Chemoembolization in Treating Patients With Primary Liver Cancer or Metastases to the Liver: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Chemoembolization in Treating Patients With Primary Liver Cancer or Metastases to the Liver (NCT00003907), a non-randomized phase 2 study. The primary outcome is Time to Progression.

2. Clinical Question

The trial’s stated purpose was to study chemoembolization in patients with these liver cancers. Tumor response and overall survival are listed as secondary outcomes. The registry record does not report results for those outcomes either. The available design description supports explaining what was intended to be measured, but not judging the observed effect or comparing treatment outcomes between the disease groups.

3. Trial Design

This was a phase II, non-randomized, parallel trial sponsored by the Eastern Cooperative Oncology Group. The registry lists enrollment of 50 and a completed status. The recorded start date is 1999-12-15, and primary completion is listed as 2011-05. These are registry fields; they do not by themselves establish the duration of follow-up for each participant.

The record describes separate disease groups: hepatocellular carcinoma and neuroendocrine hepatic metastases. Both group descriptions specify chemoemulsion containing doxorubicin, mitomycin and cisplatin, followed by embolization. The procedure was to be repeated separately for each involved lobe, beginning within 8 weeks of the last lobar chemoembolization. The trial is listed as having no masking. Allocation was non-randomized, so the registry explicitly cautions that the groups are not a like-for-like control comparison.

The primary outcome was time to progression, assessed every 3 months for 2 years, then every 6 months for 3 year. Tumor response and overall survival are listed as secondary outcomes. Although the record identifies the intended outcomes and assessment schedule, it does not report numerical outcome results. The analysis details, including any adjustment for differences between disease groups, are not reported in the registry record.

The primary outcome population is described as all eligible and treated patients. The registry record does not provide further details about eligibility counts, exclusions after enrollment, or the number included in an analysis. It also does not report a protocol-based sample-size justification or a statistical analysis plan. The design should therefore be read as a non-randomized phase II study with a defined time-to-event endpoint, rather than as evidence from a randomized treatment comparison.

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: All eligible and treated patients

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 Time to Progression, measured over Assessed every 3 months for 2 years, then every 6 months for 3 year..

The primary endpoint, time to progression, measures the interval from embolization until documented disease progression. The record defines progression as a significant increase in lesion size or the appearance of new metastatic lesions. Its measurement criteria include at least a 25% increase in the area of malignant lesions greater than 2 cm² or in the sum of lesion products at an organ site; at least a 50% increase in the product of diameters when only one lesion is measurable and it was less than or equal to 2 cm² at treatment initiation; at least a 25% increase in the sum of liver measurements below the costal margins and xyphoid; or new malignant lesions.

In general, a time-to-event endpoint can incorporate both whether an event occurred and how long it took to occur. The registry specifies that patients without documented progression were censored at the time of the last documented disease evaluation or the last treatment ended, whichever was more recent. Censoring means that the event time is not observed during the available follow-up; it does not establish that progression never occurred.

The record gives a median as the summary measure and a 90% confidence interval as the stated measure of uncertainty. In general, a confidence interval conveys uncertainty around an estimated quantity. However, the median and interval values are not reported in the registry record. The endpoint definition and planned assessment schedule are available, but the reported result needed to describe the observed progression time is not.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a median time-to-event measure?

In general, time-to-event analysis concerns the time from a defined starting point until a specified event. Here, the registry defines the starting point as embolization and the event as documented disease progression. A median is a summary of the distribution of event times, rather than an average of all observed times. It can be used when some participants have incomplete follow-up, provided the analysis method accounts for censoring. The record states that the reported parameter was the median, but it does not report the median value.

What does censoring mean?

In general, censoring records that the event time is not known beyond a participant’s available observation. The registry says that patients without documented progression were censored at the more recent of the last documented disease evaluation or the last treatment ended. A censored observation contributes information up to its stated censoring point, but it does not mean the event was absent indefinitely. The record does not describe how many patients were censored, the distribution of follow-up, or any additional handling of incomplete observations; those details are not reported in the registry record.

How should the confidence interval be understood?

In general, a confidence interval communicates statistical uncertainty around an estimate. Its width is affected by the amount and pattern of available information and by the analysis method. The registry identifies a 90% confidence interval for the primary median. That label alone does not reveal the interval’s limits or whether the estimate is precise. Since the actual interval is not reported in the registry record, its width and the precision of the primary estimate cannot be evaluated.

Why does the non-randomized design matter?

In general, random assignment can help balance known and unknown participant characteristics between groups. A non-randomized design does not provide that protection. In this trial, the groups are distinguished by disease type, and the record explicitly says the arms are not a like-for-like control comparison. Therefore, a difference between group outcomes, if one were reported, would not on its own establish that the intervention caused the difference. The registry record does not report adjusted comparisons or a statistical plan for addressing group differences.

What can be learned from the analysis entries?

The record lists three outcome measures and no analysis records. This does not furnish estimates for the primary outcome or the listed secondary outcomes. In general, an outcome definition specifies what is measured, whereas an analysis result reports what the data estimate. The record provides the former for time to progression, but not the latter. A numerical effect, uncertainty interval, or inferential test cannot be described without reported results.

What are planning sample sizes?

In general, a sample-size calculation is a planning exercise based on assumptions about a future study; it is not an observed treatment effect and is not a measure of observed power. The registry’s planning note says its illustrations are computed from the observed effect rather than the trial’s own design, using a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. It says the target effect is a stated fraction of the observed effect, and that time-to-event illustrations give the total number of events using the Schoenfeld method; other illustrations give participants per arm. These are planning assumptions, not evidence that the trial enrolled or analyzed those planned numbers. The registry record does not provide numerical planning outputs or the underlying observed effect in the registry record described.

What does over-dispersion mean in general?

For count outcomes, over-dispersion means that observed variability is greater than a model expects under its assumptions. If ignored, uncertainty estimates can be too small, making results appear more precise than warranted. This is a general statistical concept, not a method identified for this trial. The registry does not report a count-model analysis or a method for handling over-dispersion, so it should not be inferred that either was used.

Why is the statistical plan important?

In general, a statistical plan describes how outcomes will be estimated, how uncertainty will be quantified, and how analysis decisions will be made. Such details help readers distinguish prespecified analyses from choices made after results are known. The registry record does not report the statistical plan, model details, or interim-analysis procedures. No particular method, adjustment, or interim review should be inferred from the endpoint label alone.

10. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when several outcomes, comparisons, or statistical tests are considered. In general, examining multiple results can increase the chance that at least one appears notable by chance; a prespecified strategy can clarify how such results should be interpreted. This record lists time to progression as the primary outcome and tumor response and overall survival as secondary outcomes. It lists three outcome measures, but no analysis records. Adjustment for multiplicity, test hierarchy, or a prespecified strategy for interpreting the outcomes is not reported in the registry record.

The primary outcome definition describes censoring for patients without documented progression: censoring occurred at the more recent of the last documented disease evaluation or the last treatment ended. In general, censoring is distinct from treating an observation as if the event occurred or as if follow-up continued indefinitely. The record does not report how many patients were censored, how much follow-up was available, or whether other missing-data procedures were used. Additional missing-data methods are not reported in the registry record.

The primary population is described as all eligible and treated patients. The registry enrollment field is 50, but the number eligible and treated, and the number included in each outcome analysis, are not reported in the registry record. The record does not describe exclusions, analysis-population rules beyond the primary population statement, or sensitivity analyses. These omissions limit assessment of how complete the outcome data were and how the analysis population related to enrollment.

The registry record also does not report interim analyses, a statistical analysis plan, or a sample-size justification for the trial’s own design. Those procedures should not be assumed from the phase II label or the endpoint definition.

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: sgot (100.0%), nausea (22.6%), bilirubin (22.2%), sgpt (19.4%), hypoalbuminemia (16.7%).

The registry lists serious adverse event figures separately for hepatocellular carcinoma and neuroendocrine hepatic metastases. For hepatocellular carcinoma, 18 were at risk and 18 were affected in the serious category; the other adverse event fields also list 18 at risk and 18 affected. For neuroendocrine hepatic metastases, the serious category lists 31 at risk and 30 affected; the other fields list 31 at risk and 31 affected. These are the registry’s recorded counts and should not be treated as a randomized comparison between groups.

The record also lists serious adverse event terms with percentages: SGOT, 100.0%; nausea, 22.6%; bilirubin, 22.2%; SGPT, 19.4%; hypoalbuminemia, 16.7%; and abdominal pain, 16.1%. These are term-level entries as recorded. The registry information described does not clarify the denominator or the detailed coding and reporting rules behind each term percentage, so they should not be reinterpreted as counts or converted into other quantities.

In general, adverse event summaries describe events recorded under the study’s reporting system; they do not by themselves establish that an event was caused by a treatment. The registry record does not provide a randomized safety comparison, and the disease groups are not like-for-like controls. The figures therefore describe recorded safety data, not a causal contrast between groups.

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

The registry record establishes that a phase II study of chemoembolization was conducted in patients with hepatocellular carcinoma or neuroendocrine hepatic metastases. It identifies the intervention components, describes the repeat procedure, and defines time to progression as the primary outcome. It also identifies tumor response and overall survival as secondary outcomes. Those details explain the trial’s question and measurement framework.

The record does not report the primary endpoint estimate, its 90% confidence interval, or results for the secondary outcomes. It therefore cannot show the observed time to progression, the precision of that estimate, or the direction and size of an outcome effect. The absence of numerical outcome results must not be interpreted as evidence of benefit, lack of benefit, or no difference.

The non-randomized design also limits what can be concluded by comparing the disease groups. They are not described as a like-for-like control comparison. Even if group-level outcome results were available, this design would not by itself isolate an intervention effect from differences between the groups. The record does not report an adjusted comparative analysis.

The adverse event counts and listed serious-event percentages describe what is recorded, but they do not establish causation or support a randomized safety comparison. The enrollment field and the safety figures are not substitutes for a reported primary endpoint analysis population or a complete account of follow-up.

In short, the registry record is informative about the study design, intervention, endpoint definition, and selected safety entries. It does not provide the outcome results needed to assess the observed effect or its uncertainty. Any conclusion about treatment effectiveness or a comparative effect would go beyond what the registry record reports.

13. Why This Trial Matters Statistically

This trial matters statistically because it illustrates the difference between a clearly defined endpoint and a reported result. The registry specifies time to progression, defines disease progression using lesion measurements and new lesions, identifies the starting point as embolization, and describes censoring. These details are essential for understanding what an eventual estimate would represent. They do not, however, supply the estimate itself.

The record also highlights why design determines the scope of interpretation. The study is non-randomized, and its groups are defined by different disease types. The registry explicitly warns against treating them as a like-for-like control comparison. Statistical comparisons cannot remove that basic design limitation unless suitable methods and assumptions are described; no such analysis is reported in the registry record.

Time-to-event outcomes require attention to incomplete follow-up. The stated censoring rule gives a starting point for understanding how observations without documented progression were handled. In general, censoring and uncertainty affect the interpretation of a median and its confidence interval. Here, those values and further follow-up details are not reported, which makes it impossible to assess precision or the role of censored observations.

Finally, the record lists secondary outcomes and safety information without analysis results for the primary and secondary efficacy outcomes. This makes clear why outcome definitions, analysis methods, and results all matter: a design can be described in detail while the evidence needed to interpret its outcomes remains unavailable in the registry record.

14. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The principal limitation for statistical interpretation is that the registry record does not report the median time to progression or its 90% confidence interval. It also does not report results for tumor response or overall survival. The outcome definitions and assessment schedule cannot substitute for observed results, so the size and precision of any outcome effect cannot be assessed from this record.

The non-randomized design and disease-specific groups limit comparative interpretation. The registry cautions that the arms are not a like-for-like control comparison. No randomized control group or adjusted comparison is described. Differences between the groups cannot be treated as an estimate of a causal treatment effect on the basis of the record.

The enrollment field is 50, but the record does not state how many participants were eligible and treated for the primary outcome analysis, how many had documented progression, or how many were censored. It provides a censoring rule but not the follow-up distribution or additional missing-data procedures. The statistical analysis plan, interim analyses, and trial-specific sample-size justification are not reported in the registry record.

The safety entries include group counts and term percentages, but the registry entries described do not specify the denominator and reporting rules for each term percentage. These entries should not be interpreted as comparative or causal estimates. The record’s planning note describes assumptions for illustrative sample-size calculations, but those planning assumptions are not trial outcome results and do not establish observed power.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Chemoembolization in Treating Patients With Primary Liver Cancer or Metastases to the Liver — NCT00003907.
• Registry record last updated 2023-07-05; status last verified 2023-06.
• Results first posted 2013-02-08.
• 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

NCT00003907 is a completed phase II, non-randomized, parallel trial sponsored by the Eastern Cooperative Oncology Group. The registry lists enrollment of 50 and describes patients with hepatocellular carcinoma or neuroendocrine hepatic metastases. Both groups received chemoemulsion with doxorubicin, mitomycin and cisplatin followed by embolization, with repeat treatment described for involved lobes.

The primary outcome was time to progression, with a stated median and 90% confidence interval. Tumor response and overall survival were listed as secondary outcomes. The record defines progression and censoring but does not report numerical efficacy results or analysis estimates. It lists serious and other adverse event counts by disease group and selected serious-event term percentages. Because allocation was non-randomized and the groups were not like-for-like controls, these entries do not support a causal comparison between groups.