Gemcitabine and Oxaliplatin as Second-Line Therapy in Treating Patients With Metastatic Colon Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Gemcitabine and Oxaliplatin as Second-Line Therapy in Treating Patients With Metastatic Colon Cancer (NCT00276861), a non-randomized phase 2 study. The primary outcome is Response Rate as Measured by RECIST Criteria.

2. Clinical Question

This is a small, single-arm phase II investigation, not a randomized comparative trial. Its clinical question is therefore descriptive: what response was recorded among enrolled patients treated with the combination, and how uncertain is that estimate? Without a reported response count or estimate, the posted results cannot answer that question quantitatively. The registry record also lists time to progression, measured using a Kaplan Meyer curve at completion of study treatment, as a secondary outcome. No numerical result for that outcome is reported in the registry record.

3. Trial Design

This was a phase II study sponsored by the University of Miami. It was non-randomized and unmasked, with a single treatment arm identified in the safety record. Gemcitabine hydrochloride and oxaliplatin were studied together as second-line therapy for metastatic colon cancer. The intended study population is also described in the registry summary as patients with metastatic or recurrent colon cancer. The registry record reports enrollment of 10 participants.

The record lists the study start as 2005-09 and primary completion as 2006-09. The current status is terminated. The reason for termination, whether the planned enrollment was reached, and whether all enrolled participants received treatment are not reported in the registry record. The duration and details of treatment, including dosing and modifications, are also not reported in the registry record.

The primary outcome was response rate as measured by RECIST criteria. Complete response or partial response under RECIST v 1.0 constituted a response, and the stated timeframe was 4 - 6 months. The registry record describes the measure as a number and gives its unit as a percentage of participants; it also identifies a 95% confidence interval as the dispersion measure. No numerical response estimate or confidence interval is reported. Time to progression measured by a Kaplan Meyer curve at completion of study treatment is listed as a secondary outcome, without a posted numerical result.

Although the record lists 2 outcome measures, it contains no analysis entries. The design does not support a randomized comparison between treatment groups: no concurrent control arm is identified, and allocation is explicitly non-randomized. Consequently, the study description supports understanding the treatment question and planned outcome measures, but it does not establish a comparative treatment effect.

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

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 Response Rate as Measured by RECIST Criteria, measured over 4 - 6 months.

The primary endpoint was response rate measured by RECIST criteria. A response was defined as complete response or partial response using RECIST v 1.0. The timeframe is recorded as 4 - 6 months, and the measure is described as a percentage of participants. The registry record specifies a 95% confidence interval as the intended way to express dispersion around the estimate. However, neither the number of participants with complete or partial response nor the response estimate and its confidence interval are reported in the registry record.

Response endpoints are used in oncology studies to describe whether tumors meet prespecified criteria for shrinkage or disappearance. In general, standardized criteria such as RECIST make response classification more consistent than an informal judgment, because response is tied to defined tumor measurements and categories. A response-rate endpoint can summarize observed tumor changes in the treated group, but it does not by itself show how outcomes compare with another treatment. In a non-randomized, single-arm study, interpretation also depends on how participants were selected, assessed, and followed. Those details are not reported in the registry record.

The record also lists time to progression as a secondary outcome, measured by a Kaplan Meyer curve at completion of study treatment. The event definition, follow-up detail, and numerical estimate are not reported in the registry record. The primary endpoint and secondary outcome therefore identify the intended questions, but the registry record does not provide numerical findings with which to answer them.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a response-rate endpoint describe?

Response rate summarizes the proportion of assessed participants whose tumors meet the specified response categories. In this study, the registry record defines response as complete response or partial response according to RECIST v 1.0. The endpoint is recorded as a percentage of participants over 4 - 6 months. The distinction between the definition and the result matters: an outcome definition explains what would count as a response, but does not indicate how many participants responded. No numerical result is reported in the registry record.

In general, standardized tumor-response criteria aim to apply consistent rules to clinical measurements. Complete and partial response are categories, not interchangeable judgments, and classification depends on the criteria and assessment process. The registry record does not describe imaging schedules, independent review, assessment procedures, or handling of unevaluable cases. Those omissions limit the ability to evaluate how response would have been assessed in practice.

How should a confidence interval be understood?

The primary outcome lists a 95% confidence interval as its dispersion measure. In general, a confidence interval communicates uncertainty around an estimate, reflecting both the observed data and the statistical method. It is not a guarantee that a particular interval contains the true value, and it does not replace the estimate itself. The actual response estimate and interval are not reported in the registry record, so their precision and compatibility with different possible response levels cannot be assessed.

For a response rate based on a small group, the number of observed responses can have a substantial influence on the estimate. In general, fewer observations provide less information about the underlying rate, and the interval is important for showing the range of uncertainty. The registry enrollment was 10, but the number included in the endpoint analysis is not reported in the registry record. It is therefore not appropriate to assume that all enrolled participants contributed to the primary endpoint.

What does the single-arm design allow?

The allocation is non-randomized, and the safety record identifies a single arm. A single-arm design can describe outcomes observed in a treated group, but it does not provide a concurrent randomized group for estimating a treatment contrast. In general, random assignment helps balance known and unknown prognostic factors between groups. Without it, observed outcomes can reflect participant selection, disease characteristics, treatment exposure, assessment, follow-up, and other influences alongside any treatment effect. A response result from this study, if available, would therefore be descriptive rather than a randomized comparative estimate.

The record’s sample-size assumptions describe planning illustrations rather than the trial’s own design. They use 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 stated fraction of the observed effect. The record further describes time-to-event illustrations as total event counts using Schoenfeld, and other illustrations as participants per arm. These assumptions do not change the actual study design: it was non-randomized and had a single arm recorded. The numerical planning illustrations are not reported in the registry record. They cannot be treated as observed power or as evidence that the study had a particular probability of detecting an effect.

What does time to progression add?

Time to progression is listed as a secondary outcome, with measurement by a Kaplan Meyer curve at completion of study treatment. In general, time-to-event methods describe how long it takes for a defined event to occur while accounting for participants whose event time is not observed during available follow-up. Such methods rely on clear event and censoring definitions, as well as information about follow-up. The registry record does not provide those details or a numerical result. A named curve method alone is not an outcome finding.

Time-to-event analysis can be affected by censoring, which occurs when the event time is not observed for a participant during the period of follow-up. In general, interpretation depends on how censoring is handled and whether the assumptions behind the method are reasonable. The event definition, censoring rules, follow-up completeness, and number of progression events are not reported in the registry record. No curve estimate or comparison can therefore be evaluated from the posted information.

Why can over-dispersion matter?

In general, over-dispersion occurs when observed variation is greater than a statistical model expects. For count outcomes, ignoring over-dispersion can make uncertainty appear smaller than it is, potentially producing intervals or tests that are too optimistic. This is a general statistical consideration, not a reported feature of this study. The registry record does not describe a count model or an over-dispersion assessment, and the primary endpoint is specified as a response rate rather than a count-model analysis. No particular over-dispersion issue should be attributed to this trial.

More broadly, a statistical method should match the endpoint, study design, and available observations. The registry record gives endpoint definitions and a confidence-interval label but does not report the analysis method, analysis population, missing-data handling, or numerical efficacy analyses. Those details are necessary to reproduce or fully appraise an estimate. In their absence, the defensible reading is limited to the registered design, outcomes, and posted safety counts.

10. Multiplicity, Missing Data and Analysis Populations

The registry record lists 2 outcome measures: response rate as the primary outcome and time to progression as a secondary outcome. It does not describe a multiplicity adjustment, a hierarchy of hypotheses, or any other plan for controlling the chance of false-positive findings across outcomes. The analysis entries are absent, so no set of inferential results is available for assessing how multiple outcomes were interpreted. Whether a multiplicity strategy was specified is not reported in the registry record.

Missing-data rules are not reported in the registry record. The record does not state how participants without response assessments, those who discontinued treatment, or those with incomplete follow-up were handled. It also does not define an analysis population, such as whether the primary result was intended to include all enrolled participants or only those with evaluable assessments. With enrollment of 10, the distinction between enrolled and analyzed participants could be important, but the analyzed count is not reported in the registry record.

The record does not report interim analyses, stopping boundaries, or a statistical analysis plan. The terminated status alone does not establish whether a formal stopping rule was used or why the study ended. These points should therefore remain unknown rather than being inferred from the small enrollment or missing results. The primary outcome’s stated timeframe and confidence-interval measure describe the endpoint specification, but they do not fill the gaps in the analysis and missing-data information.

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 (10.0%), infection (10.0%).

The safety record identifies a single arm with 10 participants at risk for both serious and other adverse events. It records 2 participants affected by serious adverse events and 10 affected by other adverse events. These are reported counts; they do not indicate the timing, duration, severity grading beyond the serious classification, or treatment relatedness of the events. Attribution to the study drugs is not reported in the registry record.

Neutropenia and infection are listed as top serious adverse event terms, each at 10.0%. The registry record does not state whether these terms refer to distinct participants or whether one participant could contribute to both terms. It also does not give detailed event counts or descriptions for each term. The listed percentages should be read as registry-reported values, not as a basis for reconstructing individual participant histories.

Deaths are not reported in the registry record. That absence should not be interpreted as evidence that no deaths occurred. Because there is no comparator arm, these safety data cannot establish whether event frequency differed from another treatment or from an untreated group. They describe the posted safety information for this small study group, with important details and context not reported in the registry record.

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

The record establishes that a phase II, non-randomized study of gemcitabine hydrochloride with oxaliplatin was conducted for second-line treatment of metastatic colon cancer, with enrollment of 10 and a terminated status. It also identifies response rate by RECIST v 1.0 as the primary endpoint and time to progression as a secondary outcome. The registry record does not report numerical efficacy results for either outcome, so the clinical outcomes cannot be quantified from the posted information.

The missing response estimate is especially consequential because the registered primary question concerns whether participants achieved complete or partial response. Without the number responding, the percentage, or the stated 95% confidence interval, no claim about the observed response level or precision is supportable. No outcome conclusion should be substituted for those absent results.

The study’s single-arm design also limits what any observed response could establish. Without randomization and a concurrent comparator, outcomes cannot be interpreted as a treatment difference. The record does not permit separating treatment-associated changes from participant selection, disease course, assessment, or other factors. The secondary time-to-progression outcome is similarly not accompanied by numerical results or follow-up details.

The posted adverse-event counts indicate that serious and other adverse events were recorded among participants. They do not establish that the study drugs caused those events, and the record does not give enough detail to characterize their course. The serious-event terms and percentages are descriptive registry entries, not comparative safety estimates.

Accordingly, this record documents the study’s question, broad design, and limited safety counts, but it does not provide the efficacy findings needed to answer the primary question. The reason for termination, analysis population, missing-data approach, and explanation for the absence of numerical outcome results are not reported in the registry record. Those gaps should remain explicit when interpreting the record.

13. Why This Trial Matters Statistically

This study is statistically notable because its registered primary endpoint is clearly defined while the corresponding numerical outcome is not reported in the registry record. That difference illustrates why an endpoint description should not be mistaken for a result. The record specifies response by RECIST v 1.0, a 4 - 6 months timeframe, and a 95% confidence interval, but does not provide the response count, estimate, or interval needed to assess the observed finding.

The design also illustrates the distinction between describing outcomes and estimating comparative effects. A non-randomized single-arm study can provide information about what was recorded among its participants, but it cannot by itself estimate how outcomes would differ under another treatment. Even a complete response-rate result would need to be interpreted with that limitation in mind. Here, the lack of numerical efficacy findings further restricts interpretation.

Enrollment of 10 and terminated status make the absence of analysis details especially salient, but neither fact explains the missing outcome results. The reason for termination, endpoint evaluability, analysis populations, and missing-data handling are not reported in the registry record. These omissions prevent readers from assessing whether the planned endpoint could be estimated reliably or how incomplete observations might have affected it.

The safety information provides another statistical lesson: counts and listed event terms are descriptive unless their definitions, attribution, timing, and comparison context are clear. The registry reports serious and other adverse-event counts in a single arm, but does not offer comparative safety results. Read carefully, the record supports statements about what it registers, while leaving important questions unanswered rather than inviting unsupported conclusions.

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 this record. The study enrolled 10 participants and is marked terminated, but the reason for termination and the relationship between actual enrollment and any intended enrollment are not reported in the registry record. Participant characteristics, disease history, and treatment exposure are not described in the posted information. This makes it difficult to judge how the enrolled group relates to the intended population of patients with metastatic or recurrent colon cancer.

The design is non-randomized, with a single arm recorded. There is no like-for-like control comparison, so the design cannot isolate a treatment contrast. The record does not describe the treatment regimen in detail, including doses, schedule, duration, adherence, or modifications. Assessment procedures, imaging schedule, and independent review for RECIST response are also not reported in the registry record.

Most importantly, no numerical primary or secondary efficacy results are reported. The number assessed, number responding, response estimate, confidence interval, time-to-progression results, event count, and follow-up summary are unavailable. The analysis population, missing-data rules, and reasons for any exclusions are not reported in the registry record. The record also does not report interim analyses, a statistical analysis plan, or multiplicity handling.

Safety reporting is limited to group-level counts and selected serious adverse event terms. Event timing, severity detail, relationship to treatment, and whether named events occurred in overlapping participants are not specified. Deaths are not reported in the registry record. These limitations constrain both efficacy and safety interpretation; they do not establish that unreported outcomes or events did not occur.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Gemcitabine and Oxaliplatin as Second-Line Therapy in Treating Patients With Metastatic Colon Cancer — NCT00276861.
• Registry record last updated 2017-02-08; status last verified 2016-12.
• Results first posted 2013-02-26.
• 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

NCT00276861 is a terminated phase II study sponsored by the University of Miami, evaluating gemcitabine hydrochloride with oxaliplatin as second-line therapy for metastatic colon cancer. The allocation was non-randomized, masking was none, and enrollment was 10. The registry record lists the start as 2005-09 and primary completion as 2006-09.

The primary outcome was response rate by RECIST criteria, defining response as complete response or partial response under RECIST v 1.0 over 4 - 6 months. A 95% confidence interval was specified, but numerical efficacy results are not reported in the registry record. Time to progression by a Kaplan Meyer curve at completion of study treatment was listed as a secondary outcome; numerical results are not reported.

The single-arm safety record lists 2 of 10 participants affected by serious adverse events and 10 of 10 affected by other adverse events. Neutropenia and infection are each listed as serious event terms at 10.0%. Deaths, the reason for termination, and analysis details are not reported in the registry record. The design provides no randomized comparator for estimating a treatment difference.