Fotemustine and Dacarbazine Versus Dacarbazine +/- Alpha Interferon in Advanced Malignant Melanoma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Fotemustine and Dacarbazine Versus Dacarbazine +/- Alpha Interferon in Advanced Malignant Melanoma (NCT01359956), a randomized phase 3 study. The primary outcome is Overall Survival (OS).

2. Clinical Question

3. Trial Design

This was a randomized, open-label Phase III study with a factorial model. The registry lists 269 participants as enrolled and describes the study as evaluating chemotherapy regimens with and without interferon in advanced or recurrent melanoma. Masking was listed as none. The study is marked completed, with a start date of 2002-04 and primary completion in 2011-02.

The four arm descriptions distinguish combination chemotherapy from single-agent dacarbazine, and treatment with interferon from treatment without interferon. The combination chemotherapy consisted of fotemustine and dacarbazine. Interferon alfa-2b was the additional drug in the interferon arms. This factorial structure can be used to ask about the chemotherapy regimen, the addition of interferon, and potentially their joint pattern. The registry record does not report a formal factorial analysis or explain how those effects were to be estimated.

The primary outcome was overall survival, with a timeframe listed as 24 months. The record defines the outcome from randomization to death from any cause, or last follow-up for living participants. Secondary outcomes include progression-free survival, overall response rate, and treatment-related toxicity, although the posted outcome detail is not equally complete for each outcome.

Although randomized allocation is recorded, details about sequence generation, allocation concealment, eligibility criteria, treatment schedules, and the intended analysis population are not reported in the registry record. The record also does not provide a sample-size justification or a full statistical analysis plan. Those omissions limit how fully the design and analysis can be reconstructed from the registry information alone.

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:

Analysis population for the primary outcome, as defined in the registry: Nine patients were lost to follow-up immediately after randomization and, in particular, for FDI + FD 132, instead of 136 randomized (4 lost pts.), for DI + D, 128 instead of 133 (5 lost pts.), for FD1 + DI 126, instead of 131 (5 lost pts.), 134 instead of 138 (4 lost pts.)

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 Overall Survival (OS), measured over 24 months.

Overall survival measures the time from randomization until death from any cause. For participants still living, the registry definition uses the date of last follow-up. This outcome includes deaths regardless of their recorded cause, so it captures survival over the follow-up period rather than response to treatment or a specific disease event.

Time-to-event outcomes are useful when participants can have different lengths of follow-up or when some remain alive at the end of observation. A participant who has not had the event by their last follow-up contributes information up to that point, rather than being treated as if the event occurred. This is called censoring. The registry says that overall-survival curves were estimated with the Kaplan-Meier method and that treatments were compared with a two-sided log-rank test.

The reported primary endpoint is median overall survival, in months, with lower and upper bounds shown for each group. A median describes the time by which half the estimated group has experienced the event; it does not describe every participant’s survival time or establish a causal difference between groups on its own. The registry lists a timeframe of 24 months, but does not explain whether this refers to a specific assessment point or the intended duration of follow-up. It also does not report a hazard ratio or a sponsor-reported test result for the primary outcome.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

For overall survival, the registry describes Kaplan-Meier estimation and a two-sided log-rank test. These are standard general methods for time-to-event data, but the record does not include the numerical result of the stated comparison. The posted medians are summaries of the estimated survival experience; without the sponsor’s comparison result, they should not be interpreted as a completed test of whether groups differ.

What does Kaplan-Meier estimation do?

In general, Kaplan-Meier estimation uses the observed event times and the follow-up information available for participants who have not yet experienced the event. It updates the estimated probability of remaining event-free as events occur. Participants who are alive at their last follow-up are censored at that time: their later outcome is unknown, but their observed follow-up still contributes information. The method relies on assumptions about censoring, including that censoring does not systematically conceal different event risks in a way that invalidates the estimate. The registry does not report diagnostics or details about those assumptions.

What does the log-rank test assess?

In general, a two-sided log-rank test compares the event patterns over follow-up between groups. “Two-sided” means that the test allows for evidence in either direction, rather than considering only a prespecified direction of difference. A test result is usually interpreted alongside an effect estimate and its uncertainty. The registry record states that this test was used, but does not report its statistic or p-value for overall survival. A method named in the record is not the same as a posted result of that method.

How should medians and bounds be read?

A median time-to-event estimate is a single point on an estimated survival curve. It is not an average, and it does not indicate how long each participant lived. The posted lower and upper bounds accompany each median, but the primary outcome description calls the dispersion “Full Range” and does not identify the bounds as confidence limits. General interpretation of a confidence interval is that it represents uncertainty around an estimate under a specified method; that interpretation should not be assigned to these bounds without an explicit label. The registry record does not describe their calculation.

What is special about the factorial structure?

In general, a factorial design organizes treatment assignments around more than one intervention factor. In this study, the arm descriptions distinguish the chemotherapy regimen and whether interferon was included. A factorial analysis might estimate an effect associated with each factor and examine whether the effect of one depends on the other. Such conclusions require an analysis that reflects the design and its assumptions. The record does not report a formal factorial model, interaction assessment, or prespecified contrast, so no such effect should be inferred solely from the arm labels.

How should the response-rate calculations be interpreted?

The registry posts overall-response counts and unadjusted comparisons for selected pairs of groups. A risk difference is a subtraction of the response proportions between two groups, with positive values indicating a larger proportion in the group designated as the comparison and negative values indicating a smaller proportion, according to the calculation’s ordering. In general, its confidence interval describes the range of effects compatible with the data under the chosen method. The posted intervals for the response-rate comparisons include values on both sides of no difference. The registry lists no sponsor-reported analyses for this outcome, so the calculations should not be mistaken for the trial’s own reported inferential analysis.

The sample-size figures are planning illustrations based on the observed response-rate effects, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. They illustrate how required group size changes when planning for 100%, 75%, or 50% of the observed effect. They are not the trial’s original design justification, do not describe the number enrolled, and are not observed power. The trial’s own sample-size assumptions are not reported in the registry record.

Over-dispersion is a general concern when outcomes vary more than a model assumes. If a model underestimates that variation, uncertainty can appear too small and tests can be too optimistic. The record does not identify an over-dispersion model or provide information showing whether it was assessed for any outcome; it is therefore a general modeling consideration, not a reported feature of this trial’s analysis.

10. Multiplicity, Missing Data and Analysis Populations

The registry record lists one primary outcome and several secondary outcomes, but does not report a multiplicity strategy. When a study evaluates multiple outcomes or treatment comparisons, general statistical practice considers how repeated testing affects the chance of obtaining apparently notable results by chance. Without a prespecified testing hierarchy or adjustment plan, secondary comparisons should be interpreted cautiously. The record does not report whether a hierarchy, adjustment, or other multiplicity procedure was planned or used.

Missing follow-up is specifically mentioned for overall survival. The population description states that nine participants were lost to follow-up immediately after randomization and then gives group-specific counts that differ from the stated randomized counts. The description’s wording is difficult to reconcile with the overall statement, so the exact accounting cannot be established from the registry record. It does not provide a complete participant flow, reasons for loss, or details about how missing follow-up was handled in the analysis.

The record reports outcome group counts of 132, 128, 126, and 134, while total enrollment is listed as 269. The relationship between those outcome counts and the total enrollment is not explained clearly. It is therefore not possible to determine from the record whether the group counts reflect a particular factorial grouping, repeated representation, or another reporting convention. The intended analysis population, such as intention-to-treat or per-protocol, is not reported in the registry record.

For time-to-event analysis, censoring at last known follow-up is part of the endpoint definition for living participants. General interpretation depends on the completeness and timing of follow-up and on the assumptions underlying censoring. The registry does not report follow-up distributions, the number of deaths, detailed censoring information, or sensitivity analyses. It also does not report how missing data were handled for the secondary outcomes.

11. 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: neutropenia (40.3%), platelets (34.3%), anaemia (19.4%).

The registry lists serious adverse event counts by arm as 14 of 62 at risk in combination chemotherapy without interferon, 27 of 67 in combination chemotherapy with interferon, 8 of 71 with single-agent dacarbazine without interferon, and 10 of 52 with single-agent dacarbazine plus interferon. The corresponding counts for other adverse events are 12 of 62, 25 of 67, 16 of 71, and 13 of 52. These are counts with the listed at-risk denominators; they should not be interpreted as rates unless the registry defines the measure that way.

The listed leading serious-event terms are Neutropenia at 40.3%, Platelets at 34.3%, and Anaemia at 19.4%. The record does not clarify in this outcome entry whether these percentages refer to participants, events, or another denominator. The groups also differ in their listed at-risk counts, so raw counts alone do not support a fair comparison of safety burden.

No death counts are posted in the adverse-event groups. The registry’s safety information should therefore be read as a limited description of reported adverse events, not a complete comparison of all harms or a basis for conclusions about the relative safety of the regimens.

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

The posted results describe overall survival medians for four treatment groups and response counts for selected comparisons. The survival medians are relatively close in scale, and their reported lower and upper bounds overlap. The registry describes Kaplan-Meier estimation and a two-sided log-rank test but does not post the test result, hazard ratio, or a sponsor-reported treatment comparison for overall survival. The primary outcome therefore cannot be interpreted as establishing a difference or equivalence among regimens from the available registry information.

General statistical interpretation distinguishes an observed estimate from the uncertainty around it. A point estimate is compatible with a range of possible underlying effects, especially when the data are limited or uncertainty is substantial. A non-significant test, if one were reported, would not by itself establish equivalence; an equivalence conclusion would require a prespecified margin and an analysis designed for that question. Neither an equivalence margin nor an equivalence analysis is reported in the registry record.

The response-rate comparisons are unadjusted calculations, and no sponsor analyses are listed for them. They offer descriptive context but do not answer the primary survival question. The four-arm factorial structure may invite questions about the separate contributions of chemotherapy regimen and interferon, but the registry does not report the relevant factorial estimates or interaction analysis. Comparing arm-level medians informally cannot substitute for those analyses.

These findings do not establish that a treatment works or does not work. They describe what the registry posts and the precision or omissions that affect interpretation. Important details, including the primary comparison result, the analysis population, the sample-size justification, and fuller missing-data accounting, are not reported in the registry record. Conclusions should remain limited to the available descriptive summaries and stated methods.

13. Why This Trial Matters Statistically

This study is statistically notable because its factorial structure raises questions about more than a simple comparison of two treatments. The arm descriptions vary both the chemotherapy regimen and the addition of interferon. A clear analysis would need to connect the estimand, or treatment effect being targeted, to those factors and to the intended comparisons. The registry identifies a factorial model but does not report how the factors were analyzed, whether their effects were considered jointly, or whether an interaction was assessed.

The primary outcome is time-to-event, which requires careful handling of follow-up and censoring rather than a simple comparison of event counts. The registry names standard general methods for estimating survival curves and comparing groups, but leaves out the numerical test result and effect estimate. This illustrates why knowing the method alone is not enough to judge the strength or precision of a treatment comparison.

The record also shows why denominator clarity matters. It lists overall enrollment and separate outcome group counts, along with a difficult-to-reconcile description of losses after randomization. Without a transparent participant flow and stated analysis population, readers cannot fully connect randomized assignment, follow-up, and the analyzed outcomes.

Finally, secondary comparisons are presented without sponsor-reported analyses or a stated multiplicity plan. Their role is therefore primarily descriptive. Together, these features make the registry a useful example of how trial design, estimand definition, missing-data reporting, and complete statistical results all shape interpretation, even when outcome summaries are available.

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.

Several limitations are specific to the available record. The primary outcome description names Kaplan-Meier estimation and a two-sided log-rank test but does not provide the test result, hazard ratio, or sponsor-reported comparison. The lower and upper bounds accompanying the medians are not clearly identified as confidence intervals, which restricts their interpretation as measures of precision.

The factorial design is recorded, but no analysis of the chemotherapy factor, interferon factor, or their interaction is reported. The arm labels and outcome group counts do not make the intended contrast structure fully clear. The registry lists enrollment of 269 and outcome counts of 132, 128, 126, and 134; it does not explain how those counts relate to total enrollment. Its loss-to-follow-up narrative is also difficult to reconcile with the accompanying group details.

The registry does not report the sample-size justification, the analysis population, randomization procedures beyond randomized allocation, or a complete statistical analysis plan. It does not report how missing data were handled for outcomes other than the general survival definition. Secondary outcome information is incomplete, and the posted response comparisons have no sponsor-reported analyses. A multiplicity strategy is not reported.

The safety entry gives counts and leading serious-event terms but does not clearly specify the denominator for the listed percentages or include death counts. These limitations concern what can be established from the registry record; they do not establish that the underlying trial lacked further analyses or documentation.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Fotemustine and Dacarbazine Versus Dacarbazine +/- Alpha Interferon in Advanced Malignant Melanoma — NCT01359956.
• Registry record last updated 2023-04-11; status last verified 2020-09.
• Results first posted 2020-11-13.
• 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

The registry record describes a completed randomized, open-label Phase III factorial study of chemotherapy regimens with and without interferon in advanced or recurrent melanoma. Enrollment is listed as 269. The primary outcome is median overall survival, measured from randomization to death from any cause or last follow-up for living participants. Kaplan-Meier estimation and a two-sided log-rank test are named as the primary analysis methods.

Four outcome groups have posted median survival estimates, but no sponsor-reported primary comparison result is included. Overall response rate is also reported with counts and selected unadjusted comparisons; no sponsor analyses are listed for those comparisons. Progression-free survival and treatment-related toxicity are identified as secondary outcomes, while the available safety details are limited. The record does not report the sample-size justification, analysis population, multiplicity strategy, or a full account resolving the stated follow-up losses and group counts.