record describes a completed phase IIa, single-group, unmasked study sponsored by MedSIR. It lists enrollment as 22 and concerns first-line patients with locally advanced or metastatic non-squamous non-small cell lung cancer, EGFR mutations, and concomitant EGFR T790M mutation at diagnosis. The intervention is osimertinib (AZD9291), 80 mg by mouth daily, until progression or unacceptable side effects, up to 78 weeks.

The primary outcome is objective response rate under revised RECIST guidelines, version 1.1, with a stated 95% confidence interval and a timeframe through study completion, from baseline up to approximately 28 months. The observed primary result and its confidence interval are not reported in the registry record. Secondary outcomes include survival, response duration, disease control, tumor shrinkage, mutation status, resistance, and safety. The safety entries include serious-event and other-event counts, a death count, and selected serious-event terms. Statistical analysis details, including missing-data methods and the analysis population, are not reported.