Study Of Ispinesib In Subjects With Breast Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study Of Ispinesib In Subjects With Breast Cancer (NCT00089973), a non-randomized phase 2 study. The primary outcome is Percentage of Participants With Overall Response Rate (ORR) Following Administration of Ispinesib.

2. Clinical Question

The study was intended to examine how breast cancer responded to ispinesib, as well as its side effects and the amount of drug in the blood at specific times after it was taken. The primary endpoint was labeled overall tumor response rate. The record says it was assessed using CT or MRI and RECIST 1.0. However, its written definition includes complete response, partial response, stable disease, and progressive disease. That wording makes the exact meaning of the named response endpoint unclear: stable disease and progressive disease are separate RECIST categories, not response categories in the usual sense. The registry record does not report the observed primary endpoint result, so the response percentage cannot be described from this record.

3. Trial Design

This was a Phase II, open-label, non-randomized, single-group study of ispinesib in subjects with advanced or metastatic breast cancer. The registry record identifies GlaxoSmithKline as the sponsor and lists the study as completed. The recorded start date was 2004-06-30 and the primary completion date was 2006-08-25. Enrollment was 50.

All participants received the study drug; the record does not describe a control group. With no randomized comparator, the study can describe outcomes observed among participants receiving ispinesib, but it cannot provide a like-for-like estimate of how those outcomes differed from outcomes under another treatment or under no study treatment. The record itself flags that the arms are not a like-for-like control comparison; its design is single-group.

The study was open label, meaning masking was listed as none. The registry record does not report whether outcome assessment was performed by reviewers unaware of other study information. The primary endpoint was assessed after cycle 2 and repeated every 2 cycles up to Cycle 10, up to 26 months. The record also lists outcomes concerning time to response, duration of response, time-to-progression, adverse events, clinical concern values, toxicity shifts, and pharmacokinetic clearance. It lists 10 outcome measures overall, but does not report statistical analyses in the analyses listing.

The stated primary analysis population consisted of participants who received at least one dose of study drug. This is a treatment-exposed population definition, rather than evidence that all enrolled participants were included regardless of treatment exposure. The registry record does not report screening counts, reasons for non-treatment or withdrawal, or details of follow-up completeness.

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: Intent to treat population consisted of all participants (n=50) who received at least one dose of study drug.

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 Percentage of Participants With Overall Response Rate (ORR) Following Administration of Ispinesib, measured over After cycle 2 and repeated every 2 cycles up to Cycle 10, up to 26 months.

The primary endpoint is labeled the percentage of participants with overall response rate following administration of ispinesib. It is a binary-outcome summary: each participant is classified according to whether the specified event or category is met, and the result is expressed as a percentage of participants. The record says response was assessed by CT or MRI using RECIST 1.0, after cycle 2 and repeated every 2 cycles up to Cycle 10, up to 26 months.

RECIST provides rules for classifying tumor changes. In the registry description, complete response means disappearance of all target lesions; partial response requires at least a 30% decrease in the sum of the longest diameters of target lesions from baseline. Progressive disease is described as at least a 20% increase from the smallest recorded sum or the appearance of new lesions. Stable disease is neither sufficient shrinkage for partial response nor sufficient increase for progressive disease.

The endpoint label and its accompanying definition do not align cleanly: the definition lists complete response, partial response, stable disease, or progressive disease as qualifying categories. Since stable disease and progressive disease are not ordinarily response categories, the precise numerator intended by the endpoint label is ambiguous. The registry record does not report the observed percentage or clarify this wording. Response endpoints can summarize tumor change, but a percentage alone does not describe how long a response lasts or establish a treatment effect relative to a comparator.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The study used a single-group, non-randomized design. That structure determines the kinds of statistical claims the results can support. A percentage within the treated group can describe how often a defined outcome was recorded in that group. Without a comparator, the percentage is not a randomized estimate of the difference attributable to treatment. General statistical interpretation of a single-group outcome should therefore focus on the observed group and the uncertainty around its estimate, while recognizing that uncertainty intervals do not remove design limitations such as confounding or selection.

What does a percentage endpoint represent?

For a binary endpoint, participants are classified as having or not having the event specified by the endpoint definition. The percentage is based on the number meeting that definition relative to the analysis population. Its meaning depends on both the numerator and the denominator, and on how participants with missing or unevaluable assessments are handled. The registry record gives the stated primary analysis population, but does not report the number meeting the primary endpoint or the missing-data rules. The response percentage therefore cannot be reconstructed from the record.

A confidence interval is a statistical range expressing uncertainty around an estimated quantity under the assumptions of the method used. For a percentage, its width is affected by the observed event count and the number of participants contributing to the estimate. It describes sampling uncertainty, not all possible sources of bias. No confidence interval or method for calculating one is reported for the primary response endpoint.

How should response classifications be understood?

The record says that tumor assessments used CT or MRI and RECIST 1.0. Its description defines complete response as disappearance of all target lesions, partial response as at least a 30% decrease in the sum of longest diameters relative to baseline, and progressive disease as at least a 20% increase from the smallest recorded sum or the appearance of new lesions. Stable disease is described as neither enough shrinkage for partial response nor enough increase for progressive disease. These categories distinguish different patterns of tumor change.

For a response-rate calculation, the categories included in the numerator must be clear. The endpoint title refers to overall response rate, but the written definition says participants achieving complete response, partial response, stable disease, or progressive disease. Since these categories do not all represent response in the usual RECIST usage, the endpoint's operational meaning is uncertain. The record does not clarify the discrepancy or state the number in each category.

What can the study design establish?

Randomization is a way to create treatment groups that are comparable in expectation, helping support causal comparisons. This study was non-randomized and single-group, so it has no randomized control group against which to estimate a treatment contrast. Even if the primary percentage had been reported, it would describe participants in this study rather than establish what would have happened to comparable participants under another treatment.

Open-label status means masking was listed as none. General statistical considerations are that knowledge of treatment assignment can influence some parts of trial conduct or assessment, depending on how outcomes are collected. Imaging-based assessments may use defined criteria, but the registry record does not state whether image review was blinded or describe any steps taken to reduce assessment bias. No assumptions about such procedures can be made from the record.

What is known about the analysis?

The record identifies the primary endpoint population as all participants who received at least one dose of study drug, with n=50. It does not provide a statistical analysis plan, the analysis method for the primary endpoint, a confidence interval method, a hypothesis-testing procedure, or details about adjustment for other factors. Those methods are not reported in the registry record. The registry listing records 0 analyses; that listing alone does not establish whether analyses were performed outside the registry or how they were conducted.

There are 10 outcome measures listed, including efficacy, safety, and pharmacokinetic outcomes. When multiple outcomes are examined, the chance of at least one apparently unusual result can increase if many tests are performed. General methods for addressing that issue include pre-specifying a primary endpoint and adjusting for multiple testing when appropriate. The registry record does not report a multiplicity strategy or which secondary outcomes, if any, were included in formal statistical testing.

Time-to-event outcomes such as duration of response or time-to-progression generally require information about event times and censoring. Censoring occurs when follow-up ends before the event is observed, so the exact event time is unknown. Methods for such outcomes account for that incomplete timing information rather than treating every participant as if the event were either observed or absent at a fixed time. The registry lists time-to-response, duration of response, and time-to-progression outcomes, but does not report their estimates, analysis methods, or handling of censoring. These are general methodological points, not claims about methods used in this study.

Safety summaries also require careful interpretation. Counts of participants with adverse events describe how many participants had recorded events, but event counts and participant counts are different quantities: a participant may experience more than one event. The registry reports participant counts for serious and other adverse events and lists selected serious-event terms with percentages. It does not report detailed event-level exposure time or a comparative safety analysis. These summaries can describe what was recorded, but by themselves cannot show that treatment caused an event or quantify a comparative risk.

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("SB-715992")
n      <- c(50)
events <- c(4)
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  SB-715992  50  4  8.0  2.2  19.2  3.2  18.8  3.3  18.9

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 10 outcome measures, spanning response, time-to-event, safety, laboratory or clinical concern values, toxicity shifts, and pharmacokinetic clearance. Considering multiple outcomes can create multiplicity: the more results examined, the greater the general possibility that some may appear notable by chance. A prespecified primary endpoint can help focus interpretation, and statistical plans may specify how secondary outcomes are handled. The registry record does not report a multiplicity adjustment, a hierarchy among secondary outcomes, or a testing strategy.

The record lists 0 analyses. This registry listing does not establish whether analyses were performed elsewhere; the methods and results of any such analyses are not reported in the registry record. The primary endpoint percentage, its uncertainty, and inferential test results are also not reported.

The primary analysis population is described as participants who received at least one dose of study drug, with n=50. The record does not report whether any treated participants lacked evaluable imaging, how missing response assessments were handled, whether participants who discontinued treatment remained in the endpoint analysis, or how follow-up loss was addressed for time-to-event outcomes. Those omissions matter because estimates can depend on which participants contribute data and how unavailable outcomes are treated.

The term intent-to-treat is used for this population, but the stated definition is based on receiving at least one dose. In a non-randomized single-group study, that definition should be read as written rather than assumed to include everyone enrolled regardless of treatment exposure. The registry record does not provide further detail about enrollment exclusions or analysis-set construction.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: febrile neutropenia (10.0%), vomiting (6.0%), neutropenia (4.0%), diarrhoea (4.0%), chest pain (4.0%).

For the SB-715992 group, the registry reports 50 participants at risk for both serious and other adverse-event summaries. It lists 23 participants affected by serious adverse events and 42 affected by other adverse events. It also lists 11 deaths. These are descriptions of the recorded study group; the record does not establish that ispinesib caused any particular event or death.

The selected serious adverse-event terms and reported percentages are febrile neutropenia, 10.0%; vomiting, 6.0%; neutropenia, 4.0%; diarrhoea, 4.0%; chest pain, 4.0%; and pyrexia, 4.0%. The record does not give further detail about event timing, duration, severity beyond the serious-event classification, or attribution to study drug. The percentages should be read as the registry's reported figures, not as comparative estimates.

Participant counts and event counts are not interchangeable: a participant may experience more than one adverse event. The registry summaries do not provide a control group, so they cannot quantify how these outcomes compare with another treatment. They describe safety events recorded in this study and should not be interpreted as proof of causation or as a complete account of every clinical consideration.

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

The registry record establishes that this was a completed, open-label Phase II study of ispinesib in females with advanced or metastatic breast cancer. It identifies the primary endpoint as a percentage-based tumor response outcome, assessed with CT or MRI using RECIST 1.0, and states that the analysis population included participants who received at least one dose. It also reports group-level adverse-event counts and selected serious-event terms.

The record does not report the observed primary response percentage, the number of participants in each response category, a confidence interval, or a statistical test. Therefore, the primary efficacy result cannot be described quantitatively from the registry record. The ambiguity between the response-rate label and the listed qualifying categories further limits interpretation of what the intended endpoint numerator was.

Even a fully reported response percentage from this study would be a single-group description, not a randomized comparison. The design has no comparator, so it cannot by itself separate treatment effects from other influences on outcomes. Open-label conduct also means masking was listed as none; the record does not say whether image assessment was blinded.

The safety counts show that serious and other adverse events, as well as deaths, were recorded. They do not establish causation, and the lack of a comparator prevents a like-for-like estimate of how frequently such events occurred relative to another treatment group. Likewise, listed secondary outcomes cannot be interpreted as results where the registry record does not report estimates or analyses.

In short, the registry supports a description of the trial question, design, endpoint criteria, and selected safety summaries. It does not provide the primary outcome result or enough analysis detail to judge its statistical precision, and it does not support a comparative treatment-effect conclusion.

14. Why This Trial Matters Statistically

This study is statistically informative as an example of why a clearly defined endpoint and an appropriate comparison structure matter. A percentage outcome appears straightforward, but its meaning depends on exactly which participants count as having the event. Here, the endpoint is labeled overall response rate while its definition includes stable disease and progressive disease as well as complete and partial response. That mismatch makes the intended event definition uncertain, even though the record names an established imaging assessment framework.

The single-group, non-randomized design also shapes what can be learned. A treated-group percentage can describe outcomes in the study population, but there is no randomized comparator to anchor a treatment contrast. Statistical precision, even if reported, would address uncertainty in the group estimate; it would not correct the lack of a comparison group or resolve other design limitations.

The record further illustrates the difference between listing outcomes and reporting results. It identifies 10 outcome measures and several kinds of efficacy, safety, and pharmacokinetic questions. Yet the primary response estimate and its uncertainty are not reported in the registry record, and the analyses listing records 0 analyses. Readers therefore cannot use this record alone to evaluate the numerical primary result or the statistical handling of secondary outcomes.

Finally, adverse-event counts are descriptive and need careful wording. Serious events and deaths are important recorded outcomes, but counts from a single treatment group do not establish attribution or comparative safety. Distinguishing observation from causal interpretation is a core statistical discipline, especially when no control group is available.

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 the registry information for this study. The primary response result is not reported in the registry record. Neither the response numerator nor a confidence interval or test result is available, so the outcome's magnitude and precision cannot be assessed. The definition also creates an endpoint ambiguity by listing stable disease and progressive disease as qualifying categories for an endpoint labeled overall response rate.

The study was non-randomized, open label, and single group, with no comparator. Consequently, the registry record does not support a like-for-like treatment comparison. The record does not describe masking of imaging assessment or other procedures that might reduce assessment bias. It also does not report a statistical analysis plan, sample-size justification, or the methods used for endpoint analysis.

The primary analysis population is defined as participants who received at least one dose, but details on missing scans, withdrawal, follow-up completeness, or handling of unevaluable outcomes are not reported in the registry record. The listed secondary outcomes are not accompanied by results or analysis details. The record lists 0 analyses, which does not establish whether analyses were done outside the registry.

Safety reporting is also limited for interpretation: the group counts and selected serious-event terms do not provide a comparator, event timing, exposure-adjusted rates, or a basis for assigning causality. These limitations concern what the registry record reports; they do not imply that additional information was not collected during the study.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Study Of Ispinesib In Subjects With Breast Cancer — NCT00089973.
• Registry record last updated 2018-02-26; status last verified 2017-08.
• Results first posted 2018-02-26.
• 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.

19. Record Summary

NCT00089973 is a completed Phase II, open-label, non-randomized, single-group study of ispinesib in females with advanced or metastatic breast cancer. The sponsor is GlaxoSmithKline. The recorded study start date was 2004-06-30, and the primary completion date was 2006-08-25. Enrollment was 50.

The primary endpoint was labeled percentage of participants with overall response rate, assessed by CT or MRI using RECIST 1.0 after cycle 2 and repeated every 2 cycles up to Cycle 10, up to 26 months. The stated primary population consisted of participants who received at least one dose, with n=50. The primary response result is not reported in the registry record, and the endpoint definition includes categories that make its intended numerator unclear.

The registry lists 10 outcome measures and 0 analyses. For the SB-715992 group, it reports 23 participants affected by serious adverse events, 42 by other adverse events, and 11 deaths. These are descriptive counts from a single treatment group and do not establish causation or a comparative treatment effect.