A Study to Evaluate and Compare the Efficacy and Safety of Alectinib Versus Crizotinib and to Evaluate the Pharmacokinetics of Alectinib in Asian Participants With Treatment-Naive Anaplastic Lymphoma Kinase (ALK)-Positive Advanced Non-Small Cell Lung Cancer (NSCLC): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Evaluate and Compare the Efficacy and Safety of Alectinib Versus Crizotinib and to Evaluate the Pharmacokinetics of Alectinib in Asian Participants With Treatment-Naive Anaplastic Lymphoma Kinase (ALK)-Positive Advanced Non-Small Cell Lung Cancer (NSCLC) (NCT02838420), a randomized phase 3 study. The primary outcome is Progression-Free Survival (PFS) as Determined by Investigator Using Response Evaluation Criteria in Solid Tumor (RECIST) v1.1.

2. Clinical Question

3. Trial Design

This was a randomized, multicenter, Phase III, open-label, parallel-group study. Participants were assigned in a 2:1 ratio to alectinib or crizotinib. The registry lists an enrollment of 187. The posted primary efficacy population was the intent-to-treat population, defined as all randomized participants. The arm records list 125 participants for alectinib and 62 for crizotinib.

The alectinib regimen was 600 mg twice daily with food. The crizotinib regimen was 250 mg twice daily with or without food. Treatment was to continue until disease progression, unacceptable toxicity, withdrawal of consent, or death. The stated focus was efficacy and safety, with pharmacokinetics of alectinib also to be evaluated.

The study was unmasked: participants and study personnel were not blinded to treatment assignment. Randomization can help balance characteristics across treatment groups on average, while open-label conduct means treatment assignment was known. The registry record does not report further detail about how randomization was implemented or whether any measures were used to limit possible effects of unblinded assessment.

The study was listed as completed. Its recorded start date was 2016-08-03, and its primary completion date was 2018-05-31. These dates describe the registry record and do not by themselves indicate when analyses were conducted or when results became available. The exact number of participating centers and the countries or sites included 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:

Analysis population for the primary outcome, as defined in the registry: The primary analysis population for efficacy was the intent-to-treat (ITT) population, defined as all randomized participants.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Progression-Free Survival (PFS) as Determined by Investigator Using Response Evaluation Criteria in Solid Tumor (RECIST) v1.1, measured over From the date of randomization to the date of the first documented disease progression or death, whichever occurred first (up to overall period of approximately 40 months).

The primary endpoint was progression-free survival (PFS), assessed by investigators using RECIST v1.1. It measured the time from randomization until the first documented disease progression or death from any cause, whichever occurred first. The registry describes the timeframe as extending from randomization through the first documented progression or death, up to an overall period of approximately 40 months.

PFS is a time-to-event endpoint: each participant’s follow-up contributes information about how long it takes for the defined event to occur. A participant who has not had a recorded event by the analysis cutoff may have incomplete event-time information; how such observations were handled is not reported in the registry record. The median PFS is the time by which half of the relevant population has had an event, as estimated from the analysis. It is not an average, and it does not describe the experience of every participant.

This endpoint is used in trials because it captures time until progression or death, rather than only whether an event occurred. The registry’s primary endpoint is investigator-assessed, so its interpretation is tied to the recorded investigator assessments and the specified response criteria. A separate secondary outcome assessed PFS by an independent review committee; that is not the primary endpoint.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a time-to-event measure. For each participant, the event is the first investigator-documented disease progression or death from any cause. The time origin is randomization. This definition combines progression and death into one endpoint: whichever occurs first determines the event time. The registry specifies RECIST v1.1 for investigator assessment of progression.

What does a median PFS estimate describe?

A median is a central point in the estimated event-time distribution. It marks the time at which the estimated proportion still free of the event reaches half. It is not the mean time to an event, and it does not say that all participants experienced an event at or near that time. A time-to-event analysis can use information from participants with different lengths of follow-up, including those whose event has not been observed by the analysis cutoff, provided the analysis method accounts for that incomplete event-time information. The registry record does not report the details of that method or the handling of censoring.

The posted crizotinib median is 11.1 months, with a 95% confidence interval of 9.1 to 13.0 months. The alectinib record gives a lower confidence limit of 20.3 months but does not report the median or upper confidence limit. These entries cannot be combined into a valid estimate of the difference between medians. A comparison of treatment groups would ordinarily require a defined comparative measure and its uncertainty; the registry record does not report a hazard ratio for this primary outcome or a complete two-arm comparison.

What does a confidence interval add?

A confidence interval describes the uncertainty around an estimated quantity under the analysis procedure. A narrower interval generally indicates greater precision than a wider interval, while an interval’s endpoints depend on the data and method used. A confidence interval is not a range containing a fixed proportion of individual participants’ outcomes. Nor should it be read as a probability statement about the true value without specifying a different inferential framework.

The reported 95% interval for crizotinib can be read as the registry’s uncertainty range for that group’s median estimate. It does not resolve the missing alectinib estimate or supply a between-group test. An incomplete interval should remain incomplete in interpretation; the missing endpoint should not be inferred from the available limit.

Why does the analysis population matter?

The registry defines the primary efficacy population as intent-to-treat: all randomized participants. An intent-to-treat approach keeps participants in the group to which they were randomized, regardless of later treatment changes, and preserves the randomized comparison as the basis for analysis. This is a general explanation of the concept, not a claim about treatment changes in this study. The record does not describe treatment switching, exclusions from the analysis, or the number of participants contributing event-time information.

Random assignment is intended to reduce systematic differences between treatment groups at the start of a study. It does not guarantee that every measured or unmeasured characteristic will be identical between groups. The registry states that assignment was randomized in a 2:1 ratio, but it does not report the sequence-generation or concealment procedures.

How should the study design inform interpretation?

The study was open-label, meaning treatment assignment was not masked. The primary endpoint involved investigator assessment, so knowledge of treatment assignment is relevant context when considering how outcomes were assessed. RECIST v1.1 provides stated criteria for evaluating response and progression, but the registry record does not report additional safeguards or procedures for assessment. This design feature does not establish that assessments were biased; it identifies a consideration when interpreting the results.

The record also lists PFS assessed by an independent review committee as a secondary outcome. Independent review and investigator assessment are distinct assessment approaches. The secondary endpoint should not be substituted for the primary endpoint, and its presence does not fill in the missing primary comparison.

What can planning illustrations establish?

Any sample-size illustrations associated with the posted results are planning exercises based on stated assumptions, not observed trial outcomes and not the trial’s original design justification. The stated illustration assumptions use a two-sided significance level of 0.05, equal allocation, a normal approximation, and effects defined as fractions of the observed effect. For time-to-event illustrations, the stated quantity is the total number of events; for other outcomes, it is participants per arm. No numerical planning values are reported in the registry record, and the registry does not report the trial’s own sample-size justification. Such illustrations cannot be treated as observed power or as evidence that the original design had a particular power.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 13 outcome measures and 12 posted secondary outcomes. The secondary outcomes include independently reviewed PFS, investigator-assessed objective response, central nervous system progression outcomes assessed using RECIST v1.1 and RANO, duration of response, overall survival, adverse events, and quality-of-life deterioration. The record does not report a multiplicity adjustment or a prespecified hierarchy for interpreting results across these outcomes. When several outcomes are examined, the chance of at least one apparently notable finding can increase; this is a general statistical consideration, not a claim about the analyses in this study.

The record identifies the intent-to-treat population as the primary efficacy population and defines it as all randomized participants. Details of analysis populations for the secondary outcomes are not reported in the registry record. The number of participants with available data for each outcome, the number of events used in the PFS analysis, and the handling of incomplete follow-up are also not reported.

Missing outcome data can affect precision and, depending on why information is missing, can also affect estimates. The registry record does not describe missing-data assumptions, imputation procedures, sensitivity analyses, or how participants without a recorded progression or death by the analysis cutoff were handled. These omissions limit assessment of the robustness of the estimates, especially given the incomplete alectinib median and confidence interval and the absence of a complete primary comparison.

Details of interim analyses, stopping rules, and the statistical analysis plan are not reported in the registry record. Their absence from the record should not be interpreted as proof that such procedures were or were not used.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: interstitial lung disease (4.8%), lung infection (3.2%), pneumonia (3.2%), blood creatine phosphokinase increased (3.2%), bradycardia (3.2%).

The registry reports serious adverse events for 19 of 125 participants in the alectinib group and 16 of 62 participants in the crizotinib group. It reports non-serious adverse events for 124 of 125 participants receiving alectinib and 62 of 62 receiving crizotinib. These are counts of participants reported in each category, not comparative effect estimates. No statistical comparison or uncertainty interval for these safety counts is reported in the registry record.

The listed serious adverse event terms and percentages are interstitial lung disease, 4.8%; lung infection, 3.2%; pneumonia, 3.2%; blood creatine phosphokinase increased, 3.2%; bradycardia, 3.2%; and hepatic function abnormal, 3.2%. The registry excerpt does not identify the treatment group associated with each listed term, so these terms should not be assigned to one group based on this information.

The record also reports 8 deaths in the alectinib group and 13 in the crizotinib group. It does not establish that these deaths were caused by treatment, nor does it provide a comparative analysis of mortality in this safety listing. Serious adverse events, non-serious adverse events, and deaths are different reported categories; their counts should not be combined or interpreted as interchangeable measures. The record does not provide event timing, exposure-adjusted rates, or detailed attribution information.

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

The registry provides a randomized comparison of alectinib and crizotinib in Asian participants with treatment-naive, ALK-positive advanced non-small cell lung cancer. For the primary endpoint, it reports a crizotinib median PFS of 11.1 months and a 95% confidence interval of 9.1 to 13.0 months. For alectinib, it reports a lower confidence limit of 20.3 months but not the median or upper confidence limit. It does not report a complete numerical comparison of the groups for this endpoint.

These figures can describe what is recorded for each arm, but they do not provide enough information to quantify the relative PFS effect or its precision. The partial alectinib entry cannot be treated as a median estimate, and comparing it directly with the crizotinib median would be statistically invalid. The missing comparative result should not be interpreted as evidence of no difference, nor does the available record justify a claim about the magnitude or certainty of a difference.

The study’s randomized design is relevant to how a treatment comparison can be interpreted, while the open-label design is important context for an investigator-assessed endpoint. The registry identifies the primary efficacy population but does not provide detailed methods for handling incomplete follow-up, missing data, or endpoint assessment. These omissions make it difficult to assess how analytic choices may have affected the posted results.

The adverse-event counts provide descriptive information about participants reported with serious and non-serious events, alongside a list of serious event terms. They do not by themselves establish causality or furnish a complete comparative safety analysis. The listed deaths likewise should not be interpreted as treatment-caused deaths based on the registry information alone.

Overall, the record documents the trial design and some outcome data, but its incomplete primary endpoint reporting limits the statistical conclusions that can be drawn from the posted results. A fuller interpretation would require the missing comparative estimate and further analysis details; those are not reported in the registry record.

13. Why This Trial Matters Statistically

This study illustrates why the completeness of outcome reporting matters as much as the choice of endpoint. PFS is a time-to-event outcome that can summarize how long participants remain without documented progression or death. Yet a group-level median alone does not quantify a treatment contrast, and an incomplete median and confidence interval for one group leave the primary comparison unresolved in the registry record.

The contrast between a stated randomized design and incomplete comparative reporting is statistically important. Randomization establishes the framework for comparing assigned groups, but readers still need a reported effect estimate, its uncertainty, and enough information about the analysis to evaluate what the data show. Without those elements, a reader cannot assess the estimated magnitude or precision of the treatment difference from the posted primary outcome.

The open-label design also makes assessment procedures relevant, because the primary endpoint was investigator-assessed. The presence of a separate independent-review PFS outcome as secondary does not replace the primary result; it highlights that assessment approach is one part of endpoint interpretation.

Safety information requires similar care. Counts of participants with reported events are useful descriptive data, but without a comparative analysis, event attribution, and fuller context, they do not establish a causal difference between treatments. The registry’s listed serious-event terms and deaths should be interpreted within those limits.

For biostatisticians and clinical researchers, this record is a practical example of distinguishing trial design from result reporting. A completed randomized trial may still have registry entries that do not support a full numerical interpretation of its primary comparison. Recognizing what is reported, what is partial, and what is not reported prevents descriptive values from being mistaken for a complete inferential result.

14. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The main trial-specific limitation in the registry record is incomplete reporting of the primary endpoint. The crizotinib median PFS and its confidence interval are present, while the alectinib median and upper confidence limit are not. A two-arm comparison is identified as missing or not estimable, and no hazard ratio is reported in the registry record. The posted primary endpoint information therefore does not support a complete estimate of the comparative effect.

The primary endpoint was assessed by investigators in an open-label study. The registry names RECIST v1.1 but does not report details about assessment procedures, review safeguards, or the handling of potential differences in assessment. The record also does not provide event counts, follow-up distribution, or the statistical approach used to estimate the medians.

Although the intent-to-treat population is defined, the record does not report detailed missing-data methods, censoring rules, sensitivity analyses, or outcome-specific analysis populations. It also does not report the statistical analysis plan, interim analyses, multiplicity procedures, or sample-size justification. The absence of these details from the registry record limits independent evaluation; it does not establish whether such procedures were used elsewhere.

The safety entries are descriptive and lack comparative estimates or uncertainty intervals. The serious-event terms are listed with percentages but are not linked to treatment groups in the record. The death counts are reported without a stated causal assessment. These limitations mean that the safety information should not be read as a complete comparative safety analysis.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Study to Evaluate and Compare the Efficacy and Safety of Alectinib Versus Crizotinib and to Evaluate the Pharmacokinetics of Alectinib in Asian Participants With Treatment-Naive Anaplastic Lymphoma Kinase (ALK)-Positive Advanced Non-Small Cell Lung Cancer (NSCLC) — NCT02838420.
• Registry record last updated 2026-06-24; status last verified 2026-06.
• Results first posted 2019-07-11.
• 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.

18. Record Summary

This completed, randomized, multicenter, Phase III, open-label, parallel-group study compared alectinib with crizotinib in Asian participants with treatment-naive, ALK-positive advanced non-small cell lung cancer. Participants were assigned in a 2:1 ratio. The registry lists enrollment of 187, with 125 participants in the alectinib arm and 62 in the crizotinib arm.

The primary outcome was investigator-assessed PFS using RECIST v1.1. The registry reports a crizotinib median of 11.1 months, with a 95% confidence interval of 9.1 to 13.0 months. For alectinib, it reports a lower confidence limit of 20.3 months but no median or upper confidence limit. A complete comparative estimate is not reported in the registry record.

Safety entries include counts of participants with serious and non-serious adverse events and deaths, plus a list of serious adverse event terms. They are descriptive and do not provide a complete comparative analysis. Missing-data procedures, multiplicity methods, and further analysis details are not reported in the registry record.