Real-world Study of Dedifferentiated Liposarcoma Patients in China: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Real-world Study of Dedifferentiated Liposarcoma Patients in China (NCT06115681), a registered phase not stated study. The primary outcome is Cohort 1: Overall Survival.

2. Clinical Question

3. Trial Design

This is a completed real-world study of DDLPS patients in China, sponsored by Boehringer Ingelheim. The registry lists enrollment as 1390. That is the registered enrollment figure; the record does not clarify whether it represents the number included in each cohort or the number contributing to the primary analysis. The study start date is 2024-03-13, the primary completion date is 2024-05-15, and the status is COMPLETED.

The study describes two patient cohorts rather than randomized treatment groups. Cohort 1 includes patients with unresectable locally advanced or metastatic DDLPS who received at least one line of systemic antineoplastic treatment by the end of cohort identification. Its index date is the start of 1L treatment. Cohort 2 includes patients who had not initiated 1L systemic antineoplastic treatment by the end of cohort identification; its index date is the initial diagnosis date during that period. The cohort descriptions therefore use different index events, and the registry record cautions against treating the cohorts as like-for-like controls.

The brief summary describes the aim as characterizing demographic and clinical features, treatment patterns, and clinical outcomes associated with current real-world treatment. Secondary outcomes include survival after initial diagnosis; total number of treatment lines; treatment types for 1L, second-line (2L), and third-line (3L) therapy; and counts of patients classified as receiving concurrent steroid therapy, immunosuppressant therapy, or hormone replacement therapy.

The registry record does not report a randomized allocation method, and the allocation field is not stated. It also does not provide a complete description of cohort identification procedures or a full account of the Cohort 2 treatment history. Part of that cohort description is incomplete. These gaps limit how precisely the study population and the timing of all treatment classifications can be reconstructed from 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: Only patients diagnosed with DDLPS, who met all the inclusion criteria of having at least 2 documented clinical visits in the National Anti-Tumor Drug Surveillance System (NATDSS) on or after January 1, 2013, and were ≥18 years at initial diagnosis, were included.

Overall Study

Analysis population for the primary outcome, as defined in the registry: All patients diagnosed with dedifferentiated liposarcoma (DDLPS) who met the inclusion criteria of having at least 2 documented clinical visits in National Anti-Tumor Drug Surveillance System (NATDSS) in China on or after January 1, 2013 and aged ≥18 years at initial diagnosis. Only patients who received at least one line of systemic antineoplastic treatment by the end of cohort identification period were included in this analysis.

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: All patients diagnosed with dedifferentiated liposarcoma (DDLPS) who met the inclusion criteria of having at least 2 documented clinical visits in the National Anti-Tumor Drug Surveillance System (NATDSS) in China on or after January 1, 2013 and were aged ≥18 years at initial diagnosis. Eligible patients were identified from January 1, 2013 to December 31, 2021 and were followed from index date until death, end of study period (December 31, 2022), or loss to follow-up.

6. Primary Endpoint

The registry lists the primary outcome as Cohort 1: Overall Survival, measured over Time from index date until the earliest record of death or end of the study period, up to 10 years..

The primary endpoint is overall survival (OS) in Cohort 1, reported as a median in months with a 95% confidence interval. The registry definition describes OS as the interval from the index date to the earliest record of death or the end of the study period, up to 10 years. For a patient with a death record, the description defines OS using the death date minus the index date or the start day of each line of therapy. For a patient without a death record, OS is censored at the last activity date before the end of the study period.

In general, overall survival is a time-to-event measure. It incorporates both whether an event was recorded and how long the patient was observed before that event or censoring. Censoring means that the event time is not observed during the available follow-up; it is not the same as recording that the event never occurred. A median summarizes the time-to-event distribution by identifying the point at which the estimated survival experience reaches its midpoint. A confidence interval conveys uncertainty around that estimate.

OS is commonly used because death is a clearly defined event in a survival analysis and can summarize outcomes over follow-up. However, interpretation depends on a clearly specified time origin, event definition, follow-up, and handling of censoring. The registry wording refers both to the index date and to the start of each line of therapy, leaving the precise analysis time origin unclear. The median and confidence interval themselves are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry classifies the primary outcome as a time-to-event measure, with median OS and a 95% confidence interval. These labels describe the intended type of summary, but the numerical estimate and interval are not reported in the registry record. The record does not name the estimator, the confidence-interval procedure, or the software or analysis details used. No particular method should therefore be attributed to the study.

What does a median survival estimate describe?

In general, survival analysis studies the time until a defined event. A median survival estimate marks the point at which the estimated survival distribution reaches its midpoint. Unlike a simple average of observed times, it can account for patients whose follow-up ends before an event is recorded. Such patients are censored: their event time is not known, but their observed time contributes information up to the point follow-up ends. Censoring is informative only under assumptions about how loss of observation relates to event risk; the registry record does not describe how those assumptions were assessed.

A 95% confidence interval is an uncertainty interval around an estimated quantity. In general, a narrower interval indicates greater statistical precision under the method's assumptions, while a wider interval indicates less precision. It does not show the range of individual patient experiences, nor does it establish that every value inside it is equally plausible. Because neither the median nor the interval limits are reported, the precision of the Cohort 1 OS estimate cannot be evaluated.

Why are the time origin and censoring rules important?

A time-to-event result is tied to its time origin. In this study, Cohort 1's index date is defined as the start date of 1L treatment, while Cohort 2's index date is the date of initial diagnosis during cohort identification. The primary endpoint description also refers to the index date or the start day of each line of therapy. That wording does not fully clarify whether the primary analysis uses one common time origin or includes time measured from multiple treatment-line starts. This is important because changing the starting point changes the time being summarized and can alter which follow-up contributes to the estimate.

Censoring also needs a clear operational definition. The record says that patients without a death record are censored at the last activity date before the end of the study period. In general, this approach requires adequate information about when patients were last observed and how the end of observation is determined. The registry does not report the number censored, the extent of follow-up, or whether censoring differed across patient groups.

What can be inferred from a cohort comparison?

In general, observational cohorts describe associations in the populations recorded; they do not automatically isolate the causal effect of treatment. Patients receiving treatment and patients not yet receiving treatment can differ in clinical status, disease course, or time under observation. Statistical adjustment can sometimes account for measured differences, but it cannot guarantee that unmeasured differences are removed. The registry expressly says the cohorts are not a like-for-like control comparison, and it does not report an adjusted treatment effect or a complete statistical analysis plan.

The cohorts also begin follow-up at different events: treatment initiation for Cohort 1 and initial diagnosis for Cohort 2. This makes a simple contrast especially difficult to interpret. A valid comparison would need to define the target question, align time origins and eligibility as appropriate, and specify how confounding and follow-up are addressed. The registry record does not provide those details, so no comparative survival conclusion is warranted.

What is the role of hazard ratios?

A hazard ratio is a relative comparison of event rates over time under a specified model. In general, its interpretation depends on the populations compared, the time origin, covariate adjustment, and model assumptions. The record's methodological description says that only sponsor-reported hazard ratios are used, but no hazard ratio estimate is listed. No hazard ratio can therefore be quoted or interpreted for this analysis. A hazard ratio would not, by itself, replace the absolute time-to-event summary or resolve the cohort comparability concerns.

How should the planning assumptions be read?

The record describes planning illustrations based on an observed effect, with a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. It says the effect to detect is a stated fraction of the observed effect, and that time-to-event illustrations use a total event count under the Schoenfeld approach. In general, such calculations illustrate the sample size or event information that might be needed under chosen assumptions; they are not observed power and do not establish the power of the completed study.

No numerical planning estimates, observed effect, or sponsor hazard ratio are reported in the registry record. The study is not described as a randomized comparison, and equal allocation is an assumption for the planning illustration, not evidence that the observed cohorts had equal sizes. The planning assumptions consequently cannot be used to infer the precision of the actual OS result or to reconstruct a missing analysis.

10. Multiplicity, Missing Data and Analysis Populations

The study lists multiple outcomes. Alongside the primary OS outcome for Cohort 1, secondary outcomes cover survival after initial diagnosis, total treatment lines, therapy types across 1L, 2L, and 3L, and counts of patients classified as receiving concurrent steroid, immunosuppressant, or hormone replacement therapy. Considering several outcomes can raise a multiplicity issue: in general, testing many hypotheses can increase the chance of at least one apparently notable finding by chance. Whether and how to address this depends on the analysis objectives and prespecified plan.

The registry record does not report a multiplicity procedure, a hierarchy of confirmatory and exploratory outcomes, or a statistical analysis plan. It lists zero analyses, and it does not give numerical results for the primary or secondary outcomes. It is therefore not possible to determine whether any adjustment was planned or applied, or whether the listed outcomes were analyzed as descriptive measures only.

Information about missing data is also limited. The OS description states that patients without a death record are censored at their last activity date before the end of the study period. The record does not state how incomplete covariates, incomplete treatment records, or patients with no later activity were handled. It does not report censoring counts, missing-data frequencies, sensitivity analyses, or an assessment of whether missingness could be related to outcomes.

The primary population description is incomplete, and no cohort-specific analysis denominators are stated. The enrollment figure alone does not identify the analysis population for each outcome. These gaps matter because results can depend on who was eligible, who had usable records, and who remained observable. The registry record does not provide enough information to assess those populations or determine how missing information affected the analyses.

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

The registry record establishes that this completed study was designed to characterize DDLPS patients in China and describes OS in Cohort 1 as its primary outcome. It also specifies how the two cohorts are defined and lists several treatment-pattern and clinical-outcome measures. These details identify the intended scope of the study, but they are not a numerical report of its findings.

The primary median OS and its 95% confidence interval are not reported in the registry record. A reader therefore cannot determine the estimated survival time, how precise the estimate was, or whether follow-up supported a stable median. The record also does not report a sponsor hazard ratio. The listed death counts are not accompanied by denominators or an explanation linking them to the primary OS analysis, so they do not supply a substitute for the missing survival estimate.

The study's two cohorts should not be interpreted as treatment and control arms in a randomized trial. Their definitions differ in treatment status and index date, and the registry record warns that they are not like-for-like controls. Even if comparative results were available, differences between the groups would not by themselves show that treatment caused an outcome difference. Patient selection, disease characteristics, timing, follow-up, and other factors could contribute to an observed association.

In general, a real-world study can describe patterns in the data source and population it includes. It cannot automatically establish how well those records represent all patients with DDLPS in China or elsewhere. The registry record does not fully describe the cohort-identification process, analysis populations, missing-data handling, or adjustment strategy, which limits assessment of generalizability and potential bias.

The defensible conclusion is consequently about the study's stated aims and methods, not about treatment benefit or harm. The record documents a completed study with a registered enrollment figure and a defined primary survival outcome, but its numerical primary result and critical analysis details are not reported in the registry record.

12. Why This Trial Matters Statistically

This study is statistically relevant because it addresses outcomes and treatment patterns in a real-world DDLPS population in China, rather than describing a randomized treatment comparison. Such observational work can provide context about who is represented in a data system, what therapies are recorded, and how outcomes are defined in routine-care data. Its usefulness depends on transparent reporting of eligibility, time origins, event ascertainment, and follow-up.

The cohort definitions make time alignment particularly important. Cohort 1 begins at 1L treatment, while Cohort 2 begins at initial diagnosis. Those are different starting events, and the registry record cautions that the cohorts are not like-for-like controls. Statistical interpretation must preserve that distinction rather than treating cohort labels as randomized arms. Clear reporting of how observation time is defined is essential for understanding survival summaries.

The study also illustrates why a stated endpoint is not the same as a reported result. The record names median OS and a 95% confidence interval, but does not give either value. Without the estimate, interval, analysis population, and follow-up information, readers cannot assess magnitude or precision. This is an important reporting limitation regardless of the study's size or completion status.

Finally, the record lists several treatment-pattern and clinical outcomes, but does not provide their numerical results or a multiplicity plan. In general, broad descriptive aims can involve many measures, and readers need to know which analyses were prespecified and how uncertainty was handled. The available record makes the study question recognizable while leaving the empirical findings and much of their statistical context unavailable.

13. 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 main limitation for interpretation is that the registry record does not report the primary OS median or its 95% confidence interval. It also lists zero analyses and provides no numerical results for the listed secondary outcomes. Consequently, the statistical magnitude and precision of the study findings cannot be assessed from the registry record.

The primary population description is truncated after its eligibility criteria begin, so the full definition of the analysis population is not available. The description states that patients had at least 2 documented clinical visits in NATDSS on or after January 1, 2013, and were aged ≥18 years at initial diagnosis, but the remaining text is incomplete. The Cohort 2 description is also cut off. These omissions prevent a complete reconstruction of eligibility and cohort classification.

The endpoint description refers to OS from the index date or the start day of each line of therapy. This wording leaves uncertainty about the time origin used for the primary analysis. The record does specify censoring at the last activity date before the study period ends for patients without a death record, but does not report follow-up duration, numbers censored, or how censoring assumptions were evaluated.

The registry lists 1390 as enrollment but does not state cohort-specific denominators or the number contributing to each outcome. It records death counts of 175 in Cohort 1 and 423 in Cohort 2 without clarifying their relation to the primary endpoint analysis or giving denominators. These entries cannot establish event rates or replace the missing survival results.

Finally, the cohorts have different index dates and treatment status, and the registry explicitly says they are not like-for-like controls. No statistical adjustment strategy, multiplicity approach, or detailed missing-data plan is reported. These design and reporting limitations preclude a causal interpretation of any potential cohort contrast.

14. Related Tutorials

15. Related Statistical Calculators

16. Sources

• ClinicalTrials.gov: Real-world Study of Dedifferentiated Liposarcoma Patients in China — NCT06115681.
• Registry record last updated 2025-07-10; status last verified 2025-07.
• Results first posted 2025-07-10.
• 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.

17. Record Summary

The registry record describes a completed real-world study of DDLPS patients in China, sponsored by Boehringer Ingelheim. It lists enrollment as 1390, a start date of 2024-03-13, a primary completion date of 2024-05-15, and status COMPLETED.

The primary outcome is median OS in Cohort 1, measured in months with a 95% confidence interval. Cohort 1 is defined by receipt of at least one systemic antineoplastic treatment line by the end of cohort identification, with the 1L start date as index. Cohort 2 had not initiated 1L treatment by that point, with initial diagnosis as index. The record cautions that the cohorts are not like-for-like controls.

Secondary outcomes concern survival after diagnosis, treatment-line counts and types, and concurrent treatment classifications. The primary estimate, interval limits, sponsor hazard ratio, cohort-specific analysis denominators, and numerical secondary results are not reported in the registry record. The primary population description and part of the Cohort 2 description are incomplete.