Combination Immunotherapy With Herceptin and the HER2 Vaccine NeuVax: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Combination Immunotherapy With Herceptin and the HER2 Vaccine NeuVax (NCT01570036), a randomized phase 2 study. The primary outcome is Disease-free Survival (DFS).

2. Clinical Question

3. Trial Design

This was a phase II, randomized, single-blinded, parallel-group study of combination immunotherapy in breast cancer. The registry reports enrollment of 275. Participants were assigned to Herceptin plus NeuVax vaccine or Herceptin plus GM-CSF alone. The study summary describes a multicenter, prospective, placebo-controlled design. Details about the sites, randomization sequence, allocation concealment, and which people were blinded are not reported in the registry record.

The target population was disease-free following standard-of-care therapy and had low or intermediate HER2 expression, described as HER2 1+ or 2+. The summary specifies node-positive disease, with node-negative disease eligible when both ER and PR were negative. It also describes screening and HLA typing because the E75 vaccine was restricted to particular HLA types. The registry summary says E75 was associated with HLA-A2 or HLA-A3 and was extended to HLA-A24 and HLA-A26. The number of participants with each HLA type and the precise eligibility rules are not reported in the registry record.

Both groups received Herceptin every 3 weeks for 1 year according to the arm descriptions. For the NeuVax arm, the first Herceptin infusion was to occur no sooner than 3 weeks and no later than 12 weeks after completion of chemotherapy or radiotherapy. The comparator description gives the same timing. The vaccine was the added component under study, while the comparator arm received GM-CSF alone with Herceptin. Further intervention details are not reported in the registry record.

The study status is completed. The registry lists a start date of 2013-05-21 and primary completion on 2018-09-28. Its primary endpoint was disease-free survival at 24 months. The registry also lists percent ejection fraction as a measure of cardiac toxicity and local and systemic toxicities as secondary outcomes. The posted results contain four outcome measures and eight analyses, but additional details about analysis timing or prespecified statistical procedures are not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

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.

6. Primary Endpoint

The registry lists the primary outcome as Disease-free Survival (DFS), measured over Disease-free survival at 24 months.

The primary endpoint was disease-free survival, assessed at 24 months. In general, disease-free survival measures the time or status of remaining free of specified disease events after treatment. In this registry record, the outcome unit is described as the percentage of participants who survived, and the listed timeframe is disease-free survival at 24 months. The registry description says participants' own physicians at the study sites determined DFS during routine follow-up screening. It describes clinical examinations, laboratory assessment, and radiographic surveillance, occurring every three months for the first 24 months after primary therapies and every six months thereafter.

The endpoint is used to describe whether participants remain disease-free during follow-up, rather than measuring a short-term change in a laboratory value. Such an endpoint can capture clinically relevant events over time. Its interpretation depends on how events and censoring are defined, how follow-up is handled, and which participants are included in the analysis. Those operational details, including the full event definition and censoring rules, are not reported in the registry record.

The posted arm values are 89.8% for Herceptin plus NeuVax vaccine and 83.8% for Herceptin plus GM-CSF alone. The record also reports a hazard ratio from Kaplan-Meier survival analysis. A percentage at a stated time and a hazard ratio convey related but different information: the percentage describes disease-free survival at the stated time, whereas the hazard ratio compares event rates over follow-up under the analysis assumptions. No confidence intervals for the arm-specific percentages are reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry reports a Kaplan-Meier survival analysis for the comparison of disease-free survival. Kaplan-Meier methods are used generally to estimate the probability of remaining free of an event over time while accounting for participants whose follow-up ends before an event is observed. Such participants are censored at their last known follow-up under the analysis conventions. The validity of the estimate depends on the appropriateness of those conventions and on assumptions about censoring. The registry record does not report the detailed censoring rules or the number of participants censored.

What does the hazard ratio summarize?

A hazard ratio compares event rates between groups across the observed follow-up period. It is not the same as a ratio of the two percentages at 24 months, and it does not directly state the chance that an individual participant will experience an event. The reported hazard ratio is 0.62. The analysis lists a 95% confidence interval from .31 to 1.25. The registry record does not state the reference-group coding explicitly, so the estimate should be described as the reported comparison rather than given an unqualified directional interpretation.

In general, a confidence interval describes the range of estimates compatible with the observed data and the statistical procedure, under its assumptions. It communicates precision as well as uncertainty. Here the interval includes 1, which is the relative-hazard value associated with equal hazards. Since the interval also spans values on both sides of 1, it does not pin down the comparison precisely. A confidence interval is not a guarantee that the true value lies within its endpoints, nor does it resolve limitations in the study design or analysis.

How should the p-value be read?

The reported p-value is 0.18. In general, a p-value describes how unusual the data, or data more extreme, would be under a specified null hypothesis and the assumptions of the test. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of an effect. A p-value should be interpreted alongside the estimate and its confidence interval. In this study, the estimate is uncertain, and the reported p-value does not provide clear evidence against the null comparison under the analysis used.

The record does not describe whether the reported p-value was adjusted for multiple comparisons, whether it was prespecified as the principal test, or which exact test produced it. It does not report a full statistical analysis plan. Therefore, further claims about testing procedures, adjustment, or prespecification cannot be made from the registry record.

What do the arm-level percentages tell us?

The posted values at 24 months are 89.8% for Herceptin plus NeuVax vaccine and 83.8% for Herceptin plus GM-CSF alone. These provide a direct description of the recorded endpoint at the stated time. They are not accompanied by arm-specific confidence intervals in the registry record. Without those intervals, their precision cannot be assessed from the percentages alone. The hazard ratio adds a comparison over follow-up, but it does not remove uncertainty about the time-specific values.

General survival-analysis methods can account for varying follow-up, but their estimates depend on the timing and completeness of observations. The registry says DFS was determined by participants' physicians through routine screening, with visits every three months for the first 24 months after primary therapies and every six months thereafter. The record does not provide numbers at risk over time, event counts, follow-up distributions, or details about how differences in follow-up were handled.

Why does over-dispersion matter for count outcomes?

In general, count outcomes can vary more than a simple probability model expects, a feature called over-dispersion. If over-dispersion is present and not accounted for, uncertainty estimates from a model may be too small. That general issue does not mean over-dispersion was found or modeled in this trial. The primary endpoint is presented as disease-free survival and analyzed with Kaplan-Meier methods, not as a count endpoint. The registry record does not report enough detail to assess whether a count model was used for any other outcome, or whether over-dispersion was evaluated.

The record classifies the endpoint as a count/other type and notes that values are missing or not estimable for a two-arm comparison. As a result, no unadjusted recalculation is reported for the two-arm endpoint comparison. The posted Kaplan-Meier analysis remains the available comparative analysis. A recalculation based on summary values would not replace the sponsor analysis in any case; in this record there is no such recalculation available to compare with it.

Finally, statistical significance and clinical meaning are distinct. A numerical estimate can be compatible with a range of possible associations, and the p-value does not determine whether an observed difference is important to patients. The posted results support describing the endpoint values and the uncertainty of the reported comparison. They do not resolve the effect with precision, and they do not supply the additional design and analysis details needed for a complete independent evaluation.

10. Analyses Reported by the Sponsor

The registry reports the sponsor's Kaplan-Meier survival analysis as a hazard ratio of 0.62, with a 95% confidence interval from .31 to 1.25 and p = 0.18. It does not report a recalculated two-arm comparison for this count/other endpoint; the record states that values are missing or not estimable for that purpose. The arm-level 24-month DFS values are 89.8% and 83.8%, but they are not sufficient to reproduce the reported survival analysis because the registry does not provide the event and follow-up details needed for that calculation.

There is no discrepancy flagged in the registry record. When a discrepancy exists, the sponsor's own analysis takes precedence over a simple unadjusted recalculation from posted summary statistics. Here, no competing recalculated result is reported, so the sponsor's Kaplan-Meier estimate is the only posted comparative analysis to interpret. Its confidence interval includes 1, and the p-value is 0.18, so the estimate remains uncertain. The registry record does not provide the full analysis plan, reference-group coding, or missing-data procedures needed to assess the analysis in greater detail.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary endpoint of disease-free survival at 24 months and also lists percent ejection fraction as a measure of cardiac toxicity and local and systemic toxicities as secondary outcomes. It reports four outcome measures and eight analyses. The multiplicity strategy, if any, is not reported in the registry record. In general, assessing several outcomes or analyses can increase the chance of observing an apparently notable result by chance if no adjustment or hierarchy is used. This is a general consideration, not evidence that a particular adjustment was or was not used in this trial.

The record does not state whether the reported hazard-ratio analysis was the sole primary comparison, whether other analyses were prespecified, or how testing across outcomes was handled. It does not report interim analyses or stopping rules. No conclusions about those elements can be drawn from the listed results.

Missing-data handling is also not reported in the registry record. The endpoint description gives a follow-up schedule and says DFS was determined by the participants' own physicians during routine follow-up screening. It does not state how missed visits, incomplete assessments, withdrawal, or loss to follow-up were handled. The number of participants censored, the reasons for censoring, and the amount of missing endpoint information are not reported.

The analysis population is not described in sufficient detail to determine whether the primary result used an intention-to-treat population, a per-protocol population, or another set of participants. The endpoint description says DFS was to be determined for all patients regardless of randomization, but that statement does not specify which participants were included in the reported hazard-ratio analysis or how deviations from assigned treatment were handled. The registry does not report these analysis-population details.

These omissions affect interpretation because the estimate may depend on the analytic population, follow-up completeness, and handling of missing information. They do not establish that the analysis was biased; they mean that readers cannot evaluate these aspects from the registry record. The record also does not report a sensitivity analysis addressing alternative assumptions for missing outcomes or censoring.

12. 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: thromboembolic event (1.5%), skin infection (0.7%), anemia (0.7%), heart failure (0.7%), colonic hemorrhage (0.7%).

The registry reports serious adverse events among 14 of 136 participants in the Herceptin plus NeuVax vaccine group and 12 of 139 participants in the Herceptin plus GM-CSF-only group. It also reports deaths of 1 and 0, respectively. These are counts in the posted safety data. They are not a standalone comparative safety analysis, and the record does not provide confidence intervals or a statistical test for these figures.

The listed serious adverse-event terms include thromboembolic event at 1.5%, and skin infection, anemia, heart failure, colonic hemorrhage, and pancreatitis at 0.7% each. The registry record does not clarify in this listing whether these terms are mutually exclusive or whether participants could contribute to more than one category. It also does not provide event timing, severity breakdowns, duration, or attribution to study treatment.

The record lists other adverse events affecting 136 participants in the vaccine arm and 139 in the comparator arm, but the corresponding at-risk fields are not reported. The listed totals alone do not establish whether an event was caused by treatment or indicate the comparative risk for a specific event. Differences in exposure time, ascertainment, and event definitions can also matter. The safety figures should therefore be read as registry-reported counts and terms, not as proof of a treatment-related effect or a complete safety profile.

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

The posted results describe disease-free survival at 24 months in a randomized comparison of Herceptin plus NeuVax vaccine against Herceptin plus GM-CSF alone. The arm values are 89.8% and 83.8%, and the sponsor-reported Kaplan-Meier analysis gives a hazard ratio of 0.62, a 95% confidence interval from .31 to 1.25, and p = 0.18. These figures convey the observed outcome values and the uncertainty around the reported comparison.

The confidence interval includes 1, and the p-value is 0.18. The estimate therefore does not establish a clear difference between the groups. The interval is compatible with a range of relative-hazard values, and the registry does not report arm-specific confidence intervals for the DFS percentages. The results should not be reduced to the point estimate alone.

The findings apply to the population enrolled in this study and to the treatment regimens described in the record. The target population had selected HER2 expression, was disease-free after standard-of-care therapy, and was subject to HLA-related screening. The registry does not report detailed subgroup results, so the results cannot be interpreted separately by HLA type, node status, or other patient characteristics from the posted information.

The results do not establish how the comparison would apply to populations outside the study criteria, nor do they establish the effect of either treatment in settings not studied. They also do not provide enough information to assess the impact of missing data, the analysis population, multiplicity, or the exact event and censoring definitions. Those details are not reported in the registry record.

Statistical uncertainty is not the same as proof of no difference, and a numerical contrast is not proof of a real difference. The responsible interpretation is that the registry reports these arm values and a comparative survival estimate, with substantial uncertainty around the estimate. The record does not support a definitive claim about the comparative effect.

14. Why This Trial Matters Statistically

This trial raises a common statistical challenge in randomized clinical research: how to interpret a time-to-event comparison when the endpoint is summarized both at a fixed follow-up time and over follow-up. The 24-month DFS percentages provide an understandable snapshot, while the Kaplan-Meier hazard ratio summarizes the comparison across the observed follow-up period. These measures answer related but distinct questions, so reading them together is important.

The confidence interval is central to interpretation. The reported hazard ratio is 0.62, but its 95% confidence interval ranges from .31 to 1.25. That interval includes the equal-hazards value and also allows estimates on either side of it. The p-value of 0.18 is consistent with the substantial uncertainty conveyed by the interval. This illustrates why a point estimate alone can give an incomplete impression of what the data support.

The record also illustrates how much a registry result depends on reporting detail. A survival estimate is easier to assess when event definitions, censoring rules, numbers at risk, analysis populations, and missing-data procedures are described. Those items are not reported in the registry record. Without them, the numerical results can be stated and interpreted cautiously, but many aspects of the analysis cannot be independently checked.

There is no recalculated two-arm result for this endpoint, so the sponsor-reported analysis is the available comparative estimate. This is a reminder that recalculation from summary statistics is not always possible, especially for survival outcomes where event timing and censoring information are important. The posted analysis should be read with its uncertainty and reporting limitations in view.

More broadly, this trial demonstrates that statistical interpretation involves more than deciding whether a p-value crosses a threshold. The estimate, interval, endpoint definition, study population, and completeness of reporting all shape what can be concluded. Here, the results describe an observed comparison but do not resolve its magnitude precisely.

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 registry record does not report a full statistical analysis plan, detailed randomization procedures, allocation concealment, or which parties were blinded. Although the study is described as single-blinded and placebo-controlled, further information about the blinding and placebo procedures is not reported. These omissions limit assessment of the potential for performance or ascertainment effects.

The primary endpoint description does not give the full operational definition of DFS, including the events counted, the rules for censoring, or how deaths and other events were classified. It says that participants' physicians determined DFS during routine follow-up, but the record does not report independent adjudication or quality-control procedures. Numbers at risk, DFS event counts, duration of follow-up, and arm-specific confidence intervals for the 24-month values are not reported.

The analysis population, protocol deviations, missing outcomes, withdrawal, and loss-to-follow-up handling are not reported. The record also does not state whether analyses were adjusted for baseline characteristics, whether proportional hazards were assessed, or which reference coding was used for the hazard ratio. Consequently, the statistical assumptions and the exact interpretation of the comparison cannot be fully evaluated.

Multiplicity procedures, interim analyses, stopping rules, and the sample-size justification are not reported in the registry record. The observed sample sizes and posted results do not reveal the original planning assumptions. No conclusions about the trial's planned power or its design rationale should be inferred from the posted endpoint values.

The target group was defined by HER2 expression, disease status, node status or hormone-receptor status, and HLA-related eligibility. The record does not provide enough results to assess whether the endpoint differed across these groups. The safety listing also lacks event timing, attribution, and detailed comparative analysis. These are trial-specific limits on the conclusions that can be drawn from the registry results.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Combination Immunotherapy With Herceptin and the HER2 Vaccine NeuVax — NCT01570036.
• Registry record last updated 2020-12-02; status last verified 2020-11.
• Results first posted 2020-12-02.
• 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.

19. Record Summary

This completed phase II randomized trial compared Herceptin plus NeuVax vaccine with Herceptin plus GM-CSF alone in people with selected HER2-expressing breast cancer who were disease-free after standard-of-care therapy. The registry reports enrollment of 275 and describes the study as single-blinded, parallel-group, and placebo-controlled.

The primary endpoint was disease-free survival at 24 months. The posted values were 89.8% for Herceptin plus NeuVax and 83.8% for Herceptin plus GM-CSF alone. The reported Kaplan-Meier hazard ratio was 0.62, with a 95% confidence interval from .31 to 1.25 and p = 0.18. The estimate is uncertain, and the record does not establish a clear between-group difference.

The registry also reports cardiac-toxicity and local and systemic toxicity outcomes. Details of the statistical plan, analysis population, missing-data handling, and several survival-analysis procedures are not reported in the registry record, limiting independent assessment.