Nivolumab, Ifosfamide, Carboplatin, and Etoposide as Second-Line Therapy in Treating Patients With Refractory or Relapsed HL: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Nivolumab, Ifosfamide, Carboplatin, and Etoposide as Second-Line Therapy in Treating Patients With Refractory or Relapsed HL (NCT03016871), a non-randomized phase 2 study. The primary outcome is Complete Response Rate by Lugano Classification.

2. Clinical Question

3. Trial Design

This is a phase II, non-randomized, unmasked study with a sequential model. Its stated focus is response-adapted second-line treatment for relapsed or refractory Hodgkin lymphoma, using nivolumab and ICE chemotherapy as a bridge to autologous hematopoietic cell transplant. The registry lists an enrollment of 78 and identifies City of Hope Medical Center as sponsor. The record gives a start date of 2017-05-08, a primary completion date of 2024-01-30, and a status of active, not recruiting.

The registry describes Cohort A and Cohort B, with both cohort titles naming nivolumab, etoposide, ifosfamide, and carboplatin. In Cohort A, the description begins with nivolumab and says that additional treatment depends in part on response after the initial treatment period. In Cohort B, the description starts with nivolumab and then describes adding etoposide and ifosfamide, with carboplatin included in the cohort title. These descriptions do not provide the complete schedules or enough detail to reconstruct every treatment decision.

The non-randomized allocation is important when interpreting any cohort comparison. Treatment assignment was not randomized, and the registry specifically cautions that the cohorts are not a like-for-like control comparison. The record does not report how participants were assigned to cohorts or whether the cohorts differed in other ways. Consequently, a difference between cohort outcomes would not, by itself, establish that one treatment strategy caused a better outcome.

The primary endpoint is a response measure, while the listed secondary outcomes include overall response, overall survival, progression-free survival, outcomes after autologous transplant, relapse or progression, and non-relapse mortality. Although the study title and outcome list refer to transplant, the record does not establish that every participant underwent transplant. Detailed eligibility criteria, cohort-assignment procedures, and the full protocol 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.

Overall Study

Analysis population for the primary outcome, as defined in the registry: 1 participant in Cohort A and 3 participants in Cohort B were not evaluable for treatment response.

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 Complete Response Rate by Lugano Classification, measured over From the initial treatment to the end of the treatment, up to 6 months..

The primary endpoint is complete response by Lugano Classification, expressed as the percentage of evaluable participants with a confirmed complete response. The registry describes confirmation using radiographic response assessment, including computed tomography and/or positron emission tomography scans. The stated timeframe runs from initial treatment through the end of treatment, up to 6 months.

Complete response is a categorical endpoint: each evaluable participant contributes to the count of those who did or did not meet the response definition. Reporting the rate describes how common confirmed complete response was in the evaluable group; it does not by itself describe how long a response lasted or whether participants remained free of progression. Those are distinct questions addressed by time-to-event outcomes listed among the secondary outcomes.

The registry says that the estimate uses a 95% Clopper-Pearson confidence interval. A confidence interval communicates uncertainty around an estimated response rate: it gives a range of values compatible with the data under the method's assumptions. Its width depends in part on how many participants contribute evaluable response information and how many meet the endpoint definition. The registry record does not report the complete-response estimate, the interval limits, or the evaluable denominator, so the magnitude and precision of the primary result cannot be assessed from this record.

One participant in Cohort A and 3 participants in Cohort B were not evaluable for treatment response. The record does not give the reasons for non-evaluability or explain in detail how missing response assessments were handled beyond defining the endpoint among evaluable patients.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is a binary response outcome. For each evaluable participant, the response assessment determines whether the participant had a confirmed complete response under the Lugano Classification. The endpoint is then expressed as a percentage of evaluable participants. This is a descriptive estimate for the analyzed group; by itself, it does not compare the treatment with a randomized control or establish a treatment effect.

What does the confidence interval describe?

The registry specifies a 95% Clopper-Pearson confidence interval for the complete-response estimate. A confidence interval reflects sampling uncertainty: if comparable studies were repeated under the method's assumptions, the interval procedure would capture the underlying response probability at its stated confidence level in the long run. It is not a guarantee that the true value lies in a particular interval, and it does not account automatically for design limitations, missing assessments, or measurement error.

The Clopper-Pearson method is an exact method for a binomial proportion. It is often used when the outcome is a count of participants with an event out of a defined denominator. Its limits are determined by the observed count and denominator, rather than by relying on a normal approximation to the proportion. The registry names the method but does not report the observed count, denominator, estimate, or confidence limits. Therefore, the interval's width and the result's precision cannot be evaluated.

Who contributes to the endpoint?

The endpoint description specifies evaluable patients. The registry notes that 1 participant in Cohort A and 3 participants in Cohort B were not evaluable for treatment response. It does not state the evaluability criteria, reasons for missing or inadequate assessment, or how the exclusions were determined. It also does not provide the cohort-specific evaluable denominators. These details matter because a percentage depends on its denominator, and exclusion of participants with unavailable outcomes can affect how representative the analyzed group is of everyone enrolled.

The assessment is based on radiographic response, including computed tomography and/or positron emission tomography scans, with the response confirmed under the named classification. The timeframe is from initial treatment to the end of treatment, up to 6 months. The registry does not provide further information on review procedures, assessor masking, or how disagreements were handled. Masking is listed as none at the trial level; whether imaging assessments were independently reviewed is not reported in the registry record.

How should cohort comparisons be read?

The trial is non-randomized and uses a sequential model. Randomization can, on average, balance measured and unmeasured prognostic factors between groups. Without randomized allocation, the groups may differ for reasons other than the treatment strategy. As a general statistical principle, an observed difference between non-randomized cohorts can reflect treatment, participant characteristics, timing, or other influences. The registry warns that the two cohorts are not a like-for-like control comparison. It does not report a detailed cohort-adjusted comparison or an analysis plan that would allow causal interpretation.

The registry lists no analyses in its analysis count. This should not be interpreted as proof that no analyses were conducted; it means the registry record does not list analysis entries. The complete-response estimate and confidence interval are also absent, so no recalculated estimate can be described from the record.

What do the planning illustrations mean?

The sample-size assumptions describe planning illustrations based on the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is expressed as a stated fraction of the observed effect. These are planning assumptions, not evidence that the trial was designed using those assumptions, and not a measure of observed power. The record does not report the trial's own sample-size justification. The numerical sample-size illustration results are not reported in the registry record.

In general, a planning sample size estimates how many participants may be needed under specified assumptions to detect a target effect with a chosen statistical power. Such a calculation depends on the assumed effect and analysis method. If the assumed effect is based on an observed estimate, uncertainty in that estimate is relevant. Planning numbers do not show that a study had a particular probability of finding a result after it was conducted; observed power is not a substitute for interpreting the estimated effect and its confidence interval.

For time-to-event outcomes, a general sample-size approach may focus on the number of events rather than only the number of participants, because information accumulates as events occur. The registry's planning description distinguishes total events for time-to-event illustrations from participants per arm for other illustrations. No numerical outputs for these illustrations are reported 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("Cohort A (Nivolumab, Etoposide, Ifosfamide, Carbop", "Cohort B (Nivolumab, Etoposide, Ifosfamide, Carbop")
n      <- c(42, 32)
events <- c(30, 28)
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  Cohort A (Nivolumab, Etoposide, Ifosfamide, Carbop  42  30  71.4  55.4  84.3  56.4  82.8  56.3  82.8
ARM  Cohort B (Nivolumab, Etoposide, Ifosfamide, Carbop  32  28  87.5  71.0  96.5  71.9  95.0  71.8  94.9

11. Multiplicity, Missing Data and Analysis Populations

The registry lists the complete-response rate as the primary outcome and includes several secondary outcomes: overall response rate, overall survival, progression-free survival, progression-free survival after autologous transplant, cumulative incidence of relapse or progression, and non-relapse mortality. It also reports eight outcome measures overall. Considering multiple outcomes can raise a multiplicity question: when many hypotheses are tested, the chance of at least one nominally significant finding can increase. Statistical approaches can address this through prespecified testing sequences or adjustments, but the registry record does not report a multiplicity plan or adjustment method.

The record lists no analyses in its analysis count and does not provide outcome estimates or significance tests for the primary or secondary outcomes. It is therefore not possible to assess whether analyses were prespecified, whether secondary outcomes were treated as exploratory, or how any multiple testing was handled. The absence of a reported analysis plan should not be filled in by assumption.

For the primary response endpoint, 1 participant in Cohort A and 3 participants in Cohort B were not evaluable. The endpoint is defined among evaluable participants, but the record does not give reasons, missingness patterns, or sensitivity analyses. It does not report whether a missing response was treated as no complete response in any alternative analysis. These omissions limit assessment of how missing response information could affect the estimate.

The record also does not fully describe the analysis populations for the secondary outcomes, the handling of censoring, or the treatment of participants who did not proceed to transplant. Those details are not reported in the registry record. This is particularly relevant when interpreting outcomes defined after autologous transplant, since the group contributing to such an outcome may differ from the full enrolled population.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: fever (5.7%), renal failure (2.9%), anemia (2.3%), febrile neutropenia (2.3%), adrenal insufficiency (2.3%).

The registry reports cohort-level safety counts. For Cohort A, 43 participants were at risk for the serious-event summary and 5 were affected; for Cohort B, 35 were at risk and 3 were affected. The record also reports 2 deaths in Cohort A and 0 in Cohort B. These are descriptive counts, not estimates of a causal effect of treatment. The groups were not randomized, and the record does not provide the timing, duration of exposure, or circumstances of each event and death.

For other events, the registry reports 42 affected among 43 at risk in Cohort A and 34 affected among 35 at risk in Cohort B. The meaning of these aggregate other-event counts should not be confused with the serious-event counts or with the frequency of any single event term.

Selected serious adverse event terms and percentages are listed as Fever, 5.7%; Renal Failure, 2.9%; Anemia, 2.3%; Febrile neutropenia, 2.3%; Adrenal insufficiency, 2.3%; and Thyroiditis, 2.3%. The registry summary does not provide a cohort breakdown for these terms. These percentages describe reported event terms and do not establish that the study treatment caused them. Comparisons with other treatments or populations are not supported by these figures alone.

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

The study asks whether a response-adapted strategy using nivolumab and ICE chemotherapy can be evaluated in relapsed or refractory Hodgkin lymphoma, with the stated aim of serving as a bridge to autologous hematopoietic cell transplant. The primary measure is confirmed complete response by Lugano Classification among evaluable participants. That endpoint captures a radiographic response status during the stated treatment period; it does not by itself measure response duration, long-term disease control, survival, or the chance of proceeding to transplant.

The registry record does not report the primary complete-response percentage or its 95% confidence interval. As a result, it does not allow readers to determine the observed rate or how precisely it was estimated. The registry also does not report numerical results for the listed secondary outcomes. No claim about the magnitude of response, survival, or progression outcomes can be made from the outcome list alone.

The non-randomized, sequential design is central to what the study can establish. It can describe outcomes observed in the enrolled cohorts, but a comparison between Cohort A and Cohort B would not be equivalent to a randomized comparison. The registry cautions that the cohorts are not a like-for-like control comparison. Even if their outcomes were reported, differences could be influenced by factors beyond the treatment strategies.

The safety summaries indicate that serious events and deaths were recorded in the cohorts, alongside selected serious adverse event terms. These figures are descriptive. They do not determine whether an event was caused by treatment, and the summary does not provide enough context to compare safety reliably between the cohorts.

The study's design and endpoint definition can inform how to frame the research question, but the absent efficacy estimates prevent a numerical assessment of the primary finding. A treatment decision or conclusion about benefit cannot be drawn from the registry information; the data support only a cautious account of the design and the reported safety counts.

14. Why This Trial Matters Statistically

This trial is statistically instructive because it combines a response-adapted treatment strategy with a non-randomized, sequential design. The endpoint is not simply a general impression of improvement: it is defined as confirmed complete response by Lugano Classification, based on radiographic assessment, and is calculated among evaluable participants. A clear endpoint definition helps establish what the study is trying to measure, while the evaluability rule makes the denominator and handling of missing assessments important to interpretation.

The registry specifies an exact confidence-interval method for the primary response estimate: the Clopper-Pearson method. That detail signals that uncertainty around a proportion was part of the endpoint reporting plan. Yet the response estimate and interval are not reported in the registry record. This illustrates why a named method is not a substitute for the numerical estimate, its denominator, and its uncertainty limits.

The study also highlights the distinction between describing cohorts and estimating a treatment effect. Because allocation is non-randomized and the registry warns against treating the cohorts as like-for-like controls, cohort outcomes cannot be read as though participants had been assigned at random. That distinction remains important even when both cohorts receive components of the same broad treatment strategy.

Finally, the registry lists several secondary outcomes alongside the primary endpoint. Multiple outcomes can provide a broader account of response, survival, progression, and transplant-related outcomes, but they also raise questions about prespecification and multiplicity. The record does not describe how those issues were handled. The combination of an explicit endpoint method, limited evaluability information, absent outcome estimates, and a non-randomized design makes disciplined interpretation especially important.

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.

The main limitation for interpreting efficacy is that the registry record does not report the primary complete-response estimate or its 95% confidence interval. It also does not report numerical results for the listed secondary outcomes. Readers therefore cannot assess the size or precision of the response finding or evaluate the longer-term outcomes named in the registry.

The primary endpoint is calculated among evaluable participants, and 1 participant in Cohort A and 3 participants in Cohort B were not evaluable. The record does not state the cohort-specific evaluable denominators, reasons for non-evaluability, or sensitivity analyses addressing missing response assessments. This leaves uncertainty about how missing outcome information was handled and how exclusions relate to the full enrolled population.

The non-randomized sequential design does not support a straightforward causal comparison between cohorts. The registry explicitly cautions that they are not a like-for-like control comparison. Detailed cohort-assignment procedures, baseline comparability, and adjusted comparative analyses are not reported in the registry record. The complete treatment schedules are also not fully described in the available cohort summaries.

The registry does not report the trial's own sample-size justification, interim analyses, a detailed statistical analysis plan, or a multiplicity strategy. Although it describes assumptions for planning illustrations, those assumptions should not be taken as the trial's design rationale or as evidence of observed power. The registry also provides only selected aggregate safety information, without event timing, exposure context, or a cohort-specific breakdown for the listed serious adverse event terms.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Nivolumab, Ifosfamide, Carboplatin, and Etoposide as Second-Line Therapy in Treating Patients With Refractory or Relapsed HL — NCT03016871.
• Registry record last updated 2026-11-02; status last verified 2026-10.
• Results first posted 2024-10-09.
• 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

This phase II study is listed as non-randomized, unmasked, and sequential, with enrollment of 78. It evaluates nivolumab with ifosfamide, carboplatin, and etoposide in patients with recurrent or refractory Hodgkin lymphoma. The primary endpoint is complete response by Lugano Classification, defined as the percentage of evaluable participants with confirmed radiographic complete response over treatment, up to 6 months.

The registry states that 1 participant in Cohort A and 3 participants in Cohort B were not evaluable for response, but does not report the complete-response estimate or confidence interval. It lists several secondary outcomes but does not report their numerical results. Cohort-level safety counts and selected serious adverse event terms are reported. The registry cautions that the non-randomized cohorts are not a like-for-like control comparison. The record lists City of Hope Medical Center as sponsor and gives a status of active, not recruiting.