This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Azacitidine and Pembrolizumab in Pancreatic Cancer (NCT03264404), a na phase 2 study. The primary outcome is Progression-Free Survival (PFS).
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT03264404 |
| Phase | Phase 2 |
| Condition | Pancreas Cancer |
| Allocation | Not applicable (single group) |
| Design model | Single Group |
| Masking | None (open label) |
| Enrollment | 36 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Susan E. Bates |
| Results posted | Yes |
2. Clinical Question
Population
People with advanced pancreatic cancer whose disease had progressed after first-line therapy. The registry record lists 36 participants enrolled; it does not report additional eligibility details in the outcome information.
Intervention
A regimen involving pembrolizumab and azacitidine. The official title describes pembrolizumab following induction with azacitidine, while the brief summary describes the two agents as a combination. The brief summary gives schedules of every 3 weeks for immune therapy and every 4 weeks for the hypomethylating agent.
Comparator
There is no separate comparator group. The study is listed as non-randomized and single-group, with no allocation to different treatment arms. Outcomes therefore cannot be interpreted as a randomized contrast between this regimen and another treatment.
Primary question
The primary endpoint is progression-free survival, defined from the first day of trial treatment to documented progression under RECIST 1.1 or death, whichever occurs first. The registry record lists a 24-month timeframe, but does not report the median progression-free survival or its 95% confidence interval.
The clinical question is whether progression-free survival and other recorded outcomes can be characterized for participants receiving this regimen in this study. The listed secondary outcomes are objective response rate, duration of response, and disease control rate. These outcomes address different aspects of tumor response and its duration, but the registry record does not report their numerical results in the information available for this analysis.
Because the study has no control group, its results would describe outcomes among enrolled participants rather than establish how outcomes compare with a different treatment or with no treatment. Differences between the participants and people treated in other settings could affect such a comparison. The study is also described as single-center and open-label. The registry record does not report further details about the enrolled population or the analysis populations used for each outcome.
3. Trial Design
This is a phase II, non-randomized, open-label, single-center study listed as completed. The registry record lists enrollment of 36 participants. It identifies a single group receiving pembrolizumab and azacitidine; there is no randomized or concurrent control arm. Allocation is listed as not applicable, consistent with the single-group design.
The official title describes pembrolizumab following induction with azacitidine in people with advanced pancreatic cancer after failure of first-line therapy. The brief summary describes the study treatment as a combination of immune therapy and a hypomethylating agent. It states that immune therapy is given every 3 weeks and the hypomethylating agent every 4 weeks. The arm description says participants receive pembrolizumab with azacitidine. These descriptions establish the study’s two agents and the stated schedules, but the registry record does not report further treatment-delivery details in the available outcome record.
Open-label means that treatment assignment is not masked from participants and study personnel. In a single-group study there is no assigned alternative regimen against which the treatment group can be compared. Knowledge of treatment can be relevant to how some outcomes are assessed or reported, although progression is defined using RECIST 1.1 and death is also included in the primary endpoint.
The primary outcome is median progression-free survival, with a 95% confidence interval, and a listed timeframe of 24 months. The secondary outcomes are objective response rate, duration of response, and disease control rate. The registry record lists four outcome measures and zero analyses. It does not report the protocol’s sample-size justification, interim analyses, statistical analysis plan, or the precise analysis populations. The study’s completed status does not, by itself, establish that every planned analysis or result is present in the registry record.
Since the design does not include a like-for-like control group, the principal statistical task is descriptive: to estimate outcomes in the treated group and convey uncertainty around those estimates. Any hypothetical planning illustration based on a two-arm comparison should be distinguished from the actual single-group design and should not be read as observed evidence from a controlled comparison.
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
| Stage | Pembrolizumab |
|---|---|
| STARTED | 36 |
| COMPLETED | 2 |
| NOT COMPLETED | 34 |
Reasons for not completing this period:
| Reason | Pembrolizumab |
|---|---|
| Death | 2 |
| Physician Decision | 1 |
| Disease Progression | 28 |
| Adverse Event | 1 |
| Off study prior to treatment | 2 |
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.
| Characteristic | Unit | Pembrolizumab |
|---|---|---|
| Participants analysed | 36 | |
| Age, Continuous | years | 64.3 (SD 9.79) |
| Sex: Female, Male: Female | Participants | 10 |
| Sex: Female, Male: Male | Participants | 26 |
| Race (NIH/OMB): American Indian or Alaska Native | Participants | 0 |
| Race (NIH/OMB): Asian | Participants | 3 |
| Race (NIH/OMB): Native Hawaiian or Other Pacific Islander | Participants | 0 |
| Race (NIH/OMB): Black or African American | Participants | 3 |
| Race (NIH/OMB): White | Participants | 28 |
| Race (NIH/OMB): More than one race | Participants | 0 |
| Race (NIH/OMB): Unknown or Not Reported | Participants | 2 |
| Region of Enrollment: United States | participants | 36 |
6. Primary Endpoint
The registry lists the primary outcome as Progression-Free Survival (PFS), measured over 24 months.
| Element | Registry entry |
|---|---|
| Result type | Median |
| Measure of spread | 95% Confidence Interval |
| Unit | months |
| Endpoint classification | Time-to-Event Analysis |
Progression-free survival (PFS) is the time from the first day of trial treatment until the first documented disease progression under RECIST 1.1 or death from any cause, whichever occurs first. The registry record defines progression using target-lesion measurements: an increase of at least 20% in the sum of diameters from the smallest sum recorded on study, with an absolute increase of at least 5 mm, or the appearance of one or more new lesions. The endpoint therefore combines progression and death into one time-to-event measure.
PFS is used in clinical research to describe how long participants remain alive without a recorded progression event. It incorporates both the timing and occurrence of events, rather than simply reporting the proportion with progression at one selected visit. For a time-to-event endpoint, some participants may not have had an event by the end of their observed follow-up. Statistical methods can account for such incomplete event times, provided the necessary participant-level event and follow-up information is available.
The registry record identifies the reported summary as a median and the uncertainty measure as a 95% confidence interval. It lists 24 months as the timeframe. However, it does not report the median PFS or the interval endpoints in the outcome information available for this analysis. The timeframe alone does not give the observed follow-up distribution or indicate how many participants experienced progression or death by that point.
Without the estimate and its confidence interval, the endpoint’s numerical result and precision cannot be evaluated. The definition explains what event time is intended to measure; it does not establish the observed outcome or show how it compares with another treatment.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record identifies a clinically defined time-to-event endpoint and a single treated group, but it does not report the primary endpoint estimate. In particular, the median PFS and its 95% confidence interval are not reported in the registry record. As a result, the magnitude of the observed PFS and the precision of its estimate cannot be described from this record. The listed 24-month timeframe is not itself a PFS result and should not be mistaken for the median or for a statement that all participants were followed for that duration.
The design also sets an important boundary on interpretation. With one group and no randomized comparator, the study can describe outcomes among participants who received the regimen, but it cannot isolate the effect of the regimen from other explanations for those outcomes. A comparison with results from another study would not have the same safeguards as a randomized, concurrent comparison: differences in eligibility, disease status, follow-up, assessment, or other features could contribute to an apparent difference. Details about these factors are not reported in the registry record.
The outcome list includes objective response rate, duration of response, and disease control rate as secondary outcomes. Numerical results for these outcomes are not reported in the registry record. Their presence in the outcome list identifies what the study planned to assess, but is not evidence of a particular response rate or duration. The record’s listing of four outcome measures and zero analyses likewise does not provide effect estimates or inferential results.
Safety information is listed for the pembrolizumab group. Serious adverse events affected 9 participants among 36 at risk; other adverse events affected 36 participants among 36 at risk; and 26 deaths are listed. The registry record also gives selected serious-event terms and percentages. These data describe recorded events in this study group, but the record does not establish that every event was caused by a study drug. Nor can the single-group figures establish whether event rates were higher or lower than they would have been under another treatment.
The sample-size assumptions associated with planning illustrations use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The illustrations are planning scenarios, not observed power or a replacement for the study’s actual design. The study itself is single-group, and the registry record does not report its own sample-size justification. These distinctions matter because a planning calculation cannot supply the missing outcome estimate, demonstrate statistical significance, or turn an uncontrolled study into a comparative trial.
Overall, the registry record supports a description of the trial design, endpoint definition, and selected safety counts. It does not provide enough primary efficacy information to characterize the observed PFS result or its precision. No claim about benefit or lack of benefit follows from the absence of a reported estimate in this record.
9. Statistical Methods Explained
Progression-free survival is a time-to-event outcome. Each participant’s event time begins at the first day of trial treatment, according to the registry definition. The event is the first documented progression under RECIST 1.1 or death from any cause, whichever occurs first. Combining these events allows the endpoint to capture time without recorded progression or death, while also making clear that the two event types are not interchangeable descriptions of clinical experience.
In a time-to-event analysis, not every participant necessarily has an observed event at the time data are summarized. A participant who has not had a recorded event by the last available assessment has a censored observation: the event time is not known, but the participant contributes information up to the point of censoring. This is a general explanation of the concept. The registry record does not report how many PFS observations were censored, the censoring rules, or the amount of follow-up for individual participants. It is therefore not possible to assess those features for this study from the record.
A median time-to-event estimate is the point at which the estimated event-free proportion reaches one half. A confidence interval expresses uncertainty around an estimate under the assumptions of the statistical method used. A 95% confidence interval is not a guarantee that the true value lies inside the interval for this particular study; rather, it is a conventional interval procedure whose long-run coverage has a 95% target when its assumptions hold. The registry record names a 95% confidence interval for median PFS but does not report the median or interval limits. Consequently, neither the central estimate nor its precision can be assessed.
The primary endpoint’s progression rule is based on RECIST 1.1. As described in the registry record, progression can be identified by a sufficiently large increase in the sum of target-lesion diameters relative to the smallest sum on study, together with an absolute increase of at least 5 mm, or by a new lesion. The definition gives a common operational rule for classifying progression. It does not describe how often assessments occurred, who reviewed them, whether review was independent, or how missing or unevaluable scans were handled; those details are not reported in the registry record.
The term “median” also differs from a simple average. In general, a median identifies the midpoint of a distribution and is less influenced by very long event times than a mean. For PFS, the median is estimated using time-to-event methods so that censored observations can be incorporated. The validity and precision of that estimate depend on the observed event and censoring information, which is not included in the posted result information. The record does not report a hazard ratio or a comparative time-to-event analysis.
The registry lists a 24-month timeframe for the primary endpoint. In general, a timeframe can identify the period associated with an outcome’s collection or assessment. It does not, by itself, specify the median follow-up, guarantee that participants were observed for the full period, or state how event times after that period were treated. Those details are not reported in the registry record. The timeframe should therefore be read as a registered outcome detail, not as a numerical result.
Because this is a single-group trial, there is no between-group treatment contrast to estimate within the trial. In general, randomization helps balance measured and unmeasured prognostic factors between groups, supporting a causal comparison under the design’s assumptions. A single treated group does not provide the same basis for separating treatment effects from natural disease history, selection, concurrent care, or other influences. These are general design considerations, not claims that any particular influence occurred in this study.
The record’s planning assumptions describe a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It also states that planning illustrations use fractions of an observed effect and that time-to-event illustrations refer to total events, while other illustrations refer to participants per arm. Such calculations are general planning tools: they show how a hypothetical design may be sized under stated assumptions. They are not observed power, and they do not show the probability that this completed study produced a particular result. Since the actual study is single-group, two-arm planning assumptions do not describe its enrolled design. The registry record does not report the trial’s own sample-size justification or the calculations used to select enrollment of 36.
Over-dispersion is a general statistical issue that arises when the variability in count data is greater than expected under a model’s assumptions. If present, it can make standard errors too small and confidence intervals too narrow unless the analysis accounts for it. The posted information does not identify an over-dispersion model or report count-model analyses. This general concept should not be taken to imply that over-dispersion was present or relevant to the study’s primary PFS analysis.
Finally, a registry outcome listing and an outcome estimate are distinct. The record names the primary outcome, its definition, summary parameter, uncertainty measure, and timeframe. Those details describe what was to be assessed and how it was to be expressed. They do not provide the data needed to reconstruct an estimate. The median PFS, interval limits, event counts for PFS, and analysis details are not reported in the registry record, so no independent numerical recalculation of the primary result is possible.
10. Multiplicity, Missing Data and Analysis Populations
The registry identifies PFS as the primary outcome and lists objective response rate, duration of response, and disease control rate as secondary outcomes. It also lists four outcome measures and zero analyses. The record does not explain how the four measures correspond to each registered outcome, nor does it provide a statistical analysis plan describing which comparisons or tests were intended.
Multiplicity is a general concern when several outcomes or analyses are examined. Testing multiple questions can increase the chance of at least one apparently positive result if each is judged separately without an appropriate strategy. Statistical plans may address this through prespecified priorities or adjustments, but the registry record does not report whether any such approach was planned or used. No multiplicity adjustment should therefore be assumed.
Information about missing data and analysis populations is also not reported in the registry record. The record does not state how participants with missing assessments, incomplete follow-up, or unevaluable response data were handled, or whether the analyses used all enrolled participants or a defined subset. For PFS, censoring is a standard part of time-to-event analysis, but the censoring details and participant-level follow-up are not reported. For response outcomes, the denominators and rules for missing or unevaluable assessments are likewise not reported.
These omissions limit reproducibility and interpretation. Without a statistical plan, analysis population, missing-data rules, and reported outcome estimates, it is not possible to determine whether the registry’s planned and reported analyses align or to assess the sensitivity of the findings to analytic choices. The completed status and enrollment count do not resolve these uncertainties.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Pembrolizumab | 9 | 36 | 25.0% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Pembrolizumab | 26 of 36 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Pembrolizumab | 36 of 36 |
Most frequent serious adverse event terms
| Event term | Organ system | Pembrolizumab |
|---|---|---|
| Dyspnea | Respiratory, thoracic and mediastinal disorders | 2 of 36 |
| Gastric hemorrhage | General disorders | 2 of 36 |
| Abdominal pain | Gastrointestinal disorders | 2 of 36 |
| Endocrine disorders | Endocrine disorders | 1 of 36 |
| Nervous system disorders | Nervous system disorders | 1 of 36 |
| Respiratory failure | Respiratory, thoracic and mediastinal disorders | 1 of 36 |
| Pneumonitis | Respiratory, thoracic and mediastinal disorders | 1 of 36 |
| Stroke | Nervous system disorders | 1 of 36 |
| Hypoxia | Respiratory, thoracic and mediastinal disorders | 1 of 36 |
| Sepsis | Infections and infestations | 1 of 36 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Pembrolizumab |
|---|---|---|
| Diarrhea | Gastrointestinal disorders | 13 of 36 |
| Constipation | Gastrointestinal disorders | 12 of 36 |
| Abdominal Pain | Gastrointestinal disorders | 9 of 36 |
| Anorexia | Metabolism and nutrition disorders | 9 of 36 |
| Fatigue | General disorders | 8 of 36 |
| Alkaline phosphatase increased | Investigations | 7 of 36 |
| Blood bilirubin increased | Investigations | 7 of 36 |
| Bloating | Gastrointestinal disorders | 7 of 36 |
| Injection site reaction | General disorders | 7 of 36 |
| Dyspnea | Respiratory, thoracic and mediastinal disorders | 6 of 36 |
Serious events with the highest arm-level rate: dyspnea (5.6%), gastric hemorrhage (5.6%), abdominal pain (5.6%), endocrine disorders (2.8%), nervous system disorders (2.8%).
The registry lists serious adverse events for the pembrolizumab group, with 36 participants at risk and 9 affected. It lists other adverse events with 36 at risk and 36 affected, and 26 deaths. These are counts in the registry record; they should be read as descriptions of events recorded in the study group, not as comparisons between treatments.
The listed serious-event terms and percentages are dyspnea, 5.6%; gastric hemorrhage, 5.6%; abdominal pain, 5.6%; endocrine disorders, 2.8%; nervous system disorders, 2.8%; and respiratory failure, 2.8%. The record presents these as selected serious adverse-event terms. It does not provide additional detail about event timing, duration, severity grades, recurrence, or overlap among participants. The percentages should not be added together to infer a total, since the record does not state whether participants could have multiple listed events.
An adverse event is not necessarily caused by the treatment. The registry figures alone do not establish causation, and a single-group study has no concurrent comparator for estimating how event counts might differ under another regimen. The death count is not, by itself, a cause-of-death breakdown or a treatment-attribution result. The registry record does not report those details in the information available for this analysis.
12. What the Result Does, and Does Not, Mean
The record describes a phase II study of pembrolizumab and azacitidine in people with advanced pancreatic cancer after first-line therapy. It establishes that the study was open-label, single-center, non-randomized, and single-group, with enrollment of 36. It also defines PFS and identifies other outcomes and selected safety counts. These are useful features for understanding what the study set out to examine and how its design constrains interpretation.
The record does not report the median PFS or its 95% confidence interval. It also does not report numerical results for objective response rate, duration of response, or disease control rate. Therefore, the posted information does not allow readers to quantify the primary efficacy outcome, assess its precision, or describe the secondary efficacy results. The absence of those estimates in the registry record is not evidence that the regimen did or did not have an effect.
A single-group result, even when fully reported, describes outcomes among participants who received the study regimen. Without a concurrent comparator, it cannot establish what would have happened to comparable participants under another treatment. A comparison with external data would require careful consideration of differences between populations and methods and would not be equivalent to randomization. No such comparison is reported in the registry record.
The safety figures document affected participants and selected serious-event categories, but they do not establish that the events were caused by a study drug. They also do not provide a controlled estimate of relative safety. The 26 deaths listed in the adverse-event record should not be interpreted as a cause-specific or treatment-attributable count without further details, which are not reported in the registry record.
Thus, the strongest reading is descriptive: the registry records the design, the intended PFS definition, the outcome categories, and selected event counts. The missing efficacy estimates, analytic details, and comparator prevent a fuller statistical assessment. Neither a favorable nor an unfavorable treatment conclusion can be drawn from the listed information alone.
13. Why This Trial Matters Statistically
This study is statistically notable because its question concerns a treatment regimen in a population described as having advanced pancreatic cancer after first-line therapy, while the design is single-group. The design determines what can be learned: outcomes can be described among participants receiving the regimen, but there is no randomized comparison to separate treatment-associated outcomes from other influences. That distinction remains important regardless of how clinically relevant the question is.
The primary outcome is time-to-event PFS, which captures time until progression or death rather than a single measurement at one visit. Its definition includes RECIST 1.1 progression criteria and death from any cause. A median and confidence interval would ordinarily convey both an estimated event-free duration and uncertainty, but neither is reported in the registry record. Naming a sound endpoint does not substitute for reporting its result.
The record also illustrates why outcome registration and result reporting are different. It lists secondary response outcomes and a 24-month timeframe, yet does not provide the corresponding numerical results in the information available for this analysis. Without estimates, denominators, and methods, readers cannot judge the size or precision of the observed outcomes or assess the influence of missing observations and analytic choices.
Finally, planning assumptions should not be confused with the trial’s actual design. The record’s planning illustrations assume equal allocation between two arms and a normal approximation, whereas the trial is single-group. Such illustrations are hypothetical design calculations, not observed power and not evidence of comparative efficacy. The trial’s own sample-size justification is not reported in the registry record.
The statistical importance of this study record lies in the contrast between a clearly named clinical question and the limits imposed by design and incomplete outcome reporting. Transparent estimates, uncertainty intervals, analysis populations, and methods are necessary for readers to move from a registered question to an interpretable 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 main trial-specific limitation is the absence of the reported primary endpoint estimate. Although the record identifies median PFS and a 95% confidence interval, neither is reported in the registry record. The listed 24-month timeframe does not fill that gap. The record also does not report numerical results for the listed secondary outcomes.
The single-group, non-randomized design has no concurrent control group. It therefore cannot support a randomized comparison of the regimen with another treatment. The study is also described as open-label and single-center. The registry record does not report enough detail to assess how these features affected outcome assessment or how representative the enrolled participants were of other populations.
Important analysis details are not reported in the registry record, including the sample-size justification, interim analyses, statistical analysis plan, analysis populations, missing-data handling, and censoring rules. The record does not provide the number of PFS events, participant-level follow-up, or the response denominators needed to assess the relevant outcome estimates. It also does not explain how the four outcome measures map to the listed outcome categories.
The safety information is limited to group-level counts and selected serious-event terms with percentages. Additional details about timing, severity, attribution, and cause of death are not reported in the registry record. These limitations mean that the safety figures cannot establish treatment causation or comparative risk. The completed status and the enrollment count do not compensate for missing outcome estimates or methodological details.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Azacitidine and Pembrolizumab in Pancreatic Cancer — NCT03264404.
- Registry record last updated 2026-01-27; status last verified 2026-01.
- Results first posted 2023-07-05.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
The registry record describes NCT03264404, titled “Azacitidine and Pembrolizumab in Pancreatic Cancer.” It is listed as a completed phase II, open-label, non-randomized, single-center, single-group study with enrollment of 36 participants. The study population is described as people with advanced pancreatic cancer after failure of first-line therapy. The regimen involves azacitidine and pembrolizumab.
The primary outcome is median PFS, defined from the first day of treatment to documented RECIST 1.1 progression or death from any cause, whichever occurs first. A 95% confidence interval and a 24-month timeframe are listed. The median and confidence interval are not reported in the registry record. Secondary outcomes are objective response rate, duration of response, and disease control rate; numerical results for these outcomes are not reported in the registry record.
The safety record lists 9 participants with serious adverse events among 36 at risk, 36 with other adverse events among 36 at risk, and 26 deaths. The trial has no control group, so its posted outcome and safety information does not provide a randomized comparison.