Safety and Efficacy Study of Nab®-Paclitaxel With CC-486 or Nab®-Paclitaxel With Durvalumab, and Nab®-Paclitaxel Monotherapy as Second/Third-line Treatment for Advanced Non-small Cell Lung Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Safety and Efficacy Study of Nab®-Paclitaxel With CC-486 or Nab®-Paclitaxel With Durvalumab, and Nab®-Paclitaxel Monotherapy as Second/Third-line Treatment for Advanced Non-small Cell Lung Cancer (NCT02250326), a non-randomized phase 2 study. The primary outcome is Kaplan Meier Estimate of Progression-Free Survival (PFS) as Assessed by the Investigator.

2. Clinical Question

The study examined second- or third-line treatment with nab-paclitaxel, either alone or combined with CC-486 or durvalumab, in people with advanced non-small cell lung cancer. Its primary question concerned progression-free survival (PFS): how long participants remained alive without documented disease progression, as assessed by the investigator. The study also evaluated safety and several secondary outcomes, including response, overall survival, treatment discontinuation, and dose intensity.

The treatment groups were nab-paclitaxel plus CC-486, nab-paclitaxel plus durvalumab, and nab-paclitaxel monotherapy. Since participants were not randomly allocated, observed differences between groups could reflect differences in who entered each group as well as treatment. The registry record does not report information that would establish that the groups were comparable at baseline or make a causal interpretation appropriate.

The clinical question therefore has two parts: what outcomes were observed within each treatment group, and how much uncertainty surrounds those outcomes? The posted information identifies the primary endpoint and its intended analysis, but does not give the PFS medians, confidence intervals, or sponsor-reported hazard-ratio estimates. It is not possible to describe the magnitude or direction of a treatment contrast from the registry record.

3. Trial Design

This was a phase 2, open-label, multicenter study with a parallel-group structure. The registry lists enrollment as 240 and allocation as non-randomized. It describes three groups: nab-paclitaxel with CC-486, nab-paclitaxel with durvalumab, and nab-paclitaxel alone. The trial status is completed. The registry lists a start date of 2015-01-07 and a primary completion date of 2017-07-17.

In the CC-486 combination group, nab-paclitaxel was described as a 100 mg/m^2 intravenous infusion on Days 8 and 15, with CC-486 200 mg orally daily on Days 1 to 14 of each 21-day treatment cycle. In the durvalumab group, nab-paclitaxel was described as a 100 mg/m2 intravenous infusion on Days 1 and 8, with durvalumab 1125 mg intravenously on Day 15 of each 21-day treatment cycle. In the monotherapy group, nab-paclitaxel was described as a 100 mg/m^2 intravenous infusion on Days 1 and 8 of each 21-day treatment cycle.

Open-label means that treatment masking was not used. The registry lists masking as none. Non-randomized allocation is particularly important when interpreting the groups: assignment may be associated with participant characteristics, clinical circumstances, or other factors. The record cautions that these arms are not a like-for-like control comparison. Consequently, a difference between group outcomes cannot be attributed to treatment alone on the basis of this design description.

The registry specifies an intent-to-treat population for the primary outcome, including all randomized or assigned participants whether or not they received study drug or had efficacy assessments. It does not report the number analyzed for each endpoint, baseline characteristics, the allocation process, or a statistical analysis plan. Sample-size justification and details of any interim analyses 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.

Before assignment: Eligible participants included those with advanced non-small cell lung cancer who had received no more than one prior containing chemotherapy regimen. Immunotherapy as a prior line of treatment was allowed. Randomization was stratified by eastern cooperative oncology group performance status, gender and the smoking status of the participant.

Pre-treatment

Treatment

Reasons for not completing this period:

Extension Period

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Intent to Treat Population included all randomized or assigned participants regardless of whether the participant received any study drug or had any efficacy assessments performed.

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: Intent to Treat (ITT) Population included all assigned participants regardless of whether the participant received any investigational product (IP) or had any efficacy assessments performed.

6. Primary Endpoint

The registry lists the primary outcome as Kaplan Meier Estimate of Progression-Free Survival (PFS) as Assessed by the Investigator, measured over From date of first dose of IP to DP; up to data cut-off date of 30 August (Aug) 2017 for nab-paclitaxel and CC-486 + nab-paclitaxel and 23 December (Dec) 2017 for Durva + nab-paclitaxel; participants were followed for PFS for up to 18 months.

The primary endpoint was the investigator-assessed Kaplan-Meier estimate of median progression-free survival. PFS was defined as the time from randomization or assignment to documented disease progression under RECIST Version 1.1, or death from any cause on or before the clinical cutoff date, whichever occurred first. Progression was based on computed tomography assessment and did not include symptomatic deterioration. Progressive disease was defined in the record as at least a 20% increase in the sum of target-lesion diameters from nadir.

Participants without documented progression who had not died were censored at their last tumor assessment on or before the applicable cutoff date. Censoring means that follow-up ended without observing the endpoint event by that time; the participant contributes information up to the censoring time, rather than being treated as if an event occurred then. The registry says participants were followed for PFS for up to 18 months.

A median summarizes the time-to-event distribution: it is the point at which the estimated proportion remaining progression-free reaches the halfway mark. It is useful when follow-up is incomplete or participants have different follow-up durations, because the Kaplan-Meier method can incorporate censored observations. A 95% confidence interval describes uncertainty around the estimated median; it is not a range guaranteed to contain the true value.

The record gives the endpoint definition and analysis type, but the median estimates and confidence intervals are not reported in the registry record. The PFS cutoff was 30 August (Aug) 2017 for nab-paclitaxel and nab-paclitaxel plus CC-486, and 23 December (Dec) 2017 for durvalumab plus nab-paclitaxel. Different cutoff dates mean the endpoint follow-up windows were not identical across those groups.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Time-to-event methods are used when the outcome is the time until an event, such as progression or death, and participants may have different lengths of follow-up. The primary outcome was investigator-assessed PFS, analyzed as a Kaplan-Meier estimate of the median. This method uses observed event times and accounts for participants whose event status is not observed through the end of their follow-up.

What does Kaplan-Meier estimation do?

In general, the Kaplan-Meier method estimates the probability of remaining free of the event over time. At each observed event time, the estimate reflects the participants still under observation immediately before that event. Censored participants contribute information up to their last known assessment and are not counted as having progressed or died at the time follow-up ends. This approach relies on the assumption that censoring is not informative in a way that systematically distorts the event experience of those remaining under observation. Whether that assumption is appropriate cannot be determined from the registry record.

The median PFS is derived from the estimated survival curve at the point where the estimated proportion remaining event-free reaches one half. The registry specifies a 95% confidence interval for the median, which conveys statistical uncertainty. A wide interval indicates less precision than a narrow one, though interpretation also depends on the observed data and analysis. The actual medians and intervals are not reported in the registry record.

What is censoring in this endpoint?

The registry defines PFS as time from randomization or assignment until disease progression according to RECIST Version 1.1 or death from any cause, whichever occurred earlier by the clinical cutoff. Participants who had neither documented progression nor death were censored at their last tumor assessment on or before the cutoff date. This makes the timing and completeness of tumor assessments relevant: the last assessment marks the point through which the participant’s progression-free status is known.

The primary population is described as intent-to-treat, including all randomized or assigned participants regardless of study-drug receipt or efficacy assessment. This definition can preserve the original assigned-group framework, but in a non-randomized study it does not create the protection against confounding that random assignment would provide. The record does not report how participants with no post-assignment assessment were handled in the actual analysis.

How should hazard ratios be understood?

In general, a hazard ratio compares event rates over follow-up between groups, conditional on the model and its assumptions. It is not the same as a ratio of median times, and it does not directly describe the probability that an individual participant will experience an event. The registry indicates that only sponsor-reported hazard ratios are used for time-to-event comparisons. Their estimates and uncertainty are not reported in the registry record. A proportional-hazards assumption, which treats the relative event rate as stable over time, cannot be confirmed from the information recorded; the statistical plan is not reported.

Why does allocation matter for comparisons?

Random allocation can, in general, balance measured and unmeasured characteristics on average. This study is listed as non-randomized, and the registry cautions that its arms are not a like-for-like control comparison. Group differences may therefore reflect factors other than the assigned treatment. Statistical adjustment can address measured factors only when those factors are collected and modeled appropriately; the record does not report the adjustment strategy or enough detail to evaluate it.

The study was open-label, with no masking. Investigator assessment of progression can involve judgment about when to assess or classify disease, although the registry specifies RECIST Version 1.1 and computed tomography for progression. The record does not report whether assessments underwent independent review or how assessment timing was standardized across groups.

What do the planning illustrations mean?

General sample-size planning asks how much information a future study might need to detect an assumed effect under stated statistical conditions. The registry’s planning assumptions describe a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is described as a stated fraction of an observed effect. For time-to-event planning, the quantity is the total number of events; other planning results are expressed as participants per arm for 80% power or other stated power levels. These are planning illustrations, not observed power and not evidence that the completed study had a particular probability of detecting an effect.

Because the study was non-randomized, an observed effect used in a planning illustration may reflect confounding as well as treatment differences. Such a calculation should not be mistaken for a retrospective validation of the design. The original sample-size justification is not reported in the registry record. No further statistical details, including model specifications, covariate adjustment, sensitivity analyses, or interim analyses, are reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The study includes a primary PFS outcome and several secondary outcomes: disease control, objective response, overall survival, treatment-emergent adverse events, discontinuation, and dose intensity measures for the study drugs. Evaluating multiple outcomes and treatment groups creates multiplicity: the more statistical questions examined, the more opportunities there are for apparently noteworthy results to arise by chance. In general, a prespecified strategy can control or describe this issue, for example by prioritizing endpoints or adjusting inference across comparisons.

The registry record does not report a multiplicity strategy, a hierarchy of hypotheses, or whether secondary analyses were considered exploratory. It also does not report the statistical analysis plan. Therefore, the level of confirmatory evidence intended for the secondary outcomes cannot be determined from the registry entry.

The primary endpoint population is described as intent-to-treat, including all randomized or assigned participants whether or not they received study drug or had efficacy assessments. The record does not report endpoint-specific analysis counts, the number of participants with missing assessments, or the handling of missing efficacy data beyond the stated censoring rule for PFS. In general, missing outcome information can affect estimates when its occurrence is related to prognosis or treatment experience; the direction and size of any such effect cannot be inferred without further information.

For PFS, participants without progression or death were censored at their last tumor assessment before the cutoff. The registry does not report the number censored, reasons for censoring, or group-specific follow-up patterns. It also specifies different clinical cutoff dates across groups, so the amount of available observation may differ. These omissions limit assessment of completeness and comparability. The record does not report interim analyses, stopping rules, or procedures for controlling repeated looks at outcomes.

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: pneumonia (11.5%), dyspnoea (5.1%), pleural effusion (5.1%), adrenal insufficiency (3.8%), lower respiratory tract infection (3.8%).

The registry’s safety summary reports serious adverse events for each treatment group. Nab-paclitaxel plus CC-486 had 79 participants at risk and 30 participants affected; nab-paclitaxel plus durvalumab had 78 at risk and 44 affected; nab-paclitaxel alone had 79 at risk and 30 affected. These are the reported group counts, not estimates of comparative treatment effects. Because allocation was non-randomized, the groups are not a like-for-like control comparison.

The registry lists pneumonia at 11.5%, dyspnoea at 5.1%, pleural effusion at 5.1%, adrenal insufficiency at 3.8%, lower respiratory tract infection at 3.8%, and pulmonary embolism at 3.8% among the top serious event terms. These percentages are reported as listed; the registry information does not specify the denominator or analysis details for each term. The terms should not be added together, since a participant may have experienced more than one event.

The record also lists deaths by group: 62 with nab-paclitaxel plus CC-486, 48 with nab-paclitaxel plus durvalumab, and 50 with nab-paclitaxel. The registry summary does not provide the context needed to interpret these as treatment-related deaths, nor does it report exposure-adjusted rates or a causal comparison. Serious adverse-event counts describe the posted safety information but do not, by themselves, establish relative safety.

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

The registry record describes a phase 2 study of three treatment strategies and identifies investigator-assessed PFS as the primary endpoint. It also defines progression, death, censoring, and the intent-to-treat population. These details clarify what the endpoint was intended to measure and which participants were intended to contribute to the primary analysis.

The record does not report the primary PFS medians, their 95% confidence intervals, or sponsor-reported hazard-ratio estimates. Consequently, it does not allow a reader to state how long participants remained progression-free in each group, how precise the estimates were, or the size and direction of a time-to-event contrast. The absence of these values is not evidence that no effect occurred; it means the registry information does not report the values needed to evaluate that question.

The treatment groups were not randomly allocated, and the registry cautions against treating them as like-for-like controls. Even when outcome estimates are available, observed differences between these groups cannot automatically be attributed to treatment. Participant selection, clinical factors, and differences in follow-up could contribute. The registry record does not report enough information to determine how much such factors influenced outcomes.

The serious adverse-event counts and listed event terms provide descriptive safety information, but they do not establish comparative risk. Counts do not account on their own for differences in follow-up or exposure, and the record does not provide a causal analysis. The listed deaths likewise should not be interpreted as treatment-related without information establishing that relationship.

In general, a confidence interval expresses uncertainty in an estimate, while a p-value addresses compatibility with a specified statistical model and null hypothesis. Neither alone resolves design limitations, missing information, multiplicity, or clinical relevance. The primary outcome estimates and statistical analysis details needed for a fuller interpretation are not reported in the registry record. No treatment recommendation follows from these registry entries.

13. Why This Trial Matters Statistically

This trial is statistically instructive because it combines a time-to-event primary outcome with non-randomized treatment allocation. Kaplan-Meier methods are designed to use unequal follow-up and censored observations, both common in progression-free survival analyses. Their interpretation depends on clear endpoint definitions, appropriate censoring, and enough information about follow-up and assessment patterns. The registry defines the endpoint and censoring approach, but does not report the estimates or the amount of censoring.

The design also demonstrates why a parallel-arm structure is not equivalent to randomized comparison. Having multiple groups permits descriptive summaries for each treatment strategy, but non-randomized allocation can leave differences between groups that are unrelated to treatment. As the registry itself notes, the arms are not like-for-like controls. This caution is central when considering any apparent contrast.

Different clinical cutoff dates are another relevant feature. The PFS cutoff dates differ by treatment group, so the follow-up windows do not align exactly. Such differences need consideration alongside event counts, censoring, and assessment schedules when comparing time-to-event outcomes. The registry does not report enough detail to assess how these factors affect estimates.

The record also lists numerous secondary outcomes alongside PFS. This illustrates the importance of prespecifying outcome priorities and handling multiplicity. Without a reported statistical analysis plan or multiplicity procedure, it is not possible to determine how the secondary findings were intended to support confirmatory claims.

Finally, planning calculations based on an observed effect are not observed power and do not replace the original study-design rationale. They can illustrate assumptions for planning another study, but the non-randomized nature of this study matters when interpreting an observed effect as an input. The registry record does not report its original sample-size justification. Together, these features show why design, endpoint reporting, uncertainty, and analysis transparency all matter alongside outcome counts.

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 limitation for interpreting efficacy is that the registry record does not report the primary PFS medians, 95% confidence intervals, or sponsor-reported hazard ratios. Although the endpoint and censoring definitions are described, the key outcome estimates needed to characterize the results are absent. The record also does not report the outcome values for the listed secondary measures.

Allocation was non-randomized, and the registry explicitly says the arms are not a like-for-like control comparison. The record does not describe the allocation process, baseline comparability, or any adjustment strategy. This prevents assessment of how participant differences may affect group contrasts. Open-label treatment is another design feature; the registry does not report independent review of progression assessments or other safeguards against assessment differences.

The PFS cutoff dates differ across groups, and the record does not provide group-specific follow-up distributions, censoring counts, or assessment schedules. This limits evaluation of whether the groups had comparable opportunity for event detection. Although follow-up was described as up to 18 months, the amount of follow-up actually observed in each group is not reported in the registry record.

The registry does not report a statistical analysis plan, multiplicity strategy, interim analyses, missing-data procedures beyond PFS censoring, or the original sample-size justification. The safety summary gives counts and selected event-term percentages, but the detailed denominator and analysis context for each listed term are not reported. Exposure-adjusted safety rates and causal comparisons are not reported in the registry record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Safety and Efficacy Study of Nab®-Paclitaxel With CC-486 or Nab®-Paclitaxel With Durvalumab, and Nab®-Paclitaxel Monotherapy as Second/Third-line Treatment for Advanced Non-small Cell Lung Cancer — NCT02250326.
• Registry record last updated 2024-09-19; status last verified 2024-08.
• Results first posted 2018-08-10.
• 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

The registry describes NCT02250326 as a completed phase 2, open-label, multicenter, non-randomized parallel-group study in advanced non-small cell lung cancer. The recorded enrollment is 240. The three groups received nab-paclitaxel with CC-486, nab-paclitaxel with durvalumab, or nab-paclitaxel alone. The primary outcome was investigator-assessed median PFS under RECIST Version 1.1, with a 95% confidence interval.

The record defines PFS and its censoring rule, and lists response, overall survival, safety, discontinuation, and dose intensity among secondary outcomes. It reports serious-adverse-event counts and selected serious event terms. It cautions that non-randomized groups are not like-for-like controls. The primary PFS estimates, uncertainty intervals, sponsor-reported hazard ratios, and further statistical analysis details are not reported in the registry record.