Combination Chemotherapy and Bevacizumab in Treating Patients With Advanced Neuroendocrine Tumors: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Combination Chemotherapy and Bevacizumab in Treating Patients With Advanced Neuroendocrine Tumors (NCT00227617), a na phase 2, phase 3 study. The primary outcome is Rate of Discontinuation Due to Adverse Events Possibly Related to Study Treatment.

2. Clinical Question

This is a safety-focused question about treatment discontinuation, not a randomized test of comparative benefit. A discontinuation endpoint captures an outcome relevant to whether participants remain on a study regimen, but it does not by itself describe every adverse event, its severity, or its cause. The record also names time to progression, overall median survival, overall time to treatment failure, and biochemical marker response as secondary outcomes. It does not provide their results in the information reported for this analysis.

The registry marks the study as terminated and explains that it ended early because of low accrual. It also notes that some study data were not transferred after the institution moved its clinical research database. These circumstances limit what can be concluded from the posted record: the registered question is identifiable, but the full outcome evidence and the intended analysis cannot be reconstructed from the available information.

3. Trial Design

The study is registered as a single-group trial with allocation listed as not applicable and masking listed as none. Enrollment was 36. The sponsor was the University of California, San Francisco. The registry lists phase 2 and phase 3, while the brief summary describes the study as phase I/II; these phase descriptions are inconsistent within the record and cannot be reconciled from the reported information.

The intervention was FOLFOX in combination with bevacizumab. The regimen description says treatment began on Day 1 and was administered every two weeks. It lists 5-fluorouracil at 2400 mg/ m2 by continuous infusion over 46-48 hours, leucovorin at 200 mg/ m2 over 2 hours, oxaliplatin at 85 mg/m2 over 2 hours, and bevacizumab at 5 mg/kg intravenously over 30-90 minutes. These are protocol details, not evidence that every enrolled participant received every scheduled dose.

The registry gives a start date of 2005-06-08 and a primary completion date of 2012-01. The study status is terminated, and the record attributes early termination to low accrual. It further says that the institution moved its clinical research database from Velos to OnCore in July 2010 and that some data were not transferred. Results are described as accurate to date, but the record does not establish that the posted data represent complete follow-up for all participants.

Because there is no control arm, the design can describe outcomes among participants in this study but cannot isolate the regimen's effect from other influences through a concurrent randomized comparison. The record does not report a statistical analysis plan, details of participant assignment beyond the single-group designation, or a sample-size justification. Those design features should not be inferred from the listed enrollment or from the outcome summaries.

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: Simon 2-stage design: 13 patients in stage 1. If at least 2 objective tumor responses were observed by 12 cycles of therapy, the cohort was to be expanded to enroll a total of 39 treated patients.

Stage 1

Stage 2

Analysis population for the primary outcome, as defined in the registry: Study terminated early due to low accrual. UCSF moved its clinical research database from Velos to OnCore in July 2010 and some of the data for this study was not transferred. Results reported accurate to date

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: Patient data was migrated at time of close to accrual and the treatment arm information was not included in data migration. As such baseline data for patients with Carcinoid Neuroendocrine Tumors, Pancreatic Neuroendocrine Tumors, Poorly Differentiated Neuroendocrine Carcinoma data has been reported across treatment arms

6. Primary Endpoint

The registry lists the primary outcome as Rate of Discontinuation Due to Adverse Events Possibly Related to Study Treatment, measured over From beginning of treatment up to 18 months; Post-study survival follow-up up to 8 years.

The primary endpoint is the rate of discontinuation due to adverse events possibly related to study treatment, with neuropathy excluded. Its unit is participants, and the endpoint description says rates were calculated as counts and percentages of patients who discontinued for the specified reason. The stated timeframe extends from the beginning of treatment up to 18 months, with post-study survival follow-up up to 8 years.

A discontinuation endpoint records whether a participant stopped treatment for a particular safety-related reason. In general, this can summarize a consequence of tolerability that matters to the delivery of a regimen. It is not interchangeable with the number of adverse events: one participant may experience more than one event, and an event does not necessarily lead to discontinuation. Nor does a discontinuation count alone describe onset, duration, severity, or the certainty that treatment caused the event.

The registry does not report the primary endpoint count or percentage in the information available for this analysis. It therefore is not possible to describe the observed discontinuation rate from the record. The endpoint description also does not give detailed rules for how investigators judged whether an event was possibly related to treatment or how neuropathy was handled in other safety summaries. Such details are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a single-group design allow?

In a single-group study, participants receive the study regimen without a concurrent control arm. Descriptive statistics can report what happened among enrolled or treated participants, if those data are available. However, there is no randomized comparison to distinguish the regimen's effect from differences in participant characteristics, the course of illness, supportive care, or other influences. This is a general limitation of single-group evidence, not a finding about a particular source of bias in this study.

Here, allocation is listed as not applicable and the model as single group. The record therefore does not support a treatment-versus-control estimate. The primary endpoint is a participant count for discontinuation due to possibly treatment-related adverse events, not a comparative effect measure. The result itself is not reported in the registry record.

What does a count or rate describe?

A count states how many participants met a defined outcome. A rate or percentage relates that count to a stated group or denominator. In general, the denominator and the definition of the event are essential to interpretation: the same count can have different meaning in groups of different sizes, and a broad event definition can combine clinically different experiences. The primary endpoint description says that counts and percentages were used, but the actual primary count and percentage are not reported in the registry record.

For adverse event summaries, event counts and affected-participant counts are not necessarily the same. A participant may experience multiple events, while an affected-participant count generally concerns whether a participant had at least one event within a category. The registry lists participants at risk and affected for serious and other adverse events, but does not explain all event-counting and coding details. The serious event terms and percentages should therefore be read as the specific reported safety summaries, not as a full event-level dataset.

Why does uncertainty matter?

In general, a confidence interval describes a range of values compatible with the observed data under a specified statistical model and procedure. It helps convey precision: a point estimate alone does not show how much it might vary because of sampling. No confidence interval for the primary discontinuation endpoint is reported in the registry record. No confidence intervals for the listed safety terms are reported either. Without estimates and their uncertainty, it is not possible to assess the precision of these outcomes from the posted information.

A p-value, when reported, measures compatibility between data and a specified null model under the assumptions of a test; it is not the probability that a hypothesis is true. No p-value or analysis result is reported for the primary endpoint. The record lists zero analyses in the outcome information summarized for the study, which means analyses are not posted in that record; it does not establish whether analyses were conducted elsewhere.

What can be said about time-to-event outcomes?

Time-to-event methods are generally used when the outcome concerns time until an event, while accounting for participants whose event has not been observed by the end of their follow-up. Their interpretation depends on clear definitions of the event, follow-up, censoring, and analysis population. Time to progression, overall median survival, and overall time to treatment failure are named as secondary outcomes. The registry information does not report their numerical results or enough method detail to assess how these outcomes were analyzed. No particular time-to-event method should be assumed.

How should the planning illustration be understood?

In general, sample-size planning illustrations describe how many participants might be needed under stated assumptions to detect a chosen effect with a chosen significance level and power. They are prospective planning calculations, not evidence about what the observed study established. The registry notes that its planning illustration uses the observed effect, a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; it also distinguishes event totals for time-to-event outcomes from participants per arm for other outcomes.

Those assumptions do not describe this study's design: the study is single group, with no allocation and no concurrent comparison arm. The illustration is not the trial's own design justification, and its results should not be interpreted as observed power. The registry does not report the original sample-size rationale or a formal justification for enrollment. Any planning illustration based on an observed effect is also conditional on that effect estimate and its uncertainty; it cannot repair missing outcome information or turn a single-group study into a randomized comparison.

Overall, the methods that can be described are the registry's stated endpoint and design, plus general principles for interpreting counts, uncertainty, and planning. The specific statistical analysis plan, handling of incomplete observations, model choices, and inferential results are not reported in the registry record.

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("Carcinoid", "PNET", "PDNEC")
n      <- c(22, 12, 2)
events <- c(3, 3, 0)
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  Carcinoid  22  3  13.6  2.9  34.9  4.7  33.3  5.0  33.6
ARM  PNET  12  3  25.0  5.5  57.2  8.9  53.2  9.1  53.8
ARM  PDNEC  2  0  0.0  0.0  84.2  0.0  65.8  0.8  70.8

11. Multiplicity, Missing Data and Analysis Populations

The registry names several outcomes: the primary discontinuation endpoint and secondary outcomes of time to progression, overall median survival, overall time to treatment failure, and biochemical marker response. It reports six outcome measures and zero analyses in the outcome information. The record does not give results for the listed secondary outcomes or explain whether they were analyzed, omitted, or not transferred.

When a study evaluates multiple outcomes, general statistical practice recognizes that the chance of at least one apparently notable result can increase if many tests are performed without accounting for multiplicity. An analysis plan may specify a primary outcome, testing order, or adjustment method. No multiplicity strategy or detailed statistical plan is reported in the registry record, so it is not possible to determine how multiple outcomes were intended to be handled. This absence should not be taken as evidence that no such plan existed outside the registry.

The record says the study terminated early due to low accrual. It also reports that the clinical research database moved from Velos to OnCore in July 2010 and that some study data were not transferred. This is direct evidence of incomplete data transfer, but the record does not identify which participants, variables, outcomes, or follow-up periods were affected. It does not describe imputation, exclusions, or sensitivity analyses. The missing-data mechanism and the consequences for any analysis are therefore not reported in the registry record.

Analysis populations are also not described. Enrollment of 36 does not establish how many participants began treatment, contributed to each endpoint, or remained under observation for the full stated timeframe. Those counts and definitions are not reported in the registry record. Without them, the denominators and completeness of endpoint-specific summaries cannot be fully assessed.

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 2% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: pain - abdomen nos (8.3%), neutrophils/granulocytes (anc/agc) (5.6%), perforation, gi - small bowel nos (5.6%), infection - other (specify,__) (5.6%), hemoglobin (2.8%).

The registry lists 36 participants at risk for serious adverse events, with 12 affected, and 36 at risk for other adverse events, with 36 affected. These are descriptive participant figures for the combined neuroendocrine tumor group. The record also lists 3 deaths. It does not report whether those deaths were considered related to study treatment, nor does it provide their timing or causes.

For serious adverse events, the listed terms and percentages are Pain - Abdomen NOS (8.3%), Neutrophils/granulocytes (ANC/AGC) (5.6%), Perforation, GI - Small bowel NOS (5.6%), Infection - Other (specify,__) (5.6%), Hemoglobin (2.8%), and Constitutional Symptoms (2.8%). These percentages should be read as reported registry summaries. The record does not clarify the denominator or event-coding details for each term, whether a participant could appear under more than one term, or whether the list is exhaustive.

Serious adverse event status and treatment-relatedness are distinct concepts. The primary endpoint concerns discontinuation for events possibly related to study treatment, excluding neuropathy; the safety figures do not establish that the events listed caused treatment discontinuation. The registry does not report the primary discontinuation count or percentage, so the safety list cannot substitute for that endpoint. These figures describe what the record reports, not a comparison with another treatment.

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

The registry record establishes that a single-group study of FOLFOX with bevacizumab enrolled 36 participants and was terminated early because of low accrual. It identifies discontinuation due to possibly treatment-related adverse events, excluding neuropathy, as the primary endpoint and describes a count-and-percentage approach. However, the primary endpoint result is not reported in the registry record. The actual frequency of discontinuation for that reason therefore cannot be stated from the posted information.

The safety summary reports 12 participants affected by serious adverse events among 36 listed at risk, 36 affected by other adverse events among 36 listed at risk, and 3 deaths. It also lists selected serious event terms with percentages. These are descriptive figures. They do not establish that an event was caused by treatment, that it led to discontinuation, or that the regimen differs from an alternative. The record does not report a concurrent comparison group or comparative estimates.

The study's named secondary outcomes include time to progression, overall median survival, overall time to treatment failure, and biochemical marker response. Their numerical results are not reported in the registry information. The statistical analysis plan, analysis populations, uncertainty estimates, and handling of missing observations are also not reported. This limits assessment of the precision and completeness of the outcome evidence.

These limits do not erase the information that is present: the design, intended primary endpoint, treatment regimen, early termination, and selected safety counts are documented. But the posted data do not support a conclusion about the regimen's efficacy or comparative safety. The database transition and loss of some data add uncertainty about completeness, while the single-group design restricts causal and comparative interpretation even if outcome summaries are available.

14. Why This Trial Matters Statistically

This trial illustrates how registry records can contain a clearly stated clinical question while leaving the principal numerical result unavailable. The primary endpoint is specific enough to identify what investigators intended to count, and the record describes its general timeframe and the exclusion of neuropathy. Yet without the endpoint count, percentage, denominator details, or uncertainty, readers cannot assess the observed result or its precision.

It also shows why study design and outcome reporting must be interpreted together. A single-group study can describe outcomes among participants, but cannot directly compare the regimen with a concurrent alternative. Even a complete set of descriptive results would not provide the same basis for comparative inference as a randomized control group. The early termination due to low accrual and the untransferred data further limit what can be learned from the posted record.

Finally, the record distinguishes neither all safety-event details nor analysis populations in enough depth to support a full statistical appraisal. The listed serious-event counts, other-event count, deaths, and selected event terms are useful as limited descriptive evidence, but not as a complete assessment of the benefit-risk balance. For biostatisticians and clinical researchers, the key lesson is to separate documented design and data from unreported methods and results. A registry entry can identify what was planned without supplying enough evidence to evaluate how the trial's central question was answered.

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.

Several limitations are specific to this study record. First, the study is marked terminated, with early termination attributed to low accrual. The record does not report how accrual compared with any target or how termination affected the planned analyses. No sample-size justification is reported, so the enrollment of 36 should not be treated as evidence that the study had adequate precision or power.

Second, the institution's clinical research database moved from Velos to OnCore in July 2010, and the registry says some data were not transferred. The record does not specify which data were lost, whether missing information concerned particular participants or outcomes, or whether the posted analyses account for the incomplete transfer. It describes results as accurate to date, but does not establish complete outcome capture or follow-up.

Third, there is an inconsistency in the phase descriptions: the registry phase listing includes phase 2 and phase 3, while the brief summary calls the study phase I/II. The available information does not explain the difference. Also, the record supplies no actual primary endpoint result, no numerical secondary outcome results, and no detailed analysis populations or uncertainty estimates.

Finally, a single-group design without allocation or masking does not provide a concurrent control comparison. The registry does not report enough detail to assess event attribution, all safety definitions, or the analysis methods. These limitations constrain interpretation of both the planned question and the posted safety information.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Combination Chemotherapy and Bevacizumab in Treating Patients With Advanced Neuroendocrine Tumors — NCT00227617.
• Registry record last updated 2023-05-22; status last verified 2019-12.
• Results first posted 2019-12-26.
• 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.

19. Record Summary

NCT00227617 is a single-group study of FOLFOX with bevacizumab in patients with advanced neuroendocrine tumors, sponsored by the University of California, San Francisco. The registry reports enrollment of 36, a start date of 2005-06-08, primary completion in 2012-01, and terminated status. The record says the study ended early due to low accrual and that some data were not transferred during a clinical research database transition.

The primary endpoint concerns participant discontinuation due to adverse events possibly related to study treatment, excluding neuropathy. Its result is not reported in the registry record. The safety summary lists 12 participants affected by serious adverse events among 36 at risk, 36 affected by other adverse events among 36 at risk, and 3 deaths. Secondary outcomes are named, but their results are not reported. The phase descriptions are inconsistent within the record.