A Study of Bempegaldesleukin (NKTR-214: BEMPEG) in Combination With Nivolumab Compared With the Investigator's Choice of a Tyrosine Kinase Inhibitor (TKI) Therapy (Either Sunitinib or Cabozantinib Monotherapy) for Advanced Metastatic Renal Cell Carcinoma (RCC): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Bempegaldesleukin (NKTR-214: BEMPEG) in Combination With Nivolumab Compared With the Investigator's Choice of a Tyrosine Kinase Inhibitor (TKI) Therapy (Either Sunitinib or Cabozantinib Monotherapy) for Advanced Metastatic Renal Cell Carcinoma (RCC) (NCT03729245), a randomized phase 3 study. The primary outcome is Objective Response Rate (ORR) Per mRECIST 1.1 by BICR in IMDC All-risk Patients and Intermediate- or Poor (I/P)-Risk Patients With Previously Untreated Advanced RCC.

2. Clinical Question

The study also aimed to compare overall survival, according to its brief summary. A registered secondary outcome concerns progression-free survival assessed by mRECIST 1.1 and blinded independent central review in the same broad risk populations. The available registry description does not give results for these outcomes, so their comparative findings cannot be interpreted from this record.

3. Trial Design

This was a phase 3, randomized, open-label, parallel-group study sponsored by Nektar Therapeutics. The registry lists an enrollment of 623 and describes the study as terminated. It records a start date of 2018-12-18 and a primary completion date of 2022-01-07. The reason for termination is not reported in the registry record.

Participants were allocated to one of two treatment strategies: bempegaldesleukin plus nivolumab, or investigator-selected monotherapy with sunitinib or cabozantinib. Random allocation means that treatment assignment was intended to be determined by chance rather than chosen by participants or clinicians. The allocation ratio and details of the randomization procedure are not reported in the registry record.

The study was open label, meaning that masking was not used for treatment assignment. The primary endpoint nevertheless specified blinded independent central review for response assessment. This distinction matters: the treatment assignment was unmasked, while the stated endpoint assessment relied on a blinded review process. Further operational details about that review are not reported in the registry record.

The primary outcome was objective response rate (ORR) by mRECIST 1.1 in both the IMDC all-risk and intermediate- or poor-risk populations, with a timeframe of approximately 32 months. The registry describes the ITT all-risk population as 623 participants and the ITT intermediate- or poor-risk population as 514. The arm-level population description is incomplete in the registry record, and no primary outcome counts or arm-level ORR estimates are available in the posted material.

Three outcome measures are listed, including the primary ORR and a secondary progression-free survival outcome. The registry record does not provide a count of posted analyses. Details such as the randomization strata, planned analysis dates, and reasons for ending the study 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

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: A total of 623 patients randomized in the ITT All-risk population, including 514 patients for the ITT I/P-risk population: * NKTR-214/nivolumab: 311 (ITT All-risk population) and 256 (ITT I/P-risk population), respectively * Sunitinib or Cabozantinib (TKI): 312 (ITT All-risk population) and 258 (ITT I/P-risk population), respectively

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: A total of 623 patients were randomized in the ITT International Metastatic Renal Cell Carcinoma Database Consortium (IMDC) All-risk population, including 514 patients for the ITT I/P-risk population: * NKTR-214/nivolumab: 311 and 256, respectively * Sunitinib or cabozantinib: 312 and 258, respectively

6. Primary Endpoint

The registry lists the primary outcome as Objective Response Rate (ORR) Per mRECIST 1.1 by BICR in IMDC All-risk Patients and Intermediate- or Poor (I/P)-Risk Patients With Previously Untreated Advanced RCC, measured over Approximately 32 months.

The primary endpoint was objective response rate, defined as the proportion of enrolled participants whose best overall response was complete response or partial response. Response was assessed using modified Response Evaluation Criteria in Solid Tumors (mRECIST) 1.1 and blinded independent central review. The outcome was specified for IMDC all-risk patients and for intermediate- or poor-risk patients, with a timeframe of approximately 32 months.

Under the registry definition, complete response means disappearance of all target lesions. Any pathological lymph nodes, whether target or non-target, had to decrease in short axis to <10 mm. Partial response means at least a 30% decrease in the sum of target-lesion diameters, using the baseline sum as the reference. ORR combines participants meeting either response category.

In general, ORR is a binary participant-level outcome: each participant is classified as having achieved a response or not. It provides a direct summary of the proportion meeting predefined response criteria. It does not by itself describe how long a response lasts or whether participants live longer. Those questions require other outcomes, such as duration of response or overall survival. The registry record does not report a single arm-level ORR comparison, response counts, or confidence intervals, so the endpoint’s comparative result and precision cannot be assessed from this record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Interpretation begins with the endpoint definition. ORR is the proportion of participants classified as responders under prespecified imaging criteria. The registry defines response as complete response or partial response and specifies mRECIST 1.1 assessment by blinded independent central review. In general, defining an endpoint before analyzing outcomes helps make the result less dependent on post hoc choices about what counts as a response. The registry record does not report further details of endpoint adjudication or the analysis procedures.

What does randomization contribute?

In general, random allocation aims to make treatment groups comparable at the start of a study, including with respect to factors that are measured and factors that are not measured. It does not guarantee that the groups will be identical, especially for any one characteristic. The record describes the study as randomized but does not report the allocation ratio, randomization method, or stratification factors. Accordingly, those design details cannot be evaluated from the registry record.

The stated analysis populations are the ITT all-risk population and the ITT intermediate- or poor-risk population. In general, an intention-to-treat population retains participants in their randomized groups for analysis, regardless of treatment received or later changes, according to the analysis definition. The registry uses ITT labels, but does not describe all population-handling rules or explain how participants with missing response assessments were treated. Those details are not reported in the registry record.

How should an ORR comparison be read?

For a binary endpoint such as response, the basic evidence would include the number of responders and the number assessed in each arm, together with an estimated between-arm effect and an interval expressing its uncertainty. Different effect measures answer slightly different questions. A risk difference describes an absolute difference in proportions, while a ratio compares the proportions relative to one another. General statistical interpretation should consider both the estimated effect and its precision, rather than relying on a significance test alone.

The record does not report arm-level response counts, estimates, confidence intervals, or a test result. No effect measure can therefore be interpreted for the primary comparison. The planned endpoint is specified for two risk populations, but no subgroup-specific outcomes are reported. Without those data, it is not possible to assess whether the estimated effect was similar across populations or whether any apparent subgroup difference could be due to random variation.

What does blinded central review mean?

In general, independent central review applies a common assessment process across study participants and can reduce the influence of knowledge about treatment assignment on outcome classification. Here, the study is described as open label, while the primary response endpoint is designated for blinded independent central review. The registry record does not report the review procedures, how disagreements were resolved, or the extent to which reviewers remained unaware of assignment. The designation describes the planned assessment approach, but does not permit an evaluation of its implementation.

General statistical reasoning also distinguishes response outcomes from time-to-event outcomes. ORR records whether a participant met response criteria within the specified assessment framework. Progression-free survival and overall survival concern time until an event. Such analyses often account for participants who have not experienced the event by the end of follow-up, using censoring methods. The registry lists a progression-free survival outcome and mentions overall survival in the study purpose, but does not provide event counts, follow-up summaries, hazard estimates, or analysis details. No interpretation of those time-to-event results is possible from the available record.

What can planning calculations tell readers?

The registered assumptions describe an illustrative sample-size calculation using a two-sided significance level of 0.05, equal allocation, a normal approximation, and an effect set as a fraction of the observed effect. In general, such calculations estimate the sample size needed under specified assumptions to detect a chosen effect with a chosen power. They are planning illustrations, not evidence that a trial had a particular probability of finding a result after the fact. They are not observed power.

The record does not report numerical outputs for these illustrations, and it does not report the trial’s own sample-size justification or target effect. Further, the primary outcome comparison itself is not reported. Consequently, the planning assumptions cannot be used to infer what effect the trial was designed to detect, how much precision it achieved, or whether the study was adequately powered. Any trial-specific sample-size rationale beyond the listed illustration is not reported in the registry record.

Why do missing details matter?

Analysis choices can affect estimates and their uncertainty. Examples include the handling of missing assessments, the definition of analysis populations, covariate adjustment, and the method used to compare response proportions. General best practice is to specify these choices in advance and report them alongside results. The registry record does not describe the primary statistical test, adjustment strategy, missing-data approach, or interim analysis plan. These absences limit reproducibility and make it impossible to determine whether an unreported analysis would match the planned approach.

When multiple outcomes or populations are examined, multiplicity can also affect interpretation. In general, testing several hypotheses increases the chance of at least one apparently positive result under a framework that treats each test separately. A prespecified testing sequence or adjustment can address that issue, depending on the plan. The registry lists multiple outcome measures and populations, but does not report a multiplicity procedure. No claim about adjustment or control of false-positive risk is warranted from this record.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists three outcome measures, including ORR as the primary endpoint and progression-free survival by mRECIST 1.1 and blinded independent central review as a secondary outcome. The study summary also states that overall survival was a principal comparison of interest. The primary endpoint is specified for both the IMDC all-risk and intermediate- or poor-risk populations. Examining more than one outcome or population can raise multiplicity questions because multiple statistical comparisons may increase the chance of a nominally notable result.

The multiplicity strategy, including any testing hierarchy or adjustment, is not reported in the registry record. It is therefore not possible to determine how the study intended to control error rates across endpoints or populations. Nor should a particular correction or testing sequence be assumed.

The registry labels the populations as ITT, but does not explain the full analysis-set definitions or how missing response assessments were handled. The approach to missing data, participants who discontinued treatment, and any sensitivity analyses is not reported in the registry record. These details matter because excluding participants or classifying missing assessments in different ways can affect a binary response estimate.

The registry reports no analyses in the outcome record and provides no primary endpoint counts, estimates, confidence intervals, or test results. Thus, no assessment of missing-data impact, multiplicity-adjusted significance, or robustness of the results can be made. These are limitations of the available reporting; they do not establish that such procedures were or were not carried out outside the registry record.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: corona virus infection (4.2%), pneumonia (2.9%), pleural effusion (2.3%), hypotension (2.3%), acute kidney injury (1.6%).

The registry lists serious adverse event figures for the two treatment groups. For the bempegaldesleukin plus nivolumab group, 113 participants are recorded as affected among 310 at risk. For the sunitinib or cabozantinib group, 91 are recorded as affected among 306 at risk. These are counts, not adjusted comparisons. The registry also records 109 deaths in the combination group and 114 in the TKI group; the details and relationship of these deaths to adverse events are not reported in the registry record.

The most frequently listed serious adverse event terms include Corona virus infection at 4.2%, Pneumonia at 2.9%, Pleural effusion at 2.3%, Hypotension at 2.3%, Acute kidney injury at 1.6%, and Anaemia at 1.6%. The record does not specify in this summary whether those percentages are calculated within an arm or across another denominator, so they should be read as registry-listed percentages rather than as a direct between-arm comparison.

In general, serious adverse event counts describe how many participants had at least one qualifying event under the study’s reporting rules. They do not show event duration, severity grades beyond the serious classification, timing, causality, or whether a participant had multiple events. The registry record does not provide those details. Differences in raw counts alone should not be interpreted as evidence of a causal safety difference, particularly without fuller exposure, follow-up, and analysis information.

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

The record establishes that a randomized, open-label phase 3 study compared bempegaldesleukin plus nivolumab with investigator-selected sunitinib or cabozantinib in previously untreated advanced renal cell carcinoma. It also establishes the primary endpoint definition, the use of blinded independent central review for response assessment, and the intended all-risk and intermediate- or poor-risk populations. Those design details clarify what comparison the trial was intended to address.

However, the registry record does not report a single arm-level primary endpoint comparison. It gives no response counts, ORR estimates, confidence intervals, or significance test for either stated population. The primary outcome’s magnitude, direction, and precision therefore cannot be determined from the available record. Absence of a reported comparison is not evidence of no difference, equivalence, or benefit.

The study summary identifies overall survival as an intended comparison, and the registry lists a secondary progression-free survival outcome. Numerical results for these outcomes are not reported in the registry record. The listed adverse-event counts and terms provide limited descriptive safety information, but do not establish comparative risk or causality without additional definitions and analyses.

The terminated status is a fact about the record, not an explanation of why the trial ended. The reason for termination, whether the planned analysis changed, and whether any other analysis was completed are not reported in the registry record. Similarly, the registry’s sample-size assumptions are described as a planning illustration, not a trial-reported design justification or post hoc measure of success.

In practical statistical terms, the record is informative about the question and design but insufficient for judging the trial’s primary efficacy result. An interpretation of comparative efficacy would require the arm-specific response data and an estimate of uncertainty, together with the analysis methods and relevant outcome reporting. Those elements are not available in the registry record.

13. Why This Trial Matters Statistically

This trial is statistically notable for setting a defined response question within a randomized comparison of a combination regimen and an investigator-choice TKI strategy. The endpoint uses explicit response criteria and blinded independent central review, while the study is otherwise described as open label. This illustrates how endpoint assessment procedures can be distinguished from masking of treatment assignment.

The endpoint is evaluated in both an IMDC all-risk population and an intermediate- or poor-risk population. Reporting results separately by population can help clarify the scope of an estimate, but it also makes the analysis plan important. The registry record does not report how multiplicity across outcomes and populations was managed or provide population-specific results. This makes it a useful example of why endpoint definitions alone are not enough for statistical interpretation.

The record also shows why status labels and outcome data should be kept separate. The study is recorded as terminated, but the reason is not reported in the registry record. Without outcome estimates, a termination label cannot establish what the comparison showed. Likewise, counts of serious adverse events are descriptive unless accompanied by adequate definitions, denominators, follow-up context, and comparative analyses.

Finally, a planning illustration based on an observed effect should not be confused with observed power or with the trial’s original design rationale. Statistical conclusions depend on the outcome data, the prespecified analysis, and the precision of the estimates. In this case, the available registry record describes the design and measurement framework but does not provide the primary numerical comparison needed to assess the result.

14. Limitations

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

The most important trial-specific limitation for interpretation is that the registry record does not report a single arm-level primary ORR comparison. Response counts, ORR estimates, confidence intervals, and a test result are not available. This prevents assessment of the size, direction, and precision of the primary endpoint result.

The study was open label, while response assessment was designated for blinded independent central review. The registry does not describe how the review was conducted or how blinding was maintained. The record also does not report the randomization procedure, allocation ratio, or stratification factors. These omissions limit evaluation of the methods, although they do not establish that the procedures were inadequate.

The registry lists two risk populations and a secondary progression-free survival outcome, and the brief summary includes overall survival as a study purpose. Numerical results for those outcomes and populations are not reported in the registry record. The multiplicity plan, missing-data handling, analysis-set details, and statistical analysis methods are also not reported. Therefore, the robustness of any treatment comparison cannot be assessed.

The study’s terminated status is recorded, but the reason is not reported in the registry record. The available serious adverse event information is limited to counts and selected terms; additional context on event definitions, timing, exposure, and attribution is not included in this summary. These limitations constrain interpretation of both efficacy and safety and should not be filled by assumptions about the protocol or study conduct.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Study of Bempegaldesleukin (NKTR-214: BEMPEG) in Combination With Nivolumab Compared With the Investigator's Choice of a Tyrosine Kinase Inhibitor (TKI) Therapy (Either Sunitinib or Cabozantinib Monotherapy) for Advanced Metastatic Renal Cell Carcinoma (RCC) — NCT03729245.
• Registry record last updated 2023-04-11; status last verified 2023-03.
• Results first posted 2023-04-11.
• 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.

18. Record Summary

NCT03729245 is a randomized, open-label, parallel-group phase 3 study of bempegaldesleukin plus nivolumab compared with investigator-selected sunitinib or cabozantinib in previously untreated advanced renal cell carcinoma. The registry lists enrollment of 623, a terminated status, a start date of 2018-12-18, and a primary completion date of 2022-01-07.

The primary endpoint was ORR by mRECIST 1.1 and blinded independent central review in IMDC all-risk and intermediate- or poor-risk populations, with a timeframe of approximately 32 months. The registry record does not report a single arm-level primary endpoint comparison or numerical results for overall survival and progression-free survival. The reason for termination, detailed analysis plan, multiplicity strategy, and missing-data approach are not reported in the registry record.