A Study of the Combination of Pemetrexed and Irinotecan Every Two Weeks in Metastatic Colorectal Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of the Combination of Pemetrexed and Irinotecan Every Two Weeks in Metastatic Colorectal Cancer (NCT00191984), a non-randomized phase 2 study. The primary outcome is Best Overall Tumor Response.

2. Clinical Question

The study asked whether a combination of pemetrexed and irinotecan, administered every two weeks, could be evaluated for efficacy and safety in people with second-line metastatic colorectal cancer. It was a phase II, non-randomized study with a single treatment group, rather than a comparison of the combination with a control treatment. The registry record gives an enrollment of 46.

Secondary outcomes listed in the registry record were duration of response, progression-free survival, time to treatment failure, and overall survival. Their numerical findings are not reported in the registry record. Consequently, the clinical question can be described from the planned outcomes and treatment group, but the magnitude and precision of the observed tumor response cannot be evaluated from the posted results summarized in the record.

3. Trial Design

This was an open, multicenter phase II study with non-randomized allocation and a single-group model. The intervention group was pemetrexed plus irinotecan. Masking was absent. These features mean that treatment assignment was not randomized and outcome assessment was not masked according to the registry record. Enrollment was 46. The record lists the study status as completed, with a start date of 2004-06 and a primary completion date of 2007-02.

The study’s stated purpose was to determine the efficacy and safety of the combination in metastatic colorectal cancer. The official title places the population in the second-line setting and specifies administration every two weeks. The primary endpoint was best overall tumor response. Its timeframe was baseline to measured progressive disease, with up to 2 years of follow-up. The primary endpoint population was defined as enrolled participants with at least one completed cycle; the number included under that definition is not reported in the registry record.

A single-group study can describe outcomes among people who received the intervention, but it does not directly estimate how those outcomes differ from outcomes under a concurrent control. Non-randomized allocation also means that this study design cannot use random assignment to balance participant characteristics between treatment groups. Because there was no control group, the posted results should be read as a description of outcomes in the treatment group, not as a randomized estimate of a treatment effect.

The record lists duration of response, progression-free survival, time to treatment failure, and overall survival as secondary outcomes. It does not report their numerical results in the record details used for this analysis. The protocol, eligibility details beyond the stated population, dose schedule, handling of treatment changes, and follow-up procedures beyond the primary endpoint description are not reported in the registry record. The rationale for the enrollment target and any interim monitoring plan are also not reported in 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: Two participants discontinued the trial before receiving study treatment. One due to entry criteria exclusion and one due to withdrawal by subject. Therefore, 44 participants received at least one dose of chemotherapy and are included in the efficacy and safety analyses.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All enrolled participants with at least one completed cycle.

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 Best Overall Tumor Response, measured over baseline to measured progressive disease (up to 2 years follow-up).

The primary endpoint, best overall tumor response, classifies each participant according to the best tumor response recorded from the start of treatment until disease progression or recurrence. The registry record says that response was assessed using RECIST criteria. It describes complete response as disappearance of all target lesions and partial response as a 30% decrease in the sum of the longest diameter of target lesions. Progressive disease is described as a 20% increase in that sum, while stable disease refers to changes that do not meet the stated response or progression criteria.

Best overall response is used in oncology studies to summarize the most favorable recorded tumor status during the specified assessment period. A participant-level category can be expressed as a count, which is consistent with the registry record’s outcome unit of participants and parameter of number. The endpoint therefore describes how many participants fall into response categories; it does not, by itself, describe how long a response lasts or how long participants remain free of progression. Those questions relate to other outcomes listed in the record, including duration of response and progression-free survival.

The registry record defines the assessment window as baseline to measured progressive disease, with up to 2 years of follow-up, and defines the primary analysis population as enrolled participants with at least one completed cycle. However, the record does not report the number of participants in each response category, the number included in the endpoint population, or a numerical estimate of response. Without those results, the endpoint’s observed distribution and its precision cannot be assessed from the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Best overall tumor response is a categorical participant-level outcome. Each participant is assigned a response category based on the best assessment recorded during the specified treatment period. RECIST supplies the tumor-measurement rules used to distinguish complete response, partial response, stable disease, and progressive disease. The registry record states the definitions for these categories, but does not report the observed counts in them.

What does a count endpoint tell the reader?

A count reports how many participants were recorded in an outcome category. To interpret a count as a rate or proportion, the relevant denominator must also be known and the population definition must be clear. Here, the primary population is described as all enrolled participants with at least one completed cycle, but the number meeting that definition is not reported in the registry record. The overall enrollment of 46 does not establish the denominator for the primary endpoint. A proportion or percentage should not be reconstructed without that endpoint denominator and the category count.

Even when a count and denominator are available, an observed proportion is an estimate rather than a perfectly precise description of the population from which the participants came. A confidence interval is a range of values compatible with the data under a specified statistical method. Its width conveys uncertainty: a narrower interval indicates greater precision, while a wider interval indicates less precision. The registry record details used for this analysis do not include an interval for best overall response or enough numerical response information to calculate one. No uncertainty interval for the primary result is therefore available to interpret.

Why does the single-group design matter?

In a single-group study, response outcomes can describe the participants who received the intervention. The design does not include a randomized control group against which those outcomes can be compared. Randomization is a method for reducing systematic differences between groups at the time treatment is assigned. Without it, observed differences from a historical expectation or another study may reflect differences in participants, assessment, follow-up, or study conduct as well as treatment. Since this record describes only one group, a causal comparison with an alternative treatment is not available.

Masking is also absent according to the record. In general, masking can limit the influence of knowledge about treatment assignment on assessment or reporting. This is particularly relevant when outcomes require clinical judgment. The registry record identifies RECIST as the response framework, but does not describe additional safeguards for response assessment or the role of independent review. Those details are not reported in the registry record.

How should safety summaries be read?

Adverse-event counts and event-term percentages describe reported safety information, but their interpretation depends on the population observed, the period of observation, and the event definitions. The registry record reports 13 participants with serious adverse events and 38 with other adverse events for the treatment group; the corresponding numbers at risk are not reported. The record also lists several serious-event terms with percentages. The analysis details needed to determine the denominator or possible overlap between event terms are not reported in the registry record. These summaries should therefore be understood as registry-reported safety descriptions, not as comparative risk estimates.

For general context, adverse-event summaries can count participants with at least one event or count event occurrences. Those quantities are not equivalent when a participant can have repeated events. The registry entries identify participant counts for serious and other adverse events, but do not detail the event-counting method. No assumption about repeated events, attribution, severity grading beyond the serious-event label, or causal relationship to treatment should be added.

What can planning calculations establish?

Planning calculations are distinct from analyses of observed outcomes. The registry record’s stated planning illustration uses a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect expressed as a fraction of the observed effect. The record describes such sample sizes as planning illustrations, not observed power. It also describes a total-event approach for time-to-event outcomes and participants per arm for other outcomes. These assumptions do not change the actual design, which was a single-group, non-randomized study.

A planning illustration asks what sample size might be needed under specified assumptions; it is not evidence that a study had a particular probability of finding an effect after completion. The observed effect on which an illustration is based, and numerical planning results, are not reported in the registry record details used for this analysis. The planning assumptions should therefore not be treated as a justification for the actual enrollment or as an estimate of the trial’s evidential strength.

Finally, the registry record reports no analyses in the posted outcome-analysis information. The statistical plan, model choice, confidence-interval method, missing-data handling, and any sensitivity analyses are not reported in the registry record. It is consequently not possible to assess whether the posted outcome was adjusted, how incomplete assessments were treated, or whether a prespecified inferential procedure was followed. General statistical principles explain what such methods can do, but cannot fill in trial-specific details that the record does not state.

10. Multiplicity, Missing Data and Analysis Populations

The registry record lists best overall tumor response as the primary outcome and lists duration of response, progression-free survival, time to treatment failure, and overall survival as secondary outcomes. It also states that there are five outcome measures, but does not report numerical analysis results for them in the registry record used for this analysis. When a study examines multiple outcomes or multiple contrasts, the chance of at least one apparently notable result can increase unless the analysis plan accounts for multiplicity. The record does not report a multiplicity strategy, a hierarchy for testing outcomes, or which outcomes were formally tested. No adjustment or testing sequence should be assumed.

The primary endpoint population is described as enrolled participants with at least one completed cycle. The number included in that population and the number with evaluable response assessments are not reported in the registry record. The handling of participants who did not complete a cycle, had missing tumor assessments, discontinued treatment, or had an assessment after treatment discontinuation is also not reported. These details affect how an observed response count would be interpreted and what population it represents.

The registry record does not report whether missing observations were excluded, imputed, treated as non-response, or handled using another approach. It also does not report sensitivity analyses examining alternative assumptions about missing outcomes. For secondary time-to-event outcomes, censoring rules and follow-up completeness are not reported in the registry record. Those omissions prevent a full evaluation of potential missing-data and follow-up bias.

Likewise, the statistical analysis plan, interim analyses, sample-size justification, and prespecified analysis population beyond the primary population description are not reported in the registry record. The absence of these details from the record does not establish what procedures were or were not used; it means they cannot be verified from the registry information.

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

The registry record establishes that this completed phase II study evaluated a two-weekly pemetrexed and irinotecan combination in a single group of people with metastatic colorectal cancer. It defines best overall tumor response using RECIST and states that the primary outcome population consisted of enrolled participants with at least one completed cycle. It also records selected safety summaries and names secondary outcomes.

These registry entries do not establish the observed response distribution. The primary response counts, analysis denominator, and numerical estimate are not reported in the registry record. Accordingly, the available information does not show how many participants had complete response, partial response, stable disease, or progressive disease, nor does it establish the precision of any response estimate. The secondary-outcome results are also not reported in the registry record.

The single-group, non-randomized design limits what can be concluded even if response results are available elsewhere. An outcome observed after treatment cannot, by itself, show what would have occurred in a comparable group receiving a different treatment. Without a randomized comparator, the study does not yield a like-for-like treatment contrast. The absence of masking is another design feature to consider when interpreting assessments, although the record does not provide details of any independent review process.

The safety figures are descriptions of events recorded for the intervention group. The registry record lists 13 participants with serious adverse events and 38 with other adverse events, as well as percentages for selected serious-event terms. Since the corresponding numbers at risk and event-analysis details are not reported, these summaries should not be converted into comparative risks or used to infer the likelihood of an event in other populations.

In short, the registry record gives useful information about the study question, design, endpoint definitions, and selected safety reporting. It does not provide enough outcome data to assess the magnitude or precision of the primary efficacy result. Neither positive nor negative conclusions about the treatment’s efficacy can be drawn from the absence of a numerical result in this record.

12. Why This Trial Matters Statistically

This trial illustrates an important distinction between defining an endpoint and reporting its result. The record gives a detailed description of best overall tumor response and identifies the participant population and assessment timeframe. Yet without response counts and a denominator, the endpoint cannot be translated into an observed response proportion or an assessment of its precision. A clear outcome definition is necessary for interpretation, but it is not a substitute for outcome data.

The design also matters statistically. A single-group study can document what happened among people receiving an intervention, but it does not establish a controlled comparison. The non-randomized allocation and absence of masking further limit the protections against bias that a randomized, masked comparison may offer. These are not merely technical labels: they shape which questions the observed data can answer.

The safety summaries highlight a related issue. Counts and percentages need a clearly defined denominator, observation period, and event-counting approach for meaningful interpretation. The record reports participant counts and selected percentages but does not report the corresponding numbers at risk or full analysis details. Accurate reporting of these elements is essential for judging how directly a safety summary applies to the treated group.

Finally, the record names several secondary outcomes but does not report numerical results or a statistical plan. Multiplicity, missing-data handling, and prespecified analysis choices can affect the interpretation of a set of outcomes. When these details are unavailable, readers should distinguish what the record states from what remains uncertain. This study is therefore statistically instructive as an example of why design, denominator reporting, endpoint results, and analysis transparency all contribute to an interpretable clinical trial record.

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 trial-specific limitation is the absence of numerical primary efficacy results in the registry record. Although the endpoint definition and analysis population are described, the response-category counts, endpoint denominator, and any estimate of uncertainty are not reported. This prevents assessment of the observed response distribution and its precision.

The design was a single-group, non-randomized study without masking. There was no randomized control group for estimating a treatment contrast. As a result, outcomes cannot be attributed to the combination by comparison with a concurrent control, and differences from other settings cannot be separated from differences in participants, assessment, or follow-up. The record’s phase II designation does not remove this design limitation.

The enrolled population numbered 46, while the primary endpoint population was defined more narrowly as enrolled participants with at least one completed cycle. The number in that endpoint population is not reported in the registry record. The record also does not report the number with evaluable response assessments or explain how missing or incomplete assessments were handled. These details are necessary to understand the denominator and potential selection effects.

Trial-specific analysis details are limited. The statistical analysis plan, sample-size justification, interim analyses, multiplicity approach, and sensitivity analyses are not reported in the registry record. The protocol-level details needed to interpret dose exposure and response-assessment procedures are also not reported in the registry record for this analysis. Secondary outcomes are listed, but their numerical results are not reported in the registry record.

Safety interpretation is constrained by the absence of numbers at risk and by unreported event-counting details. The listed event percentages cannot be treated as comparative estimates, and the relationship between participants and event terms is not described. These limitations concern what can be established from the registry record; they do not establish that the trial investigators did not conduct additional analyses or collect further data.

14. Related Tutorials

15. Related Statistical Calculators

16. Sources

• ClinicalTrials.gov: A Study of the Combination of Pemetrexed and Irinotecan Every Two Weeks in Metastatic Colorectal Cancer — NCT00191984.
• Registry record last updated 2011-01-05; status last verified 2011-01.
• Results first posted 2009-06-15.
• 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 record describes NCT00191984, a completed, open, multicenter phase II study of pemetrexed plus irinotecan every two weeks in second-line metastatic colorectal cancer. Allocation was non-randomized, masking was absent, and the study used a single-group model. Enrollment was 46; the recorded study dates are 2004-06 for start and 2007-02 for primary completion.

Best overall tumor response was the primary outcome, assessed from baseline to measured progressive disease, with up to 2 years of follow-up, among enrolled participants with at least one completed cycle. The record defines response categories using RECIST but does not report numerical primary efficacy results. Secondary outcomes listed are duration of response, progression-free survival, time to treatment failure, and overall survival; their numerical findings are not reported in the registry record.

The safety summary lists 13 participants with serious adverse events and 38 with other adverse events. The record does not report the corresponding numbers at risk. It also lists percentages for selected serious-event terms. The statistical analysis plan and details of missing-data handling are not reported in the registry record.