Immunotherapy of HLA-A2 Positive Stage II-IV Melanoma Patients: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Immunotherapy of HLA-A2 Positive Stage II-IV Melanoma Patients (NCT01308294), a non-randomized phase 1, phase 2 study. The primary outcome is Ex Vivo Frequency of Melan-A Specific CD8+T Cells.

2. Clinical Question

3. Trial Design

The registry describes a Phase 1 and Phase 2, parallel, non-randomized, unmasked study of vaccination in melanoma. The study began in 2010-06, had primary completion in 2014-04, and is recorded as terminated. Enrollment is listed as 16. The sponsor was Centre Hospitalier Universitaire Vaudois.

Three intervention groups are described. Group 1 was planned to receive two vaccine injections in one limb, with the injection sites separated by 5 cm. The arm description says 9 patients were initially planned. Group 2 was planned to receive the same vaccine in two syringes injected into distinct limbs; its description also says 9 patients were initially planned. Group 3 was planned to receive the vaccine without the MHC class II peptide MAGE-A3, and the description says 9 patients were initially planned, but no patients could be enrolled before premature termination. The detailed arm descriptions in the registry record end partway through their description of vaccine components.

These design details matter when interpreting comparisons. The record explicitly notes that allocation was not randomized and that the groups are not a like-for-like control comparison. It does not describe a randomized comparator or establish that the groups can be used to estimate a causal effect of injection location or vaccine composition. The Group 3 comparison is especially unavailable because that group had no enrolled patients.

The registry lists Tumor Response as a secondary outcome, but numerical tumor-response results are not reported in the registry record. The record lists seven outcome measures and zero analyses, but does not provide numerical results for the primary immune endpoint. It also does not report a detailed statistical analysis plan, sample-size justification, interim analysis plan, or reasons for termination beyond the Group 3 statement that premature termination prevented enrollment there.

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.

Before assignment: An unexplained high percentage, i.e. about 90% of the 11 prescreened patients were HLA-A2 negative. We assume that this was due to the increasing ethnic diversity of patients. The consequence of the time delay was that some of the peptides expired definitively.

Cycle 1

Cycle 2

Cycle 3

Analysis population for the primary outcome, as defined in the registry: No patients were included in the group 3

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: Due to premature trial termination, no patients could be enrolled in the Group 3.

6. Primary Endpoint

The registry lists the primary outcome as Ex Vivo Frequency of Melan-A Specific CD8+T Cells, measured over Melan-A specific CD8+ T cells were measured in PBMC collected at the end of Cycle 1 (Week 7), end of Cycle 2 (Week 25), end of Cycle 3 (Week 40), and Follow-up (6 to 18 months after the end of Cycle 3)..

The primary endpoint measured fold change in the frequency of Melan-A-specific CD8+ T cells in peripheral blood mononuclear cells (PBMC). The cells were measured directly ex vivo, without prior in vitro expansion, using multicolor flow cytometry with Melan-A ELA tetramers. Measurements were specified at the end of Cycle 1 (Week 7), the end of Cycle 2 (Week 25), the end of Cycle 3 (Week 40), and follow-up, 6 to 18 months after the end of Cycle 3.

For each time point, the registry defines fold change as the frequency at that time point divided by the frequency at baseline. It defines a significant T cell response as at least 2-fold change compared with the pre-immunotherapy frequency. This gives a biological interpretation to the endpoint: it tracks change in the measured frequency relative to each participant’s baseline rather than reporting only a post-vaccination level.

As a general statistical point, a fold-change measure can be sensitive to the baseline value, and a summary across participants does not show the individual response pattern. The registry lists the endpoint summary as a median with standard deviation, but numerical values and participant-level distributions are not reported in the registry record. It is therefore not possible to describe the observed magnitude, variability, or timing of the immune response from the posted results.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies a median as the summary measure for the primary endpoint and standard deviation as its dispersion measure. However, it does not report the numerical summary values. The explanations below describe general statistical concepts and do not imply that a particular analysis was conducted in this study.

What does the fold change represent?

The endpoint is a ratio of a measured cell frequency at a follow-up time point to that participant’s baseline frequency. A value equal to the baseline would represent no change on that ratio scale; a larger value would indicate a higher measured frequency relative to baseline, and a smaller value would indicate a lower frequency. The registry defines a significant T cell response using a threshold of at least 2-fold change. That operational threshold is distinct from statistical significance, which ordinarily refers to evidence evaluated using an inferential procedure such as a test or confidence interval. No such inferential result is reported in the registry record.

Using each participant’s baseline as a reference can help express change relative to starting frequency, but ratio measures can be affected by the size and variability of the baseline values. A small baseline can make a ratio sensitive to modest absolute changes. The record does not provide baseline frequencies or participant-level values, so the effect of baseline variation cannot be assessed.

What do a median and standard deviation tell readers?

In general, the median is the middle of an ordered set of observations and is often useful when data are skewed or contain extreme values. The standard deviation describes spread around a mean, not around the median. Pairing a median with a standard deviation is possible as a reporting choice, but it does not provide the same interpretation as a median with an interquartile range. Without the numerical values or the underlying distribution, readers cannot determine whether the endpoint data were skewed, how dispersed they were, or whether the chosen summary captured the observed pattern well.

A summary value alone would also not show how many participants contributed at each time point or how many had an increase, a decrease, or little change. Repeated measurements at the specified cycle endpoints and follow-up could be analyzed in several ways, but the registry record does not report a statistical model, a within-participant method, or the handling of incomplete repeated measurements. No particular method should be inferred.

How should comparisons between groups be understood?

Random assignment is commonly used to balance measured and unmeasured participant characteristics across comparison groups, on average. This study was non-randomized, and the registry itself warns that the arms are not a like-for-like control comparison. As a general matter, observed differences between non-randomized groups may be associated with the intervention, but may also reflect differences in participant characteristics, enrollment timing, or other factors. The record does not report adjusted comparisons that might address measured differences, and adjustment would not by itself eliminate all potential confounding.

The parallel design means that the interventions were organized into separate groups rather than described as a crossover sequence. But the record does not provide enough detail to establish the specific analytic contrast intended among all groups. Group 3 had no enrolled participants, preventing an observed comparison for the planned vaccine without MAGE-A3. The registry does not report numerical primary endpoint results for Group 1 or Group 2 either.

What would a confidence interval add?

As a general statistical concept, a confidence interval gives a range of values compatible with the data and an estimation procedure, expressing uncertainty about an effect estimate. A narrower interval indicates greater precision under that procedure; a wider interval indicates greater uncertainty. It does not show that every value in the interval is equally likely, and its interpretation depends on the analysis assumptions. The registry record reports no confidence interval for the primary endpoint, safety outcomes, or tumor response. The precision of these outcomes therefore cannot be evaluated from the posted results.

Why can over-dispersion matter?

As a general concept, over-dispersion occurs when observed variability exceeds what a statistical model assumes. If a model does not account for that extra variation, standard errors can be too small and inferential results can appear more precise than warranted. This is relevant to many biological measurements, but the registry does not report a model for the immune endpoint or information to assess over-dispersion in these data. It would be inappropriate to claim that over-dispersion occurred or that a particular correction was used.

What do the planning illustrations mean?

The registry’s planning assumptions describe illustrations based on the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The intended effect to detect is described as a stated fraction of the observed effect. These are planning illustrations, not observed power. They do not establish that the trial was designed around those assumptions, and the registry record does not report a sample-size justification from the original protocol. The numerical planning results are not reported in the registry record. The record also states that time-to-event planning quantities, when present, refer to total events, while other planning quantities refer to participants per arm; no tumor-response analysis or time-to-event result is reported in the posted results.

10. Multiplicity, Missing Data and Analysis Populations

The primary endpoint includes measurements at the end of Cycle 1 (Week 7), the end of Cycle 2 (Week 25), the end of Cycle 3 (Week 40), and follow-up 6 to 18 months after the end of Cycle 3. Considering several time points can raise a multiplicity issue: when many outcomes or comparisons are tested, the chance of at least one apparently unusual result can increase. General approaches include prespecifying a primary time point, adjusting for multiple testing, or presenting analyses as exploratory. The registry record does not report a multiplicity strategy, a hierarchy among time points, or a statistical analysis plan.

The registry lists seven outcome measures, including Tumor Response as a secondary outcome, but numerical results are not reported for the primary immune endpoint or tumor response. It lists zero analyses. No p-values, confidence intervals, or group comparisons are reported in the registry record. It is therefore not possible to assess whether any multiplicity adjustment was made or to distinguish prespecified analyses from exploratory ones.

Missing-data handling is also not reported in the registry record. The record does not state how many participants contributed measurements at each scheduled time point, why any measurements might be absent, or whether missing measurements were excluded, imputed, or addressed through a longitudinal model. The termination and absence of Group 3 enrollment make completeness and analysis-population definitions important, but the registry does not describe those populations or give a participant flow account.

The registry record does not report whether analyses followed an intention-to-treat, per-protocol, or other population definition. It also does not state how participants who discontinued vaccination or lacked follow-up measurements were handled. Without those details, the reader cannot establish which participants contributed to each outcome or how missing information may have affected the estimates.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: upper limb fracture (10.0%).

The safety record reports serious adverse event counts separately by group. Group 1 lists 10 participants at risk and 1 with a serious adverse event. Group 2 lists 6 at risk and 0 affected. Group 3 lists 0 at risk and 0 affected, consistent with the statement that no patients were included in that group. The record reports zero deaths in each group. These figures are descriptive and should be read with their group-specific denominators; the groups were not randomized and are not described as like-for-like controls.

The record also lists Upper limb fracture among the top serious adverse event terms at 10.0%. It does not report the term’s context, timing, attribution, or a group-specific breakdown in the information available for this analysis. A serious adverse event count is not the same as a judgment that an event was caused by vaccination. Causality and event details are not reported in the registry record.

Other adverse event counts are listed as 10 affected among 10 at risk in Group 1, 6 among 6 in Group 2, and 0 among 0 in Group 3. These are recorded as other adverse events, distinct from the serious adverse event counts. The small group sizes and lack of randomization limit comparative interpretation. Zero recorded events in a group do not establish that the underlying risk is zero, and the record does not provide uncertainty intervals or a detailed safety analysis.

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

The registry record describes a study designed to examine whether vaccination with tumor antigenic peptides, IMP321/LAG-3, and Montanide could induce an immune response in melanoma patients, while assessing safety and tolerability. Its primary endpoint was a fold change in ex vivo Melan-A-specific CD8+ T cell frequency relative to baseline. The registry defines a response threshold, but does not report the numerical endpoint results or how many participants met that definition.

Accordingly, the posted results do not establish the size, precision, or consistency of an immune response. They also do not establish whether changes differed across the scheduled time points or between groups. No inferential results for the primary endpoint are reported in the registry record. The existence of an endpoint definition or a planned measurement schedule should not be mistaken for a reported finding.

The design does not support a straightforward causal comparison of the planned injection approaches. Allocation was non-randomized, and the record warns that arms are not a like-for-like control comparison. The third group, which would have received a vaccine without MAGE-A3, had no enrolled patients. It is therefore not possible to use the record to compare outcomes with versus without that peptide.

The safety counts describe recorded events in the groups with participants at risk, including serious adverse events and deaths. They do not establish that vaccination caused or did not cause an event, and the absence of recorded deaths does not establish the absence of other risks. The record does not report clinical context, timing, attribution, or uncertainty estimates for these safety outcomes.

Finally, the registry lists tumor response as a secondary outcome but does not report numerical tumor-response findings. This record can be used to understand the study’s intended question, endpoint definition, design, and reported safety counts. It cannot, from the posted results alone, support conclusions about clinical benefit, lack of benefit, or comparative efficacy.

13. Why This Trial Matters Statistically

This study illustrates why endpoint definitions and numerical outcome reporting are both essential. The registry specifies a biologically focused endpoint, describes the laboratory measurement, gives the baseline-relative fold-change calculation, and defines a threshold for a significant T cell response. Those details make the intended measurement understandable. Without the actual values, participant counts at each time point, and uncertainty estimates, however, readers cannot judge the magnitude or precision of any observed change.

The record also shows how design features constrain what a study can answer. A non-randomized structure may be suitable for describing outcomes, but it does not ensure comparable groups. The registry explicitly warns against treating the groups as like-for-like controls. Group 3’s lack of enrollment means that one planned variation in vaccine composition has no observed group outcomes to interpret.

The scheduled measurements at multiple times raise useful questions about longitudinal analysis and multiplicity. The registry identifies the measurement schedule, but does not report a statistical analysis plan, handling of missing observations, or a multiplicity approach. These omissions prevent readers from determining how the investigators planned to distinguish the primary analysis from additional comparisons.

The safety record offers an additional lesson in reading denominators. Counts should be interpreted alongside the number at risk in each group and the nature of the event category. Here, the recorded groups are small, one planned group has no participants, and the study is non-randomized. Descriptive counts can characterize what the registry reports, but they cannot provide a precise comparative safety estimate by themselves.

More broadly, this record distinguishes a trial’s stated purpose from the evidence available in its results. It documents intended assessment of immune response, safety, tolerability, and tumor response, but numerical primary and secondary outcome results are not reported in the registry record. For statistical interpretation, that gap is consequential: readers need estimates, denominators, variability, and analysis details to assess what the observed data support.

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 registry record has important limitations specific to this study. It reports enrollment of 16 but does not provide a complete participant flow or explain the allocation of all enrolled participants across groups. Safety counts list 10 at risk in Group 1 and 6 in Group 2, while Group 3 had none. The arm descriptions say 9 patients were initially planned for each group, including Group 3, but do not explain the difference between those plans and the reported safety denominators.

The detailed descriptions of the vaccine components for Groups 1 and 2 are truncated in the registry record. Group 3 is described as omitting the MHC class II peptide MAGE-A3, but no participants were enrolled there due to premature trial termination. This leaves the planned vaccine-composition comparison unobserved.

The primary endpoint results are not reported numerically. There are no medians, standard deviations, participant counts at each measurement, or results for the defined response threshold. The record also does not report numerical tumor-response results despite listing Tumor Response as a secondary outcome. It lists zero analyses, and does not provide confidence intervals, p-values, group comparisons, or adjusted estimates.

The non-randomized design and lack of masking further limit interpretation, particularly for comparisons between groups. The registry itself says the arms are not a like-for-like control comparison. Missing-data procedures, analysis populations, sample-size justification, statistical methods, multiplicity handling, interim analyses, and the reason for overall termination are not reported in the registry record. These gaps prevent a detailed assessment of bias, precision, and the completeness of outcome analyses.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Immunotherapy of HLA-A2 Positive Stage II-IV Melanoma Patients — NCT01308294.
• Registry record last updated 2020-06-11; status last verified 2020-05.
• Results first posted 2020-03-20.
• 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

NCT01308294 is a non-randomized, unmasked, parallel Phase 1 and Phase 2 study of vaccination in HLA-A2 positive Stage II-IV melanoma patients. It was sponsored by Centre Hospitalier Universitaire Vaudois, began in 2010-06, had primary completion in 2014-04, and is recorded as terminated. Enrollment is listed as 16.

The planned groups varied injection location and vaccine composition. Group 3 had no enrolled patients because of premature termination. The primary endpoint measured fold change from baseline in ex vivo Melan-A-specific CD8+ T cells at scheduled cycle visits and follow-up. The numerical primary and tumor-response results are not reported in the registry record. Safety counts are reported by group, with zero deaths recorded in each group. The record cautions that the non-randomized groups are not like-for-like controls.