Safety and Efficacy of APX005M With Gemcitabine and Nab-Paclitaxel With or Without Nivolumab in Patients With Previously Untreated Metastatic Pancreatic Adenocarcinoma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Safety and Efficacy of APX005M With Gemcitabine and Nab-Paclitaxel With or Without Nivolumab in Patients With Previously Untreated Metastatic Pancreatic Adenocarcinoma (NCT03214250), a randomized phase 1, phase 2 study. The primary outcome is Phase 1b Primary Safety Outcome.

2. Clinical Question

This is a safety-and-efficacy study, but the primary outcome described in the registry record is a safety outcome for phase 1b. It asks how many participants in the defined population experienced adverse events, serious adverse events, or dose-limiting toxicities during the specified safety follow-up period. This is a descriptive question about observed events; it is not, by itself, a single comparison of one treatment against another.

The study is randomized and has five listed treatment groups, spanning combinations with different APX005M dose levels and the presence or absence of nivolumab. The registry record does not set out which efficacy comparison should be treated as the main clinical contrast. Accordingly, the clinical question is best read as an evaluation of safety and activity across the listed combinations, rather than as a clearly specified single superiority question.

3. Trial Design

The registry describes an open-label, multicenter, randomized, parallel phase 1b/2 clinical study. It enrolled 129 participants with previously untreated metastatic pancreatic adenocarcinoma and is listed as completed. The study began on 2017-07-21, and the primary completion date is 2021-02-11. The registry title describes APX005M with gemcitabine and nab-paclitaxel, with or without nivolumab.

Five treatment groups appear in the posted safety data: gemcitabine, nab-paclitaxel, and nivolumab; gemcitabine, nab-paclitaxel, and APX005M at 0.1 mg/kg; the same backbone with APX005M at 0.3 mg/kg; and the nivolumab-containing backbone with APX005M at 0.1 mg/kg or 0.3 mg/kg. These group descriptions identify the regimens, but the registry record does not report the allocation ratio, randomization procedure, stratification factors, or the planned statistical contrasts.

Open-label means participants and study personnel were not masked to assigned treatment, according to the registry listing. The parallel design means the listed groups are treatment groups rather than a crossover sequence, but the record does not provide further details about treatment assignment or the analysis plan. Although the study includes phase 1b and phase 2, the posted primary outcome description is specifically labelled phase 1b. The record defines a dose-limiting-toxicity-evaluable population for that outcome; it does not explain how the phase 2 efficacy population was selected or analyzed.

The registry record does not report the protocol’s sample-size justification, any interim-analysis plan, or detailed stopping rules. The listed enrollment describes the study as a whole; it should not be interpreted as the number in any one treatment group. Arm-level denominators in the safety data are shown separately in the posted results.

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.

Only the first 6 of 7 groups are shown.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The DLT-Evaluable Population is defined as all Phase 1b patients who received at least 2 doses of NP, at least 2 doses of Gem, and 1 dose of APX005M during Cycle 1.

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: The Phase 1b Safety Population is defined as all patients who were enrolled in Phase 1b and who received at least 1 dose of any study drug. The Phase 2 Efficacy Population is defined as all patients who were randomized in Phase 2 and who received at least 1 dose of any study drug.

Only the first 6 of 8 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Phase 1b Primary Safety Outcome, measured over Initiation of study drug (or informed consent for SAEs) through 100 days after the last dose of study drug or initiation of a new systemic anti-cancer therapy with a maximum exposure of 34.3 months..

The phase 1b primary safety outcome is the number and percentage of participants with adverse events, serious adverse events, and dose-limiting toxicities. Its stated timeframe runs from initiation of study drug, or informed consent for serious adverse events, through 100 days after the last dose of study drug or initiation of a new systemic anti-cancer therapy, with a maximum exposure of 34.3 months. The dose-limiting-toxicity-evaluable population includes phase 1b patients who received at least 2 doses of nab-paclitaxel, at least 2 doses of gemcitabine, and 1 dose of APX005M during Cycle 1.

General statistical context: safety endpoints are used to describe the types and frequency of unwanted events observed during treatment and follow-up. Counts indicate how many participants experienced an event, while percentages put those counts in relation to an at-risk or evaluable group. The denominator matters: percentages from groups of different sizes should not be read as if they had equal precision. A serious adverse event is a distinct category from an adverse event overall, so the corresponding participant counts need not match.

The registry’s primary outcome combines several event categories and does not reduce them to one comparison. The registry entries do not give a single summary result for all components of that endpoint or a dose-limiting-toxicity count. The group-level safety figures should therefore be read as reported safety summaries, not as a complete statistical analysis of the primary endpoint.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies the study as randomized, open-label, and parallel, but it does not report the detailed statistical analysis plan. The explanations below distinguish what the design information establishes from general statistical principles; they are not claims about methods the investigators used.

What randomization contributes

General statistical context: randomization assigns participants to treatment groups using a chance-based process. When implemented as intended, it can help balance measured and unmeasured baseline characteristics across groups on average. It does not guarantee that groups will be identical in a particular study, and it does not replace a clear analysis plan. Here, the record says allocation was randomized but does not describe the sequence generation, concealment, allocation ratio, or any stratification. Those details are not reported in the registry record.

The five listed groups include differing combinations and APX005M dose levels. A comparison between any two groups would address a particular contrast; it would not automatically answer every possible question about the combinations. The registry record does not specify which pairwise comparisons were primary, whether comparisons were planned across all groups, or how they were to be interpreted together.

What the event counts represent

General statistical context: for a participant-level event outcome, a count is the number of participants recorded with the event, and a denominator identifies the group being summarized. A percentage expresses the count relative to that denominator. Counts and percentages are descriptive; they do not alone quantify uncertainty or establish a treatment difference. A confidence interval is a range of values calculated using a statistical procedure to show the uncertainty around an estimate. No confidence intervals for these safety counts are included in the registry record.

Precision depends in part on the amount of information in a group. The safety denominators vary across the listed groups, including groups with 7 and 8 participants at risk. A percentage from a smaller group will generally be less stable than the same percentage from a larger group. This is a general principle, not a calculation of the uncertainty for any specific group in this study.

Serious adverse events and other adverse events are reported as separate summaries. These categories describe whether participants were recorded as affected; they do not state how many separate episodes each person had or how long an event lasted. A participant could have multiple events, so participant counts are not necessarily counts of all event episodes. The record does not provide episode-level data in the information.

Why open-label status matters

General statistical context: when treatment assignment is known, expectations or knowledge of treatment can affect some aspects of event recognition, reporting, and assessment. This does not show that such effects occurred in this study. It is a consideration when interpreting reported outcomes, particularly outcomes that involve judgment. The registry identifies the study as having no masking, but the registry entries do not describe independent outcome review or procedures intended to limit assessment bias.

What is missing from the analysis description

The registry record does not report the model used for safety or efficacy analyses, the method for calculating uncertainty, or the planned approach to adjusted or unadjusted comparisons. It also does not report a statistical hypothesis, significance threshold for the trial’s analyses, or a rationale for the enrollment target. Without those details and the corresponding results, the counts should not be recast as inferential tests.

The sample-size section on this page is a planning illustration based on an observed effect, not a reconstruction of the trial’s design or an observed-power calculation. Its stated assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is a stated fraction of the observed effect. Such illustrations show how planning assumptions relate to a hypothetical future study. They do not show the probability that this completed study detected an effect, and they do not establish that the trial was originally designed with those assumptions.

General statistical context: a p-value, when used, measures how compatible data are with a specified statistical model and null hypothesis; it is not the probability that a treatment works or that the null hypothesis is true. An estimate and its uncertainty are often more informative than a p-value alone. Since inferential results are not reported in the information here, neither a p-value nor a confidence interval should be inferred from the displayed counts.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 8 outcome measures and 3 analyses. Secondary outcomes include objective response rate in the dose-limiting-toxicity-evaluable and efficacy populations, duration of response in both populations, disease control rate, and progression-free survival. The primary outcome is a phase 1b safety outcome with several categories: adverse events, serious adverse events, and dose-limiting toxicities.

General statistical context: when a study examines multiple endpoints, treatment groups, time points, or pairwise comparisons, the chance of obtaining at least one apparently notable result can increase if each is tested separately. Multiplicity procedures can control or describe this issue, but their choice depends on the questions and the planned hierarchy of tests. The registry record does not report a multiplicity adjustment, testing hierarchy, or which efficacy outcome should be treated as definitive. The listed number of outcomes and analyses alone is not enough to determine how multiplicity was handled.

The registry entries also do not report a missing-data strategy, rules for censoring time-to-event outcomes, or how participants who did not meet an analysis-population definition were handled. The phase 1b dose-limiting-toxicity-evaluable population is defined, but details of the efficacy population are not. It is therefore not possible to infer that all randomized participants were included in each analysis or to assess how missing observations may have affected the results.

These omissions matter when interpreting any later comparison: the endpoint definition, analysis population, missing-data method, and multiplicity approach all influence what an estimate means. Each is not reported in the registry record in 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: pyrexia (37.5%), septic shock (25.0%), atrial fibrillation (14.3%), bacteraemia (14.3%), device related infection (14.3%).

The safety listing gives, for each treatment group, the number at risk and the number of participants recorded with serious adverse events and with other adverse events. For example, the gemcitabine, nab-paclitaxel, and nivolumab group lists 18 serious adverse events among 36 at risk. The APX005M 0.1 mg/kg combination without nivolumab lists 4 among 7, while its nivolumab-containing counterpart lists 6 among 8. The other listed groups and their counts are described and shown in the tables and plots.

These are participant counts, not counts of individual event episodes. The number of participants recorded with an adverse event overall can include people who also had a serious adverse event; the categories should not be assumed to be mutually exclusive. The separate deaths column reports 25, 5, 29, 8, and 27 deaths across the listed groups, respectively. Those figures do not establish cause or treatment attribution.

The registry also lists serious-event terms with percentages, including pyrexia at 37.5% and septic shock at 25.0%. The denominator for these percentages is not reported in the registry record. They should be read as reported term percentages, not recalculated as arm-level risks. Differences in group sizes and the absence of uncertainty intervals also limit direct comparisons.

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

The posted results show that adverse events and serious adverse events were recorded in each of the listed treatment groups, with group-specific counts and denominators. They also show the reported death counts and a list of serious-event terms with percentages. These observations describe what the registry recorded; they do not establish that a particular event was caused by a study treatment.

The counts do not by themselves demonstrate that one combination is safer or less safe than another. The groups have different denominators, and the registry entries include no confidence intervals or statistical comparisons. A difference in observed counts may reflect underlying risk, chance variation, group composition, or other factors; the available information does not separate these explanations. The listed deaths also cannot be interpreted as treatment-related deaths without causes and attribution information, which are not reported.

The study’s stated purpose includes efficacy evaluation, and the registry lists response, disease control, duration of response, and progression-free survival outcomes. However, numerical results for these efficacy measures are not present in the registry record analyzed on this page. This means the page cannot assess the size or precision of an efficacy effect. It is not evidence for or against efficacy; it is a limit on what these posted data can support.

Likewise, the safety summaries do not answer every question about the primary endpoint. The endpoint includes adverse events, serious adverse events, and dose-limiting toxicities, but a complete combined outcome result or dose-limiting-toxicity count is not included. Readers should distinguish the available descriptive safety summaries from a prespecified inferential comparison. The registry information supports careful description of the reported groups, not a conclusion about comparative benefit or harm.

13. Why This Trial Matters Statistically

This trial illustrates why a randomized label does not, by itself, tell the reader what comparison a study can establish. The five listed groups vary in the presence of nivolumab and APX005M and in APX005M dose level. Each contrast addresses a different question. Without a reported testing hierarchy or statistical plan, the registry’s treatment-group structure should not be converted into an assumed single primary comparison.

It also demonstrates the importance of distinguishing the overall enrollment from the denominators for a specific endpoint. The study enrolled 129 participants, while the posted safety groups have different at-risk counts. The phase 1b primary safety population has its own dose-based eligibility definition. Those distinctions affect which participants a particular summary describes, and they matter when interpreting percentages or comparing groups.

There is a useful statistical separation between describing what happened and estimating how confidently groups differ. The safety counts provide a record of participants with reported events. An inferential comparison would additionally require an analysis method and an assessment of uncertainty. Those details are not reported. The registry’s open-label designation is also relevant context when considering how outcomes were observed and reported, though it does not establish that reporting was biased.

Finally, the planning illustrations should be read as hypothetical design aids. They are not observed power and do not explain how the completed study’s enrollment was chosen. Together, these features make the study a useful example of reading trial design, safety denominators, outcome definitions, and analysis information as separate parts of the evidence.

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 available record gives useful design and group-level safety information, but leaves important details unavailable. It does not report the detailed randomization method, allocation ratio, stratification factors, analysis plan, or protocol-based sample-size justification. Although the study is listed as randomized and parallel, those labels do not resolve how the planned treatment contrasts were defined or tested.

The safety results are aggregate counts. Individual-level information, event timing, event duration, recurrence, baseline characteristics, and causes of death are not reported in the registry record. The listed serious-event term percentages have no stated denominator in this information. These gaps limit interpretation of the clinical context and prevent a full assessment of comparability across groups.

The primary safety outcome includes several components, but the registry entries do not provide a single complete result for all components or a dose-limiting-toxicity count. The registry lists efficacy outcomes but does not include their numerical results. Confidence intervals, p-values, adjusted comparisons, missing-data procedures, and multiplicity handling are also not reported in the registry record. Consequently, the posted safety counts cannot support a precise estimate of treatment differences, and the efficacy outcomes cannot be assessed from this information.

The enrollment of 129 is a study-level figure, not an endpoint-specific denominator. The five group denominators in the safety listing vary, and the phase 1b evaluable population has an explicit treatment-exposure definition. The record does not explain the analysis-population rules for each other outcome. These limitations are specific to interpreting this registry information and should not be mistaken for proof that the underlying study lacked further protocol or analysis documentation.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Safety and Efficacy of APX005M With Gemcitabine and Nab-Paclitaxel With or Without Nivolumab in Patients With Previously Untreated Metastatic Pancreatic Adenocarcinoma — NCT03214250.
• Registry record last updated 2022-12-23; status last verified 2022-12.
• Results first posted 2022-08-11.
• 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

This is a completed, randomized, open-label, multicenter, parallel phase 1b/2 study of combinations involving APX005M, gemcitabine, nab-paclitaxel, and, in some groups, nivolumab for previously untreated metastatic pancreatic adenocarcinoma. The registry lists 129 participants, a start date of 2017-07-21, and a primary completion date of 2021-02-11.

The phase 1b primary outcome counts participants with adverse events, serious adverse events, and dose-limiting toxicities over the stated safety follow-up period. The posted safety summaries report group denominators, serious and other adverse-event counts, and deaths. Efficacy outcomes are listed, but their numerical results are not included in the information.

The registry record does not report the detailed statistical plan, multiplicity approach, missing-data methods, or sample-size justification. The available counts are descriptive and do not establish comparative treatment effects.