Veliparib With or Without Radiation Therapy, Carboplatin, and Paclitaxel in Patients With Stage III Non-small Cell Lung Cancer That Cannot Be Removed by Surgery: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Veliparib With or Without Radiation Therapy, Carboplatin, and Paclitaxel in Patients With Stage III Non-small Cell Lung Cancer That Cannot Be Removed by Surgery (NCT01386385), a randomized phase 1, phase 2 study. The primary outcome is Maximum Tolerated Dose of Veliparib When Given Concurrently With Standard Carboplatin/Paclitaxel and Radiotherapy, Determined According to Incidence of Dose Limiting Toxicity (DLT) (Phase I).

2. Clinical Question

3. Trial Design

This is a phase I/II, randomized, single-masked, parallel study sponsored by the National Cancer Institute. The registry record lists enrollment of 53 and a start date of 2011-06-20. Primary completion is recorded as 2019-05-26, and the status is active, not recruiting. These record-level details do not establish how many participants were included in each particular analysis; arm-specific denominators differ across posted outcomes.

The phase I component is dose finding. Its primary endpoint concerns dose-limiting toxicity attributed to the study regimen during radiotherapy or within the specified period after radiotherapy. The posted dose-finding results give counts for the 40mg, 80mg, and 120mg cohorts. The endpoint description defines several qualifying toxicities, including specified severe hematologic and non-hematologic events and treatment delays. It also lists exceptions to the toxicity criteria. These definitions determine what counts as an event for the endpoint; they are not interchangeable with all adverse events recorded elsewhere.

The phase II component is described as randomized and placebo-controlled. One arm receives veliparib with 3D conformal radiation therapy, carboplatin, and paclitaxel; the other receives placebo with the same named chemoradiotherapy components. The posted phase II objective response results include 18 participants in the veliparib arm and 13 in the placebo arm. The registry record does not report allocation concealment, the identity of the masked party, or a detailed randomization procedure.

The study also lists overall survival and serious adverse events among its phase II secondary outcomes. The record includes serious adverse event counts across phase I and phase II treatment and consolidation groups. A complete statistical analysis plan, protocol deviations, and analysis-population rules are not reported in the registry record. Therefore the endpoint-specific results should be read in light of their own populations and definitions, not treated as one unified analysis.

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: In phase I, 21 participants were enrolled (8 on 40mg cohort, 7 on 80mg cohort and 6 on 120mg cohort separately). All 21 participants met the eligibility criterial. 14 of 21 participants later went on to consolidation therapy. In phase II, 32 participants were enrolled. 1 was ineligible and thus excluded from analysis. 18 of 31 eligible participants were randomized to veliparib arm and 13 participants to the placebo arm. 23 of 31 eligible participants later on went on to consolidation therapy

Only the first 6 of 8 groups are shown.

Phase I 40mg: Concurrent CRT

Reasons for not completing this period:

Phase I 40mg: Consolidation Therapy

Phase I 80mg: Concurrent CRT

Reasons for not completing this period:

Phase I 80mg: Consolidation Therapy

Analysis population for the primary outcome, as defined in the registry: Eligible participants evaluable for dose limiting toxicity assessment

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: All eligible participants

6. Primary Endpoint

The registry lists the primary outcome as Maximum Tolerated Dose of Veliparib When Given Concurrently With Standard Carboplatin/Paclitaxel and Radiotherapy, Determined According to Incidence of Dose Limiting Toxicity (DLT) (Phase I), measured over 9 weeks.

The primary endpoint is the maximum tolerated dose of veliparib given concurrently with carboplatin, paclitaxel, and radiotherapy, determined according to dose-limiting toxicity. The measure is a count of qualifying toxicities among participants evaluable for this assessment, with a timeframe of 9 weeks. Toxicities are graded using the National Cancer Institute Common Terminology Criteria for Adverse Events version 4.0. The record requires that a dose-limiting toxicity be considered probably, possibly, or definitely related to the study regimen and occur during radiotherapy or within 2 weeks after radiotherapy ends.

The endpoint uses a defined toxicity rule rather than treating every reported symptom or adverse event as a dose-limiting toxicity. Its criteria include specified severe radiation effects, blood-count abnormalities, nausea or vomiting, treatment delays, and other non-hematologic toxicities, with stated exceptions. This makes the outcome relevant to dose finding: the objective is to identify a dose in the context of tolerability under the regimen and assessment rules. In general, a dose-limiting toxicity endpoint can guide which dose is taken forward, but the registry record does not report the protocol's decision algorithm or its final maximum tolerated dose determination.

The posted results list one dose-limiting toxicity in the 40mg cohort, one in the 80mg cohort, and none in the 120mg cohort. These are counts, not a complete account of all safety events. The record does not report a formal uncertainty estimate for these counts or a final dose-selection conclusion.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Different endpoints in this trial require different statistical summaries. The primary dose-finding outcome is a count of dose-limiting toxicities, while the phase II response outcome is a proportion. A count reports how many events were observed under the endpoint definition. A proportion places the number meeting a binary outcome in relation to the participants assessed for that outcome. These quantities are not directly comparable, and the phase I dose cohorts should not be interpreted as if they were the randomized phase II arms.

What does the confidence interval add?

A confidence interval conveys uncertainty around an estimate. In repeated use of the same statistical procedure, the stated confidence level describes how often intervals constructed this way would cover the underlying quantity. For an individual trial, the interval shows which values are compatible with the data and method, rather than guaranteeing that the true value falls inside it. Wider intervals indicate less precision. The response estimates have 95% confidence intervals, and the interval for the risk difference spans values below and above no difference. That breadth is important: the point estimate alone gives an incomplete account of the evidence.

The phase II response proportions are 56% for the veliparib arm and 69% for the placebo arm, each with its own 95% confidence interval. The registry record also contains an unadjusted recalculation of the risk difference and an unadjusted p-value of 0.44. A p-value measures how unusual data at least as discrepant as those observed would be under a specified null model and its assumptions. It is not the probability that the null hypothesis is true, nor a measure of the size or importance of an effect. Because this is an unadjusted recalculation rather than a reported sponsor analysis, it should be regarded as a descriptive statistical aid, not as a substitute for a prespecified analysis.

Why distinguish the posted and recalculated results?

Arm-level summary results and a calculated comparison can be useful for different purposes. The arm-level estimates describe what was posted for each group. A recalculated contrast expresses their difference on a chosen scale. Here, the recalculated risk difference is unadjusted, meaning it does not account for covariates or other adjustments. The registry record reports no sponsor analysis for this outcome comparison. There is no discrepancy flag in the posted information, but that does not turn the recalculation into the trial's formal analysis. A sponsor's own analysis would take precedence if a discrepancy were identified.

The risk difference is an absolute measure: it expresses the contrast between the response proportions on the percentage-point scale when represented as proportions. The confidence interval around it is more informative than the point estimate alone because it indicates the range of compatible contrasts. Since that interval spans zero, it does not isolate a direction of difference. This does not prove equivalence; equivalence requires a design and margin suited to that question, and such a margin is not reported in the registry record.

What do the planning illustrations mean?

The response comparison also has planning illustrations based on the observed effect. These are not observed power and are not the trial's original sample-size justification. The stated assumptions are a two-sided significance level of 0.05, equal allocation, and a normal approximation; the effect to detect is a stated fraction of the observed effect. For planning for 100% of the observed effect, the illustration gives 196 participants per arm for 80% power and 262 participants per arm for 90% power. For planning for 75% of the observed effect, the corresponding figures are 343 and 458 participants per arm. For planning for 50% of the observed effect, they are 755 and 1,010 participants per arm.

These illustrations show how required sample size can increase when a smaller effect is the target, under the specified assumptions. They are prospective planning calculations based on the observed effect, which itself is uncertain. They should not be read as evidence that the trial was designed to detect those effects, or as a probability that the trial detected a true effect. The registry record does not report the actual design assumptions, sample-size justification, or planned statistical power.

Why can event counts be difficult to compare?

Counts depend on the number of participants observed, the outcome definition, and the observation window. A count of dose-limiting toxicities in evaluable participants is not the same as an all-cause adverse event count or a serious adverse event count. The primary endpoint has a 9-week timeframe and explicit attribution and grading rules. The safety record separately lists groups with participants at risk, serious events, other events, and deaths. Those measures have distinct definitions and should not be merged into one rate or ranking without appropriate denominators and a common analysis plan.

In general, small groups produce uncertain estimates, especially for uncommon events. Over-dispersion matters most in count models when event counts vary more than a model's assumed distribution allows; it can make standard errors too small if ignored. The primary results are reported as counts, and the registry record does not report a count model, an over-dispersion assessment, or inferential testing for those dose cohorts. No such model should be assumed.

What cannot be established from the record?

The registry record does not report a detailed statistical analysis plan, multiplicity adjustment, missing-data method, interim analysis plan, or a full definition of the analysis populations for every endpoint. These omissions matter because endpoint estimates depend on who was included, how missing outcomes were handled, and whether analyses were prespecified. Without those details, results are best interpreted as posted descriptive findings, with the recalculated response contrast kept distinct from sponsor-reported analyses. General statistical principles can explain the role of intervals, contrasts, and planning calculations, but they cannot fill in trial-specific methods that are not reported.

10. Multiplicity, Missing Data and Analysis Populations

The record identifies a primary dose-finding endpoint and several secondary outcomes, but does not report a multiplicity strategy. Multiplicity arises when investigators test or estimate several outcomes, doses, subgroups, or comparisons. With more opportunities to find an apparently unusual result, the chance of at least one false-positive result can increase. A prespecified hierarchy or statistical adjustment can address some of this concern. Without a reported plan, secondary results should be treated as exploratory or descriptive rather than as a collection of independent confirmatory tests.

The registry record also does not report an interim analysis plan or stopping boundaries. It is therefore not possible to determine whether accumulating data were examined during the trial in a way that affected decisions or inference. No interim procedure should be presumed from the phase I/II design.

Missing-data methods and analysis-population definitions are also not reported in sufficient detail. The primary endpoint refers to participants evaluable for dose-limiting toxicity, while the phase II response result gives arm-specific denominators. The record does not explain all exclusions, the handling of participants without an assessable response, or whether the response analysis followed an intention-to-treat principle. It also does not give a complete population definition for each safety group. These details can affect the interpretation of estimates and their generalizability to all randomized or enrolled participants.

In general, missing outcomes can bias estimates when the chance of being observed is related to outcome or treatment. Complete-case summaries can be adequate under some assumptions, but those assumptions need support; the record does not describe a missing-data strategy. Thus the posted denominators should be read as endpoint-specific analysis counts, not as proof that every enrolled participant contributed to every result. The available information does not permit a reconstruction of the planned analysis or an assessment of all departures from it.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: sepsis (25.0%), lung infection (20.0%), atrial fibrillation (16.7%), atrial flutter (16.7%), febrile neutropenia (14.3%).

The registry lists serious adverse event counts separately for phase I dose cohorts, phase II arms, and consolidation groups. For example, in the Phase II Arm I group, 2 serious adverse events are reported among 17 participants at risk; in Phase II Arm II, 3 are reported among 13 at risk. The separate consolidation groups have their own denominators and counts. These are group-specific safety summaries, not a single comparison across a common population and observation window.

The record also lists the most frequent serious event terms with percentages, including sepsis at 25.0%, lung infection at 20.0%, atrial fibrillation at 16.7%, atrial flutter at 16.7%, febrile neutropenia at 14.3%, and esophageal perforation at 14.3%. The registry record does not provide enough detail to infer how those percentages were calculated across all listed groups, so they should be read as posted rather than recalculated or combined.

Serious adverse event counts do not establish causation. Nor do the listed deaths indicate that a death was caused by the study regimen. The record reports deaths by group but does not provide an analysis attributing them to treatment. These summaries describe safety events recorded in the registry; differences between groups may reflect group size, follow-up, treatment phase, or other factors. They should not be treated as a randomized comparison unless a corresponding analysis is reported.

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

The posted results answer limited, endpoint-specific questions. For phase I, they report dose-limiting toxicity counts for three veliparib dose cohorts under a defined toxicity assessment. For phase II, they report objective response proportions in the randomized veliparib and placebo arms. These results are not interchangeable, and the dose-finding counts do not establish the phase II comparative effect.

The response estimates are uncertain. The confidence intervals are broad, and the unadjusted recalculated risk-difference interval includes values in both directions around no difference. The unadjusted p-value is 0.44. Together, these findings do not establish a clear between-arm difference, but they also do not prove that the treatments are equivalent or that no difference exists. The trial's small outcome-specific groups limit precision.

The primary dose-finding counts do not by themselves identify a maximum tolerated dose. The registry record does not report the final dose decision, the dose-selection algorithm, or uncertainty around that decision. Likewise, the listed serious adverse events and deaths cannot be interpreted as treatment-caused events from the information reported. The record lists overall survival as a secondary outcome but does not report its result in the information available.

In general, a trial result should be understood in relation to its design, endpoint definition, analysis population, and uncertainty. Here, important details about missing data, multiplicity, and the full statistical plan are not reported in the registry record. The posted information therefore supports descriptive interpretation and cautious comparison, not a broad conclusion about all outcomes or all patients with this condition. It does not establish that veliparib improves outcomes, nor does it establish that it does not.

13. Why This Trial Matters Statistically

This study illustrates why the endpoint and trial phase matter when interpreting clinical results. A phase I dose-finding endpoint is designed to characterize dose-limiting toxicity under specified treatment conditions. It is not an efficacy comparison. The phase II randomized component asks a different question by comparing a veliparib-containing regimen with a placebo-containing regimen on outcomes such as objective response.

The record also demonstrates the importance of uncertainty around estimates. The response proportions alone might appear to suggest a difference, but their confidence intervals and the unadjusted contrast show substantial imprecision. Readers should distinguish the observed point estimate from the range of effects compatible with the data. A nonsignificant p-value is not evidence of equivalence, and a point estimate is not a definitive treatment effect.

Finally, the study highlights how registry reporting affects interpretation. The dose-finding counts, phase II response results, and safety summaries are available, but the record does not report a full analysis plan, multiplicity approach, missing-data methods, or final dose-selection result. Recognizing what is and is not documented is part of sound biostatistical reading. The information can be useful as a record of measured outcomes while still leaving important inferential questions unresolved.

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.

Several limitations are specific to the available record. The phase I endpoint is reported as small cohort-level counts without uncertainty estimates or a stated final maximum tolerated dose. The dose-finding decision rule is not reported, so the posted counts cannot be used to reconstruct how the selected dose was determined.

The phase II objective response denominators are limited, and the confidence intervals are broad. The record gives no sponsor-reported analysis for this response comparison; the available contrast and p-value are unadjusted recalculations. The registry does not provide an analysis plan that would clarify whether this comparison was prespecified, how it was intended to be analyzed, or how multiple outcomes were handled.

The record describes a phase I/II study with different treatment components and populations across phases. Safety groups include treatment and consolidation cohorts, and their counts should not be combined or directly compared without consistent populations and follow-up definitions. Although deaths are listed, attribution to treatment is not reported. The record does not provide all endpoint-specific analysis-population rules, missing-data procedures, or details of follow-up for the posted outcomes.

Overall survival is listed as a secondary outcome, but its result is not reported in the registry record. Detailed eligibility criteria, allocation procedures, the identity of the masked party, and a complete statistical analysis plan are also not reported. These gaps limit assessment of bias, precision, and the extent to which the results can be generalized. They do not invalidate the posted observations, but they constrain the conclusions that can be drawn from them.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Veliparib With or Without Radiation Therapy, Carboplatin, and Paclitaxel in Patients With Stage III Non-small Cell Lung Cancer That Cannot Be Removed by Surgery — NCT01386385.
• Registry record last updated 2026-10-20; status last verified 2026-10.
• Results first posted 2021-07-12.
• 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.

18. Record Summary

The registry record describes a phase I/II study of veliparib with radiation therapy, carboplatin, and paclitaxel for unresectable stage III non-small cell lung cancer. It is randomized, single-masked, and parallel in phase II, with a placebo-containing comparator. The sponsor is the National Cancer Institute.

The primary endpoint is the number of dose-limiting toxicities during a 9-week period. The posted phase I counts are one in the 40mg cohort, one in the 80mg cohort, and none in the 120mg cohort. The registry also reports phase II objective response estimates of 56% and 69%, with broad 95% confidence intervals. The unadjusted recalculated comparison is imprecise and does not establish a clear difference.

Serious adverse events are listed by treatment and consolidation group. Overall survival is named as a secondary outcome, but its result is not reported in the registry record. The final dose-selection conclusion, detailed analysis plan, and several population and missing-data methods are not reported.