LDK378 in Crizotinib naïve Adult Patients With ALK-activated Non-small Cell Lung Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for LDK378 in Crizotinib naïve Adult Patients With ALK-activated Non-small Cell Lung Cancer (NCT01685138), a na phase 2 study. The primary outcome is Overall Response Rate (ORR) by Investigator Assessment.

2. Clinical Question

3. Trial Design

This was a phase II, multicenter, single-arm, open-label study of LDK378 in crizotinib-naïve adults with ALK-activated non-small cell lung cancer. The registry describes a two-stage study and records enrollment of 124. Allocation was not applicable because participants were not assigned between treatment arms. The study status is completed.

Participants took oral LDK378 750 mg once daily. Treatment could continue until unacceptable toxicity, discontinuation at the discretion of the investigator or patient, initiation of a new anticancer therapy, or death. The protocol summary also says treatment could continue beyond progressive disease defined by RECIST if the investigator judged there to be clinical benefit. Participants who stopped study medication without progression continued tumor assessments until investigator-assessed progression.

The primary endpoint was investigator-assessed overall response rate (ORR), defined as the percentage with a best overall response of complete or partial response under RECIST 1.1. Assessments were scheduled every 8 weeks, beginning on the first day of treatment, until permanent discontinuation of study drug, for up to 5 years. The Full Analysis Set consisted of patients who received at least one dose.

The record lists additional outcomes, including ORR assessed by a blinded independent review committee, duration and time to response, disease control, and intracranial response. The registry record does not report the primary endpoint result, its confidence interval, or the detailed analysis plan. The study dates recorded are 2012-12-20 for start and 2018-01-22 for primary completion.

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 Full Analysis Set (FAS) consisted of all patients who received at least one dose of study drug.

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: The Full Analysis Set (FAS) consisted of all patients who received at least one dose of study drug.

6. Primary Endpoint

The registry lists the primary outcome as Overall Response Rate (ORR) by Investigator Assessment, measured over every 8 weeks (i.e. every 2 cycles; cycle = 28 days), starting from the first day of treatment with LDK378 until permanent discontinuation of study drug up to 5 years.

Overall response rate measures the percentage of participants whose best overall tumor response met the study definition of complete response or partial response. Under the stated RECIST 1.1 criteria, complete response requires disappearance of target lesions and, for pathological lymph nodes counted as target lesions, reduction of the short axis to < 10 mm. Partial response requires at least a 30% decrease in the sum of target-lesion diameters relative to baseline. These definitions specify which observed changes count as a response; they do not describe how large or durable any response was.

ORR is used as a summary of the proportion of participants who achieve a defined tumor response during the assessment period. It can be informative in a single-arm study because it describes observed responses without requiring a concurrent control group. However, it does not by itself establish that treatment caused the responses, quantify a comparative benefit, or capture how long responses lasted. Duration of response was listed as a separate secondary outcome.

The primary endpoint was assessed by the investigator. The registry also lists ORR assessed by a blinded independent review committee as a secondary outcome, which is a distinct assessment source. The observed investigator-assessed ORR and its 95% confidence interval are not reported in the registry record, so the magnitude and precision of the primary result cannot be described from that record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The statistical information in the registry record is enough to identify the main endpoint and its intended population, but not enough to reproduce or fully assess the primary analysis. The result and confidence interval are not reported in the registry record. The analysis method used to calculate the estimate, the confidence-interval method, and details of any adjusted analyses are also not reported in the registry record.

What does a percentage endpoint describe?

For a binary response endpoint, each participant is classified according to whether the defined event occurred. ORR counts participants whose best overall response met the complete- or partial-response definition, then expresses that count as a percentage of the relevant analysis population. This gives a readily interpretable summary of response occurrence. It does not preserve all the detail of tumor change, and it does not describe response duration.

The Full Analysis Set is defined as all patients who received at least one dose of study drug. That definition identifies the population for the primary endpoint in the registry. It does not, by itself, explain how participants with missing or unevaluable tumor assessments were handled, whether every treated participant was included in the endpoint denominator, or what rules were applied to assessments after treatment discontinuation. Those details are not reported in the registry record.

Why does the confidence interval matter?

An estimate from a study population is subject to sampling variability: a different set of eligible participants could yield a different percentage. A confidence interval gives a range of values compatible with the estimate under the method’s assumptions. A narrower interval generally indicates greater precision, while a wider interval signals greater uncertainty. The interval does not remove the limitations of a single-arm design, and it is not a guarantee about the value in every future population.

The primary outcome is described as having a 95% confidence interval, but neither that interval nor the percentage is reported. The interval construction method is also not reported in the registry record. It is therefore not possible to check the calculation or to judge the precision of the estimate. No confidence-interval method should be inferred from the endpoint description alone.

How should a single-arm estimate be read?

A single-arm study reports what happened among participants who received the study treatment. It can estimate the observed response proportion in the specified population, subject to measurement and missing-data considerations. Without a concurrent control group, it does not directly estimate a contrast between treatment and control. Differences from historical experience would require an appropriate external comparison and assumptions that are not specified in the registry record.

Open-label investigator assessment can also involve judgment in determining response. This does not mean an assessment is invalid; it means the assessment source and design should be considered when interpreting the estimate. The record lists blinded independent review committee ORR as a secondary outcome, but does not report its result or explain how discrepancies between assessment sources were handled.

What do the planning illustrations represent?

The registry describes planning illustrations based on the observed effect, using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The illustrated effect is a stated fraction of the observed effect. Such calculations are hypothetical design exercises. They are not observed power, do not show that the single-arm study had two randomized arms, and do not establish the precision or validity of the trial’s result.

Because the primary observed effect is not reported in the registry record, the illustrations cannot be independently interpreted from the record’s outcome information. A planning calculation also depends on its assumptions. In general, a sample-size calculation is sensitive to the target effect, error-rate choices, allocation, and statistical method. Those considerations describe the general role of planning calculations; they do not supply missing details about this study’s own design rationale.

Why can over-dispersion matter in other settings?

General statistical context: over-dispersion occurs when observed variability is greater than a model assumes. If ignored, standard errors can be too small and confidence intervals can be too narrow, making results appear more precise than warranted. The registry record does not report that an over-dispersion model was used, and this explanation should not be read as a description of the trial’s 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("LDK378 (Ceritinib)")
n      <- c(124)
events <- c(84)
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  LDK378 (Ceritinib)  124  84  67.7  58.8  75.9  59.1  75.3  59.1  75.3

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 11 outcome measures, including the investigator-assessed primary ORR and secondary outcomes assessed by the investigator or a blinded independent review committee. These outcomes address different aspects of response and disease control. Examining multiple endpoints can increase the chance that at least one result appears noteworthy by chance. Statistical methods can address multiplicity, for example by defining a hierarchy or adjusting error rates, but the registry record does not report a multiplicity strategy or identify which secondary outcomes were confirmatory.

The registry records 0 analyses. This is a registry field and does not establish that no analyses were conducted; details of analyses are not reported in the registry record. The primary endpoint result and confidence interval are also absent. No conclusions about statistical significance or the planned interpretation of secondary outcomes can be drawn from the outcome list alone.

The Full Analysis Set is defined as all patients who received at least one dose of study drug. This gives a stated analysis population for the primary endpoint, but it does not explain the treatment of missing assessments, unevaluable responses, withdrawals, or assessments collected after treatment discontinuation. Missing-data assumptions and sensitivity analyses are not reported in the registry record. These details matter because excluding participants with unavailable assessments, or assigning them a response status, can affect an estimated response percentage.

The registry describes follow-up for tumor assessment after medication discontinuation in the absence of progression, and assessment of the primary endpoint every 8 weeks until permanent discontinuation, up to 5 years. The record does not specify how follow-up interruptions, incomplete assessments, or different durations of observation were handled analytically. It is therefore not possible to judge from the record whether missing outcome information could have influenced the reported 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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: nausea (4.0%), pneumonia (4.0%), pericarditis (3.2%), lung infection (3.2%), vomiting (2.4%).

The registry lists 124 participants at risk for serious adverse events and 50 affected. For other adverse events, it lists 124 at risk and 123 affected. These are recorded counts, not comparative estimates. Because the study had no control arm, the figures cannot show how event frequency differed from an alternative treatment or from the background risk in a similar population.

The listed serious adverse event terms include nausea and pneumonia at 4.0% each; pericarditis and lung infection at 3.2% each; and vomiting and hyperglycaemia at 2.4% each. These percentages describe the entries in the registry record. The record does not report further event-level context, such as the timing, duration, severity distribution, or treatment attribution for each term. The listed terms should not be treated as a complete account of every safety outcome.

The record also lists 56 deaths. It does not state in the provided safety fields whether those deaths were attributed to treatment, disease, or another cause. A count of deaths in a trial is not, by itself, evidence that treatment caused them. Interpretation also depends on follow-up and the population observed, details that are not sufficient to establish comparative risk in this single-arm record.

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

The registry record supports a description of the study’s question and design: a phase II, single-arm evaluation of oral LDK378 in crizotinib-naïve adults with ALK-rearranged non-small cell lung cancer. It defines investigator-assessed ORR as the primary endpoint and explains the RECIST 1.1 response criteria. It also identifies the Full Analysis Set and the schedule for tumor assessments.

It does not report the observed primary endpoint percentage or its 95% confidence interval. The record therefore does not support a numerical conclusion about how many participants had a complete or partial response, or how precise an ORR estimate was. The endpoint definition should not be mistaken for evidence that a response was observed at any particular rate.

Even with a reported response estimate, the single-arm, open-label design would describe outcomes among treated participants rather than establish a causal treatment effect. Without a concurrent comparator, the study cannot distinguish treatment effects from other possible explanations for observed outcomes. The registry record also does not report the detailed analysis plan, multiplicity approach, or missing-data handling needed to assess those aspects of the results.

The safety counts and listed serious adverse event terms describe events recorded in the registry, but do not establish that treatment caused them or show how event occurrence compares with another treatment. The listed secondary outcomes could add context about response assessment, duration, and intracranial outcomes, but their results are not reported in the registry record.

Planning illustrations are calculations for hypothetical study designs, not evidence about observed treatment outcomes and not observed power. Their assumptions include equal allocation between two arms, whereas this trial was single-arm. They should not be interpreted as a substitute for a missing result or as a comparison performed within this trial.

14. Why This Trial Matters Statistically

This study illustrates why the distinction between an endpoint definition and an endpoint result matters. The registry gives a specific definition of investigator-assessed ORR, including the response criteria and assessment schedule, but does not report the primary estimate or its confidence interval. Statistical interpretation depends on both: a clear endpoint says what was to be measured, while the estimate and its uncertainty show what the study observed and how precisely it was measured.

The single-arm design is also central to interpretation. It permits description of outcomes among treated participants, but it does not create a randomized control comparison. This distinction is important when reading response outcomes, safety counts, or planning calculations. None should be recast as a comparative treatment effect when the study had no comparator arm.

The record identifies several kinds of outcome assessment, including investigator and blinded independent review committee assessments, as well as response duration, disease control, time to response, and intracranial response. Those measures can address complementary questions, but their results and the approach to multiple outcomes are not reported in the registry record. A list of endpoints is not evidence that each was analyzed or that a particular statistical interpretation was planned.

Finally, the record’s sample-size assumptions describe hypothetical planning calculations, not observed power. Keeping design illustrations separate from observed findings helps prevent a common interpretive error: treating a calculation based on an effect as if it were evidence about the precision or success of the completed study. Here, the missing primary result limits what can be concluded from the registry record, regardless of the endpoint’s clinical relevance.

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 for statistical interpretation is that it does not report the primary ORR estimate or its 95% confidence interval. Without these, the primary result’s magnitude and precision cannot be assessed. The record defines the endpoint, but the definition alone cannot answer the study’s empirical question.

The single-arm, open-label design limits causal and comparative interpretation. There is no concurrent control group to separate outcomes associated with treatment from other influences, and investigator assessment may involve judgment. Although blinded independent review committee ORR is listed as a secondary outcome, its result and any comparison between assessment sources are not reported.

Further limitations arise from missing analysis details. The registry record does not report the statistical method for the primary estimate, the confidence-interval method, multiplicity handling, missing-data rules, or sensitivity analyses. It defines the Full Analysis Set as patients who received at least one dose, but does not fully explain how unevaluable or unavailable assessments were treated. These omissions prevent a complete assessment of analysis choices and their possible effects.

The safety fields list counts and selected serious adverse event terms, but do not supply enough context to establish treatment attribution or comparative risk. The registry also lists multiple secondary outcomes without reporting their results. Finally, the planning illustrations describe a hypothetical two-arm approach, not the actual single-arm design. These limitations concern what can be inferred from the registry record and should not be mistaken for evidence about unreported results.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: LDK378 in Crizotinib naïve Adult Patients With ALK-activated Non-small Cell Lung Cancer — NCT01685138.
• Registry record last updated 2019-03-26; status last verified 2019-03.
• Results first posted 2019-02-12.
• 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

NCT01685138 is a completed phase II, multicenter, single-arm, open-label study of oral LDK378 in crizotinib-naïve adults with ALK-rearranged non-small cell lung cancer. Participants received 750 mg once daily. The registry records enrollment of 124 and defines the primary endpoint as investigator-assessed ORR under RECIST 1.1 in the Full Analysis Set.

The registry record describes complete and partial response criteria and tumor assessment every 8 weeks, up to 5 years. It lists secondary outcomes including independent-review ORR, duration and time to response, disease control, and intracranial response. The observed primary ORR and its 95% confidence interval are not reported in the registry record. The record also lists serious and other adverse event counts and selected serious-event terms, without establishing treatment attribution or a comparative safety estimate.