A Study of Neoadjuvant Chemotherapy Plus Nivolumab Versus Neoadjuvant Chemotherapy Plus Placebo, Followed by Surgical Removal and Adjuvant Treatment With Nivolumab or Placebo for Participants With Surgically Removable Early Stage Non-small Cell Lung Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Neoadjuvant Chemotherapy Plus Nivolumab Versus Neoadjuvant Chemotherapy Plus Placebo, Followed by Surgical Removal and Adjuvant Treatment With Nivolumab or Placebo for Participants With Surgically Removable Early Stage Non-small Cell Lung Cancer (NCT04025879), a randomized phase 3 study. The primary outcome is Event-Free Survival (EFS) by BICR.

2. Clinical Question

This question combines the period before surgery with the period after surgery. An event can occur before surgery if disease progresses or worsens in a way that precludes surgery, or if attempted surgery is abandoned because the tumor is unresectable or disease has worsened. Events can also arise after surgery, or when surgery does not occur. Death due to any cause is included. The registry states that participants who do not undergo surgery for a reason other than progression are considered to have an event at RECIST 1.1 progression or death.

The comparison therefore concerns the full treatment strategy, not nivolumab in isolation from chemotherapy or from the subsequent treatment phase. The posted result is a time-to-event comparison, while pathologic complete response and major pathological response provide complementary measures of tumor response in the secondary outcomes.

3. Trial Design

This is a randomized, double-blind, parallel-group phase 3 trial sponsored by Bristol-Myers Squibb. The registry lists the condition as carcinoma, non-small-cell lung. The title identifies a comparison of neoadjuvant chemotherapy plus nivolumab with neoadjuvant chemotherapy plus placebo, followed by surgery and adjuvant nivolumab or placebo. The brief summary describes the objective as examining whether periadjuvant immunotherapy will prolong event-free survival.

The registry records a start date of 2019-11-05 and a primary completion date of 2023-07-26. Its status is active, not recruiting. Enrollment is listed as 461. The primary outcome analysis identifies all randomized participants as its population, with 229 in the nivolumab plus standard-of-care chemotherapy arm and 232 in the placebo plus standard-of-care chemotherapy arm. Those outcome denominators need not match denominators used for other outcomes, such as safety.

Randomization is intended to create groups that are comparable on average at the start of a trial, while double masking can reduce the influence of treatment knowledge on participant care and outcome assessment. Those are general features of randomized, masked studies; the registry record does not describe the randomization procedure or how masking was maintained. BICR means blinded independent central review, and the primary outcome description specifies that progression and recurrence were assessed in that manner according to RECIST 1.1.

The intervention list includes nivolumab, placebo, and chemotherapy drugs. The arm titles describe the regimens at a broad level, and the official title specifies the resectable stage II-IIIB population. The registry record does not provide more detail in the arm descriptions. It also does not report the original design assumptions used to choose enrollment, the planned allocation ratio, or a sample-size justification. Those omissions limit what can be inferred about the original planning from the posted outcome results.

Several secondary outcomes are named, including overall survival, pathologic complete response, major pathological response, and adverse events. Detailed posted numerical results in the record include pathologic complete response and major pathological response, along with adverse-event counts. Numerical overall-survival results 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.

Randomization

Reasons for not completing this period:

Treatment

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All randomized participants

5. Baseline Characteristics

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

6. Primary Endpoint

The registry lists the primary outcome as Event-Free Survival (EFS) by BICR, measured over From randomization to disease progression, worsening, recurrence, or death due to any cause (up to approximately 44 months).

Event-free survival is the primary endpoint. It measures the time from randomization until the first qualifying event described in the registry: progression or worsening that prevents surgery; an attempted operation abandoned because the tumor is unresectable or disease has worsened; progression or recurrence after surgery; progression or recurrence when surgery does not occur; or death due to any cause. Progression and recurrence are assessed by blinded independent central review under RECIST 1.1. For participants who do not undergo surgery for a reason other than progression, the registry considers progression or death to be an event.

A time-to-event endpoint is useful when the timing of an event matters and participants may have different lengths of follow-up. It can incorporate both whether an event occurs and when it occurs, rather than treating all participants as if they had identical observation time. General statistical context: participants without a recorded event by the analysis time may contribute information up to their last known event-free time, depending on the analysis and censoring rules. The registry record does not report the detailed censoring rules or missing-data procedures.

The event definition spans more than tumor progression alone: it includes surgical feasibility, recurrence, and death. That makes the endpoint relevant to the overall treatment pathway but also means its result cannot identify which event component accounts for the comparison. The reported median and hazard ratio describe different aspects of the endpoint: the median is a time point, while the hazard ratio compares event rates over follow-up under the model used. Neither alone describes every participant's clinical course.

7. Results Posted in the Registry

• Nivolumab + SOC Chemotherapy vs Placebo + SOC Chemotherapy: hazard ratio 0.58 (95% CI 0.43–0.78), reported p < 0.001 (as reported in the registry).

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Pathologic Complete Response (pCR) Rate

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Placebo + SOC Chemotherapy: 4.7% (11/232) versus 25.3% (58/229) on control; risk difference -20.6 percentage points (95% CI -26.8 to -14.3), risk ratio 0.19 (0.10–0.35); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically lower for Placebo + SOC Chemotherapy than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Major Pathological Response (MPR) Rate

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Placebo + SOC Chemotherapy: 12.1% (28/232) versus 35.4% (81/229) on control; risk difference -23.3 percentage points (95% CI -30.8 to -15.8), risk ratio 0.34 (0.23–0.50); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically lower for Placebo + SOC Chemotherapy than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The posted secondary outcome pattern is directionally consistent: both pathological response measures have higher arm-level percentages in the nivolumab plus chemotherapy group than in the placebo plus chemotherapy group. These measures concern response in pathological assessment and complement event-free survival by describing a different aspect of the treatment course. They do not establish which components of the primary composite event account for the time-to-event result, nor do they replace follow-up for progression, recurrence, or death.

The registry also names overall survival and several adverse-event outcomes among its secondary outcomes, but numerical overall-survival results are not reported in the registry record. Adverse-event results are posted separately and should be interpreted as safety descriptions rather than evidence about efficacy. The presence of an outcome in the registry's list does not mean a numerical result is available for every listed measure.

Unadjusted secondary comparisons carry less weight for confirmatory claims when multiple outcomes are examined. General statistical explanation: testing many hypotheses increases the chance that at least one result will appear unusually strong by chance, unless the testing strategy accounts for multiplicity. The registry record does not report a multiplicity adjustment or hierarchy for the secondary outcomes. Their estimates and intervals remain useful for describing the observed data, but their inferential status cannot be fully judged without that analysis plan.

Pathological response and event-free survival may be related in a biological or clinical context, but a relationship between outcomes does not establish that one mediates the other. The registry results alone do not quantify mediation or establish that a response measure predicts an individual participant's later outcome. The secondary results therefore add context while leaving the primary endpoint as the main posted time-to-event comparison.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Overall Survival (OS)
• Pathologic Complete Response (pCR) Rate
• Major Pathological Response (MPR) Rate
• The Number of Participants With Adverse Events (AEs)
• The Number of Participants With Serious Adverse Events (SAEs)
• The Number of Participants With Select Adverse Events (AEs)

9. Statistical Interpretation

10. Statistical Methods Explained

Time-to-event analysis. Event-free survival records time from randomization to a qualifying event. Unlike a simple proportion, it uses the timing of events and can accommodate participants whose observation ends before an event is documented. General statistical explanation: such participants are often censored at their last known event-free time, so their partial follow-up can still inform the analysis. The registry record does not report the detailed censoring rules, the number censored, or the way missing follow-up was handled.

The median is the time by which half of the estimated event-time distribution has experienced an event, under the method used to estimate that distribution. It is not an average follow-up time, nor does it mean that each participant experiences an event at that time. The registry posts a median for the placebo arm but no numerical median for the nivolumab arm; it reports a lower confidence bound of 28.94 months and no upper bound for that arm. A missing median value should not be replaced by an estimate inferred from the other statistics.

Hazard ratio and Cox model. The reported hazard ratio compares the instantaneous event hazards between the two groups over follow-up, conditional on remaining event-free until that point. The estimate of 0.58 is for nivolumab plus standard-of-care chemotherapy versus placebo plus standard-of-care chemotherapy. A value below 1 indicates a lower estimated hazard in the first-listed group under the model. It is not a direct estimate of a difference in event-free time or of an absolute risk difference.

The registry identifies a Cox proportional-hazards parameter and reports a 95% confidence interval of 0.43 to 0.78. General statistical explanation: the Cox model estimates a relative hazard while allowing the underlying baseline hazard to vary over time. The proportional-hazards assumption says that the relative hazard is sufficiently stable over the analysis period. If that assumption is not reasonable, a single hazard ratio may not describe the time pattern well. The record does not report a diagnostic assessment, alternative time-varying estimates, or model covariates.

The log-rank method is reported for the primary comparison. General statistical explanation: the log-rank test compares the event-time experience of groups across follow-up, using the ordering and timing of events. Its p-value assesses how compatible the observed comparison is with a null hypothesis of no difference under the test's assumptions; it is not the probability that the null hypothesis is true. The reported p-value is < 0.001. The registry does not describe an adjusted analysis or additional testing details.

Confidence intervals. A confidence interval communicates uncertainty around an estimate under the statistical procedure used. Repeated use of the same procedure in comparable settings would produce intervals with the stated coverage property. It is not a range containing a fixed percentage of individual participants' outcomes. For the hazard ratio, the posted interval excludes 1, which is the null value for a ratio. The interval still allows a range of effect magnitudes; it does not make the point estimate exact.

Pathological response outcomes. Pathologic complete response and major pathological response are binary outcomes reported as percentages with 95% confidence intervals. The posted percentages are 25.3% and 4.7% for pathologic complete response, and 35.4% and 12.1% for major pathological response, in the nivolumab and placebo arms respectively. A binary percentage summarizes the fraction meeting the outcome definition, while its confidence interval reflects sampling uncertainty in that estimate. The registry also reports sponsor analyses for these outcomes, including differences and odds ratios.

Recalculated comparisons for the secondary binary outcomes are simple unadjusted comparisons from the posted summary statistics. They do not incorporate the trial's full individual-level data or any unreported adjustment. The registry's sponsor analyses and the simple recalculations answer related but not necessarily identical statistical questions. A risk difference is an absolute contrast between proportions; an odds ratio compares odds rather than probabilities. An odds ratio should not be read as a risk ratio, particularly when outcomes are not rare. No discrepancy is flagged for the posted comparisons.

Over-dispersion and model fit. General statistical context: over-dispersion occurs when observed variability is greater than a model assumes. It can make standard errors too small and confidence intervals too narrow if not addressed. It is especially relevant to count or clustered data, but the registry does not report an over-dispersion analysis for this time-to-event endpoint. More broadly, the posted summary does not provide enough model diagnostics to assess fit independently.

Safety counts. The adverse-event results are counts of participants with events among the stated at-risk denominators, alongside the registry's list of serious adverse-event terms. These are descriptive safety summaries. A count alone does not establish that an event was caused by the assigned treatment, and comparisons can be affected by exposure time, event definitions, and reporting practices. The record does not provide enough detail to assess those factors fully.

11. Confidence Intervals and Precision

The primary hazard-ratio estimate is 0.58, with a 95% confidence interval from 0.43 to 0.78. The interval communicates uncertainty in the estimated relative hazard. Its limits are below 1, the null value for a ratio, while still spanning more than one possible effect magnitude. The reported p-value is < 0.001. Statistical evidence against a null comparison does not remove uncertainty about the size of the association or its applicability beyond the studied population and follow-up.

The arm-level median estimates have different reporting completeness. For placebo plus standard-of-care chemotherapy, the median is 18.43 months, with a 95% confidence interval from 13.63 to 28.06 months. For nivolumab plus standard-of-care chemotherapy, the registry reports a lower confidence bound of 28.94 months but no numerical median or upper confidence bound. This makes a direct median-to-median comparison unavailable from the posted values. The missing values are not grounds for extrapolating a median.

Confidence intervals for the secondary response percentages also indicate uncertainty. Pathologic complete response is reported as 25.3% with a 95% interval of 19.8% to 31.5% in the nivolumab arm, and 4.7% with a 95% interval of 2.4% to 8.3% in the placebo arm. Major pathological response is 35.4% with a 95% interval of 29.2% to 41.9%, and 12.1% with a 95% interval of 8.2% to 17.0%, respectively. Each interval pertains to its own outcome and arm estimate.

General statistical explanation: interval width depends on the amount of information and the variability of the data, as well as the method used. A narrow interval indicates greater precision under that method, not freedom from bias or uncertainty about study design. The registry record does not provide all details needed to assess the modeling, censoring, or missing-data choices that contribute to the reported precision.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are illustrations based on the observed effect, not a reconstruction of the trial's original design. For event-free survival, the figures are total event counts across groups. For the binary response outcomes, the figures are participants per arm. They are not observed participant counts or event totals from the trial.

The listed planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The planning effect is set to the stated fraction of the observed effect. The time-to-event illustrations use the Schoenfeld approach and express the requirement as total events; the other illustrations express the requirement as participants in each arm.

The 80% power column describes the planning size that, under those assumptions and the selected effect, would give an 80% chance of rejecting the null hypothesis if that effect were the true effect. The 90% power column expresses the corresponding planning size for a 90% chance. These are prospective planning concepts, not a measure of what the completed trial actually achieved.

The rows consider planning for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. The smaller-effect scenarios require more information: more total events for the time-to-event calculation or more participants per arm for the binary outcomes. That pattern reflects the general statistical principle that smaller differences are harder to distinguish from random variation.

This is not observed power. The registry record does not report the original sample-size justification or the design assumptions used when the trial was planned. The displayed calculations are illustrative and should be read as conditional on their stated assumptions, not as proof that a particular number of participants or events is sufficient in another study.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

d is the total number of events needed (Schoenfeld's formula) with 1:1 allocation and proportional hazards.

Worked calculation for this trial

1. Hazard ratio HR = 0.580, so ln(HR) = -0.5447.
2. d = 4 × (1.96 + 0.842)² / [ln(0.580)]² = 105.8, rounded up to 106 events in total for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
hr <- 0.58                              # observed hazard ratio
events_needed <- function(power, frac = 1) ceiling(round(4 * (z_a + qnorm(power))^2 / log(hr^frac)^2, 6))   # Schoenfeld, 1:1 allocation
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: events = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(events_needed(0.80, f), big.mark = ","), format(events_needed(0.90, f), big.mark = ",")))

# Simulation: given d total events under proportional hazards, events in the treatment arm are Binomial(d, hr / (1 + hr)).
# The score (log-rank type) statistic is (D1 - d / 2) / sqrt(d / 4).
set.seed(4025879)
d    <- events_needed(0.80)
reps <- 20000
D1 <- rbinom(reps, d, hr / (1 + hr))
z  <- (D1 - d / 2) / sqrt(d / 4)
cat(sprintf("Simulated power with %s events: %.3f (%d replicates)\n", format(d, big.mark = ","), mean(abs(z) > z_a), reps))

Output from running the code

100% of observed effect: events = 106 (80% power), 142 (90% power)
 75% of observed effect: events = 189 (80% power), 252 (90% power)
 50% of observed effect: events = 424 (80% power), 567 (90% power)
Simulated power with 106 events: 0.763 (20000 replicates)

The simulation uses the fact that, with d events in total and proportional hazards, the number of events in the treatment arm is binomial. It is an approximation that conditions on the total number of events, and it should land close to the target power.

14. Analyses Reported by the Sponsor

The primary time-to-event estimate is the sponsor's reported analysis: a Cox proportional-hazards hazard ratio of 0.58, with a 95% confidence interval from 0.43 to 0.78 and p< 0.001, using the log-rank method. The primary endpoint is treated as a time-to-event outcome, and the posted analysis uses the sponsor's reported hazard ratio rather than a separate recalculation from summary statistics. The median values are reported separately and should not be mistaken for the hazard ratio.

For pathologic complete response and major pathological response, the registry also reports sponsor analyses of differences and odds ratios. Simple unadjusted recalculations from posted arm summaries are a separate kind of analysis and may differ in estimand or method from the sponsor's analysis. The record does not flag a discrepancy between the posted sponsor results and the recalculated summaries. The sponsor's reported analyses should therefore be interpreted as the trial's reported inferential results, while the recalculations are descriptive checks based on aggregate information, not replacements for the trial analysis.

The registry does not provide detailed methods for all secondary sponsor analyses, including adjustment choices or the handling of missing data. Their estimates should be read with that limitation in mind, and the secondary outcomes should not be treated as interchangeable with the primary endpoint.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary event-free survival outcome and multiple secondary outcomes, including overall survival, pathological response, and adverse events. It does not report a statistical testing hierarchy, multiplicity adjustment, or rules for interpreting secondary p-values. When several hypotheses are tested, the chance of at least one false-positive result can increase. General statistical methods to address this include prespecifying an order of testing or adjusting the decision threshold, but the record does not say which, if any, were used.

The record does not report the detailed handling of missing outcome data, censoring, withdrawals, or participants who lacked an evaluable pathological assessment. It states that all randomized participants form the primary endpoint population and explains how a participant who does not undergo surgery for a reason other than progression is considered to have an event at progression or death. It does not give a full analysis-population definition or describe additional rules for incomplete follow-up.

The primary endpoint uses BICR assessment according to RECIST 1.1 for progression and recurrence. The registry does not report adjudication details beyond that description, nor does it provide a sensitivity analysis for alternative event definitions. It also does not report whether the proportional-hazards assumption was assessed or what steps were taken if that assumption was not met.

These omissions do not invalidate the posted estimates, but they constrain independent evaluation of their robustness. They also matter when considering how much weight to place on secondary findings: the record does not describe how multiplicity, missing data, or analysis choices were handled across the full set of outcomes.

16. 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 (4.3%), malignant neoplasm progression (2.6%), acute kidney injury (2.2%), haemoptysis (2.2%), pneumonitis (2.2%).

The registry reports serious adverse events for 109 of 228 participants at risk in the nivolumab plus standard-of-care chemotherapy group, and 83 of 230 participants at risk in the placebo plus standard-of-care chemotherapy group. Other adverse events are reported for 214 of 228 and 212 of 230 participants, respectively. The record also lists deaths of 40 and 48 participants in the two groups. These are descriptive counts using the stated denominators; they are not adjusted comparisons and should not be interpreted as proof that treatment caused or prevented an event.

The listed serious adverse-event terms include pneumonia at 4.3%, malignant neoplasm progression at 2.6%, acute kidney injury at 2.2%, haemoptysis at 2.2%, pneumonitis at 2.2%, and colitis at 1.8%. These entries describe reported terms and percentages, not a complete account of every possible safety outcome. The registry record does not report exposure-adjusted rates or enough detail to evaluate differences in observation time, severity gradations, or attribution.

Safety counts answer a different question from event-free survival and pathological response. They should be read alongside the definitions and denominators stated for each outcome, without treating raw counts as direct measures of comparative risk.

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

The primary result describes an association between treatment assignment and the timing of a broad event-free survival outcome in the randomized population. The reported hazard ratio is below 1, and its 95% confidence interval is also below 1. The result is statistically informative under the reported analysis, but it does not by itself establish the size of benefit for every participant or predict an individual outcome.

The endpoint includes progression or worsening that can prevent surgery, surgical abandonment in specified circumstances, recurrence or progression after surgery or without surgery, and death due to any cause. It therefore reflects multiple clinical events across the neoadjuvant and adjuvant treatment pathway. The analysis cannot, from the summary result alone, show which component contributed most to the observed comparison. It also cannot isolate the effect of nivolumab from chemotherapy or from the full sequence of assigned care.

The posted median and hazard-ratio statistics are not interchangeable. The hazard ratio summarizes a relative comparison over follow-up under the Cox model; the median is a point in the estimated event-time distribution. The registry does not provide a numerical median for the nivolumab arm, so a direct comparison of the two posted medians is not available. The confidence interval conveys uncertainty but does not address every possible source of bias or model misspecification.

The pathological response results add evidence about a different outcome domain, while the safety results describe reported adverse events. Neither outcome domain alone establishes the long-term overall-survival effect. Numerical overall-survival results are not reported in the registry record. Nor does the posted summary settle questions about treatment choice, individual suitability, or outcomes in populations outside the registry's resectable stage II-IIIB description. It is a statistical account of the reported trial comparison, not medical advice.

18. Why This Trial Matters Statistically

This trial is statistically notable because it evaluates a treatment strategy across both preoperative and postoperative phases, while using event-free survival to capture events along that pathway. The endpoint definition connects disease progression, surgical feasibility, recurrence, and death rather than reducing the comparison to a single short-term response measure. That breadth can make the outcome clinically informative, but it also makes interpretation dependent on the precise event rules.

The randomized, double-blind design supports a structured comparison between the nivolumab-containing regimen and the placebo regimen, with standard-of-care chemotherapy in both arms. The primary analysis reports a hazard ratio and confidence interval, allowing readers to assess both the estimated relative comparison and its statistical precision. The use of BICR under RECIST 1.1 for progression and recurrence is also important to understanding how events were assessed.

Pathological complete response and major pathological response provide complementary views of treatment response. Their direction is consistent with the primary result, but they are not substitutes for the time-to-event endpoint, and the absence of a reported multiplicity strategy limits how strongly multiple secondary findings can be presented as confirmatory.

The record also illustrates why summary statistics should be read with care. The placebo median is reported, while the nivolumab median is not numerically stated; the hazard ratio supplies a separate time-to-event comparison. Safety counts use their own denominators and should not be confused with efficacy results. Taken together, the trial's statistical value lies in its randomized comparison and reported time-to-event estimate, with important interpretive limits where analysis details are not reported.

19. 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 registry record does not report the original sample-size justification or the design assumptions used when enrollment was planned. Planning illustrations based on observed effects cannot establish what the original investigators anticipated. The record also does not provide a detailed statistical analysis plan, the full set of analysis populations, or the handling of missing data and censoring.

The primary endpoint result is reported as a Cox proportional-hazards estimate, but diagnostics for the proportional-hazards assumption are not reported in the registry record. The event-free survival definition combines several events, and component-specific results are not reported. The numerical median for the nivolumab arm is not reported, and its upper confidence bound is also not reported, so the posted summaries do not support a direct median comparison.

Although several secondary outcomes are named, numerical overall-survival results are not reported in the registry record. The record does not report how multiplicity was handled across the outcomes or whether secondary analyses followed a prespecified hierarchy. This limits the strength of confirmatory interpretation for the secondary findings.

Safety results are presented as counts and selected serious adverse-event terms. Exposure-adjusted rates, detailed event attribution, and further information about the duration of safety observation are not reported in the registry record. The arm descriptions also provide limited regimen detail. These limitations constrain independent assessment of model assumptions, outcome completeness, and the comparability of outcome-specific denominators.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study of Neoadjuvant Chemotherapy Plus Nivolumab Versus Neoadjuvant Chemotherapy Plus Placebo, Followed by Surgical Removal and Adjuvant Treatment With Nivolumab or Placebo for Participants With Surgically Removable Early Stage Non-small Cell Lung Cancer — NCT04025879.
• Registry record last updated 2026-08-26; status last verified 2026-08.
• Results first posted 2024-07-23.
• 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.

23. Record Summary

The registry describes a randomized, double-blind, parallel-group phase 3 study of nivolumab plus standard-of-care chemotherapy versus placebo plus standard-of-care chemotherapy in resectable stage II-IIIB non-small cell lung cancer. The primary outcome is event-free survival assessed from randomization, with progression and recurrence evaluated by BICR under RECIST 1.1.

The reported primary analysis gives a hazard ratio of 0.58, with a 95% confidence interval of 0.43 to 0.78 and p< 0.001. The placebo-arm median is 18.43 months; a numerical nivolumab-arm median is not reported. Pathological response results and safety counts are also posted. Numerical overall-survival results and several analysis details, including the multiplicity approach and missing-data procedures, are not reported in the registry record.