Atezolizumab Immunotherapy in Patients With Advanced NSCLC: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Atezolizumab Immunotherapy in Patients With Advanced NSCLC (NCT03102242), a na phase 2 study. The primary outcome is Disease Control Rate (DCR) After 12 Weeks Induction.

2. Clinical Question

The clinical question is about activity during an induction period in a selected stage III population, rather than about a direct treatment comparison. The registry describes a broader treatment objective involving immunotherapy bracketing chemoradiation and consolidation, while the listed intervention details focus on induction atezolizumab. The record does not give enough detail to reconstruct every component or decision rule in that broader sequence. Nor does it report a comparative estimate that could establish how outcomes differed from another strategy. The primary endpoint result is not reported in the registry record, so the central question cannot be answered from the posted outcome information alone.

3. Trial Design

This was a Phase II, single-group trial sponsored by Alliance Foundation Trials, LLC. The registry lists enrollment as 64 and the study status as completed. The recorded start date is 2017-11-01, and the primary completion date is 2020-06-01. The condition was carcinoma, non-small-cell lung, and the study title and eligibility description focus on unresectable stage IIIA and IIIB disease among patients eligible for chemoradiotherapy with curative intent.

The intervention was atezolizumab. The treatment description specifies induction immunotherapy at 1200 mg IV every 21 days for 4 cycles, with restaging after cycle 2 and cycle 4. The registry description includes a treatment decision for patients with progression at the post-cycle 2 assessment, but the recorded description is incomplete, so the full set of subsequent treatment rules is not reported in the registry record. The stated broader objective concerns anti-PD-L1 therapy around standard chemoradiation and consolidation therapy.

Masking is recorded as none. Allocation is recorded as not applicable, consistent with the single-group design. There is no randomized control arm in the record, so the study is not designed as a head-to-head efficacy comparison. A single-group study can describe outcomes among enrolled participants, but without a concurrent comparator it cannot separate the intervention's association with outcomes from other influences, including the underlying disease course and subsequent care.

The registry lists the primary endpoint as disease control rate after 12 weeks of induction. Secondary outcomes include objective response rate, median progression-free survival, progression-free survival at 12 and 24 months, median overall survival, and overall survival at 12 and 24 months. Numerical outcome findings for these measures are not reported in the registry record. Detailed eligibility criteria, the complete treatment protocol, the statistical analysis plan, and the rationale for the enrollment target are also 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 Disease Control Rate (DCR) After 12 Weeks Induction, measured over 12 weeks.

The primary endpoint is disease control rate after 12 weeks of induction immunotherapy, expressed as a proportion of participants. In general, a proportion endpoint describes how many participants in a defined group meet an outcome criterion. Disease control is intended to capture a category of outcome broader than objective response alone, but the response categories or assessment rules included in this trial's definition are not reported in the registry record. Without those rules, the endpoint cannot be independently reconstructed from individual participant outcomes.

The registry describes the study's primary objective as determining whether an anti-PD-L1 treatment strategy around standard chemoradiation and consolidation therapy is worthy of further investigation. A disease-control measure can be useful in an early-phase study because it summarizes whether disease met the protocol's control criteria at a specified assessment point. It is not, by itself, a comparison with an alternative treatment and does not establish what would have happened to the same participants under another strategy.

The record specifies a 90% confidence interval for the primary endpoint, but does not report the disease-control count, estimated proportion, or interval limits. Consequently, neither the observed level of disease control nor its statistical precision can be described from the registry record. The missing definition and numerical result also prevent assessment of how the endpoint was applied in practice.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry record describes a binary-style primary outcome: disease control at a defined assessment point, summarized as a proportion of participants. In general, a binary outcome assigns each participant to an outcome category, such as meeting or not meeting a prespecified criterion. The proportion is interpretable only when the criterion, assessment time, and denominator are clear. In this trial, the endpoint time is listed as 12 weeks, but the exact disease-control criteria and the number of participants included in the primary analysis are not reported in the registry record.

What does a confidence interval describe?

In general, a confidence interval expresses the uncertainty around an estimated quantity under the statistical method used to construct it. For a proportion, it gives a range of values compatible with the observed data and the method's assumptions. A narrower interval indicates greater precision, while a wider interval indicates less precision. An interval does not eliminate uncertainty, and it is not a guarantee that the true value lies within its limits. The registry specifies a 90% confidence interval for the primary endpoint, but it does not report the interval or the estimate it would accompany. The method used to calculate the interval is also not reported in the registry record.

Because the primary outcome count and estimate are absent, no confidence interval can be interpreted from the posted information. Nor can the interval's width be used to assess precision. The stated confidence level alone is not an outcome finding: it describes a planned or reported way of expressing uncertainty, not the observed amount of disease control.

Why does the single-group design matter?

In general, a single-group study estimates outcomes within the enrolled group. It does not directly estimate a difference between that group and a concurrent control group. A randomized comparison helps balance, on average, factors that might affect outcomes in either arm; this study has no randomized comparison. As a result, the observed outcome in a single group can be influenced by participant characteristics, disease course, assessment practices, and later care. Those possibilities limit causal interpretation, even if a numerical endpoint estimate is available.

The registry records allocation as not applicable and identifies the model as single group. It also records no masking. These design characteristics should be considered when interpreting subjective assessments or safety reports, although the registry does not describe assessment procedures or the degree of subjectivity in the endpoint. Further details about independent review or blinded assessment are not reported in the registry record.

What can an endpoint proportion establish?

In general, a proportion is a descriptive summary of the fraction of a defined analysis group meeting a criterion. Its meaning depends on how the outcome is classified and which participants are counted. If some participants lack an assessment, different missing-data rules can produce different denominators or classifications. Such choices should be prespecified and transparent. The disease-control definition, missing-data rule, and primary analysis population for this trial are not reported in the registry record. The proportion itself is also absent, so there is no estimate to interpret even descriptively.

The record lists objective response rate as a secondary outcome in addition to disease control rate. These are distinct endpoint labels, but the operational definitions and assessment criteria for either are not reported. In general, response and disease control summaries depend on specified criteria and assessment timing; labels alone do not establish what response categories were counted.

How should survival endpoints be understood?

In general, time-to-event outcomes describe the time until a defined event, while accounting for the fact that some participants may not have had the event by the end of observation. Their interpretation depends on the event definition, the start point, follow-up, censoring rules, and statistical method. The registry lists median progression-free survival, progression-free survival at 12 and 24 months, median overall survival, and overall survival at 12 and 24 months. It does not report numerical estimates, event definitions, follow-up details, censoring methods, or analysis procedures for these outcomes. No survival result or precision estimate can therefore be assessed from this record.

What is known about the analysis plan?

The registry lists no analysis entries. The statistical analysis plan, including any hypothesis test, estimation method, covariate adjustment, or interim analysis, is not reported in the registry record. The record states that the primary endpoint has a 90% confidence interval, but does not identify the calculation method. The original sample-size justification is also not reported in the registry record.

In general, a planning illustration based on an observed effect is not an observed-power result and does not describe the study's original design rationale. Such an illustration depends on its assumptions, including the effect selected for detection and the allocation and statistical method. The registry's planning assumptions describe a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the target effect expressed as a fraction of an observed effect. These are assumptions for a planning illustration, not evidence that the trial itself had two arms: the trial is recorded as single group. The observed endpoint effect and numerical planning outputs are not reported in the registry record, so no planning sample size can be interpreted.

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("Arm A (Treatment)")
n      <- c(62)
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  Arm A (Treatment)  62  0  0.0  0.0  5.8  0.0  5.8  0.0  5.7

11. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary endpoint and several secondary outcomes: objective response rate, median progression-free survival, progression-free survival at 12 and 24 months, median overall survival, and overall survival at 12 and 24 months. When multiple outcomes are examined, the chance of at least one apparently notable result can increase if each is assessed separately without adjustment. In general, multiplicity procedures can control or describe this issue, but no such procedure is reported in the registry record. The record also contains no analysis entries, so it does not show which outcomes were analyzed or how any multiple testing was handled.

Missing-data methods are not reported in the registry record. This is important for a proportion endpoint because the result depends on how participants without a 12-week assessment are treated and which denominator is used. The record does not state how missing endpoint assessments, withdrawals, or other unavailable outcome data were handled. It also does not give the number of participants assessed for the primary endpoint.

Analysis-population definitions are not reported in the registry record. The enrollment figure is 64, but the record does not identify whether analyses used all enrolled participants, participants who began treatment, or another defined group. In general, different analysis populations can answer different questions and yield different estimates. The record does not report whether these populations were prespecified or whether sensitivity analyses were conducted. The lack of these details limits assessment of the endpoint's denominator, the robustness of any estimate, and the interpretation of the secondary 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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: lung infection (12.5%), pneumonitis (7.8%), sepsis (6.2%), colitis (4.7%), fever (4.7%).

The registry reports safety information for Arm A (Treatment), with 64 participants at risk for both serious and other adverse events. Serious adverse events affected 34 participants, and other adverse events affected 59 participants. The record also reports 18 deaths. These are separate posted counts; the registry information does not establish that the event categories are mutually exclusive or explain how the death count relates to adverse-event attribution.

The listed serious adverse-event terms and percentages are lung infection, 12.5%; pneumonitis, 7.8%; sepsis, 6.2%; colitis, 4.7%; fever, 4.7%; and infusion related reaction, 4.7%. The record does not provide detailed event-level information, the denominator for each listed percentage, or a comparison group. It also does not report whether each event was considered related to atezolizumab. The percentages should therefore be read as registry-listed safety summaries, not as comparative estimates or proof of causation.

In general, serious adverse events are classified according to seriousness criteria, whereas severity describes intensity; the terms are not interchangeable. The registry record does not provide enough detail to evaluate the grading system or collection procedures used. The counts describe reported events in this treatment group and do not, on their own, establish the cause of an event or the relative safety of a different treatment strategy.

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

The record establishes that this was a completed Phase II, single-group study of induction atezolizumab in patients with unresectable stage IIIA and IIIB non-small-cell lung cancer eligible for chemoradiotherapy with curative intent. It identifies disease control rate after 12 weeks as the primary endpoint and specifies a 90% confidence interval. It also gives enrollment and selected safety counts. These registry entries describe the trial's design and what was recorded, but they do not supply the primary endpoint result.

Without a disease-control count, proportion, and interval, it is not possible to characterize the observed primary outcome or its precision. The secondary outcome labels likewise do not establish what was observed: numerical findings for response, progression-free survival, and overall survival are not reported in the registry record. An endpoint name is not a result, and the existence of a planned time point does not establish an outcome at that time.

The design does not include a concurrent comparator. Consequently, even if outcome estimates were available, they would describe the enrolled treatment group rather than a randomized contrast between atezolizumab and another strategy. The registry record does not support a causal estimate of the intervention's effect relative to an alternative. It also does not report enough information to reconstruct the complete treatment pathway or the analytic methods.

The safety counts and event terms provide descriptive information for the treatment group. They do not establish whether events were caused by treatment, how risks compare with another strategy, or how the observed counts would generalize beyond the enrolled group. The record's safety summaries should not be interpreted as a complete account of all possible harms.

Thus, the registry information supports a limited account of the study's question, design, and selected safety findings. It does not provide the numerical efficacy results needed to determine the observed disease-control rate or the listed secondary outcomes. The absence of those results should be treated as missing information, not as evidence of benefit, lack of benefit, or no difference.

14. Why This Trial Matters Statistically

This trial illustrates how the design and reporting of an early-phase clinical study shape the questions that statistics can answer. A single-group Phase II design can describe outcomes in a defined treatment cohort and can motivate further investigation, which matches the registry's stated objective of deciding whether the strategy was worthy of further study. It cannot, by itself, estimate a randomized treatment contrast. That distinction remains important even when an endpoint is clearly named.

The primary endpoint is a proportion at a specified assessment time, and the registry identifies a confidence interval as part of its reporting framework. In general, an estimate and its interval are needed together: the estimate describes the observed outcome, while the interval conveys statistical uncertainty. A confidence level without the result does not show either the outcome's magnitude or its precision. Here, the registry record names the endpoint but does not report its numerical result.

The trial also lists response and survival outcomes, which address different aspects of outcome measurement and follow-up. Each requires clear definitions, denominators, and analysis methods for meaningful interpretation. The registry record does not report numerical findings or those analytic details. This makes the distinction between an outcome's inclusion in a registry and an interpretable result especially relevant.

Finally, safety counts need context. Counts from one treatment group can describe what was recorded among enrolled participants, but without comparative data they cannot establish relative safety or causation. Clear attribution, definitions, and analysis populations help readers understand such summaries. This record offers selected counts, while leaving those details unreported. The statistical lesson is that the design, outcome definitions, analysis population, uncertainty estimates, and completeness of reporting jointly determine what conclusions the evidence can support.

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 specific to this trial follow from the registry information. The primary disease-control result is not reported, although disease control rate after 12 weeks is identified as the primary endpoint and a 90% confidence interval is specified. The disease-control criteria, endpoint denominator, and number assessed are also not reported in the registry record. This prevents evaluation of both the estimated outcome and the precision with which it was measured.

The study is recorded as single group, with allocation not applicable and no masking. There is no concurrent control arm, so outcomes cannot be interpreted as a randomized comparison. The registry record does not report an external benchmark or a method for comparing results with an outside reference. The completeness of the treatment description is another limitation: the listed regimen and restaging schedule are available, but the record's description of subsequent decisions is incomplete.

Although the registry lists secondary response and survival outcomes, it does not report numerical findings, follow-up detail, event definitions, or analysis methods for them. It also contains no analysis entries. Multiplicity handling, missing-data rules, analysis populations, and the statistical analysis plan are not reported in the registry record. The original sample-size justification and any interim analysis plans are likewise not reported.

The safety information includes event counts and selected serious adverse-event terms, but the record does not provide detailed attribution, event-level context, or a comparator. These limits prevent comparative safety interpretation and do not support causal attribution from the counts alone. Together, these gaps leave the design and selected safety summaries describable, while limiting interpretation of efficacy, precision, and analytic robustness.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Atezolizumab Immunotherapy in Patients With Advanced NSCLC — NCT03102242.
• Registry record last updated 2023-05-24; status last verified 2023-04.
• Results first posted 2023-05-24.
• 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 describes a completed Phase II single-group trial of induction atezolizumab in patients with unresectable stage IIIA and IIIB non-small-cell lung cancer eligible for chemoradiotherapy with curative intent. Enrollment is listed as 64. The intervention description specifies 1200 mg IV every 21 days for 4 cycles, with restaging after cycle 2 and cycle 4. The primary endpoint is disease control rate after 12 weeks, with a 90% confidence interval specified.

The primary numerical result and the listed secondary outcome results are not reported in the registry record. The registry lists 34 participants with serious adverse events, 59 with other adverse events, and 18 deaths. It provides selected serious adverse-event terms and percentages but no comparative safety arm. Analysis methods, missing-data handling, analysis populations, and the full statistical plan are not reported. Because the study has no concurrent control group, its design does not provide a randomized treatment comparison.