Cyclophosphamide and Vaccine Therapy With or Without Trastuzumab in Treating Patients With Metastatic Breast Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Cyclophosphamide and Vaccine Therapy With or Without Trastuzumab in Treating Patients With Metastatic Breast Cancer (NCT00971737), a randomized phase 2 study. The primary outcome is Toxicity as Assessed by Number of Grade 3 or 4 Adverse Events.

2. Clinical Question

The central comparison is the addition of trastuzumab to cyclophosphamide and vaccine therapy. Because the registry record describes a randomized, open-label, parallel phase II study, allocation was randomized, but participants and study staff were not masked. The posted primary values are event counts: 0 in the cyclophosphamide-and-vaccine-only arm and 2 in the arm that also received trastuzumab. These values describe the reported endpoint, but do not by themselves establish whether the groups differ beyond what could occur by chance. A comparative estimate, uncertainty interval, and statistical test are not reported in the registry record.

The stated population is narrowly defined by metastatic disease and lack of HER-2/Neu over-expression. The registry record does not report enough participant-level information to assess how well the enrolled participants represent other clinical populations. It also does not describe the rationale for selecting the combination or specify a formal hypothesis for the primary comparison. The purpose can therefore be read as an exploratory comparison of toxicity and immune-related outcomes, rather than as a definitive estimate of comparative benefit.

3. Trial Design

This was a randomized, open-label, parallel phase II trial sponsored by Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins. The registry lists an enrollment of 63 and a completed status. The listed study start is 2009-07 and the primary completion is 2016-03. Randomization assigns participants to one of the two treatment groups; the parallel design means the groups receive their assigned regimens rather than crossing over between them. Open-label means masking was not used.

One group received cyclophosphamide and an allogeneic GM-CSF-secreting breast cancer vaccine. The other received those treatments with trastuzumab. The registry describes cyclophosphamide as an intravenous infusion over 30 minutes on day -1 and vaccination as an intradermal dose on day 0. Trastuzumab in the combination arm was given intravenously over 30-90 minutes on day -1. Courses were described as repeating every 4-6 weeks for 3 courses in the absence of disease progression or unacceptable toxicity. The combined-treatment description states that a fourth vaccination occurred at 6-8 months. The fourth-vaccination timing in the vaccine-only description is incomplete in the registry record.

The primary outcome was the number of grade 3 or 4 adverse events meeting the stated toxicity definition, with a timeframe of 3 years. The record also lists immune priming in vaccine-site biopsies, enumeration of CD8+ T cells specific for hTERT by ELISPOT, and characterization of the T-cell memory pool before and after vaccination. Seven outcome measures are listed, but the registry record does not report a statistical analysis plan or sample-size justification. The clinical outcome data needed to interpret the immune measures 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.

Before assignment: 3 subjects were withdrawn prior to receiving intervention (2 due to development of new medical problems, 1 due to anxiety and non-compliance).

Overall Study

Analysis population for the primary outcome, as defined in the registry: Data was not evaluable in 2/30 participants from the cyclophosphamide and vaccine-only arm.

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: 63 subjects were deemed eligible for this study. 2 were taken off-study prior to vaccination due to the development of new medical problems. 1 subject was taken off due to anxiety and non-compliance. 60 research subjects went on to receive at least 1 vaccination, where 30 subjects were randomized to Arm A and 30 subjects were randomized to Arm B.

6. Primary Endpoint

The registry lists the primary outcome as Toxicity as Assessed by Number of Grade 3 or 4 Adverse Events, measured over 3 years.

The primary endpoint is a count of qualifying toxicity events over 3 years. It includes grade 3 or 4 nonhematologic toxicity, except alopecia, and any grade 4 hematologic toxicity, as defined by NCI CTCAE v3.0. A higher count would indicate more events meeting that definition during the stated timeframe, but a count does not show how many distinct participants experienced them. A participant could contribute more than one event, and the record does not give event-level detail to determine whether that occurred.

Counting more severe adverse events is one way to describe treatment tolerability in a clinical trial. It focuses attention on events that meet a specified severity threshold, rather than combining all reported symptoms or laboratory findings into one measure. The exclusion of alopecia and the separate criterion for hematologic toxicity are part of the endpoint definition and affect which events are included.

The registry reports 0 qualifying events for cyclophosphamide and vaccine only and 2 for cyclophosphamide, vaccine, and trastuzumab. These are event counts, not rates or risks, and the registry does not report a confidence interval or a formal comparison for them. Data were not evaluable for 2/30 participants in the vaccine-only arm, while the posted endpoint count for that arm has a denominator of 28. The record does not explain the reason the data were not evaluable. As a result, the count should be read alongside the stated denominator and missingness rather than as a complete description of every randomized participant's experience.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does an event count describe?

In general, a count endpoint records how many events meeting a definition occurred in a group during a specified period. Here the definition is based on toxicity grade and type, and the timeframe is 3 years. A count is distinct from a participant-level outcome such as the number or proportion of people with at least one event. If one participant experiences multiple qualifying events, an event count can include each event; if the registry instead applies another counting convention, that convention is not reported in the registry record.

Counts also differ from rates. A rate relates events to time at risk, while the posted primary values are simple counts. The registry does not report person-time or a rate analysis. Counts can be useful as a straightforward summary, but their interpretation depends on how many people contributed data, how long they were observed, and how consistently events were captured. The posted arm denominators are 28 and 30, and data were not evaluable for 2/30 participants from the vaccine-only arm. Details about follow-up completeness and reasons for non-evaluability are not reported in the registry record.

Why define severity and event type?

In general, clinical trials can use standardized grading systems to categorize adverse-event severity. A pre-specified threshold narrows the endpoint to events considered sufficiently severe for the study's purpose. The registry defines qualifying events as grade 3 or 4 nonhematologic toxicity other than alopecia, or any grade 4 hematologic toxicity, using NCI CTCAE v3.0. This specific definition means that the primary count is not a count of every adverse event. It also means events outside the stated severity and type criteria do not contribute to the primary endpoint.

The registry's primary values are 0 and 2. Those counts are not accompanied by a distribution of event types, onset, duration, recurrence, or attribution to treatment. Therefore, readers cannot determine from the primary values whether the events had similar characteristics or whether any participant experienced more than one qualifying event. The sponsor's analysis details are not reported in the registry record.

How should a randomized comparison be read?

In general, random allocation is intended to balance prognostic factors between treatment groups, which supports a causal comparison when the trial is conducted and analyzed appropriately. The record identifies the study as randomized, but the method used to generate the random sequence and conceal assignment is not reported in the registry record. The design is parallel, so the reported groups represent different assigned regimens. The study is open-label, with no masking.

Even in a randomized trial, observed outcomes vary by chance. An estimate of the difference between groups is usually interpreted with an uncertainty interval, such as a confidence interval. In general, a confidence interval describes a range of values compatible with the data and the assumptions of the analysis; it is not a guarantee that the true effect lies within that range. The registry does not report a confidence interval or a formal statistical test for the primary count. The observed event counts alone cannot show how precise a between-group comparison would be.

For event counts, analysis choices may depend on whether events are repeated within participants, how much follow-up each participant contributes, and whether counts vary more than expected under a simple count model. In general, over-dispersion means the observed variability in counts exceeds what a model assumes; if present, it can make standard errors and inferences from an inadequately chosen model unreliable. The registry record does not report a count-data model, an over-dispersion assessment, or any adjustment for follow-up time. These are general methodological considerations, not claims about analyses performed in this trial.

What does the open-label design imply?

In general, masking can reduce the chance that knowledge of treatment assignment influences care, reporting, or outcome assessment. This trial is listed as having no masking. The primary endpoint depends on adverse-event identification and grading, so knowledge of assignment could be relevant to ascertainment in principle. The registry record does not describe the event-collection procedures or whether an independent group reviewed event grades. It is not possible to determine from the posted information whether open-label status affected the reported counts.

The registry defines the endpoint using NCI CTCAE v3.0, which provides a standardized grading framework. A standard definition can improve consistency, but it does not by itself describe who assessed events or how assessments were performed. The record also does not report whether the endpoint was assessed using the same procedures in both arms. These details matter when interpreting a small number of events, because each counted event can have a substantial influence on the descriptive total.

What can be inferred from the posted analyses?

No analyses are listed in the registry record. The record provides the arm-level primary counts and denominators, but does not report a treatment-effect estimate, standard error, confidence interval, p-value, or adjusted analysis. In general, inferential statistics combine an estimated contrast with a measure of uncertainty, allowing readers to distinguish the observed pattern from the precision with which it is known. Without those reported quantities, the result should remain descriptive.

The registry describes a sample-size illustration based on an observed effect and specified assumptions, but no resulting planning values are included in the record. In general, sample-size calculations are design tools: they show how many participants might be needed under assumptions about the effect, significance level, allocation, and analysis. They are not observed power and do not add evidence to the outcomes already observed. The trial's own sample-size justification and statistical design assumptions are not reported in the registry record.

Three immune-related secondary outcomes are named, while seven outcome measures are listed overall. The registry record does not report results for the immune outcomes or analyses addressing them. In general, when several outcomes are examined, interpretation may be affected by multiplicity: the chance of observing at least one apparently unusual result can rise as more comparisons are made. Whether or how the trial addressed this issue is not reported in the registry record. No adjustment should be assumed.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists seven outcome measures and identifies three secondary immune-response outcomes: immune priming in in-vivo vaccine-site biopsies, enumeration of CD8+ T cells specific for hTERT by ELISPOT, and characterization of the T-cell memory pool before and after vaccination. The record does not report results for those secondary outcomes. The statistical plan, including whether one outcome was designated as the main focus for inference or whether secondary comparisons were adjusted for multiplicity, is not reported in the registry record.

In general, testing multiple outcomes or multiple comparisons can increase the chance that at least one result appears notable by chance. Methods such as defining a hierarchy of outcomes or adjusting significance thresholds can address some multiplicity concerns, but no such method should be attributed to this study without documentation. The record lists no analyses, so it is not possible to assess how outcomes were analyzed or whether any multiplicity strategy was used.

For the primary endpoint, the registry states that data were not evaluable for 2/30 participants in the cyclophosphamide-and-vaccine-only arm. The posted endpoint denominator for that arm is 28; the denominator for the trastuzumab arm is 30. The reasons for missing or non-evaluable data, the timing of those cases, and any method used to handle missingness are not reported in the registry record. The record also does not identify a full analysis population, such as whether all randomized participants were included in the primary analysis.

In general, missing outcome data can affect comparisons if the likelihood of missingness is related to treatment or outcome. The available information does not establish why data were unavailable or whether missingness differed in a way that would alter interpretation. The total enrollment is 63, while the primary endpoint denominators are 28 and 30; an accounting of all enrolled participants by arm and analysis status is not reported in the registry record. These gaps limit assessment of completeness and potential bias.

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 non-serious adverse event terms

The registry's serious-adverse-event figures list 30 participants at risk in each arm and 0 affected by serious adverse events in each arm. No serious adverse-event terms are listed, and deaths are reported as 0 in both groups. These are the posted serious-event summaries; they do not replace the primary endpoint, which counts a separately defined set of toxicity events.

The record also reports 30 participants at risk and 30 affected by other adverse events in each arm. The specific other-event terms and their severity are not included in these figures. They should not be read as meaning that every participant had a serious event: the serious-event count is separately recorded as 0 in each arm. The record does not provide the detail needed to describe the range, duration, or clinical importance of the other adverse events.

In general, a zero count in a study group means no events were recorded in that group under the registry's reporting category; it does not establish that an event is impossible. The figures also need to be interpreted in light of the number followed and the completeness of reporting. The serious-event denominators are 30 per arm, while the primary endpoint has a denominator of 28 in the vaccine-only arm and notes that data were not evaluable for 2/30 participants. The registry record does not explain this difference or provide additional follow-up detail.

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

The posted primary result is a descriptive comparison of qualifying toxicity-event counts: 0 with cyclophosphamide and vaccine only, and 2 with cyclophosphamide, vaccine, and trastuzumab. The definition is limited to grade 3 or 4 nonhematologic toxicity, excluding alopecia, or any grade 4 hematologic toxicity. The counts apply to the stated 3-year timeframe. They do not tell readers how many people experienced an event, whether events recurred in one person, or how events were distributed over time.

The results do not establish that adding trastuzumab caused the difference in observed event counts. The record does not report a formal comparison or uncertainty interval, and the number of events is small. Nor do the counts establish that the regimens have equal toxicity. A conclusion of equivalence would require an analysis designed to assess equivalence or noninferiority and adequate precision; those methods are not reported in the registry record.

These primary data also do not answer whether either regimen improves cancer outcomes. The trial's listed secondary outcomes concern immune responses, but their results are not reported in the registry record. No efficacy result is included in the registry record described by the registry entry. The endpoint should therefore be interpreted as a narrow toxicity summary rather than a complete assessment of benefits and harms.

The trial was randomized, which can support comparison between groups, but it was open-label. The registry does not report the randomization procedure, allocation concealment, a detailed accounting of participants, or the analysis population. Data were not evaluable for 2/30 participants in the vaccine-only arm. Without the reasons for that missingness and a reported analysis of uncertainty, the strength and precision of any comparative conclusion cannot be determined from the posted result.

In general, a trial result is most informative when its endpoint definition, participant flow, analysis methods, and uncertainty are all available. The registry gives a defined primary endpoint and arm-level values, but important elements for interpreting a comparative treatment effect are absent. The appropriate reading is therefore limited: the registry records 0 and 2 qualifying events in the respective groups, while leaving the comparative uncertainty and broader clinical implications unresolved.

13. Why This Trial Matters Statistically

This trial illustrates why outcome definition matters in biostatistics. Its primary endpoint is not a general measure of all toxicity, but a count governed by specified severity and event-type rules. Excluding alopecia and including any grade 4 hematologic toxicity defines exactly which events contribute. Readers who treat the result as a participant count or as a summary of every adverse event would be answering a different question from the one registered.

It also highlights the distinction between a descriptive result and an inferential comparison. The posted arm-level counts are clear, but a comparative conclusion requires information about uncertainty and analysis. In general, confidence intervals help show how precisely an effect is estimated, while a statistical test evaluates data against a specified null hypothesis under stated assumptions. Neither is reported for this primary endpoint. A small count without an interval should not be mistaken for a precise estimate.

Missing outcome data are another important part of interpreting trial results. The record notes that primary endpoint data were not evaluable for 2/30 participants in the vaccine-only arm and reports a denominator of 28 for that arm. The reasons and handling are not described. This makes participant accounting and missing-data methods more than administrative details: they help readers judge whether a reported count reflects the intended comparison.

Finally, the trial lists immune-response measures alongside toxicity. In general, mechanistic or immune outcomes can help characterize biological responses, but they do not automatically answer clinical efficacy questions. The registry lists three secondary immune outcomes without reporting their results. Together, the endpoint and reporting gaps demonstrate why careful statistical interpretation distinguishes what was counted, who contributed data, what comparisons were performed, and what conclusions remain unsupported.

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 primary endpoint reporting is limited to arm-level event counts and denominators. The registry does not provide event descriptions, event timing, recurrence, severity details beyond the qualifying definition, or the number of participants experiencing one or more qualifying events. The event count therefore cannot be converted into a participant-level risk from the posted information. A rate accounting for follow-up time is also not reported.

Data were not evaluable for 2/30 participants in the cyclophosphamide-and-vaccine-only arm, with no reason or missing-data method reported. The primary endpoint denominator for that arm is 28, while the denominator for the trastuzumab arm is 30. Although total enrollment is listed as 63, a complete accounting of participants by randomized arm, treatment received, follow-up, and analysis status is not reported in the registry record.

No primary statistical analysis is listed. The record does not report a comparative effect estimate, confidence interval, p-value, adjustment for covariates, or the analysis model. It also does not report the statistical analysis plan, sample-size justification, randomization method, or allocation concealment. The trial is open-label, and the procedures for adverse-event collection and grading are not described.

The registry lists secondary immune outcomes but does not report their results. The primary outcome timeframe is 3 years, but further detail about follow-up completeness is not reported. The record's vaccine-only arm description has an incomplete timing statement for the fourth vaccination, whereas the combined-arm description states 6-8 months. These reporting limitations prevent a detailed assessment of protocol delivery and outcome completeness. They do not demonstrate that the trial was conducted improperly; they identify what cannot be established from the registry record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Cyclophosphamide and Vaccine Therapy With or Without Trastuzumab in Treating Patients With Metastatic Breast Cancer — NCT00971737.
• Registry record last updated 2019-04-24; status last verified 2019-04.
• Results first posted 2019-03-12.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

18. Record Summary

This completed randomized, open-label, parallel phase II trial enrolled 63 participants with metastatic breast cancer that does not over-express HER-2/Neu. It compared cyclophosphamide plus an allogeneic GM-CSF-secreting breast cancer vaccine with the same regimen plus trastuzumab.

The primary endpoint was the count over 3 years of qualifying grade 3 or 4 toxicity events. The posted counts were 0 in the cyclophosphamide-and-vaccine-only arm, with a denominator of 28, and 2 in the arm that also received trastuzumab, with a denominator of 30. The registry notes that data were not evaluable for 2/30 participants in the vaccine-only arm.

Three secondary immune-response outcomes are listed, but their results are not reported in the registry record. No statistical analyses are listed. Comparative uncertainty, missing-data handling, and a detailed participant accounting are not reported in the registry record.