Study of Nivolumab Given Sequentially With Ipilimumab in Subjects With Advanced or Metastatic Melanoma (CheckMate 064): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Nivolumab Given Sequentially With Ipilimumab in Subjects With Advanced or Metastatic Melanoma (CheckMate 064) (NCT01783938), a randomized phase 2 study. The primary outcome is Percentage of Participants With Treatment-Related Grade 3-5 Adverse Events (AEs) During the Induction Period (Period 1 and 2).

2. Clinical Question

The study compares treatment order, not the presence versus absence of either drug: both groups received nivolumab and ipilimumab sequentially. The primary endpoint is a safety outcome, so it addresses the frequency of participants with at least one qualifying adverse event during induction. It does not, by itself, describe the full range, duration, or severity distribution of all adverse events.

The registry also names investigator-assessed response rate at Week 25, duration of response, and rate of progression as secondary outcomes. Numerical results are posted for response rate at Week 25, but numerical results for duration of response and rate of progression are not reported in the registry record. The central statistical question is therefore the comparison of the induction-period safety percentages, with the response result providing a secondary outcome rather than replacing the primary safety question.

3. Trial Design

This was a randomized, open-label, parallel-group phase 2 study of nivolumab given sequentially with ipilimumab in participants with advanced or metastatic melanoma. The registry reports enrollment of 138. It lists the study status as completed, a start date of 2013-04-30, and a primary completion date of 2015-04-03. Randomization is reported; the method used to generate the allocation sequence and whether allocation was concealed are not reported in the registry record.

The two groups differed in treatment order. One received nivolumab followed by ipilimumab, and the other received ipilimumab followed by nivolumab. The registry describes nivolumab as an intravenous solution at 3 mg/kg every 2 weeks during induction, up to 6 doses, and describes continued dosing in the nivolumab-first group. It describes ipilimumab as an intravenous solution at 3 mg/kg every 3 weeks, up to 4 induction doses, in the ipilimumab-first group. The full subsequent regimen details for the latter group are not reported in the registry record.

The trial was not masked. Participants and study personnel therefore knew which sequence was assigned, a feature relevant to outcomes involving investigator assessment and to decisions made during care. Randomization can support a comparison between assigned groups, but it does not remove the possible influence of open-label assessment or ensure that every participant followed the same treatment exposure.

The registry identifies Bristol-Myers Squibb as sponsor. It names safety and efficacy as the study purpose and lists several secondary outcomes, but the original sample-size justification, planned analysis methods, analysis populations beyond the primary endpoint description, and any interim-analysis plan 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: Of the 140 randomized,138 treated because 1 experienced adverse event unrelated to study drug and 1 no longer met study criteria.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All treated 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 Percentage of Participants With Treatment-Related Grade 3-5 Adverse Events (AEs) During the Induction Period (Period 1 and 2), measured over From Day 1 to up to Week 25.

The primary endpoint is the percentage of treated participants who had at least one treatment-related grade 3–5 adverse event during the induction period, identified as Period 1 and Period 2. The timeframe is Day 1 to up to Week 25. The registry defines the event as an adverse event with onset on or after the first day of Induction Period 1 and no later than discontinuation from Induction Period 2. An adverse event may be a new unfavorable symptom, sign, or disease, or worsening of a pre-existing condition; the record notes that it may or may not have a causal relationship with treatment. For this endpoint, treatment-related means a relationship to study drug assessed as certain, probable, possible, or missing.

Grades 3–5 cover severe events, potentially life-threatening or disabling events, and death, respectively. The endpoint counts participants with at least one qualifying event, rather than counting every event or expressing the total burden of adverse events. It is used to summarize the occurrence of more serious treatment-related events over the specified induction period. Because it is a participant-level percentage, it does not show whether events were repeated, how long they lasted, or how many different event types an individual experienced.

The primary outcome is explicitly defined for all treated participants. The registry does not report a separate statistical analysis plan explaining how incomplete event follow-up or treatment discontinuation was handled beyond the endpoint's stated time window.

7. Results Posted in the Registry

• Ipilimumab Followed by Nivolumab: 51.4% (36/70) versus 64.7% (44/68) on control; risk difference -13.3 percentage points (95% CI -29.6 to 3.0), risk ratio 0.79 (0.60–1.06); unadjusted two-proportion z-test p = 0.114.

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.

Investigator-Assessed Response Rate at Week 25

• 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:
  • Ipilimumab Followed by Nivolumab: 30.0% (21/70) versus 54.4% (37/68) on control; risk difference -24.4 percentage points (95% CI -40.4 to -8.4), risk ratio 0.55 (0.36–0.84); unadjusted two-proportion z-test p = 0.004.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically lower for Ipilimumab Followed by Nivolumab than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary outcomes broaden the trial’s focus beyond the primary induction-period safety endpoint. The registry lists investigator-assessed response rate at Week 25, duration of response, and rate of progression. Numerical results are posted for response rate, but numerical results for duration of response and rate of progression are not reported in the registry record.

The broad pattern of the two outcomes with posted numerical comparisons is not a simple safety-and-response advantage for one sequence. The nivolumab-first group has the higher reported response percentage, while it also has the higher reported percentage for the primary treatment-related grade 3–5 adverse-event outcome. These are different outcomes with different meanings. Response is an efficacy-related measure; the primary endpoint is a safety measure. Their directions should be considered together rather than treating either as a complete summary of the sequence comparison.

The Week 25 response result is a secondary finding, and the posted comparison is unadjusted. When several outcomes are examined, the chance of finding an apparently notable result by chance can increase. A multiplicity procedure can address that concern, but whether one was planned or used is not reported in the registry record. The response result therefore carries less confirmatory weight than a result from a clearly prespecified primary analysis with a reported testing strategy.

The registry names duration of response and progression rate, but the absence of numerical results means their contribution to the overall pattern cannot be assessed from the registry record. In particular, the response-rate result alone does not describe how long responses lasted or how progression outcomes compared. The posted secondary information is thus partial and should be read as supportive descriptive evidence rather than a complete account of all listed secondary endpoints.

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

• Investigator-Assessed Response Rate at Week 25
• Investigator-Assessed Duration of Response (DOR)
• Investigator-Assessed Rate of Progression

9. Statistical Interpretation

10. Statistical Methods Explained

The statistical results are most readily read as comparisons of participant-level percentages. For each sequence, the primary endpoint classifies a treated participant as having or not having at least one qualifying treatment-related grade 3–5 adverse event during induction. This endpoint is binary at the participant level even though a participant may experience multiple adverse events. The percentage therefore summarizes the proportion meeting the endpoint definition, not the number of events in total.

What does a risk difference describe?

A risk difference subtracts one group’s event proportion from the other’s. The posted primary risk difference is -0.133 for ipilimumab followed by nivolumab relative to nivolumab followed by ipilimumab. Its negative sign indicates that the reported proportion is lower in the second sequence. The interval from -0.296 to 0.030 expresses uncertainty about the contrast. Since it includes zero, the data do not provide a clear separation between the groups on this measure under the posted unadjusted comparison. An interval containing zero does not prove that the true difference is zero; it indicates that the data are compatible with no difference as well as with differences on either side of zero.

The posted unadjusted p-value for this comparison is 0.114. A p-value describes how incompatible the observed data are with a specified null hypothesis, under the assumptions of the test. It is not the probability that the null hypothesis is true, and it does not measure clinical importance. The registry does not report the test specification, adjustment covariates, or a prespecified testing hierarchy. The p-value should therefore be read as the posted unadjusted result, not as evidence of a broader analysis plan.

How do relative measures differ?

A risk ratio compares the event proportions by division, while an odds ratio compares the odds of an event. Both express a relative contrast rather than an absolute difference. For the primary outcome, the posted risk ratio is 0.795, with an interval from 0.596 to 1.06; the posted odds ratio is 0.578, with an interval from 0.292 to 1.144. These measures use different scales, so their values should not be interpreted as interchangeable. The registry does not report adjusted versions of these estimates. Relative measures can also appear more or less dramatic than an absolute contrast depending on the baseline event proportion.

What is the role of a confidence interval?

In general, a confidence interval is a range of estimates compatible with the data and the method used to construct it. It conveys precision as well as direction: a narrower interval indicates less statistical uncertainty than a wider interval, all else equal. It is not a guarantee that the true effect lies within the particular interval. The primary comparison's interval crosses zero, whereas the posted Week 25 response comparison interval does not. Those interval patterns should be interpreted in light of the outcomes' roles: response rate is secondary, and the registry does not report an adjustment for examining multiple outcomes.

There is a notable reporting inconsistency in the primary arm-level result. The stated 64.7% estimate lies outside its listed interval of 37.6% to 62.4%. The registry record does not explain the discrepancy. The separate group-level and comparison estimates should not be silently reconciled or substituted for one another; the inconsistency reduces confidence in interpreting that arm-level interval.

Why do event-count models matter?

General statistical note: count outcomes can be analyzed with models that account for the variability of counts. When observed variation is greater than a model’s assumptions allow, over-dispersion can make standard errors too small and intervals too narrow if it is ignored. That issue is not the basis of the primary endpoint here, which is a percentage of participants with at least one event rather than an event count. The registry does not report a model for repeated event counts.

What can be concluded from the reported analysis?

The record posts arm-level percentages and an unadjusted comparison for the primary outcome, as well as an unadjusted comparison for Week 25 response. It does not report a statistical analysis plan, a covariate-adjusted analysis, or a formal multiplicity procedure. The results can therefore be described as unadjusted comparisons between randomized sequences, while the degree to which they reflect prespecified analyses is not reported in the registry record. Randomization is a design feature, but it does not supply missing details about the statistical methods or resolve the inconsistency in the primary arm-level interval.

11. Confidence Intervals and Precision

The precision of an estimate is reflected in the width of its interval: a wider interval leaves more uncertainty about the underlying value. For the primary comparison, the risk-difference interval extends from -0.296 to 0.030. It crosses zero and includes both a negative contrast and a small positive contrast on the stated scale. The posted unadjusted p-value is 0.114. Together, these results do not give a precise basis for distinguishing the sequences on the primary endpoint.

The interval listed for the primary endpoint in the nivolumab-followed-by-ipilimumab arm is 37.6% to 62.4%, while the point estimate is 64.7%. Because the reported estimate falls outside its interval, the arm-level interval cannot be interpreted as a conventional interval surrounding that estimate without further clarification. No explanation is reported in the registry record. The interval for the ipilimumab-followed-by-nivolumab arm is 31.1% to 55.3% around its reported estimate of 51.4%.

For the Week 25 response outcome, the registry explicitly labels the arm-level intervals as 95% confidence intervals: 40.0% to 68.4% for nivolumab followed by ipilimumab, and 17.7% to 45.8% for ipilimumab followed by nivolumab. The unadjusted risk-difference interval is -0.404 to -0.084. These intervals convey uncertainty in the secondary result, but do not account for the fact that multiple outcomes were considered.

Intervals are not measures of clinical importance, and their interpretation depends on the analysis method. The construction method for the primary interval estimates is not reported in the registry record. The open-label design and the lack of reported details about missing data and analysis populations further limit interpretation of 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 participants per arm for the binary outcomes in this trial. They are not total event counts. The planning illustration assumes a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. It uses the observed effect as a reference and considers planning for 100%, 75%, or 50% of that effect. These are illustration assumptions; the original trial’s sample-size assumptions and justification are not reported in the registry record.

For the primary safety outcome, planning for 100% of the observed effect gives 216 participants per arm for 80% power and 289 participants per arm for 90% power. Planning for 75% of the observed effect gives 380 participants per arm for 80% power and 508 participants per arm for 90% power. Planning for 50% of the observed effect gives 844 participants per arm for 80% power and 1,129 participants per arm for 90% power.

For the Week 25 response outcome, the corresponding figures are 64 and 84 participants per arm for planning at 100% of the observed effect; 115 and 154 participants per arm for planning at 75%; and 262 and 351 participants per arm for planning at 50%, with the 80% power figure stated first and the 90% power figure second in each case. A smaller target effect requires more participants in these illustrations because a less pronounced contrast is harder to distinguish from random variation.

In plain language, the 80% power column represents the planned chance of detecting the specified effect under the stated assumptions if that effect were present; the 90% column uses a higher target chance of detection. This is not observed power. The figures are planning illustrations based on observed effects, not evidence about the probability that the completed trial detected an effect.

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

n is the number of participants in each arm, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.647; treatment proportion p₁ = 0.514; average p̄ = 0.581.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6978 = 1.3677.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6915 = 0.5820.
4. n = (1.3677 + 0.5820)² / (-0.1328)² = 215.6, rounded up to 216 participants per arm 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
p0 <- 0.647058823529412                          # control proportion
p1 <- 0.514285714285714                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(1783938)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 216 (80% power), 289 (90% power)
 75% of observed effect: n per arm = 380 (80% power), 508 (90% power)
 50% of observed effect: n per arm = 844 (80% power), 1,129 (90% power)
power.prop.test: 215.6 
Simulated power at n = 216 per arm: 0.797 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when multiple outcomes, comparisons, time points, or analyses are examined. Without an appropriate prespecified approach, interpreting each result as if it were the sole test can overstate how persuasive an isolated unadjusted p-value is. This trial lists several secondary outcomes, and the registry posts a numerical comparison for response rate at Week 25. It does not report whether the secondary outcomes were organized into a testing hierarchy, whether any adjustment was made, or how the primary and secondary hypotheses were intended to be interpreted together. The response p-value is therefore best treated as unadjusted.

The registry reports no analyses in its analysis listing. The statistical analysis plan, including any rules for selecting models, handling multiple comparisons, or conducting interim analyses, is not reported in the registry record. The absence of reported analyses does not establish that no analyses were conducted; it means the registry record does not document them.

The primary endpoint population is described as all treated participants. That is the only explicit analysis-population definition in the record. The treatment of participants who were randomized but not treated, who discontinued, or who lacked complete event follow-up is not reported in the registry record. The outcome definition specifies an event window ending at discontinuation from Induction Period 2, but it does not describe a missing-data method or how incomplete observation was handled.

For the secondary endpoints, the analysis populations and missing-data procedures are not reported in the registry record. Duration of response and rate of progression are named, but their numerical results are not posted. Details about censoring rules for time-to-event outcomes, if applicable, are also not reported. These gaps limit assessment of whether the posted results represent the intended estimands and how robust the findings are to incomplete follow-up.

15. 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: malignant neoplasm progression (31.4%), colitis (21.4%), fatigue (14.3%), diarrhoea (11.8%), dehydration (10.0%).

The registry lists serious adverse events affecting 59 of 68 participants in the nivolumab-followed-by-ipilimumab group and 57 of 70 in the ipilimumab-followed-by-nivolumab group. It lists other adverse events affecting 68 of 68 and 68 of 70 participants, respectively. These counts describe participants with reported events; they are not event totals, and the categories should not be combined.

The listed death counts are 35 in the nivolumab-first group and 48 in the ipilimumab-first group. The registry record does not report the cause of death or clarify the relationship of these deaths to study treatment in this summary. The listed serious-event terms include malignant neoplasm progression, colitis, fatigue, diarrhoea, dehydration, and urinary tract infection, with percentages reported for those terms. The record does not establish that these categories are mutually exclusive, so the term percentages should not be added.

These broader safety listings differ from the primary endpoint, which specifically counts participants with at least one treatment-related grade 3–5 adverse event during induction. Serious adverse events are not necessarily identical to that primary endpoint’s definition. The registry does not report enough detail in these listings to reconcile the categories or event periods.

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

The results describe what was recorded for two randomized treatment sequences in this phase 2 study. For the primary safety endpoint, the posted percentages are 64.7% for nivolumab followed by ipilimumab and 51.4% for ipilimumab followed by nivolumab. The unadjusted comparison has a risk difference of -0.133, with an interval that includes zero and an unadjusted p-value of 0.114. This does not establish a clear difference between the sequences on the primary endpoint. The inconsistent interval listed for the first group adds uncertainty to interpretation of its arm-level precision.

The secondary response result is numerically higher in the nivolumab-first group, but it is an unadjusted secondary comparison and the registry does not report a multiplicity strategy. It cannot be treated as a definitive overall ranking of the treatment sequences. The response finding also needs to be considered alongside the primary safety outcome, which has a different direction, as well as the absence of posted numerical results for duration of response and rate of progression.

These results do not establish that either sequence is universally preferable, nor do they determine what would happen outside the study population or under other schedules. A randomized design supports a comparison of assigned groups, but open-label treatment, incomplete reporting of analysis methods, and uncertainty about missing-data handling remain relevant. The record also does not provide the detail needed to interpret the listed death counts in relation to treatment or the primary endpoint.

Statistical uncertainty is not the same as proof of no difference, and a nominally notable secondary p-value is not the same as a confirmed treatment effect. The most defensible reading is limited to the outcomes and comparisons actually posted: a primary safety comparison without clear separation, a secondary response contrast favoring the nivolumab-first sequence numerically, and important gaps in reported analysis and other secondary outcome results.

17. Why This Trial Matters Statistically

This trial illustrates why treatment sequence comparisons need more than a single outcome. Both groups received nivolumab and ipilimumab, but in a different order. The primary endpoint focuses on treatment-related grade 3–5 adverse events during induction, while the posted secondary response outcome addresses a different aspect of treatment experience. The numerical directions differ in their implications: the sequence with the higher response percentage also has the higher primary adverse-event percentage. Neither outcome alone captures the full comparison.

It also illustrates the value and limits of interval estimates. The primary risk-difference interval includes zero, indicating uncertainty about the direction and magnitude of the contrast. Meanwhile, the arm-level interval reported for the 64.7% estimate does not contain that estimate. Recognizing this internal inconsistency is important: a number can be quoted accurately while still requiring caution about what it communicates.

Finally, the record demonstrates why registry transparency matters. The study is identified as randomized and open-label, but the original sample-size justification, statistical analysis plan, multiplicity handling, and missing-data methods are not reported. Those details are needed to judge how the posted comparisons relate to the trial’s intended confirmatory questions. The sample-size illustrations based on observed effects cannot fill that gap or serve as evidence of the trial’s actual design rationale.

18. 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.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

The registry record has several limitations specific to interpreting this trial. The study was open-label, so participants and investigators knew the assigned sequence. This can matter particularly for investigator-assessed response and for clinical decisions that influence treatment exposure or outcome assessment. The record does not describe safeguards for reducing assessment bias.

The primary arm-level result includes a point estimate of 64.7% with a listed interval of 37.6% to 62.4%, which does not contain that estimate. The registry does not explain the inconsistency. This makes the interval difficult to use as a precision summary for that arm. Although the posted primary comparison provides a separate risk-difference interval, the record does not describe the method used to construct these intervals.

Reporting of secondary outcomes is incomplete: numerical results are available for response rate at Week 25, but not for duration of response or rate of progression. The record does not report multiplicity handling, a full statistical analysis plan, missing-data methods, or detailed analysis populations beyond the primary endpoint’s all-treated description. It also does not explain how the listed death counts relate to treatment or define the serious-adverse-event listings in enough detail to align them with the primary endpoint.

The registry reports enrollment of 138, but that figure does not itself establish the rationale or adequacy of the trial’s original sample size. Its original design assumptions are not reported in the registry record. These reporting gaps constrain statistical interpretation even when the posted percentages and comparisons are taken at face value.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Study of Nivolumab Given Sequentially With Ipilimumab in Subjects With Advanced or Metastatic Melanoma (CheckMate 064) — NCT01783938.
• Registry record last updated 2021-08-13; status last verified 2021-07.
• Results first posted 2016-04-11.
• 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.

22. Record Summary

The registry describes a completed, randomized, open-label, phase 2 parallel-group study of nivolumab and ipilimumab in advanced or metastatic melanoma. The comparison was between nivolumab followed by ipilimumab and ipilimumab followed by nivolumab. Registry enrollment was 138.

The primary endpoint was the percentage of treated participants with at least one treatment-related grade 3–5 adverse event during induction, from Day 1 to up to Week 25. The posted unadjusted comparison did not show clear separation between sequences. A secondary Week 25 response comparison was also posted, while numerical findings for duration of response and rate of progression are not reported in the registry record.

The record does not report the original sample-size justification, statistical analysis plan, multiplicity approach, or missing-data methods. The primary arm-level interval inconsistency and these reporting gaps warrant caution when interpreting precision and the broader implications of the comparisons.