Trial of Melaxin Cancer Vaccine Plus Bacillus Calmette-Guerin (BCG) to Treat Malignant Melanoma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Trial of Melaxin Cancer Vaccine Plus Bacillus Calmette-Guerin (BCG) to Treat Malignant Melanoma (NCT00671554), a na phase 1, phase 2 study. The primary outcome is Safety as Measured by Number of Participants With Unexpected Adverse Events or Unexpected Laboratory Results..

2. Clinical Question

In practical terms, this is a small, uncontrolled safety-focused study record, not a comparative test of treatment effects. Its central question concerns whether unexpected safety findings were observed among participants receiving Melaxin with BCG. Because there is no control group, any events cannot be compared with an alternative treatment or with outcomes under a different study condition within this trial.

The registry states that expected adverse events included injection site reactions, fever, chills, and arthralgias. It reports no unexpected or uncommon adverse events such as disseminated sepsis, and says clinical laboratory results were within expected ranges, including an increase in the number of IFN gamma expressing T-cells. Those statements describe the record’s reported safety findings; they do not establish a comparative effect or provide detailed participant-level results.

3. Trial Design

The registry classifies the study as Phase I and Phase II, open-label, and single group. No allocation procedure is listed, and masking is recorded as none. These design features mean that the study record does not describe randomized assignment, blinded assessment, or a concurrent comparator arm. The sponsor is Prisma Health-Upstate. Enrollment is listed as 3, and the status is terminated.

The registered intervention combines Melaxin, an autologous cellular vaccine, with BCG injections. The arm description specifies four 1 ml doses of 250,000 dendritomas administered subcutaneously at 4 week intervals, together with a separate subcutaneous injection containing 1 million Colony Forming Units of BCG. If injection site ulceration occurred, the description says the BCG dose would be decreased by 50% in subsequent dosing. The registry does not report how many participants received each scheduled dose or whether the dose adjustment was used.

The study began in 2008-04, and primary completion is listed as 2009-02. Its primary outcome is the number of participants with unexpected adverse events or unexpected laboratory results over the timeframe from first vaccine to 18 months after the last injection. Tumor response measured by RECIST criteria and progression-free survival are listed as secondary outcomes. The record lists 2 outcome measures, but does not report outcome analyses for them.

The registry record does not report the reason for termination, details of the protocol changes, the intended or achieved follow-up for each participant, or whether enrollment reached a planned target. It also does not report a sample-size justification or a statistical analysis plan. The design information supports describing the study structure, but not inferring why it ended or what its planned statistical procedures were.

Because there is no randomized comparison, differences between this group and any outside population would not be like-for-like evidence from this trial. A single-group design can describe observations among participants who received the intervention, but it cannot separate the intervention’s association with an outcome from other influences using an internal control group.

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

Reasons for not completing this period:

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 Safety as Measured by Number of Participants With Unexpected Adverse Events or Unexpected Laboratory Results., measured over From first vaccine to 18 months after the last injection.

The primary endpoint counts participants with unexpected adverse events or unexpected laboratory results. The unit is participants, and the parameter is recorded as a number. The stated observation timeframe runs from the first vaccine to 18 months after the last injection. This is a safety endpoint: it is intended to capture whether findings outside those expected by the study team were reported during the specified period.

The registry description identifies expected events as injection site reactions, fever, chills, and arthralgias anticipated with vaccine therapy. It gives disseminated sepsis as an example of an unexpected or uncommon adverse event and states that no such unexpected or uncommon events occurred. It also states that clinical laboratory results on all participants were within expected ranges, including an increase in the number of IFN gamma expressing T-cells. These statements are the posted narrative result; the record does not provide a participant-by-participant listing of primary endpoint events or laboratory values.

In general, a safety endpoint is useful because it directs attention to harms and abnormal findings during treatment and follow-up, rather than focusing only on a potential benefit. A count of participants is different from a count of events: one participant may experience more than one event, while the endpoint as titled asks how many participants had at least one qualifying finding. The record does not define the operational rules used to classify an event or laboratory result as unexpected, nor does it specify how missing follow-up was handled.

The secondary outcomes address tumor response measured by RECIST criteria and progression-free survival. Their inclusion indicates that tumor outcomes were of interest, but no analysis results for those outcomes are reported in the registry record. The primary safety description therefore should not be interpreted as a result about tumor response or progression-free survival.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What the design allows. The study is listed as single group, with no allocation and no masking. In general, an uncontrolled design describes outcomes among people receiving an intervention but lacks a concurrent group to estimate what outcomes might have occurred under another condition. Random allocation, when used, helps make comparison groups similar on average with respect to measured and unmeasured factors. There is no randomized comparison in this registry record, so a treatment contrast cannot be estimated from the registered design.

Open-label means the registry does not describe masking of treatment assignment. In general, awareness of treatment can influence reporting, evaluation, or classification of outcomes, especially when outcomes require judgment. The registry does not report the procedures used to assess adverse events or laboratory findings, or whether assessors were independent of treatment delivery. It would be inappropriate to assume such procedures were or were not used beyond the listed masking status.

Counting participants and events. The primary endpoint is expressed as the number of participants with unexpected adverse events or unexpected laboratory results. A participant count answers how many people experienced a qualifying finding, not how many separate findings occurred. In general, event counts and participant counts can differ if a person has multiple events. The record also lists serious and other adverse-event summaries; these should not automatically be treated as the primary endpoint because the primary endpoint has its own definition and the registry does not report a numerical analysis connecting the summaries.

The registry does not report a confidence interval for the primary count. In general, a confidence interval communicates uncertainty in an estimate under a specified statistical method. For rare or uncommon events, a small study can leave considerable uncertainty even when few or no unexpected events are reported. The registry record gives enrollment of 3, but it does not provide an interval estimate, a model, or a formal statistical test for the primary endpoint. No new uncertainty estimate can be recovered from the narrative alone.

Event classification and follow-up. The stated timeframe is from first vaccine to 18 months after the last injection. In general, a defined follow-up window clarifies when observations count toward an endpoint. Interpreting a safety count also depends on what qualifies as an event, how unexpectedness is determined, how severity and seriousness are classified, and how follow-up is collected. Those operational definitions and collection procedures are not reported in the registry record. The arm description includes a possible BCG dose decrease following injection site ulceration, but the record does not say whether that occurred.

The posted narrative identifies injection site reactions, fever, chills, and arthralgias as expected adverse events. It describes no unexpected or uncommon adverse events such as disseminated sepsis and says the laboratory results were within expected ranges, including an increase in IFN gamma expressing T-cells. In general, a statement that observed laboratory results were within expected ranges does not substitute for measured values, reference ranges, or a prespecified rule for classifying abnormality. Those details are not reported in the registry record.

Secondary outcome methods. Tumor response measured by RECIST criteria and progression-free survival are listed as secondary outcomes. In general, a response criterion supplies rules for categorizing tumor response, while progression-free survival concerns time until a specified progression event or death, subject to the outcome definition and analysis rules. The registry does not provide the detailed outcome definitions, assessment schedule, censoring rules, or any analysis results for these outcomes. No tumor response or progression-free survival effect can therefore be interpreted from the entry.

Multiplicity and model choices. Multiplicity arises in general when several outcomes or statistical tests are considered, because the chance of a misleading positive finding can increase across repeated testing. The record lists a primary safety outcome and secondary tumor outcomes, but does not report a multiplicity strategy, a testing hierarchy, or any analyses. It is not possible to say whether adjustment was planned or applied. Likewise, the statistical analysis plan, including model choice and any covariate adjustment, is not reported in the registry record.

Planning calculations. The registry’s sample-size planning description refers to an illustration based on an observed effect, two-sided significance level 0.05, equal allocation between two arms, and a normal approximation. It says that the effect to detect is a stated fraction of the observed effect, with time-to-event calculations using total events and other calculations using participants per arm. These are planning assumptions, not observed power. However, the record does not report outcome analyses or an observed effect for the endpoints, and the trial itself is listed as single group. Thus the planning description does not establish the study’s original sample-size rationale or a valid comparative sample-size target for this design. The actual sample-size justification is not reported in the registry record.

In general, over-dispersion occurs when variability is greater than a statistical model assumes. It matters because standard errors and uncertainty intervals can be too small if the model understates variability. The registry does not report a model for the primary endpoint, a dispersion assessment, or an analysis that would make over-dispersion relevant to a particular result. This is a general methodological consideration, not a claim about an analysis conducted in this study.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Melaxin")
n      <- c(3)
events <- c(0)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Melaxin  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2

11. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary safety outcome and secondary outcomes of tumor response measured by RECIST criteria and progression-free survival. Considering several outcomes can raise multiplicity concerns: in general, repeated statistical testing can increase the chance of at least one apparently notable result even when no true effect is present. The registry record does not report an adjustment procedure, a testing hierarchy, or the statistical analysis plan. It is therefore not possible to determine whether multiplicity was considered or how it would have been handled.

The record lists 2 outcome measures and 0 analyses. It provides a narrative description of safety findings, but does not provide statistical analyses for the primary endpoint or the secondary outcomes. No p-values, confidence intervals, model-based estimates, or hypothesis tests are reported. The absence of such quantities should not be read as evidence that a test was nonsignificant; there is no reported test result to interpret.

Missing-data methods and analysis populations are also not reported in the registry record. The entry does not explain how participants with incomplete follow-up, missing laboratory results, or unavailable tumor assessments would be handled. It does not identify an intention-to-treat, per-protocol, safety, or other named analysis population. Although the primary endpoint timeframe is specified, the record does not state how follow-up duration varied across participants or how incomplete observation affected the reported summaries.

In general, missingness can matter when the availability of data is related to outcomes or participant characteristics. Describing who was included in each analysis and how missing observations were treated helps readers assess the scope of an estimate. Here, no analysis populations or missing-data procedures are reported, so the representativeness and completeness of outcome summaries cannot be assessed from the registry record.

The study’s terminated status does not explain the missing analyses. The reason for termination, whether planned analyses were completed outside the registry, and whether results were unavailable for other reasons are not reported in the registry record.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: brain metastasis (66.7%), injection site reactions (66.7%), death due to disease progression (33.3%), post-operative meningitis (33.3%), post-operative wound infection (33.3%).

The adverse-event summary lists 3 participants at risk for serious adverse events and 2 affected. For other adverse events, it lists 3 at risk and 3 affected. These are participant-level summary fields in the registry record; they do not state how many total event episodes occurred or whether one participant had more than one event. The serious-event terms are listed with percentages: Brain Metastasis, 66.7%; injection site reactions, 66.7%; Death due to disease progression, 33.3%; Post-Operative meningitis, 33.3%; and Post-operative wound infection, 33.3%.

In general, a serious adverse event is a classification of an event’s seriousness, not a statement that the intervention caused it. The record does not report causal attribution for the listed terms. It also does not give participant-level overlap, event dates, duration, outcomes other than the named terms, or enough detail to determine whether each term represents a distinct person or event episode. The deaths field for the serious-event group is not reported in the registry record as a numeric value. The phrase Death due to disease progression appears as a serious-event term, but it does not supply a separate death count.

The primary endpoint’s narrative says that no unexpected or uncommon adverse events occurred and that laboratory results were within expected ranges. This statement should not be read as saying that no adverse events of any kind occurred: the registry separately lists serious and other adverse-event summaries. Nor should the serious-event list be treated as a direct numerical result for the primary endpoint, whose definition concerns unexpected adverse events or unexpected laboratory results.

With enrollment of 3, the summaries describe a very small study group. In general, small groups provide limited information about the frequency and range of safety events, particularly events that are uncommon. The registry does not report uncertainty intervals around these percentages, so their precision cannot be assessed from an interval estimate.

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

The record documents a terminated, open-label, single-group study of Melaxin with BCG in Stage IV malignant melanoma, with enrollment of 3 participants. The primary endpoint concerned the number of participants with unexpected adverse events or unexpected laboratory results. The posted narrative states that no unexpected or uncommon adverse events occurred, gives disseminated sepsis as an example, and says laboratory results were within expected ranges, including an increase in IFN gamma expressing T-cells.

That narrative is a descriptive safety statement, not a comparative estimate. There was no control arm, no randomized allocation, and no masking listed. Consequently, the registry record cannot show how the observed findings compare with outcomes under another treatment or without the study intervention. It also cannot determine whether a reported event was caused by Melaxin, BCG, the disease, or another factor; causal attribution is not reported.

The record separately lists serious and other adverse-event summaries, including Brain Metastasis, injection site reactions, Death due to disease progression, Post-Operative meningitis, and Post-operative wound infection. These entries do not negate or replace the primary endpoint narrative; the categories and reporting context differ. The registry does not provide enough detail to reconcile each event with the primary endpoint definition or to establish attribution.

Tumor response measured by RECIST criteria and progression-free survival are named as secondary outcomes, but analyses for them are not reported in the registry record. Therefore, this entry does not establish an estimate of tumor response or progression-free survival. A safety narrative alone cannot answer those efficacy questions.

In general, a result should be interpreted in light of its design, sample size, endpoint definition, uncertainty, and completeness of follow-up. Here, no inferential analyses or uncertainty intervals are reported, and the reason for termination is not reported. The record supports a limited description of what was entered as observed safety findings; it does not support a conclusion that the intervention is effective, ineffective, safe, or unsafe in a broader population.

14. Why This Trial Matters Statistically

This trial matters statistically because its design and registry reporting illustrate the distinction between describing outcomes and estimating comparative effects. A single-group study can record events among people who received an intervention, but without a concurrent comparator it cannot estimate a treatment contrast within the trial. The absence of random allocation and masking also limits the kinds of bias protections that a randomized, blinded comparison may provide.

The small enrollment of 3 makes the scope of the safety description especially important. In general, a small sample can reveal that particular findings were recorded, but it offers limited information about the range and frequency of uncommon events. A point summary alone does not show its precision; an uncertainty interval or other statistical characterization would be needed to communicate that uncertainty. The registry does not report such an interval or a formal analysis of the primary endpoint.

The entry also distinguishes a primary safety question from secondary tumor outcomes. It names RECIST response and progression-free survival but reports no analyses for either. This illustrates why outcome names in a registry are not evidence that results were obtained or reported, and why a safety narrative cannot stand in for an analysis of tumor outcomes.

Finally, the registry contains both a narrative statement about unexpected events and separate serious and other adverse-event summaries. These data types need careful interpretation: event categories, participant counts, and causal attribution are not interchangeable. The record does not explain the classification or attribution procedures, so a precise reconciliation is not possible.

The statistical lesson is not that the study’s findings settle a treatment question. Rather, it is that study design, endpoint definitions, analysis reporting, and uncertainty determine what readers can infer. Here, the record supports only a restricted descriptive reading and leaves important questions about analysis, follow-up, and termination unanswered.

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.

Enrollment was 3, and the study has no comparator group. This limits the descriptive breadth of the findings and prevents an internal treatment comparison. The open-label status and absence of masking are also relevant design features; the registry does not report whether outcome assessment included other safeguards against subjective classification.

The study is marked terminated, but the reason for termination is not reported in the registry record. The record does not say whether the study stopped for enrollment, operational, safety, or other reasons, and no such explanation should be inferred from its status.

The registry’s safety narrative is not accompanied by detailed participant-level records, event timing, laboratory values, event definitions, or uncertainty intervals. The serious and other adverse-event summaries do not provide causal attribution or explain whether the listed terms overlap within participants. The deaths field is not reported as a numeric value, although a serious-event term refers to death due to disease progression.

The primary endpoint has a stated timeframe, but the record does not report actual follow-up by participant, handling of incomplete observations, or analysis populations. It also does not report a statistical analysis plan, sample-size justification, interim-analysis procedures, or the reason for the planning assumptions described in the sample-size information. Those omissions prevent assessment of whether the analyses followed an intended plan.

Although tumor response measured by RECIST criteria and progression-free survival are listed as secondary outcomes, results and analysis methods for those outcomes are not reported. The registry also does not provide enough detail to determine how the secondary outcomes were defined or assessed. These limits are specific to what the registry record documents; they do not establish what may have been collected or analyzed outside the record.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Trial of Melaxin Cancer Vaccine Plus Bacillus Calmette-Guerin (BCG) to Treat Malignant Melanoma — NCT00671554.
• Registry record last updated 2014-06-04; status last verified 2014-05.
• Results first posted 2014-06-04.
• 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

The registry record describes a Phase I and Phase II, open-label, single-group study of Melaxin with BCG for Stage IV malignant melanoma. Allocation is listed as NA, masking as none, enrollment as 3, and status as terminated. The sponsor is Prisma Health-Upstate. The listed start is 2008-04, and primary completion is 2009-02.

The primary outcome is the number of participants with unexpected adverse events or unexpected laboratory results, from first vaccine to 18 months after the last injection. The registry narrative reports no unexpected or uncommon adverse events and says clinical laboratory results were within expected ranges. Tumor response measured by RECIST criteria and progression-free survival are listed as secondary outcomes, but outcome analyses are not reported in the registry record.

The adverse-event summary lists serious and other events among the 3 participants at risk. The record does not report a numerical death count, causal attribution, analysis populations, missing-data methods, or a statistical analysis plan. The reason for study termination is also not reported in the registry record. Overall, the entry offers a limited descriptive safety account without comparative or inferential outcome results.