Osimertinib for NSCLC With EGFR Exon 20 Insertion Mutation: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Osimertinib for NSCLC With EGFR Exon 20 Insertion Mutation (NCT03414814), a na phase 2 study. The primary outcome is Objective Response Rate.

2. Clinical Question

The clinical question is descriptive rather than a direct comparison of treatments. It asks whether objective tumor responses were observed in the enrolled group under the study’s assessment approach. The registry record says the participants’ disease had progressed on standard chemotherapy, which defines the study population described in the brief summary. It does not report detailed eligibility criteria or explain how participants were selected beyond the stated disease and mutation characteristics.

The primary outcome is objective response rate (ORR), based on complete response (CR) or partial response (PR) as the best response. That endpoint captures documented tumor shrinkage meeting the stated response criteria. It is not, by itself, a measure of how long a response lasted, how long participants lived without progression, overall survival, symptoms, or quality of life. Progression-free survival and overall survival are listed as secondary outcomes, along with serious adverse events.

The primary endpoint’s result is not reported in the registry record. Consequently, the registry information does not answer how many participants had a confirmed response or how precisely any response proportion was estimated. The clinical question and the trial design can be described, but the observed efficacy result cannot be reconstructed from the record.

3. Trial Design

This was a phase II, single-group study of osimertinib in NSCLC with an EGFR exon 20 insertion mutation. The registry describes it as non-randomized and multicentre. Allocation is recorded as NA, masking as none, and the model as single group. The study enrolled 15 participants and is recorded as completed.

Participants received osimertinib at 80mg orally once daily. Because there was no concurrent control group, the design does not provide a randomized comparison with another treatment or with no study treatment. It is therefore suited to describing outcomes observed in this enrolled group, but it cannot isolate treatment effects from other explanations for the outcomes. Differences in participant characteristics, the natural course of disease, prior treatment, and assessment practices cannot be balanced by random allocation in a single-group design.

The registry lists the primary outcome as investigator-assessed, confirmed objective response by RECIST version 1.1. Its timeframe runs from initiation until the first documented progression, unacceptable toxicities, or withdrawal, whichever came first, with the registry describing the period as up to about 29months. The primary outcome population is described as the proportion of participants among those administered 80mg osimertinib whose best response was CR or PR. The record does not give a separate evaluable population or explain how response assessments were handled when follow-up was incomplete.

The study began on 2018-01-04, had a primary completion date of 2021-08-03, and has completed status in the registry. The sponsor is Seoul National University Hospital. These dates and status describe the registry entry; they do not establish when every participant was assessed or whether every planned analysis was completed. The registry record does not report a protocol-level statistical plan, sample-size justification, interim analyses, or the results for the primary outcome.

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

Analysis population for the primary outcome, as defined in the registry: The proportion of patients who had complete reponse (CR) and partial reponse (PR) as the best response among 80mg osimertinib administered 15 patients.

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: Patients with NSCLC harboring EGFR exon 20 insertions

6. Primary Endpoint

The registry lists the primary outcome as Objective Response Rate, measured over Until study completion, from date of initiation until the date of first documented progression, unacceptable toxicities or withdrawl, whichever came first. (upto about 29months).

The primary endpoint is objective response rate, defined in the registry as the proportion of participants whose best response was a complete response or partial response. The response was investigator-assessed, confirmed, and based on RECIST version 1.1. The endpoint therefore reduces tumor assessment to whether a participant met the stated criteria for either of those response categories during the specified follow-up period.

As a binary outcome, each participant contributes to the response count according to whether the endpoint criterion was met. The corresponding proportion requires both a numerator—the number with a qualifying response—and a denominator—the relevant number assessed. Although the registry describes the intended population and reports total enrollment, it does not report the observed response count, the response proportion, or a separate denominator for participants with evaluable assessments. The primary result is therefore not reported in the registry record.

In general, objective response is used to describe observable tumor reduction under a defined assessment framework. It can provide a direct summary of whether a response occurred, but it does not by itself describe response duration, time to progression, survival, symptoms, or treatment-related harms. Those are distinct outcomes and should not be inferred from an ORR result. The registry lists progression-free survival and overall survival as secondary outcomes, but their results are not reported in the registry record.

The endpoint is based on investigator assessment; the registry does not describe independent central review. It also does not provide the schedule of tumor assessments or details about handling missing, unevaluable, or unconfirmed assessments. Those details matter when interpreting a response proportion because they determine which observations can count and how participants without a recorded assessment are treated.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What kind of outcome is the primary endpoint?

The primary endpoint is a binary participant-level outcome: a participant either met the response definition or did not. The response definition includes confirmed CR or PR as the best response, assessed by investigators according to RECIST version 1.1. In general, a binary endpoint can be summarized as a count of participants meeting the criterion and a proportion calculated using a defined analysis population. The registry describes the intended proportion but does not report the response count, the resulting proportion, or an endpoint-specific analysis population. No estimated response rate can therefore be interpreted from this entry.

What would a confidence interval convey?

In general, a confidence interval gives a range of values compatible with the observed data under a specified statistical method. For a response proportion, it helps show how uncertain the estimated proportion is, rather than treating the observed value as exact. Its width depends on the amount of information in the data and the method used. The registry record does not report a confidence interval for objective response, and it does not identify a method for estimating one. No interval should be inferred from enrollment alone.

Why does the study structure matter?

This is a single-group, non-randomized study, with no concurrent comparator. In general, random allocation helps make treatment groups comparable with respect to measured and unmeasured characteristics, on average. Without a randomized control group, a study’s observed outcomes cannot be interpreted as a randomized treatment contrast. Participant selection, disease course, prior treatment, and care during follow-up may all contribute to observed outcomes. A single-group result can describe what happened in the enrolled group, but it cannot by itself distinguish treatment effects from these other influences.

Masking is recorded as none. In general, masking can reduce the opportunity for knowledge of treatment assignment to influence care, outcome assessment, or reporting. Because no masking is recorded, the registry does not describe a blinded assessment process. The primary outcome is investigator-assessed, and the record does not indicate that an independent review was used. This is relevant context for interpreting an assessment-based endpoint; it does not show that any assessment was biased.

What is the role of the planning illustration?

The sample-size assumptions describe planning illustrations based on the observed effect, not the trial’s own design. They specify a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is a stated fraction of the observed effect. For time-to-event outcomes, the stated convention is a total number of events using Schoenfeld; other outcomes are expressed as participants per arm. These assumptions are not evidence that this single-group study was designed with a comparator or that it enrolled participants according to those planning scenarios.

In general, planning illustrations ask how much information a hypothetical design might require under specified assumptions. Their results depend on the assumed effect, statistical method, and design. They are not observed power, and they do not establish the probability that the study’s result is correct. Here, the registry does not report a protocol sample-size justification or the observed primary endpoint effect needed to assess the planning exercise. The illustration should therefore be kept separate from the trial’s actual design and from any efficacy conclusion.

What does over-dispersion mean in general?

For count outcomes, a statistical model may assume that variability follows a particular relationship to the average count. Over-dispersion means that the observed counts vary more than that model allows. If it is ignored, uncertainty estimates may be too small and tests may appear more precise than warranted. This general issue is relevant to some count-data analyses, but the primary endpoint in this study is defined as a participant response proportion, not a count rate. The registry record does not report a dispersion parameter or an analysis model for the primary endpoint, so no conclusion about over-dispersion in this study is possible.

What analysis results are recorded?

The registry records zero analyses. It does not report a primary endpoint estimate, a p-value, a confidence interval, or analysis details such as an estimation method. This absence should be read as a limitation of the registry entry rather than evidence that investigators did not conduct analyses. Without posted results, statistical significance, precision, and sensitivity to analysis choices cannot be evaluated. The registry record also does not report interim analyses or explain whether outcome analyses followed a prespecified statistical analysis plan.

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(15)
events <- c(0)
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  15  0  0.0  0.0  21.8  0.0  20.4  0.2  20.6

11. Multiplicity, Missing Data and Analysis Populations

The registry lists objective response rate as the primary outcome and lists serious adverse events, progression-free survival, and overall survival as secondary outcomes. It also records four outcome measures. Testing or estimating several outcomes can create multiplicity concerns. In general, when many statistical hypotheses are tested, the chance of at least one apparently notable result can increase unless the analysis plan addresses that issue. The registry record does not report a multiplicity adjustment, a hierarchy among outcomes, or a prespecified testing strategy. It is therefore not possible to determine how multiple outcomes were handled.

The record does not report missing-data methods for the primary endpoint or the secondary outcomes. It does not explain how participants who withdrew, had no post-baseline assessment, or had an unevaluable response were included in the denominator. The endpoint timeframe refers to follow-up until progression, unacceptable toxicities, or withdrawal, whichever came first, but the record does not say how these events affected the efficacy analysis. In general, missing outcome data can affect both the estimated proportion and its uncertainty, particularly when the reason data are absent relates to a participant’s outcome.

Analysis populations are also not described beyond the primary outcome’s reference to patients administered osimertinib. The record does not state whether all enrolled participants were included in the primary analysis, whether any participants were excluded, or how protocol deviations were addressed. No imputation method or sensitivity analysis is reported. These omissions limit the ability to assess whether the endpoint definition was applied consistently or how robust any result might be to assumptions about missing observations. Because the primary result itself is not reported in the registry record, neither missing-data handling nor multiplicity can be assessed against an observed estimate.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: pneumonia (13.3%), colitis (6.7%), azotemia (6.7%), hyperglycemia (6.7%), general weakness (6.7%).

The registry records 15 participants at risk for serious adverse events, with 8 affected. It also records 15 participants at risk for other adverse events, with 15 affected, and 2 deaths. These are counts from the osimertinib group. They are not comparative estimates because the study had no control group.

The serious adverse event terms listed include pneumonia at 13.3%, and colitis, azotemia, hyperglycemia, general weakness, and acute urinary retention at 6.7% each. The registry presents these as terms associated with serious adverse events. It does not provide enough detail to determine whether participants experienced more than one listed term, how events were timed, how long they lasted, or whether investigators judged them related to treatment. Percentages for individual terms should not be added or treated as mutually exclusive categories.

In general, adverse-event counts describe events recorded under a particular study’s definitions and follow-up procedures. A serious adverse event is not interchangeable with any adverse event, and a death count alone does not establish cause. The registry record does not provide a comparator, detailed event narratives, exposure-adjusted rates, or causal attribution for these counts. They should be read as descriptive safety information for this enrolled group, not as a causal estimate or a direct comparison with another treatment.

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

The registry record establishes that a completed phase II study enrolled 15 participants in a single-group, non-randomized evaluation of osimertinib at 80mg once daily. It identifies a primary endpoint of investigator-assessed, confirmed objective response according to RECIST version 1.1. It also lists safety outcomes, progression-free survival, and overall survival. These details clarify what the study was designed to examine.

The record does not report the primary endpoint result. It therefore does not establish how many participants had a confirmed CR or PR, the resulting response proportion, or the uncertainty around that proportion. It also does not report the secondary outcome results. No claim about the magnitude, statistical significance, or precision of an efficacy effect can be made from the endpoint definition or enrollment information alone.

Even if a response result were available, the single-group, non-randomized design would constrain what it could mean. In general, outcomes observed in a treated group do not provide the same evidence as a randomized comparison. Without a concurrent control group, the observed course cannot be separated reliably from factors such as participant characteristics, disease progression, prior treatment, and other care. The study can describe its enrolled group, but it does not furnish a like-for-like treatment contrast.

The serious adverse event and other adverse event counts provide descriptive safety information, including 8 participants with serious adverse events and 2 deaths. They do not by themselves establish causation or comparative safety. The registry’s event counts lack the contextual detail needed to determine attribution or compare risks against another group.

Thus, the record supports a description of the study population, intervention, endpoint definition, design, and selected safety counts. It does not support a conclusion about the primary response result, a causal treatment effect, comparative efficacy, or comparative safety. The distinction is not a judgment about the intervention; it reflects what can and cannot be interpreted from the results entered in the registry record.

14. Why This Trial Matters Statistically

This trial is statistically informative as an example of how design determines the kinds of conclusions a result can support. The study is phase II and single-group, with no random allocation or concurrent comparator. That structure means the central statistical task would be to describe outcomes in the treated participants, rather than estimate a randomized difference between treatments. Design details are essential context for interpreting any outcome estimate, not just technical annotations.

The primary endpoint is a binary response outcome with a clear stated basis: confirmed CR or PR as best response, assessed by investigators using RECIST version 1.1. In general, response proportions are easy to communicate, but their interpretation depends on a clear numerator, denominator, assessment process, and treatment of missing observations. The registry record defines the intended measure but does not report its result or analysis population. That gap demonstrates why a well-described endpoint is not a substitute for posted outcome data.

Precision is also central. An observed proportion is an estimate, and an uncertainty interval helps readers judge how much information supports it. The registry record does not report the response estimate or its interval. Enrollment alone cannot fill that gap. Similarly, planning illustrations based on an observed effect are not observed power and should not be mistaken for the original design justification. The record does not report a protocol-level sample-size justification.

Finally, safety counts and efficacy outcomes answer different questions. The registry records adverse events and deaths, but without a control group or attribution details those observations remain descriptive. Keeping the design, endpoint, uncertainty, missingness, and safety context distinct is fundamental to a careful statistical reading. In this case, the registry supports that methodological discussion while leaving the central efficacy result unreported.

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 most consequential limitation of the registry record is that it does not report the primary endpoint result. The number of participants with confirmed objective response, the estimated response proportion, and its uncertainty are absent. Results for progression-free survival and overall survival are also not reported in the registry record, despite being listed as secondary outcomes. This prevents assessment of the observed efficacy outcomes and their precision.

The single-group, non-randomized design has no concurrent control condition. Consequently, it cannot provide a randomized comparison, and any observed outcomes would be descriptive of the enrolled group. The record also reports no masking. Since the primary endpoint was investigator-assessed, the absence of a described independent review is relevant; however, the registry does not establish whether any assessment bias occurred.

The record does not provide detailed eligibility criteria, participant characteristics, the number completing follow-up, or a distinct analysis population. It does not explain how missing or unevaluable response assessments were handled, how protocol deviations were treated, or whether sensitivity analyses were conducted. The timing and schedule of tumor assessments are not described beyond the stated endpoint timeframe. These gaps limit evaluation of outcome ascertainment and the denominator underlying any intended response proportion.

Statistical methods are not documented in sufficient detail to assess an analysis plan. The registry records zero analyses and does not report a confidence interval, p-value, interim analysis, multiplicity approach, or protocol-level sample-size justification. The record also does not provide detailed attribution or timing for safety events. These limitations concern the information available in the registry record; they should not be read as evidence that unreported procedures or analyses were not conducted elsewhere.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Osimertinib for NSCLC With EGFR Exon 20 Insertion Mutation — NCT03414814.
• Registry record last updated 2025-02-11; status last verified 2025-01.
• Results first posted 2025-02-11.
• 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

NCT03414814 is a completed phase II, multicentre, single-group, non-randomized study sponsored by Seoul National University Hospital. It enrolled 15 participants with locally advanced or metastatic NSCLC and an EGFR exon 20 insertion mutation whose disease had progressed on standard chemotherapy. Participants received osimertinib 80mg orally once daily.

The primary outcome was investigator-assessed, confirmed objective response by RECIST version 1.1, defined as CR or PR as best response. The primary result is not reported in the registry record. Serious adverse events affected 8 of 15 participants; other adverse events affected 15 of 15, and 2 deaths were recorded. Progression-free survival and overall survival are listed as secondary outcomes, but their results are not reported. The record lists four outcome measures and zero analyses.