Targeting Microenvironment and Cellular Immunity in Sarcomas Weekly Trabectedin Combined With Metronomic Cyclophosphamide: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Targeting Microenvironment and Cellular Immunity in Sarcomas Weekly Trabectedin Combined With Metronomic Cyclophosphamide (NCT02805725), a non-randomized phase 1, phase 2 study. The primary outcome is Phase I: Number of Patients Who Experienced Dose-Limiting Toxicities (DLTs).

2. Clinical Question

3. Trial Design

This was a completed phase I/II study sponsored by Institut Bergonié. The registry records a start date of 2015-12 and a primary completion date of 2020-04. Enrollment was 50. The study was non-randomized, had no masking, and used a single-group model. These features describe the overall study structure; they do not supply a randomized comparison of treatment against a control.

The phase I component examined four dose-escalation cohorts. Trabectedin was administered intravenously as a 3-hour infusion on days 1, 8 and 15, every 4 weeks. The cohort doses were 0.30 mg/m2, 0.40 mg/m2, 0.50 mg/m2 and 0.60 mg/m2. CP was described as twice daily, 50 mg x 2, on a week-on/week-off schedule. The record also describes a phase II regimen of trabectedin 0.50 mg/m2 with CP using the same stated administration schedule.

The stated purpose was to assess efficacy and safety after determining the maximum tolerated dose in phase I. The registry record does not report the dose-escalation decision rules, the maximum tolerated dose determination, or the rationale for selecting the phase II regimen. Those details should not be inferred from the cohort doses alone.

Because allocation was non-randomized and the study is described as single-group, the phase I cohorts do not constitute a randomized comparison. They were organized around different trabectedin doses, but the registry does not report a formal comparative analysis across doses. Phase II used one listed regimen, with no concurrent randomized comparator recorded. This design can describe observed outcomes in the treated groups, but causal comparisons with an untreated or alternative-treatment group are not available from this 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.

Phase I (Dose escalation part)

Reasons for not completing this period:

Phase II

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: 4 patients were not evaluable for MTD/DLT assessment.

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 Phase I: Number of Patients Who Experienced Dose-Limiting Toxicities (DLTs), measured over During the first cycle (28 days).

The primary phase I endpoint was the number of participants who experienced dose-limiting toxicities during the first cycle, a 28-day period. It is a count endpoint: the intended quantity is how many participants meet the registry’s DLT definition, rather than a measurement of symptom severity averaged across all participants.

The registry defines a DLT as a treatment-related adverse event considered at least possibly related, assessed using CTCAE V4.0, that occurs during the first cycle and meets specified severity or duration criteria. These include any grade-4 toxicity, with a stated exception for vomiting without maximal symptomatic or prophylactic treatment; certain grade-3 non-haematological or haematological toxicities lasting > 7 days; grade 4 neutropenia with fever; and grade > 2 thrombocytopenia with bleeding. The definition excludes events attributed to disease, disease progression, inter-current illness or concomitant medications.

In a dose-escalation study, a DLT count is used to describe serious early toxicity at the tested dose levels. It can inform dose selection, but the count alone does not describe later toxicity, overall tolerability, or efficacy. The registry record does not report the observed DLT counts, so the primary endpoint cannot be interpreted quantitatively from the posted results.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The statistical information in the registry supports a distinction between the registered study design, descriptive counts, and analyses that would require reported outcome results. The study is listed as phase I/II, non-randomized, unmasked and single-group. The primary outcome is a participant count for DLTs in phase I; named secondary outcomes include objective response in phase I and median overall survival and median progression-free survival in phase II. No analysis results are reported in the registry record.

What does a DLT count describe?

A binary event outcome classifies each evaluable participant as having or not having the specified event during a defined observation period. Summing the participants who meet the definition gives a count. For this study, the event is a DLT and the observation period is the first cycle, defined as 28 days. The count is interpretable only alongside the number of participants assessed and the rule used to decide whether an event qualifies.

The registry’s DLT criteria combine event severity, duration, attribution and timing. For example, the definition includes specified grade-3 toxicities lasting > 7 days and certain grade-4 events. Attribution also matters: events must be considered at least possibly treatment-related, and the definition excludes events attributed to listed alternative causes. Applying a complex definition consistently requires clinical classification; a raw count does not show how individual events were adjudicated.

The record notes 4 patients who were not evaluable for DLT assessment because they did not complete the assessment period, with 2 in cohort 3 and 2 in cohort 4. Those exclusions matter because a participant without a completed assessment period may not have the same opportunity for an event to be observed as someone who completed it. The registry does not report how those participants were handled in any endpoint analysis, and it does not post the cohort-specific DLT counts.

How should uncertainty be understood?

General statistical explanation: a confidence interval expresses the range of values reasonably compatible with observed data under a specified statistical method. For a binary event, an interval around an event proportion can communicate uncertainty that a count alone cannot. Small groups generally yield less precise estimates than larger groups, but the exact precision depends on the observed data and method. No confidence intervals for DLTs are reported in the registry record, and there is not enough posted endpoint information to reconstruct them without making assumptions.

A p-value, when used, addresses how compatible data are with a specified null model; it is not the probability that a treatment works or that the null hypothesis is true. The registry record reports 0 analyses and supplies no primary endpoint p-value. There is therefore no reported statistical test to interpret. General statistical explanation: tests are particularly difficult to interpret without a clearly stated comparison, analysis population, and prespecified model. The registry does not report those details for a DLT analysis.

Why does the design affect inference?

Random allocation can help balance measured and unmeasured participant characteristics between treatment groups. This study is listed as non-randomized and single-group, so that protection against selection differences is absent. The phase I dose cohorts differ by the planned trabectedin dose, but the record does not describe random assignment among those doses. A difference in observed outcomes across cohorts could reflect factors other than dose, including differences in the participants entering each cohort.

General statistical explanation: a control group gives a contemporaneous reference for outcomes under another condition. Without one, a single-group study can describe events observed among its participants but cannot directly estimate a contrast against a randomized control. The registry explicitly notes that its arms are not a like-for-like control comparison. Accordingly, no comparative effect should be inferred from the cohort summaries.

What about planning illustrations?

The registry’s stated planning assumptions describe an illustration based on an observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation; the effect to detect is described as a stated fraction of the observed effect. These are not reported as the trial’s own design justification. Any participant counts calculated under such assumptions are planning illustrations, not observed enrollment requirements for this trial and not observed power. The record does not report the protocol’s sample-size justification.

General statistical explanation: power is a design property calculated under assumptions about an effect, variability, sample size and analysis method. It is not a measure of whether a completed study has demonstrated an effect. Planning calculations based on an observed effect are especially dependent on that effect estimate and should not be treated as an independent confirmation. The registry provides no inferential outcome estimates from which to assess the precision of the clinical results.

Why are over-dispersion and model choice relevant in general?

General statistical explanation: over-dispersion occurs when observed variability exceeds what a chosen probability model expects. For count or event data, ignoring over-dispersion can make uncertainty appear smaller than it is. Model choice also depends on the outcome structure, follow-up, and comparison being estimated. This record does not report a fitted model or enough endpoint results to assess over-dispersion, and no model should be inferred from the study title or design description.

For time-to-event outcomes such as overall survival and progression-free survival, medians summarize the time at which an estimated event distribution reaches its midpoint, subject to the analysis method and follow-up. General statistical explanation: a median alone does not convey the full shape of the time-to-event distribution or the precision of the estimate; censoring and follow-up also affect interpretation. The registry lists these outcomes but does not report their estimates, confidence intervals, event counts, or analysis details.

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("Cohort 1: Trabectedin 0.30 mg/m2 IV + CP", "Cohort 2: Trabectedin 0.40 mg/m2 IV + CP", "Cohort 3: Trabectedin 0.50 mg/m2 IV + CP", "Cohort 4: Trabectedin 0.60 mg/m2 IV + CP")
n      <- c(3, 3, 5, 5)
events <- c(0, 0, 0, 2)
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  Cohort 1: Trabectedin 0.30 mg/m2 IV + CP  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Cohort 2: Trabectedin 0.40 mg/m2 IV + CP  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  Cohort 3: Trabectedin 0.50 mg/m2 IV + CP  5  0  0.0  0.0  52.2  0.0  43.4  0.4  45.9
ARM  Cohort 4: Trabectedin 0.60 mg/m2 IV + CP  5  2  40.0  5.3  85.3  11.8  76.9  11.8  77.7

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 6 outcome measures. The named measures include the phase I DLT count, phase I objective response, phase II median overall survival, and phase II median progression-free survival. The registry record does not report the numerical results for these outcomes, and it does not provide detailed definitions or results for every registered measure in the registry record recorded. It also lists 0 analyses.

Multiplicity arises when several outcomes or statistical tests are examined. General statistical explanation: testing many outcomes can increase the chance of at least one apparently notable result arising by chance, so a statistical plan may specify a primary outcome, a testing hierarchy, or an adjustment approach. The registry identifies a primary phase I endpoint, but multiplicity procedures, any hierarchy among outcomes, and any interim analyses are not reported in the registry record. No particular approach should be assumed.

Missing outcome information limits interpretation. Four patients were not evaluable for DLT evaluation because they did not complete the assessment period: 2 in cohort 3 and 2 in cohort 4. The record does not give the reasons for non-completion, report the DLT results for evaluable participants, or describe how non-evaluable participants were handled in an analysis. It does not report missing-data procedures for the other outcomes.

The analysis populations are not reported in the registry record. In particular, the record does not specify which participants were included in each outcome analysis, how follow-up was defined, or how participants with incomplete assessments were treated for secondary endpoints. These omissions prevent assessment of whether the analyses followed an intention-to-treat, evaluable-participant, or other approach.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: chest pain (33.3%), diarrhea (33.3%), anemia (33.3%), pleural effusion (20.0%), febrile pancytopenia (20.0%).

The registry lists safety counts separately for each cohort, with participants at risk and participants affected. For serious adverse events, the recorded figures are 1 of 3 in cohort 1, 1 of 3 in cohort 2, 2 of 5 in cohort 3, 2 of 5 in cohort 4, and 20 of 30 in phase II. For other adverse events, the recorded figures are 3 of 3, 3 of 3, 5 of 5, 5 of 5, and 30 of 30, respectively. The death counts listed for those groups are 3, 2, 4, 4 and 20, respectively.

These entries are descriptive, and the registry record does not report an attribution linking each event or death to study treatment. It also does not specify a shared observation period for these safety figures. The groups are not randomized control arms, so their counts should not be treated as direct comparative evidence. Differences in group size also affect how raw counts read; counts alone do not provide a standardized risk estimate or its precision.

The registry’s listed serious-event terms include Chest pain, Diarrhea and Anemia, each at 33.3%, and Pleural effusion, Febrile pancytopenia and Worsening of GGT increased, each at 20.0%. The denominator for these percentages is not reported in the registry record, so they should be read as the listed percentages rather than converted into event counts or compared across cohorts.

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

The record establishes that a phase I/II study evaluated trabectedin with metronomic cyclophosphamide in advanced pretreated soft-tissue sarcomas. It describes the phase I dose cohorts, the phase II regimen, the DLT definition, and selected safety counts. It does not report the primary DLT count, efficacy estimates, or inferential analyses. As a result, the posted information does not quantify the primary safety endpoint or the registered efficacy outcomes.

The absence of posted endpoint estimates is not evidence that no events occurred, nor does it indicate that an outcome was favorable or unfavorable. It means the registry record does not provide the numerical results needed to assess those outcomes. The study’s listed deaths and adverse-event counts are descriptive safety entries; without clear event attribution and follow-up details, they should not be interpreted as proving that a study drug caused or did not cause an event.

General statistical explanation: even a precisely estimated outcome in a single-group study would describe the participants treated in that study, while a causal treatment effect requires a credible comparison. Here, the design is explicitly non-randomized and single-group, and the registry cautions that the arms are not like-for-like controls. The dose cohorts therefore do not supply randomized evidence that changing the trabectedin dose caused any observed difference.

The registry’s phase I objective response and phase II survival outcomes could, if reported with appropriate methods and uncertainty, describe outcomes among participants. Their results are not reported in the registry record. The record also does not report a protocol sample-size rationale, interim analysis plan, or complete statistical analysis plan. Those missing details limit evaluation of how the study was designed to answer its clinical questions.

Overall, the record documents the trial structure and some safety observations, but it does not allow a statistical conclusion about the primary DLT outcome, response, survival, or comparative treatment effects. It should not be read as establishing that the regimen is effective or ineffective.

14. Why This Trial Matters Statistically

This trial is statistically notable because its primary phase I question is tied to a defined early observation window and a clinically specified toxicity rule. The endpoint is a participant count during the first 28 days, and the criteria combine severity, duration, attribution and timing. Such definitions illustrate why safety outcomes require both a clear event rule and transparent reporting of who was evaluable.

The study also shows the interpretive limits of dose cohorts without randomization. Four trabectedin dose levels were used in phase I, followed by a phase II regimen at 0.50 mg/m2 with CP. These groups may describe experience at the studied doses, but the non-randomized single-group design does not make them equivalent comparison groups. Statistical interpretation must keep that design distinction in view.

The registry’s four non-evaluable patients for DLT assessment are important because completeness of the assessment period affects what can be known about early toxicity. The record gives their cohort distribution, but not the primary DLT counts or the handling of those participants in an analysis. That makes transparent reporting of denominators and analysis populations central to interpreting the endpoint.

Finally, the record illustrates the difference between registration and results reporting. It identifies primary and secondary questions and lists safety observations, but it does not report the numerical primary endpoint result or analyses. Statistical judgment depends not only on design labels, but on event counts, denominators, uncertainty, analysis methods and missing-data handling. Several of these are not reported in the registry record, so conclusions must remain correspondingly limited.

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 principal limitation for statistical interpretation is that the registry record does not report the primary endpoint result: the number of participants with DLTs during the first 28-day cycle. It notes 4 patients who were not evaluable for that assessment, but does not give cohort-specific DLT counts or an analysis population. The primary question therefore cannot be answered from the posted results.

The record does not provide numerical results for the named secondary outcomes, including objective response, median overall survival, or median progression-free survival. It also reports 0 analyses. Confidence intervals, p-values, time-to-event event counts, and model details are not reported in the registry record.

The study was non-randomized, unmasked, and single-group. There is no like-for-like randomized control comparison. The phase I cohorts used different trabectedin doses, but the record does not report random allocation or a formal comparative method. The cohort summaries cannot establish causal dose effects.

Further design and analysis details are not reported in the registry record, including the sample-size justification, dose-escalation decision rules, interim analyses, multiplicity procedures, missing-data methods, and analysis populations. The safety entries do not specify a common observation period or establish event attribution. The denominator behind the listed serious-event percentages is also not reported. These limitations leave the posted safety data descriptive and restrict assessment of both precision and comparability.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Targeting Microenvironment and Cellular Immunity in Sarcomas Weekly Trabectedin Combined With Metronomic Cyclophosphamide — NCT02805725.
• Registry record last updated 2026-01-14; status last verified 2026-01.
• Results first posted 2023-03-23.
• 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

NCT02805725 is a completed, non-randomized, unmasked, single-group phase I/II study sponsored by Institut Bergonié, with enrollment of 50. The registry records a start date of 2015-12 and primary completion in 2020-04. Participants had advanced pretreated soft-tissue sarcomas and received trabectedin with metronomic cyclophosphamide.

Phase I examined trabectedin doses of 0.30 mg/m2, 0.40 mg/m2, 0.50 mg/m2 and 0.60 mg/m2 with CP. The phase II regimen used trabectedin 0.50 mg/m2 with CP. The primary phase I endpoint was the number of participants with DLTs during the first 28-day cycle. Four patients were not evaluable for that assessment.

The registry lists 6 outcome measures and 0 analyses. It does not report the primary DLT count or numerical efficacy results. Safety counts are listed by cohort, but event attribution and a common observation period are not reported.