Effects of Single Agent Niraparib and Niraparib Plus Programmed Cell Death-1 (PD-1) Inhibitors in Non-Small Cell Lung Cancer Participants: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Effects of Single Agent Niraparib and Niraparib Plus Programmed Cell Death-1 (PD-1) Inhibitors in Non-Small Cell Lung Cancer Participants (NCT03308942), a non-randomized phase 2 study. The primary outcome is Stage 2: Cohort 1A and Cohort 2A: Objective Response Rate.

2. Clinical Question

3. Trial Design

This was a Phase 2, multicenter, open-label, parallel study with non-randomized allocation. It was designed to evaluate niraparib alone and in combination with PD-1 inhibitors in locally advanced and metastatic non-small cell lung cancer. Overall enrollment was 53. The registry record lists the study as completed, with a start date of 2017-09-29 and primary completion on 2020-05-04.

The study had two stages and several clinically defined cohorts rather than randomized arms. In Stage 1, Cohort 1 included participants with high PD-L1 expression, TPS >= 50%, and no prior systemic chemotherapy or PD-1/PD-L1 inhibitor treatment; they received niraparib plus pembrolizumab. Cohort 2 included similarly untreated participants whose tumors had PD-L1 expression from 1% to 49%, and also received niraparib plus pembrolizumab. Cohort 3 included previously treated participants with squamous non-small cell lung cancer and received niraparib alone.

Stage 2 Cohorts 1A and 2A were intended for participants matching the high-expression and 1%-to-49% PD-L1 groups, respectively, without prior systemic chemotherapy or PD-1/PD-L1 inhibitor treatment. They were assigned niraparib plus TSR-042 (Dostarlimab). The registry record says no participant was enrolled in Cohort 2A and that data were not collected for it.

The study’s non-randomized structure is important when reading outcomes. Participants were separated by disease characteristics and treatment history, and the record does not describe a common control group. Comparisons across cohorts therefore mix differences in eligibility and treatment with any differences in outcomes. The registry record does not report a protocol-level sample-size justification, interim-analysis plan, or detailed statistical analysis plan.

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: A total of 69 participants were screened, of which 16 failed screening and 53 participants (41 in Stage 1 and 12 in Stage 2) were enrolled into the study. Data was not collected for Cohort 2A of Stage 2 as the Sponsor decided not to open this Cohort to enrollment.

Stage 1 (Up to a Maximum of 45 Months)

Reasons for not completing this period:

Stage 2 (Up to a Maximum of 29 Months)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Modified intent-to-treat (mITT) Population comprised of all participants who received any study drug and did not withdraw consent prior to having at least one post-Baseline tumor 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: Data was not collected for Cohort 2A of Stage 2 as no participant was enrolled into this Cohort of the study. Safety Population comprised of participants who received at least one dose of either study medications.

6. Primary Endpoint

The registry lists the primary outcome as Stage 2: Cohort 1A and Cohort 2A: Objective Response Rate, measured over Up to a maximum of 17 months.

The primary endpoint is objective response rate (ORR) for Stage 2 Cohorts 1A and 2A. It is defined as the percentage of participants in the mITT population with a confirmed best overall response of complete response or partial response, assessed by the investigator using RECIST v1.1. Complete response requires disappearance of all target lesions; any pathological lymph nodes must be <10 mm in the short axis. Partial response requires at least a 30% decrease in the sum of target-lesion diameters relative to the baseline sum. The stated assessment period is up to a maximum of 17 months.

ORR is a binary-outcome summary: each participant is classified according to whether a confirmed response occurred, and the endpoint is the proportion meeting that definition. It can describe observed tumor response in a defined population over the assessment period. The registry record specifies a binomial exact method for the 95% confidence interval, which communicates uncertainty around the estimated response percentage.

The endpoint result itself is not reported in the registry record. In addition, no participant entered Cohort 2A, so that cohort contributes no collected data. The record does not report whether the primary endpoint was analyzed only in Cohort 1A or how the empty cohort was handled in the reported endpoint analysis.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a percentage, formed by classifying participants according to whether they met the confirmed-response definition. This is a binary-outcome setting: each participant either meets the endpoint definition or does not. The registry record specifies an exact binomial method for the 95% confidence interval. General explanation: a confidence interval expresses the range of values compatible with the observed data under the method’s assumptions. It reflects sampling uncertainty; it is not a probability statement that the true response percentage lies within a particular interval after the interval has been calculated.

What does an exact binomial interval describe?

General explanation: binomial methods are suited to counts of participants classified into one of two outcome categories, such as response and no response. An exact interval uses the binomial probability model to obtain interval limits without relying on a normal approximation for the outcome proportion. Its width depends on the amount of information in the analyzed group and on the observed response count. A narrow interval indicates greater statistical precision than a wide interval, but precision does not remove systematic bias or make a non-randomized comparison causal.

The confidence interval method is specified, but the interval and the underlying ORR are not reported in the registry record. The endpoint description says the assessment was performed by the investigator according to RECIST v1.1, with confirmation required. It does not report details of independent review, adjudication, or the timing and frequency of tumor assessments beyond the stated maximum assessment period of 17 months.

Why does the analysis population matter?

The primary endpoint description names the mITT population. General explanation: an analysis population defines which participants contribute to an estimate, and its definition can affect both the estimate and its precision. For a response percentage, the denominator is the number of participants included in the analysis, while the numerator is the number meeting the response definition. The registry record does not give a detailed mITT definition or report the endpoint numerator, denominator, or estimate. It states that Cohort 2A had no enrolled participant and that data were not collected for that cohort. How that empty cohort was reflected in the endpoint analysis is not reported in the registry record.

General explanation: missing outcome data can matter when participants with unavailable assessments differ from those with observed assessments. The direction and size of any resulting bias depend on why data are missing and how the analysis handles them. The registry record does not describe a missing-data strategy for the other cohorts, or state how absent or unevaluable tumor assessments were treated. No assumptions about those procedures can be made from the endpoint definition alone.

What does non-randomized allocation imply?

General explanation: random allocation helps balance measured and unmeasured characteristics between groups on average. In a non-randomized study, group membership may instead reflect clinical eligibility or other factors, so outcome differences can be influenced by those factors as well as by treatment. Here, cohorts had different characteristics and regimens, including different prior treatment histories, PD-L1 expression groups, and histology criteria. Accordingly, comparisons across cohorts are descriptive rather than direct randomized estimates of treatment effects. The registry record itself notes that the arms are not a like-for-like control comparison.

The record describes the study as open-label, meaning masking was none. General explanation: when treatment assignment is known, expectations or knowledge of treatment may influence some aspects of care, reporting, or assessment. The endpoint was investigator-assessed, so the absence of masking is relevant context. The record does not report procedures intended to reduce assessment bias, such as independent review.

How should planning calculations be read?

The stated planning assumptions are 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 the observed effect. General explanation: a sample-size calculation estimates how many participants would be needed under specified assumptions to detect a target effect with a chosen design criterion. Changing the target effect or assumptions changes the planning result. The registry record’s planning illustration is not the trial’s own design justification, and the result values are not reported in the registry record for this analysis. It should not be described as observed power. Observed power is not a substitute for interpreting an estimated effect and its confidence interval.

For a percentage outcome, the normal approximation is a simplified model for planning, whereas the reported endpoint method is an exact binomial interval. These methods serve different purposes: a planning approximation estimates a required sample size under assumptions, while a confidence interval summarizes uncertainty around observed data. General explanation: approximation quality can depend on the amount and distribution of information; sparse outcomes can make simple normal approximations less reliable. No additional trial-specific assumptions about expected response rates, target effects, or cohort sizes are reported in the registry record.

Over-dispersion is a general concern when observed variability in grouped outcomes exceeds what a simple model assumes. It can make standard errors too small if ignored. The primary endpoint is described as a participant-level binary response with an exact binomial interval; the registry record does not report a model that addresses over-dispersion. Therefore, over-dispersion should be understood as a general statistical consideration rather than a stated feature of this analysis.

Finally, the record lists 32 outcome measures, while no analysis entries are reported. That distinction matters: outcome definitions describe what may be measured, whereas analysis reporting documents how results were estimated and compared. The absence of reported analysis entries does not establish that no analyses were conducted; it means detailed analysis results and procedures are not available in the registry record.

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("Stage 2 (Cohort 1A): Niraparib + TSR-042 (Dostarli")
n      <- c(11)
events <- c(1)
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  Stage 2 (Cohort 1A): Niraparib + TSR-042 (Dostarli  11  1  9.1  0.2  41.3  1.6  37.7  2.1  38.5

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists 32 outcome measures, including the primary response endpoint and secondary adverse-event outcomes. When many outcomes or cohort-specific analyses are considered, multiplicity can increase the chance of observing a nominally notable result by chance. General explanation: multiplicity procedures may control error across a defined family of tests, or results may be presented as exploratory without such adjustment. The record does not report a testing hierarchy, multiplicity adjustment, or a plan for interpreting multiple comparisons. It also reports no analysis entries, so the handling of multiple endpoints cannot be assessed from analysis results.

The primary endpoint is assigned to Stage 2 Cohorts 1A and 2A, but the record says that no participant was enrolled in Cohort 2A and that data were not collected for it. This is a structural absence of data for that cohort, not an observed zero response rate. The record does not explain whether the endpoint was analyzed in Cohort 1A alone, pooled across eligible Stage 2 participants, or handled in another way.

The endpoint description names an mITT population, but its membership criteria are not reported in the registry record. The record also does not report how missing tumor assessments, unevaluable responses, withdrawals, or other incomplete outcome data were handled in the remaining cohorts. General explanation: the treatment of missing outcome data can affect both the denominator and the estimated response percentage. Without the analysis population definition and missing-data rules, the scope of the primary estimate cannot be fully evaluated.

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: respiratory failure (33.3%), small intestinal obstruction (33.3%), dyspnoea (19.0%), pneumonia (17.6%), anaemia (11.8%).

The adverse-event summaries give cohort-specific counts of participants at risk and counts with serious or other adverse events. In Stage 1 Cohort 1, 11 of 17 participants were reported with serious adverse events and 17 of 17 with other adverse events. In Stage 1 Cohort 2, the corresponding counts were 14 of 21 and 20 of 21. In Stage 1 Cohort 3, they were 1 of 3 and 3 of 3. In Stage 2 Cohort 1A, they were 7 of 12 and 12 of 12. Stage 2 Cohort 2A had no enrolled participants and no event counts.

These are descriptive participant counts, not comparisons adjusted for differences in cohort composition or duration of exposure. The cohorts were not randomized like-for-like groups. The record also lists deaths in the adverse-event summaries: 11, 16, 1, and 6 for Stage 1 Cohorts 1, 2, and 3 and Stage 2 Cohort 1A, respectively; Cohort 2A has 0. The cause and treatment attribution of these deaths are not reported in the registry record.

The listed serious-event terms include Respiratory failure and Small intestinal obstruction at 33.3% each, Dyspnoea at 19.0%, Pneumonia at 17.6%, and Anaemia and Mental status changes at 11.8% each. The denominator and cohort distribution for these term percentages are not reported in the registry record, so they should not be treated as cohort-specific rates.

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

The registry record establishes the study question, cohort structure, endpoint definition, and selected adverse-event counts. It does not report the primary objective response rate or its 95% confidence interval. As a result, the response finding itself cannot be characterized for magnitude or precision from the record. Missing endpoint results are not equivalent to a result of zero, and they do not establish either benefit or lack of benefit.

Even if cohort-specific outcome estimates were available, the non-randomized design would require cautious interpretation. Participants entered cohorts defined by factors including tumor PD-L1 expression, prior treatment, histology, and study stage. The regimens also differed: some cohorts received niraparib with pembrolizumab, another received niraparib alone, and Stage 2 cohorts were assigned niraparib with TSR-042 (Dostarlimab). The cohorts therefore do not provide a like-for-like control comparison. Differences in outcomes could reflect differences in participants, treatment history, assessment, or other factors, in addition to any treatment effect.

The response endpoint is a defined measure of confirmed tumor response under investigator assessment using RECIST v1.1. It is not, by itself, a measure of how long a response lasts, how long participants live, or how participants feel. Those outcomes are not the primary endpoint described in the record. The maximum assessment period is up to 17 months, and the endpoint definition should be interpreted within that stated window.

The adverse-event counts describe the number of participants with events in each cohort. They do not establish that niraparib or a combination drug caused an event, and they are not a randomized comparison of safety. The record does not report exposure-adjusted rates or detailed event attribution in the registry record. The cohort counts should therefore be read as descriptive safety information with limited comparative meaning.

Overall, the record documents a completed Phase 2 study and its planned clinical question, but its central posted primary outcome is not reported. The design and the missing endpoint result constrain statistical conclusions. No treatment recommendation follows from these registry entries.

14. Why This Trial Matters Statistically

This study is statistically notable because it combines several clinically distinct cohorts within a non-randomized Phase 2 design. That structure makes the definition of each cohort essential to understanding any outcome: PD-L1 expression, prior systemic therapy, prior PD-1/PD-L1 treatment, histology, stage, and regimen vary across the study. A pooled or cross-cohort summary could obscure those differences unless its population and purpose were clearly specified.

The primary endpoint has a clear participant-level definition: confirmed complete or partial response under investigator assessment using RECIST v1.1. The record also names an exact binomial method for its confidence interval. Together, these details identify the intended outcome and its uncertainty method. Yet the endpoint result and interval are not reported in the registry record, making it impossible to assess the observed response level or its precision from the posted information.

The empty Stage 2 Cohort 2A illustrates why enrollment and analysis populations matter. The record explicitly states that no participant entered that cohort and that no data were collected there. This must not be read as a measured outcome, and it leaves the scope of the Stage 2 endpoint analysis unclear because the handling of the cohort in the analysis is not reported.

Safety information also requires attention to denominators and context. The record gives participant counts by cohort, while listed serious-event term percentages do not have a stated denominator in the described details. Moreover, differing cohort eligibility and non-random allocation limit direct comparisons. These are common statistical interpretation issues: summaries can be numerically precise yet still fail to support causal conclusions if the design does not make groups comparable.

Finally, the record lists many outcome measures but no analysis entries. This highlights the distinction between registering what will be measured and reporting how outcomes were analyzed. Complete statistical interpretation depends on both the design and the presentation of estimates, uncertainty, analysis populations, and missing-data handling.

15. Limitations

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

Several limitations are specific to this study record. First, the primary objective response rate and its 95% confidence interval are not reported. Without those results, the primary endpoint’s magnitude and precision cannot be evaluated. The record names the mITT population but does not define it in detail or give the number analyzed for the primary endpoint.

Second, Stage 2 Cohort 2A had no enrolled participant, and the record states that data were not collected for that cohort. The analysis handling of the empty cohort is not reported. This limits interpretation of a primary endpoint described as covering Cohorts 1A and 2A.

Third, allocation was non-randomized and cohorts differed in eligibility and regimen. The record itself cautions that the arms are not a like-for-like control comparison. Cohort outcome contrasts would therefore be vulnerable to confounding and should not be interpreted as randomized treatment effects.

Fourth, the study was open-label and the primary response assessment was investigator-based. The registry record does not report independent assessment or other measures to address possible assessment bias. Details of tumor-assessment timing, missing assessments, and handling of unevaluable responses are also not reported.

Finally, although the record lists 32 outcome measures, it reports no analysis entries. Multiplicity methods, the detailed statistical plan, and missing-data procedures are not reported in the registry record. The serious-event term percentages are listed without a denominator in the described details, and the causal attribution of deaths and other events is not given. These gaps constrain both efficacy and safety interpretation.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Effects of Single Agent Niraparib and Niraparib Plus Programmed Cell Death-1 (PD-1) Inhibitors in Non-Small Cell Lung Cancer Participants — NCT03308942.
• Registry record last updated 2023-10-16; status last verified 2023-10.
• Results first posted 2021-07-07.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

The registry record describes NCT03308942, a completed, multicenter, open-label Phase 2 study with non-randomized, parallel cohorts and overall enrollment of 53. It evaluated niraparib alone and in combination with PD-1 inhibitors in participants with locally advanced or metastatic non-small cell lung cancer. Stage 1 combination cohorts used pembrolizumab; Stage 2 combination cohorts used TSR-042 (Dostarlimab). Stage 1 Cohort 3 received niraparib alone.

The primary endpoint was objective response rate in the Stage 2 mITT population, based on confirmed complete or partial response assessed by the investigator using RECIST v1.1, with a stated maximum assessment period of 17 months and a binomial exact 95% confidence interval. No participant enrolled in Stage 2 Cohort 2A, and the primary endpoint result is not reported in the registry record. The record also lists cohort-level adverse-event counts, but the non-randomized design does not support direct like-for-like comparisons across cohorts.