Open Label Extension In Cancer Patients: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Open Label Extension In Cancer Patients (NCT00830180), a non-randomized phase 2 study. The primary outcome is Change From Parent Study (A4091003 [NCT00545129]) Baseline in Brief Pain Inventory-Short Form (BPI-sf) Average Pain Score for at Weeks 4, 8, 16, 24, 40, 48, 56 and 64.

2. Clinical Question

This question concerns change in a participant-reported pain measure over time during an open-label extension. The record identifies assessment points at Weeks 4, 8, 16, 24, 40, 48, 56 and 64, as well as the parent-study and extension-study baselines. The questionnaire asks about pain during the 24 hour period before evaluation, and the average pain score is on a scale from 0 to 10, with lower scores indicating less pain.

The study was not a randomized comparison of treatment groups. It therefore does not estimate a treatment contrast against a concurrent control group. The stated aim was to evaluate safety and efficacy among people from the parent study who wished to receive open-label therapy. The registry record does not report the numerical changes, uncertainty estimates, or statistical tests needed to assess the primary question quantitatively.

3. Trial Design

This was a Phase 2, non-randomized, open-label safety extension study. Its official title identifies tanezumab and cancer patients with pain due to bone metastases. The brief summary says participants had taken part in the double-blind Study A4091003 and wished to receive open-label therapy. The intervention is named Anti-NGF AB; the safety group is identified as Tanezumab 10 mg (A4091029).

The registry lists enrollment as 41, a study start date of 2009-10-29, a primary completion date of 2013-02-14, and a completed status. These are registry-level design and administrative details. Enrollment should not be assumed to equal the number evaluable for every outcome or assessment: the primary outcome description distinguishes participants evaluable for the outcome overall from those evaluable at each assessment.

Non-randomized allocation and absence of masking matter for interpretation. There is no concurrent randomized arm described as a control, and participants and study personnel were not masked. Changes observed over time can describe the participants’ recorded pain scores during the extension, but this design by itself cannot separate treatment effects from other influences on those scores. The registry record does not report a control-group comparison.

The primary measure was scheduled repeatedly from the extension-study baseline through Week 64, with the parent-study baseline also identified in the measure title and timeframe. The record gives no further details of visit windows, treatment discontinuation, or how measurements outside scheduled visits were handled. It also does not report a statistical analysis plan or numerical primary-outcome results. Those omissions limit what can be concluded about the analysis and its precision.

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: The Intent to Treat (ITT) Population: all participants who received at least day 1 dose of IV study medication during Study A4091029. Here 'overall number of participants analyzed signifies participants evaluable for this outcome measure and 'number analyzed' signifies participants evaluable at each specified category of specified time point.

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: The intent to treat (ITT) analysis set included all participants who received at least 1 of IV study medication during study.

6. Primary Endpoint

The registry lists the primary outcome as Change From Parent Study (A4091003 [NCT00545129]) Baseline in Brief Pain Inventory-Short Form (BPI-sf) Average Pain Score for at Weeks 4, 8, 16, 24, 40, 48, 56 and 64, measured over Baseline of Study A4091003, Baseline (A4091029) and Weeks 4, 8, 16, 24, 40, 48, 56 and 64 of Study A4091029.

The primary outcome is change in the Brief Pain Inventory-Short Form average pain score. The BPI-sf is a self-administered questionnaire about pain severity and its impact on daily functions during the 24 hour period before evaluation. It contains 5 questions. Questions 1 to 4 assess pain at its worst, least, average, and right now; responses use an 11-point numeric rating scale anchored at 0 for no pain and 10 for pain as bad as you can imagine.

The registry identifies the average pain score as the primary measure. Its possible values range from 0 to 10, with lower scores indicating less pain. The reported change is oriented so that a negative score indicates a decrease in pain and a positive score indicates an increase. Because the measure asks participants to report their own experience, it captures a patient-reported aspect of the condition rather than an observer’s rating.

Repeated assessment can describe how scores vary during the extension. The record names the parent-study baseline, the extension-study baseline, and assessments at Weeks 4, 8, 16, 24, 40, 48, 56 and 64. The registry does not report the resulting mean changes or their standard deviations. Consequently, the direction, size, and precision of the observed changes cannot be assessed from the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes the primary outcome as a mean continuous measure with standard deviation. The following explains these statistical concepts generally; the registry record does not report the detailed methods used for this study’s outcome analyses.

What does a mean change describe?

For a continuous outcome, a mean is the arithmetic average of the observed scores or changes among the participants included in an analysis. A mean change compares a participant’s score at a follow-up assessment with a designated baseline score, and then summarizes those changes across the analyzed group. The baseline reference matters. The registry identifies both the parent-study baseline and the extension-study baseline in the outcome title and timeframe, but does not provide numerical results showing how each reference was used in the reported analysis. The exact analysis specification is not reported in the registry record.

A group mean is a summary, not a description of every participant’s experience. Participants may have changes in different directions or of different magnitudes even when the group mean is positive or negative. The registry does not report individual-level results or the distribution of changes. It is therefore not possible to tell how much individual responses varied or whether the mean would represent a typical participant’s change.

What does a standard deviation describe?

In general, a standard deviation describes how spread out observed values are around their mean. A larger standard deviation indicates greater variation among the observations, while a smaller one indicates less variation. It does not directly quantify how precisely the group mean has been estimated. Precision depends on the variability and on how many observations contribute to the particular estimate; the registry record does not report the outcome-specific counts or standard deviations at each visit.

A standard deviation should not be read as a confidence interval or as a range containing every participant’s score. The BPI-sf average pain score is bounded between 0 and 10, so its distribution is restricted by the scale limits. The registry gives the scale anchors and states that lower scores indicate less pain, but does not provide distributions with which to assess the shape or spread of observed scores.

What would a confidence interval add?

Generally, a confidence interval gives a range of values compatible with an estimate under a specified statistical model and sampling procedure. It helps readers assess precision: a narrower interval indicates greater precision than a wider interval, all else equal. An interval does not remove bias from an unsuitable design, measurement process, or missing-data mechanism. No confidence interval for the primary outcome is reported in the registry record.

A p-value, if reported, would address compatibility between data and a specified null hypothesis under the assumptions of a test. It would not by itself show the size or clinical importance of a change, prove that a treatment caused a change, or convey precision as fully as an estimate with an interval. The registry record reports no primary-outcome p-value or test result.

Why does the design affect the analysis?

In a randomized comparison, random allocation can help balance measured and unmeasured factors between treatment groups on average. This extension was non-randomized and open-label, and no concurrent comparator is described. Analyses of change within the treated group can describe the observed trajectory, but cannot by themselves estimate what would have happened to the same participants under a different treatment or without the study treatment. Open-label awareness may also influence reports of a subjective outcome. These are general design considerations, not claims that a particular bias occurred in this study.

The repeated assessments are observations over time. General methods for repeated measures can account for correlation among scores from the same participant, but the registry record does not state whether such a method was used. Nor does it report whether the analysis compared each follow-up with one or both baselines, modeled change over time, or used another approach. No model specification, covariates, or standard error method is reported in the registry record.

How should planning calculations be understood?

In general, sample-size planning uses assumptions about a target effect, variability, allocation, and statistical testing to estimate the number of participants needed for a future study. A planning illustration is not an observed treatment result and is not observed power. The stated planning assumptions for any such illustration are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the target effect expressed as a fraction of the observed effect. The record describes these as planning assumptions rather than the trial’s own design. It does not report the trial’s sample-size justification or a numerical planning result in the registry record.

The description also distinguishes participant counts from total event counts for time-to-event planning calculations. That distinction is general: a participant count describes people allocated to an arm, while an event count refers to outcome events used in a time-to-event calculation. It does not establish that this extension used a time-to-event primary analysis. The primary outcome described in the record is continuous, and the record reports zero analyses.

Over-dispersion is a general concern when observed variability exceeds what a statistical model assumes. If a model understates variability, standard errors may be too small and confidence intervals may be too narrow. This concept is relevant to interpreting models only when their assumptions and data warrant it; the registry record does not identify an over-dispersion model or report whether over-dispersion was assessed. No inference about that issue can be made for this study.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 12 outcome measures and identifies the BPI-sf average pain score as the primary outcome. It reports zero analyses. The record does not state how many comparisons, time points, or outcomes were included in any statistical testing, or whether multiplicity adjustments were planned or used. Having repeated assessment points and multiple outcomes can create multiple opportunities for statistical findings; in general, multiplicity procedures address the increased chance of false-positive findings when many hypotheses are tested. The registry record does not report a procedure or a statistical analysis plan, so the handling of multiplicity is not reported in the registry record.

Missing outcome data can affect estimates when participants do not provide measurements at every assessment. The primary outcome population is described as participants who received at least day 1 dose of IV study medication during Study A4091029. The record distinguishes the overall number evaluable for the outcome from the number evaluable at each assessment, but does not give those counts or describe why observations might be missing. It does not state how missing values were handled, whether an imputation method was used, or whether sensitivity analyses were conducted.

The registry’s population definition should not be taken to mean every enrolled participant contributed to every outcome assessment. Enrollment is reported as 41, while evaluability may vary by outcome and visit. The record does not report outcome-specific denominators or participant flow for the primary endpoint. As a result, the amount and pattern of missingness, and its possible influence on the estimated trajectory, cannot be assessed from the registry record.

These omissions do not show that a particular analysis was inappropriate; they mean that its choices and consequences cannot be evaluated from the registry information. The analysis population, missing-data methods, and multiplicity approach are not reported in the registry record.

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: anaemia (9.8%), prostate cancer (7.3%), disease progression (7.3%), pathological fracture (4.9%), metastatic neoplasm (4.9%).

The safety summary identifies a Tanezumab 10 mg (A4091029) group, with 41 participants at risk in each listed summary. It reports 23 participants with at least one serious adverse event and 18 with another adverse event. These are counts of participants in the reported categories, not estimates of a comparative treatment effect. The registry record gives no concurrent control group for comparison.

The listed serious-event terms include Anaemia at 9.8%, Prostate cancer at 7.3%, Disease progression at 7.3%, Pathological fracture at 4.9%, Metastatic neoplasm at 4.9%, and Prostate cancer metastatic at 4.9%. The record does not clarify whether these terms overlap within participants, so they should not be added to infer a total. A serious adverse event is a reporting category; seriousness is not simply a rating of symptom intensity. That distinction is general and does not establish the cause of any event.

The deaths value is not reported in the registry record. It should not be interpreted as zero deaths. The listed events and percentages describe safety reporting in this group; without a comparison group or further event details, they do not establish causation or a comparative safety profile.

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

The registry record establishes the study context: a Phase 2, non-randomized, open-label extension involving cancer patients with pain due to bone metastases who had participated in a parent study and wished to receive open-label therapy. It also identifies the BPI-sf average pain score, its scale, and scheduled assessments. Those details explain what the primary outcome was intended to measure, but they are not outcome findings.

The primary numerical results are not reported in the registry record. There is no reported mean change, standard deviation, confidence interval, or p-value from which to describe the size, direction, or precision of change in pain scores. This absence does not mean that scores remained unchanged, that no effect occurred, or that the study found no difference. It means the posted record does not supply the results needed to reach those conclusions.

Even if a within-group change estimate were available, the design would need to be considered. Because allocation was non-randomized and the study was unmasked, and because no concurrent comparator is described, a change over time would not alone identify a causal treatment effect. In general, a comparison against an appropriate control can help distinguish treatment-related change from other changes over time. The registry does not report such a comparison for this extension.

The safety counts provide a separate descriptive view of reported events in the Tanezumab 10 mg group. They do not establish that treatment caused an event, and the record does not provide a control group for judging relative event frequency. The death count is not reported in the registry record. The event summaries should therefore be read as reported safety information rather than as a causal or comparative conclusion.

In short, the record describes the research question and selected safety counts, but does not report the primary efficacy results or the analysis details needed for a statistical assessment of that outcome. Neither a positive nor a negative efficacy conclusion can be drawn from the absence of those results.

13. Why This Trial Matters Statistically

This study is statistically informative as an example of an extension design where the main question is descriptive change over time rather than a randomized comparison. The registry identifies a participant-reported pain measure and repeated assessments, which makes clear that outcome interpretation depends on the baseline reference, the participants contributing data at each visit, and the analysis of repeated observations.

The design also illustrates why a measure’s definition and its causal interpretation are separate issues. The BPI-sf average pain score has explicit scale anchors and a stated direction: lower scores indicate less pain, and negative change indicates a decrease. That clarity helps readers understand a reported estimate. It does not, by itself, show that a change was caused by treatment. Non-randomized allocation, open-label assessment, and the absence of a concurrent comparator limit causal conclusions.

The record also highlights the importance of complete reporting. Without numerical outcome estimates, variability, uncertainty intervals, and analysis details, readers cannot judge the size or precision of a change or assess how missing measurements and repeated testing were handled. These are central elements of statistical interpretation, not merely technical details.

Finally, the safety summary demonstrates that event counts and event-term percentages should be interpreted descriptively when no comparison is available. They can report what was recorded in a study group, while leaving causation and relative risk unanswered. The registry record’s value is in defining the study and making some safety information visible, while its missing outcome analyses impose clear limits on inference.

14. Limitations

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

The registry record has several trial-specific reporting gaps. It does not include numerical results for the primary pain outcome, despite identifying the outcome measure and its assessment schedule. Mean changes, standard deviations, confidence intervals, p-values, and outcome-specific participant counts are not reported in the registry record. The primary outcome’s direction and precision therefore cannot be described.

The extension was non-randomized and open-label, with no concurrent comparator described. Any within-group changes would be vulnerable to influences that cannot be separated from treatment exposure using the design information in the record. Because the endpoint is participant-reported, awareness of treatment is also a general interpretive consideration; the record does not establish whether or to what extent it affected responses.

The record does not report how the parent-study baseline and extension-study baseline were incorporated into the primary analysis. It also does not specify the statistical model, handling of repeated assessments, missing-data strategy, sensitivity analyses, or multiplicity procedures. These details are needed to evaluate how the estimates were constructed and how robust they might be.

Safety reporting is limited to selected counts and event terms. The death count is not reported in the registry record, and the record does not establish whether listed serious-event terms overlap among participants. Without a comparator and more detailed event context, these figures cannot establish causality or relative safety. These limitations concern what can be concluded from the registry record and should not be taken as evidence that an unreported result was absent.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Open Label Extension In Cancer Patients — NCT00830180.
• Registry record last updated 2021-04-05; status last verified 2021-02.
• Results first posted 2021-02-02.
• 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

This completed Phase 2 study was a non-randomized, open-label safety extension of tanezumab in cancer patients with pain due to bone metastases. The brief summary describes participants who took part in the double-blind Study A4091003 and wished to receive open-label therapy. Enrollment was 41; the study start date was 2009-10-29 and the primary completion date was 2013-02-14.

The primary outcome was change in BPI-sf average pain score, assessed from the parent-study and extension-study baselines and at scheduled extension-study visits through Week 64. The scale ranges from 0 to 10, with lower scores indicating less pain. Numerical primary-outcome results and statistical analyses are not reported in the registry record.

The safety summary lists 23 participants with a serious adverse event and 18 with another adverse event among 41 at risk. Selected serious-event terms and percentages are reported, while the death count is not reported in the registry record. The non-randomized, unmasked design and absence of a concurrent comparator constrain causal and comparative interpretation.