A Study Osimertinib in Patients With Stage 4 Non-small Cell Lung Cancer With Uncommon EGFR Mutations: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study Osimertinib in Patients With Stage 4 Non-small Cell Lung Cancer With Uncommon EGFR Mutations (NCT03434418), a na phase 2 study. The primary outcome is Objective Response Rate as Assessed by the Investigator Using Response Evaluation Criteria In Solid Tumors (RECIST 1.1).

2. Clinical Question

This clinical question is about whether tumor measurements met predefined response criteria in the study population. Under the endpoint definition, complete response involves disappearance of all target lesions and reduction of pathologic lymph nodes to the stated short-axis threshold. Partial response requires the stated decrease in the sum of the longest diameters of target lesions compared with baseline. These definitions make the outcome more explicit than a general judgment that a tumor improved.

The study record describes a population selected by cancer stage, EGFR mutation type, and prior treatment history. It does not give a response count or an estimated response proportion in the registry information available for this analysis. Therefore, the clinical question can be stated, but the recorded information does not show how many participants met the response definition. With no comparator group, any response observed in the study would describe outcomes among participants receiving osimertinib; it would not by itself establish how outcomes compare with another treatment or with no treatment.

3. Trial Design

This is a phase II, single-group study of osimertinib in patients with stage 4 non-small cell lung cancer and uncommon EGFR mutations. The registry lists Duke University as sponsor, enrollment of 17, a start date of 2018-06-30, primary completion on 2022-10-12, and status as COMPLETED. The study is not recorded as randomized: allocation is NA. Masking is recorded as NONE.

A single-group design means participants receive the named intervention without a concurrent control arm in this study. The primary outcome is objective response rate assessed by the investigator using RECIST 1.1, with a timeframe of up to 4 years. Other listed outcomes concern progression-free survival using RECIST 1.1, participants with adverse events using CTCAE Version 4.03, and overall survival followed through telephone or medical-record review.

The brief study description specifies uncommon mutations in EGFR exon 18 G719X, exon 20 S7681, or exon 21 L861Q, and states that participants had not previously received tyrosine kinase inhibitor treatment. The registry record does not provide a detailed protocol description, eligibility criteria beyond those summarized, dosing schedule, visit schedule, or a full account of how outcome assessments were conducted.

Because there is no randomized comparison, the study can describe outcomes in its enrolled group but cannot make a like-for-like comparison against a control group within the trial. A comparison with an external group would require additional information and methods not reported in the registry record. The record lists enrollment of 17, but does not explain whether that figure is a planned target, the final number enrolled, or the number included in each outcome analysis. The analysis populations and outcome-specific denominators 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:

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 as Assessed by the Investigator Using Response Evaluation Criteria In Solid Tumors (RECIST 1.1), measured over Up to 4 years.

The primary endpoint is objective response rate assessed by the investigator using RECIST 1.1. The registry describes the outcome as a count of participants: those with either a complete response or a partial response to the study drug. Its timeframe is up to 4 years. The registry record does not report the number of participants who achieved either response category.

RECIST provides operational rules for classifying changes in measurable tumors. In this record, complete response is defined as disappearance of all target lesions, together with reduction in the short-axis measurement of all pathologic lymph nodes to ≤10 mm. Partial response is defined as a ≥30% decrease in the sum of the longest diameter of target lesions compared with baseline. These criteria distinguish response categories using specified measurements rather than an undefined impression of improvement.

In general, a response endpoint can describe whether a participant crossed a prespecified threshold over a follow-up period. It does not, by itself, describe how long a response lasted, whether disease later progressed, or how survival compared with a different treatment. Those are separate questions addressed by other outcomes, such as progression-free survival and overall survival, which are also listed in this study record. Their results are not reported in the registry information available for this analysis.

The endpoint is relevant to the stated study question because it directly counts participants meeting the study's tumor-response definition. Interpretation still depends on the completeness and timing of assessments, the population analyzed, and the precision of the estimated response frequency. Those details and a response estimate are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a single-group design estimate?

In general, a single-group study describes outcomes among participants who receive the study intervention. For a binary endpoint such as whether a participant met the response definition, an analysis can report the number meeting the definition and the number assessed, and may estimate a response proportion with an interval expressing uncertainty. No primary response count, denominator for the response analysis, proportion, or interval is reported in the registry record for this study.

A single-group result is not a treatment contrast. A randomized comparison estimates the difference between groups under the study's allocation process; this study has no concurrent comparison group. Consequently, a response outcome in the enrolled group cannot separate the intervention's contribution from other influences on observed outcomes. This limitation is about design, not about whether the response classification itself is useful.

What do the RECIST categories contribute?

In general, prespecified response criteria help make outcome classification more consistent. RECIST 1.1 is named for the primary endpoint and for the listed progression-free survival outcome. The registry's response description distinguishes complete response from partial response using lesion disappearance, lymph-node measurement, and change from baseline in target-lesion diameters. Applying such rules still requires tumor assessments and decisions about evaluability. The registry information available for this analysis does not state how many participants had assessable measurements, how missing assessments were handled, or whether response was confirmed at a later assessment.

The endpoint records a count of participants with complete or partial response, rather than a continuous change in tumor measurement. A count can be clinically interpretable, but its precision depends on the analyzed denominator and number of outcome events. Those response-analysis details are not reported in the registry record. No uncertainty interval or inferential result for the primary endpoint is listed.

How should time-to-event outcomes be distinguished?

Progression-free survival and overall survival are time-to-event outcomes, unlike the primary count of participants with response. In general, time-to-event methods account for both the time until an event and participants whose event time is not observed during follow-up. Such participants may be censored, meaning their observed follow-up contributes information up to a point without treating them as having had the event at a known time. This explanation is general; the registry information does not describe the specific analysis methods, censoring rules, or estimates used for these outcomes.

The record names progression-free survival assessed using RECIST 1.1 and overall survival ascertained through telephone or medical-record review. It does not report median times, event counts for these outcomes, survival curves, hazard estimates, confidence intervals, or analysis populations. The listed outcome definitions alone do not establish what the results were.

What does a confidence interval communicate?

In general, a confidence interval gives a range of values compatible with an estimate under a statistical procedure and its assumptions. A narrow interval indicates greater statistical precision than a wide interval, but neither width nor location can correct a biased design or incomplete outcome collection. An interval should be interpreted together with the endpoint, denominator, and analysis population. No confidence interval for the primary response endpoint is reported in the registry record.

A p-value, when reported, addresses compatibility between data and a specified null model under its assumptions; it is not a measure of clinical importance or the probability that a treatment works. The registry record lists zero analyses, and no p-value or other primary-endpoint inferential result is reported. No conclusion about statistical significance can be drawn from an absent estimate.

What is over-dispersion, and does it apply?

In general, over-dispersion describes variability in count data that exceeds what a chosen probability model expects. It can matter when an analysis uses a count model, because standard errors may then be too small if extra variability is ignored. This study's primary endpoint is described as a count of participants with response, but the registry information does not identify a count-regression model or report an analysis that would permit assessment of over-dispersion. It would be inappropriate to assert that over-dispersion was present or that a particular model was used.

How should the planning illustration be read?

The planning illustration is described as a calculation based on an observed effect, using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It considers the effect to detect as a stated fraction of the observed effect. For time-to-event outcomes, the illustration is described as reporting a total number of events; for other outcomes, it reports participants per arm. These are planning assumptions, not observed trial results or observed power.

Those assumptions do not describe this study's actual design: the trial is single-group, with no allocation to two arms. The registry record does not report the protocol's sample-size justification, and the planning illustration should not be taken as evidence that the trial was designed or conducted with those assumptions. Any sample sizes from such calculations are hypothetical planning illustrations, not counts of participants observed in this study. The primary response estimate used as a basis for the illustration is not reported in the registry information available for this analysis.

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(17)
events <- c(8)
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  17  8  47.1  23.0  72.2  26.2  69.0  26.0  69.2

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists a primary response outcome and additional outcomes concerning progression-free survival, adverse events, and overall survival. When a study evaluates multiple outcomes or multiple analyses, multiplicity can arise: the chance of at least one apparently unusual result may increase when many statistical tests are conducted. In general, a prespecified hierarchy or adjustment method can help explain how multiple results are interpreted. The registry information does not report a multiplicity strategy, testing hierarchy, or adjustment method. That absence does not establish that no such plan existed; it means the method is not reported in the registry record.

The record lists zero analyses. It does not give a primary response result, an analysis population, an outcome-specific denominator, or a missing-data approach. It is therefore not possible to determine whether all enrolled participants were included in each endpoint analysis or how participants with unavailable assessments were treated. For the survival outcomes, censoring rules and follow-up completeness are also not reported in the registry record.

In general, missing outcomes can affect both the number of participants contributing to an analysis and the interpretation of its estimate. Treating missing outcomes as if they were known, excluding participants without explanation, or using a model-based method can lead to different results. No particular missing-data method can be attributed to this study from the registry information. Likewise, no intention-to-treat, per-protocol, safety, or other analysis population is specified. The enrollment figure of 17 should not automatically be assumed to be the denominator for every reported outcome.

The registry record does not report an interim-analysis plan, sample-size justification, or statistical analysis plan. Those items should be treated as unreported rather than inferred from the study's phase or completion status.

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: respiratory failure (5.9%), thromboembolic event (5.9%).

The registry lists the osimertinib group with 17 participants at risk for the serious-adverse-event measure and 2 affected. For other adverse events, it lists 17 at risk and 17 affected. It also lists 12 deaths. These are distinct registry entries and should be read according to their stated categories; the record does not provide enough detail to infer the timing or relationship between these entries.

The top serious adverse-event terms are respiratory failure and thromboembolic event, each listed as 5.9%. These percentages are reported as written. The registry information does not specify whether the terms represent participants, events, or another reporting unit, nor does it provide additional context such as event timing or attribution. The counts for serious and other adverse events should not be treated as proof that osimertinib caused those events.

In general, safety counts are most interpretable when the reporting period, event definitions, severity grading, participant exposure, and attribution rules are clear. The record names CTCAE Version 4.03 for the adverse-event outcome, but does not include detailed event-level listings or a description of exposure time. The death count is not accompanied by a cause or an attribution statement in the information available for this analysis. These figures describe the posted safety information; they do not establish comparative risk because the study has no control group.

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

The study addresses an important statistical question: how often participants in a defined group met an objective tumor-response criterion while receiving osimertinib. It enrolled 17 participants and is recorded as a completed phase II single-group study. The outcome definition is explicit about complete and partial response, and the record also identifies safety and survival outcomes.

However, the registry record does not report the primary response count or estimate. It therefore does not show how many participants met the response definition, how precise an estimate would be, or whether responses were observed for a particular duration. The record also does not provide results for progression-free survival or overall survival in the information available for this analysis. The absence of those results should not be interpreted as a finding of no response or no effect.

Even if a response result were available, the single-group design would limit causal interpretation. Without a concurrent randomized comparison, observed outcomes cannot be directly contrasted with outcomes under another treatment within this trial. Differences between participants in this study and people treated in other settings could also influence an external comparison. A single-group study can describe its enrolled population, but the registry information does not support a like-for-like treatment comparison.

The safety entries are descriptive as well. The record reports serious adverse events affecting 2 participants, other adverse events affecting 17 participants, and 12 deaths. It does not establish that the study drug caused these outcomes, and it provides no control group for comparing their frequency. The serious-event terms and percentages should be interpreted within the limits of the reporting detail available.

In statistical terms, the record identifies the question and several outcome definitions, but leaves key results and analysis details unreported. It supports a description of the study design and posted safety counts; it does not, on its own, support a quantified efficacy conclusion, a treatment comparison, or an account of the uncertainty around the primary endpoint.

14. Why This Trial Matters Statistically

This study is statistically notable because it focuses on a specific clinical population: stage 4 non-small cell lung cancer with uncommon EGFR mutations, including exon 18 G719X, exon 20 S7681, or exon 21 L861Q. The brief description also identifies participants without previous tyrosine kinase inhibitor treatment. Clearly defining the population matters because results from one group do not automatically describe people with different mutation types or treatment histories.

The primary endpoint uses RECIST 1.1 and names explicit response criteria. That is useful from a measurement perspective: a reader can see what the study counted as complete or partial response rather than relying on an unspecified clinical judgment. The record also identifies related outcomes, including progression-free survival, adverse events, and overall survival, which represent different dimensions of outcome and require distinct interpretations.

At the same time, this study illustrates why outcome definitions and outcome results must be considered separately. The registry identifies objective response as the primary question but does not report the response count or an estimate in the information available for this analysis. A well-defined endpoint cannot substitute for a posted result. Likewise, a completed status does not itself communicate an effect estimate, precision, or comparison.

The single-group design is also a central statistical feature. It can describe outcomes among the participants treated, but it does not create a randomized control comparison. For readers, recognizing this distinction prevents descriptive findings from being mistaken for evidence of a comparative treatment effect. The reported safety counts are similarly descriptive and do not establish causation or comparative risk.

Finally, the record makes clear how much interpretation depends on analysis details. The analysis populations, missing-data methods, multiplicity approach, and primary statistical result are not reported. Identifying these gaps is part of a careful statistical reading, especially when a small enrolled group and a specialized population make transparent reporting important.

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.

Several limitations are specific to interpreting this registry record. First, the primary endpoint result is not reported in the information available for this analysis. There is no response count, denominator for the response analysis, estimated response frequency, or uncertainty interval. This prevents a quantitative summary of the primary outcome.

Second, the study has a single-group design, with allocation recorded as NA and masking recorded as NONE. There is no concurrent control group for estimating a randomized contrast. The registry record itself notes that this is not a randomized comparison and that arms are not a like-for-like control comparison. Any observed outcomes would therefore be descriptive of the enrolled group rather than a controlled comparison within the study.

Third, key analysis and follow-up details are not reported in the registry information. These include the analysis populations, outcome-specific denominators, missing-data handling, response-assessment completeness, survival censoring rules, and the statistical analysis plan. The listed timeframe for the primary endpoint is up to 4 years, but the record does not provide individual follow-up information or the timing of response assessments.

Fourth, the registry lists safety counts and serious-event terms but does not provide event-level detail, timing, or causality attribution. The 12 deaths are not described by cause or relation to treatment in the information available for this analysis. Safety frequencies also lack a concurrent comparator.

Finally, the brief study description names specific mutation types and prior-treatment history, but the information available does not give further details about eligibility or the distribution of participants across those mutation categories. Generalization beyond the defined study population should therefore be cautious. The record lists enrollment of 17, but does not state how many participants contributed to each outcome analysis.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study Osimertinib in Patients With Stage 4 Non-small Cell Lung Cancer With Uncommon EGFR Mutations — NCT03434418.
• Registry record last updated 2023-10-17; status last verified 2023-10.
• Results first posted 2023-10-17.
• 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 II, single-group study sponsored by Duke University. It lists enrollment of 17, a study start date of 2018-06-30, and primary completion on 2022-10-12. The intervention is osimertinib; allocation is NA and masking is NONE.

The study concerns stage 4 non-small cell lung cancer with uncommon EGFR mutations: exon 18 G719X, exon 20 S7681, or exon 21 L861Q. The brief description states that participants had not previously received tyrosine kinase inhibitor treatment. The primary outcome is investigator-assessed objective response using RECIST 1.1, with a timeframe of up to 4 years.

The registry also lists progression-free survival, adverse events, and overall survival. It reports serious adverse events affecting 2 participants, other adverse events affecting 17 participants, and 12 deaths; respiratory failure and thromboembolic event are each listed at 5.9% among the top serious adverse-event terms. The primary response result, outcome analysis populations, and statistical analysis details are not reported in the registry record.