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 Study of CAP1-6D in Patients With Locally Advanced or Surgically Resected Pancreatic Adenocarcinoma (NCT00203892), a randomized phase 1, phase 2 study. The primary outcome is Maximum T Cell Response From Baseline.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00203892 |
| Phase | Phase 1, Phase 2 |
| Condition | Pancreatic Adenocarcinoma |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | None (open label) |
| Enrollment | 23 |
| Arms in the primary outcome | 3 |
| Lead sponsor | University of Chicago |
| Results posted | Yes |
2. Clinical Question
The clinical question was whether immunization with modified CEA peptide CAP1-6D could produce an immune response in patients with pancreatic cancer who had received chemotherapy and radiation therapy. The primary outcome addressed that question through a laboratory measure of T-cell frequency, comparing each participant’s peak on-treatment measurement with their baseline measurement. The registry describes the study population as patients with locally advanced or surgically resected pancreatic adenocarcinoma.
Population
Patients with locally advanced or surgically resected pancreatic adenocarcinoma. The primary outcome analysis included the 14 patients who received at least 3 vaccine doses and had ELISPOT measurements at baseline and after the 3rd cycle.
Intervention
Modified CEA peptide vaccine at 10mcg, 100 mcg, or 1000mcg, with Montanide ISA-51 and sargramostim (GM-CSF) 250mcg. Each dose group followed the same stated schedule.
Comparator
The registry reports a Wilcoxon (Mann-Whitney) comparison of the 10mcg group with the 1000mcg group. The 100 mcg group also has a posted outcome summary, but no comparison involving that group is reported in the registry record.
Primary question
What was the maximum change from baseline in T-cell frequency, measured as spots per 10^4 CD8+ cells, in each dose group, and how did the reported comparison between the 10mcg and 1000mcg groups evaluate?
This is an immune-response question, not a question about tumor response, survival, or comparative clinical benefit. Those outcomes are not identified as the primary endpoint in the registry record. The randomized, open-label, parallel-group design assigned participants among dose groups, but the record does not report the randomization procedure or allocation concealment. The posted primary analysis is restricted to a subset defined by receiving vaccine and having specified ELISPOT measurements; it is not an analysis of all 23 enrolled participants.
3. Trial Design
The registry describes a randomized pilot study with Phase 1 and Phase 2 listed. It used a parallel-group design and no masking. Enrollment was 23. The study is marked completed, with a start date of 2003-04 and primary completion in 2009-06. The sponsor was the University of Chicago.
Participants were assigned to one of three modified CEA peptide dose groups: 10mcg, 100 mcg, or 1000mcg. Each vaccine also contained Montanide ISA-51 and sargramostim (GM-CSF) 250mcg. In each arm, the vaccine was administered on Day 1 of each 14 day cycle, at the proximal thigh, until progressive disease or dose-limiting toxicity, for a maximum of 24 cycles. The registry does not report the number randomized to each arm or the allocation sequence.
The primary outcome was maximum T-cell response from baseline, measured by ELISPOT. Blood collection was scheduled at baseline and every 4 weeks for the first 8 cycles; after the eighth cycle, a sample was collected at disease progression. The primary outcome analysis population was limited to patients who received at least 3 doses and had ELISPOT measurements at baseline and after the 3rd cycle. This analysis population was 14 patients, while the registry’s enrollment figure is 23.
A secondary outcome concerned evidence of dose-limiting toxicities of immunization. The record lists two outcome measures and two analyses, but one comparison is identified as unavailable or not estimable for another endpoint. Eligibility details beyond the population description, reasons for excluding participants from the primary analysis, and the randomization and allocation procedures are not reported in the registry record. Nor does the record explain how the pilot study’s phase labels relate to the dose-group 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 | A: CEA Peptide 10mcg | B: CEA Peptide 100 mcg | C: CEA Peptide 1000mcg |
|---|---|---|---|
| STARTED | 7 | 10 | 6 |
| Received at Least 3 Doses of CEA Vaccine | 5 | 5 | 4 |
| COMPLETED | 5 | 8 | 6 |
| NOT COMPLETED | 2 | 2 | 0 |
Reasons for not completing this period:
| Reason | A: CEA Peptide 10mcg | B: CEA Peptide 100 mcg | C: CEA Peptide 1000mcg |
|---|---|---|---|
| Physician Decision | 1 | 2 | 0 |
| Death | 1 | 0 | 0 |
Analysis population for the primary outcome, as defined in the registry: The analysis population for the primary outcome included the 14 patients who received at least 3 doses of the CEA vaccine and had a ELISPOT at least at baseline and after the 3rd cycle.
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 19 patients who received at least 1 dose of the CEA vaccine are included in the baseline analysis population
| Characteristic | Unit | A: CEA Peptide 10mcg | B: CEA Peptide 100 mcg | C: CEA Peptide 1000mcg | Total |
|---|---|---|---|---|---|
| Participants analysed | 5 | 8 | 6 | 19 | |
| Age, Continuous | years | 51 (27 to 71) | 57.5 (50 to 86) | 61.5 (47 to 77) | 60 (27 to 86) |
| Sex: Female, Male: Female | Participants | 4 | 4 | 5 | 13 |
| Sex: Female, Male: Male | Participants | 1 | 4 | 1 | 6 |
| Region of Enrollment: United States | participants | 5 | 8 | 6 | 19 |
| Performance Status: 0 | participants | 2 | 4 | 4 | 10 |
| Performance Status: 1 | participants | 3 | 4 | 2 | 9 |
6. Primary Endpoint
The registry lists the primary outcome as Maximum T Cell Response From Baseline, measured over baseline and every 4 weeks on treatment.
| Element | Registry entry |
|---|---|
| Result type | Median |
| Measure of spread | Full Range |
| Unit | spots per 10^4 CD8+ cells |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint is the maximum change from baseline in T-cell frequency, expressed as spots per 10^4 CD8+ cells. T-cell frequency was measured using an ELISPOT assay. For each participant, the endpoint was calculated as the peak value observed on treatment minus the baseline value. A positive value therefore indicates an increase from baseline; a negative value indicates a peak below baseline.
The registry planned measurements at baseline and every 4 weeks for the first 8 cycles, followed by a sample at disease progression after the eighth cycle. The reported primary outcome analysis required an ELISPOT measurement at baseline and after the 3rd cycle, as well as receipt of at least 3 vaccine doses. The analysis consequently describes this defined subset rather than everyone enrolled.
For an immunization study, a laboratory measure of T-cell frequency is relevant to the stated question of whether vaccination can produce an immune response. It is nevertheless a biological measure, not a direct measure of tumor control, symptoms, or survival. The registry reports the outcome as a median with a full range for each dose group. A range describes the observed spread of values, while the median gives a central summary; neither by itself indicates how precisely a group comparison is estimated.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The posted primary outcome summaries are medians of each participant’s maximum change from baseline in T-cell frequency. The 10mcg group had a median of 10.5 spots per 10^4 CD8+ cells, with a full range from 0.0 to 155.0. The 100 mcg group had a median of 51.75, with a range from -16.0 to 432.5. The 1000mcg group had a median of 270.6, with a range from 72.25 to 377.2. These are descriptive summaries for the analyzed participants, not adjusted treatment effects.
The registry reports a Wilcoxon (Mann-Whitney) comparison between the 10mcg and 1000mcg groups, with p = 0.028. This is evidence of a difference in the ranked outcome distributions under that test. It should not be read as the probability that the null hypothesis is true, nor as a measure of the size or clinical importance of an effect. The record does not post a confidence interval or an estimated between-group effect for this comparison.
The medians increase across the three listed dose groups, but the reported inferential comparison is specifically between the 10mcg and 1000mcg groups. The registry does not report a test of trend across dose levels or a comparison involving the 100 mcg group. The observed ordering is therefore descriptive; it does not establish a dose-response relationship. The within-group ranges are broad, and the analyzed groups are small, with 5, 5, and 4 participants respectively. Those features make the summaries sensitive to the values of individual participants and limit precision.
The primary analysis included 14 patients selected by post-randomization conditions: receipt of at least 3 doses and availability of ELISPOT measurements at baseline and after the 3rd cycle. The record does not report outcomes for participants outside that analysis population or explain their exclusion. As a result, the posted summaries do not describe an intention-to-treat comparison of all enrolled participants. A randomized assignment can support a fair comparison when analyzed according to the assigned groups, but restricting analysis based on treatment received and measurement availability can affect that comparability.
The findings concern an ELISPOT immune-response measure in a pilot study. They do not establish a clinical benefit or a safety advantage. The registry reports no serious adverse events in the listed at-risk groups, alongside other adverse events, but interpretation of safety is limited by the small group sizes and the absence of additional detail about event types and follow-up. Overall, the registry presents a suggestive difference in the specified endpoint for the reported pairwise comparison, with substantial uncertainty about its size, reproducibility, and relevance to clinical outcomes.
9. Statistical Methods Explained
The primary outcome is a change score: each participant’s highest measured on-treatment T-cell frequency minus their baseline frequency. A positive score means the peak value was above baseline. The group summaries use medians and full ranges, rather than means and standard deviations. Medians are often useful when measurements are skewed or when extreme values could pull a mean, while a full range makes the observed minimum and maximum visible. A range is not a confidence interval: it describes observed values, not the precision of an estimated group effect.
What does the Wilcoxon comparison assess?
The registry reports the Wilcoxon (Mann-Whitney) method for comparing the 10mcg and 1000mcg groups. In general, this rank-based method orders the observations across groups and evaluates whether values from one group tend to rank higher or lower than values from the other. It does not require the same normal-distribution model used by a conventional two-sample t-test. However, it still depends on having independent observations and an outcome that can be meaningfully ordered. When the shapes of the group distributions differ, a rank-test result is not simply a test of whether the two medians differ.
The registry gives p = 0.028 for this comparison. In general, a p-value describes how incompatible the observed data, or more extreme data, are with a specified null model, assuming that model and the test assumptions hold. It does not measure the probability that the null hypothesis is true. It also does not show how large the difference is or whether the difference matters clinically. The registry does not provide a confidence interval or a direct effect estimate for the pairwise contrast, so the size and precision of that contrast cannot be read from the p-value alone.
How should the summaries and ranges be read?
The posted medians and full ranges give useful descriptions of the analyzed groups. The 10mcg group’s median was 10.5, the 100 mcg group’s was 51.75, and the 1000mcg group’s was 270.6 spots per 10^4 CD8+ cells. The ranges include both a negative value in the 100 mcg group and positive values in all groups. These summaries show variation in individual responses, but the registry does not post participant-level observations. They cannot reveal the detailed distribution, whether a few values strongly influence the ranks, or how stable the medians would be in a larger sample.
General statistical interpretation distinguishes within-group spread from uncertainty about a between-group contrast. A broad range signals that observed values vary; uncertainty about a contrast is usually assessed using an interval estimate or an appropriate model. No confidence interval for the reported contrast is posted. The group summaries should therefore be treated as descriptive, and the p-value as the reported test result, rather than combining them into an unreported effect estimate.
What can randomization and masking contribute?
Random allocation can reduce systematic differences between groups at the point of assignment, on average, when implemented and analyzed appropriately. The registry identifies the design as randomized, but details of sequence generation and allocation concealment are not reported in the registry record. The design is unmasked. In general, lack of masking can influence decisions about treatment or outcome assessment, though the record does not describe whether such influence occurred. ELISPOT is a laboratory assay, but the registry does not report masking of laboratory personnel or other assay-quality procedures.
The primary outcome analysis is not described as including everyone enrolled. It includes patients who received at least 3 doses and had measurements at baseline and after the 3rd cycle. In general, excluding participants based on post-randomization treatment exposure or measurement availability can undermine the balance created by randomization. The record does not give reasons for missing measurements, group-specific exclusions, or an analysis that retains participants according to randomized assignment. These omissions limit assessment of potential selection bias.
What do the planning assumptions mean?
The registry’s planning illustration is described as using the observed effect, with the target effect set to a stated fraction of that observed effect, equal allocation, a two-sided significance level of 0.05, and a normal approximation. In general, such a calculation asks how many participants per arm might be needed under those assumptions. It is a planning exercise, not evidence that the trial had a particular probability of detecting an effect and not an observed-power result. The specific planning counts are not reported in the registry record. Because the calculation is anchored to an observed pilot-study effect, it should be treated cautiously: estimates from small groups can be unstable, and an effect observed in a pilot may not be a reliable planning target.
The posted inferential result is one pairwise Wilcoxon comparison. The record does not explain whether this was prespecified, whether any adjustment for multiple testing was planned, or whether other analyses were conducted. Those details matter when interpreting a p-value, particularly if several outcomes or comparisons informed the analysis. The registry does not report enough statistical-plan detail to determine that context.
10. Analyses Reported by the Sponsor
| Method | Estimate | 95% CI | P-value |
|---|---|---|---|
| Wilcoxon (Mann-Whitney) | not reported | 0.028 |
The sponsor’s posted analysis compares the 10mcg group with the 1000mcg group using a Wilcoxon (Mann-Whitney) test and reports p = 0.028. That result is the primary inferential statement available for the reported pairwise comparison. The group-level medians and ranges describe the outcome values; they are not a substitute for the sponsor’s test or an estimate of the contrast.
No recalculated between-group effect estimate or confidence interval is reported in the registry record, and no separate recalculation is available to set alongside the sponsor’s result. The registry identifies the test but does not provide its detailed implementation or a confidence interval for the comparison. The sponsor’s own analysis therefore remains the relevant posted analysis, rather than a numerical reanalysis of the group summaries.
The reported p-value concerns the specified comparison, not all three dose groups collectively. It does not establish a dose trend, and it does not quantify the difference in T-cell response. The 100 mcg group’s outcome summary is descriptive in the posted record; a sponsor analysis comparing that group with either of the other groups is not reported. No discrepancy between the reported and recalculated results is identified in the registry record.
11. Multiplicity, Missing Data and Analysis Populations
The registry posts one inferential result for the primary outcome: a Wilcoxon (Mann-Whitney) comparison of the 10mcg and 1000mcg groups. It does not report a test involving the 100 mcg group, an overall comparison across all three groups, or a dose-trend test. The record does not state whether the reported comparison was selected in advance or whether it was one of several comparisons considered.
Multiplicity arises when multiple outcomes, comparisons, or analyses are evaluated. In general, the more opportunities there are to obtain a small p-value, the more important it is to understand whether and how the analysis plan addressed that issue. The registry does not report a multiplicity adjustment or explain which analyses were included in a family of tests. The reported p = 0.028 should therefore be read as the result for the stated comparison, without assuming that it was adjusted for other analyses.
The primary analysis population consists of the 14 patients who received at least 3 doses and had an ELISPOT measurement at baseline and after the 3rd cycle. The overall enrollment is 23. The registry does not report the disposition of participants who were not included in the primary outcome analysis, the number with missing outcome measurements by group, or the reasons for missing data. It also does not say whether missing values were imputed or whether a sensitivity analysis assessed their potential impact.
In general, analyzing only participants with sufficient treatment exposure and available measurements can select a subset that differs from the randomized population. The extent and direction of any resulting bias cannot be determined without the missingness and disposition details. The record does not report an intention-to-treat analysis for the primary outcome, nor does it describe a prespecified statistical analysis plan, interim analyses, or a sample-size justification. Those matters are not reported in the registry record; their absence from the record is not evidence that a particular procedure was or was not used.
12. 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 |
|---|---|---|---|
| A: CEA Peptide 10mcg | 0 | 5 | 0.0% |
| B: CEA Peptide 100 mcg | 0 | 8 | 0.0% |
| C: CEA Peptide 1000mcg | 0 | 6 | 0.0% |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| A: CEA Peptide 10mcg | 3 of 5 |
| B: CEA Peptide 100 mcg | 2 of 8 |
| C: CEA Peptide 1000mcg | 2 of 6 |
Most frequent non-serious adverse event terms
| Event term | Organ system | A: CEA Peptide 10mcg | B: CEA Peptide 100 mcg | C: CEA Peptide 1000mcg |
|---|---|---|---|---|
| Pain | General disorders | 2 of 5 | 2 of 8 | 1 of 6 |
| Fatigue | General disorders | 2 of 5 | 2 of 8 | 1 of 6 |
| Constipation | Gastrointestinal disorders | 1 of 5 | 1 of 8 | 1 of 6 |
| Diarrhea | Gastrointestinal disorders | 1 of 5 | 1 of 8 | 0 of 6 |
| Anorexia | Metabolism and nutrition disorders | 1 of 5 | 0 of 8 | 1 of 6 |
| Nausea/vomiting | Gastrointestinal disorders | 0 of 5 | 1 of 8 | 0 of 6 |
| Induration | Skin and subcutaneous tissue disorders | 0 of 5 | 1 of 8 | 0 of 6 |
The adverse-event record lists serious events by dose group. In the 10mcg group, 0 of 5 at-risk participants had a serious adverse event; in the 100 mcg group, 0 of 8 did; and in the 1000mcg group, 0 of 6 did. No serious adverse-event terms are listed. Death counts are not reported in the registry record.
Other adverse events were reported for 3 of 5 participants in the 10mcg group, 2 of 8 in the 100 mcg group, and 2 of 6 in the 1000mcg group. The registry does not provide event-level descriptions in these figures, so they do not indicate which events occurred, their severity, timing, or relationship to vaccination.
Zero reported serious events in these groups is not proof that serious harm cannot occur. The listed at-risk groups are small, and the record does not provide detailed follow-up or additional safety analyses. These counts describe what was posted; they do not establish comparative safety across doses.
13. What the Result Does, and Does Not, Mean
The registry’s primary outcome addresses whether measured T-cell frequency changed from baseline after immunization. It reports group medians and ranges for the maximum change, along with a Wilcoxon (Mann-Whitney) p-value for the comparison between the 10mcg and 1000mcg groups. The posted medians are ordered from lower to higher across the listed doses, and the reported test indicates a difference in ranked outcomes for its specified comparison.
That result does not establish that increasing the dose produces a consistent or predictable increase in immune response. Only one pairwise comparison is reported, and the registry does not report a trend test or a comparison involving the 100 mcg group. The ranges show substantial variability in observed values, while no confidence interval or between-group effect estimate is posted to quantify the contrast’s precision.
The endpoint is a laboratory measure. An increase in T-cell frequency is not itself evidence of improved tumor control, longer survival, or better quality of life. The registry record does not present those outcomes as the primary question, and the immune-response finding should not be extended to them. Nor does the adverse-event information establish that one dose is safer than another.
The result also applies to a restricted analysis population: participants who received at least 3 doses and had the required ELISPOT measurements. It should not be assumed to describe every person enrolled or every person who might be eligible for a future study. The registry does not report enough information about exclusions, missing measurements, or analysis planning to resolve those concerns.
In short, the posted results describe an immune-response measure and a reported difference between two dose groups in the selected analysis population. They do not, by themselves, establish clinical efficacy, a dose-response relationship, or comparative safety. Interpretation should remain specific to the endpoint, groups, analysis population, and statistical comparison recorded.
14. Why This Trial Matters Statistically
This study illustrates the distinction between a biological endpoint and a clinical outcome. A trial can measure a change in a laboratory marker that is relevant to its scientific question without showing that the marker corresponds to a change in disease course or patient experience. Keeping that distinction clear prevents an immune-response result from being interpreted as a clinical benefit.
It also highlights why the analysis population matters. Randomization is intended to support comparisons between assigned groups, but the primary endpoint summaries are restricted to participants who received at least 3 doses and had measurements at specified times. That restriction is important context for interpreting both the descriptive values and the pairwise p-value. The registry does not provide enough participant-flow or missing-data information to assess how the subset relates to the randomized groups.
The three dose groups make descriptive comparison possible, but the inferential result reported is limited to the lowest and highest dose groups. A sequence of medians should not be mistaken for a formal dose-response analysis. Statistical reporting is most informative when the comparison, effect estimate, uncertainty interval, analysis population, and handling of multiple tests are all clear. The registry posts the test and p-value, but not a confidence interval or a detailed analysis plan.
Finally, small pilot groups can be useful for learning how an endpoint behaves and identifying questions for later studies, while still yielding imprecise estimates. Planning illustrations based on an observed pilot effect are assumptions for future study design, not evidence of observed power. These features make the record a useful example of how to separate what a trial measured, what its analysis tested, and what the reported data can support.
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 primary outcome analysis includes 14 patients, drawn from an enrollment of 23, and requires both receipt of at least 3 vaccine doses and ELISPOT measurements at baseline and after the 3rd cycle. The registry does not report why other enrolled participants were not included, how many lacked measurements, or whether exclusions differed by randomized group. This limits assessment of selection and missing-data bias.
The group summaries are based on 5, 5, and 4 analyzed participants. Such small groups provide limited information about the distribution of responses, and individual observations can have a strong influence on medians, ranges, and rank-based tests. The full ranges are wide, but the registry does not post participant-level values or a confidence interval for the comparison.
The only reported inferential comparison is between the 10mcg and 1000mcg groups. There is no posted comparison involving the 100 mcg group, no overall three-group test, and no dose-trend analysis. The record does not report whether these analyses were planned in advance or how multiple comparisons were addressed. It also does not report the detailed statistical analysis plan, allocation concealment, or whether outcome assessors were masked.
The endpoint is a laboratory measure of T-cell frequency rather than a reported clinical outcome. The registry does not provide an estimated contrast with an uncertainty interval, nor enough detail to assess assay variability or the handling of repeated measurements. Safety reporting is also limited: serious-event counts and other-event counts are available, but event descriptions and death counts are not reported in the registry record. These limitations constrain interpretation of the observed results without determining what the study might have found under a different analysis.
16. Related Tutorials
17. Related Statistical Calculators
18. Sources
- ClinicalTrials.gov: Study of CAP1-6D in Patients With Locally Advanced or Surgically Resected Pancreatic Adenocarcinoma — NCT00203892.
- Registry record last updated 2014-04-16; status last verified 2014-03.
- Results first posted 2014-03-14.
- 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.
19. Record Summary
The registry record describes a completed randomized, open-label, parallel pilot study of modified CEA peptide CAP1-6D in pancreatic adenocarcinoma. It lists Phase 1 and Phase 2, enrollment of 23, and three dose groups: 10mcg, 100 mcg, and 1000mcg. Each vaccine included Montanide ISA-51 and sargramostim (GM-CSF) 250mcg.
The primary endpoint was maximum change from baseline in ELISPOT-measured T-cell frequency. The analysis population included 14 patients meeting the stated treatment and measurement criteria. Group medians and full ranges were posted, along with a Wilcoxon (Mann-Whitney) comparison of the 10mcg and 1000mcg groups, p = 0.028.
These results concern an immune-response measure and a restricted analysis population. The registry does not report a confidence interval for the comparison, a formal dose-trend test, or detailed information about missing data and multiplicity. Serious adverse-event counts were zero in the listed groups; counts for other adverse events were also reported.