AZD2171 to Treat Prostate Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for AZD2171 to Treat Prostate Cancer (NCT00436956), a na phase 2 study. The primary outcome is Percent Probability of Participants With 6-month Progression-free Survival (PFS).

2. Clinical Question

In practical terms, the study asked whether participants receiving AZD2171, with prednisone included for one cohort, remained free of progression or death over the specified follow-up period. The design does not establish a randomized comparison between AZD2171 alone and AZD2171 with prednisone. Any difference between those cohorts would be difficult to separate from differences in who entered each cohort, treatment context, or other factors not described in the registry record. The record states that the cohorts were analyzed together for response, rather than presenting them as comparative groups.

3. Trial Design

This was a phase II study of AZD2171 in metastatic androgen-independent prostate cancer. The registry record gives the overall enrollment as 59, describes the model as single group, and lists allocation as not applicable and masking as none. The study is marked completed. Its recorded start date is 2006-10-16, and its primary completion date is 2010-02-01. The sponsor is the National Cancer Institute.

The arm description identifies two cohorts: 35 participants received 20 mg AZD2171 orally daily, and 23 received prednisone 10 mg orally daily with 20 mg AZD2171. These cohort counts are reported as written; the record does not explain their relationship to the overall enrollment count. It says that prednisone was added to relieve toxicity and that cohort response data were analyzed together. This is not a randomized, head-to-head comparison of the regimens.

The study description outlines daily oral treatment in 28-day cycles. Planned assessments included blood testing, magnetic resonance imaging to evaluate blood flow, optional tumor biopsies, clinic visits, and imaging. The brief description is truncated during its account of imaging, so the full schedule is not reported in the registry record. The interventions list also includes dynamic contrast-enhanced MRI and prednisone.

Eligibility required metastatic androgen-independent prostate cancer and prior docetaxel treatment. The registry describes the intended population as men 18 years of age and older. The primary endpoint was a six-month progression-free survival measure. Secondary outcomes listed include adverse events, grade 2 and grade 3 toxicities, median overall survival, median progression-free survival, and response by RECIST. Detailed eligibility criteria, treatment-discontinuation rules, follow-up procedures, and the full statistical analysis plan are not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: One participant was not evaluable owing to the development of a cord compression on day 2 of therapy and was subsequently removed from the trial. Since the prednisone was added to relieve toxicity \& outcome data is based on response, the cohorts were analyzed together in terms of response. No suggestion prednisone significantly altered outcomes.

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 cohorts are grouped together because the second cohort added prednisone to minimize toxicity/side effects.

6. Primary Endpoint

The registry lists the primary outcome as Percent Probability of Participants With 6-month Progression-free Survival (PFS), measured over 6 months.

The primary endpoint is titled the percent probability of participants with 6-month progression-free survival. Progression-free survival is generally used to describe time until disease progression or death; it can summarize whether participants remain alive without documented progression over a defined period. As a general statistical concept, this kind of endpoint combines information about disease control and survival, rather than measuring tumor response alone.

The registry description defines the endpoint as the proportion of subjects who progress or die by 6 months after the start of combined therapy, while the title refers to the probability of progression-free survival at 6 months. Those formulations point in opposite directions: one describes progression or death, and the other describes being free of progression or death. The record does not resolve this wording inconsistency. It says the probability over time was estimated using the Kaplan-Meier method.

Assessment was described as using PSA consensus criteria and RECIST. The record defines RECIST response categories using changes in target lesions and also refers to PSA decline or progression. These criteria are intended to make disease status assessment more systematic, although the operational details of adjudication and the handling of conflicting PSA and imaging findings are not reported in the registry record.

No numerical six-month estimate, confidence interval, or event count for the primary endpoint is reported in the registry record. Therefore, the endpoint definition can be described, but its observed value and precision cannot be evaluated from the record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry says that the primary progression-free survival probability over time was determined using the Kaplan-Meier method. As a general statistical explanation, Kaplan-Meier estimation uses the observed times to an event while accounting for participants whose event status is not known through the full follow-up period. Such participants are censored at their last known follow-up under the method’s assumptions; censoring is not itself counted as progression or death. The registry record does not report the number censored, timing of censoring, or the assumptions used to support the analysis.

A probability estimated at a specified time point is different from a simple count of participants with an event. The estimate can incorporate follow-up of different lengths, whereas a raw event count does not describe when events occurred or how long participants were observed. In general, a confidence interval expresses statistical uncertainty around an estimate: a narrower interval indicates greater precision, while a wider interval indicates less precision. No primary endpoint confidence interval is reported in the registry record, so its precision cannot be assessed.

The endpoint definition refers to PSA consensus criteria and RECIST. RECIST is a standardized approach to classifying changes in measured target lesions. The registry describes complete response as disappearance of all target lesions, partial response as a decrease in the sum of longest diameters relative to baseline, stable disease as neither sufficient shrinkage nor sufficient increase, and progressive disease as a specified increase or appearance of new lesions. The record also refers to PSA decline or PSA progression. How the criteria were combined when they did not agree, and how assessment was reviewed, are not reported in the registry record.

As a general point, outcome definitions and assessment timing affect what an endpoint captures. Imaging-based progression can depend on scan timing and interpretive rules, while PSA-based criteria reflect a laboratory marker. The registry describes monthly MRI for blood-flow evaluation and other study procedures, but its brief description is truncated and does not give the complete tumor-assessment schedule. The full timing and adjudication rules for the primary endpoint are therefore not available in the registry record.

The design is important to the statistical interpretation. A single-group study describes outcomes among participants receiving the study regimen, but it does not by itself separate treatment effects from the natural course of disease, participant selection, or other influences. Random allocation can help balance known and unknown prognostic factors between comparison groups; this study is not recorded as randomized. The two cohorts also differed by prednisone use, and the registry says responses were combined. A between-cohort contrast should not be treated as a randomized treatment comparison.

For general context, over-dispersion occurs when observed variation in data is greater than a statistical model expects. It can make standard errors and uncertainty estimates unreliable if the model does not allow for that extra variation. The registry record does not specify a count model or report an over-dispersion assessment for the primary endpoint. The endpoint is described using Kaplan-Meier estimation, and no additional model details are reported.

The sample-size illustrations use assumptions stated in the registry record: two-sided significance level 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect expressed as a fraction of the observed effect. The record describes these as planning illustrations, not the trial’s own design. For time-to-event calculations, it identifies the total number of events as the planning quantity; other calculations are expressed as participants per arm for 80% power or another stated planning target. These are hypothetical planning quantities, not observed sample sizes or observed power. Because the trial itself is single-group and the illustrations assume equal allocation between two arms, those calculations should not be mistaken for the study’s actual design or its evidential result.

The registry record reports no sponsor statistical analysis result and does not provide enough detail to establish the planned handling of missing observations, covariates, interim looks, or multiplicity. Such details can affect estimates and uncertainty, so their absence limits independent evaluation of the analysis.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary six-month progression-free survival endpoint and secondary outcomes covering adverse events, grade 2 and grade 3 toxicities, median overall survival, median progression-free survival, and RECIST response. The record reports seven outcome measures and zero analyses. It does not describe a multiplicity strategy, such as whether interpretation of secondary outcomes was adjusted for examining several outcomes. The planned statistical hierarchy or any distinction between confirmatory and exploratory outcomes is not reported in the registry record.

Missing-data handling is also not described in enough detail to evaluate. The endpoint population description says one participant was not evaluable after developing cord compression on day 2 of therapy and was removed from the trial. The text then notes that cohorts were analyzed together for response, but the description is truncated. It does not explain whether the unevaluable participant contributed to other outcomes, how incomplete follow-up was handled, or whether any sensitivity analyses were conducted.

The registry record does not report the primary endpoint’s analysis denominator, number of events, censoring details, or confidence interval. Nor does it provide a full analysis-population definition or a statistical analysis plan. These omissions prevent assessment of how missing outcome information and exclusions might affect the estimated progression-free survival probability. They also make it difficult to determine whether the reported outcome population followed the original eligibility and treatment definitions consistently.

11. 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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: death nos (67.8%), fatigue (45.8%), hypertension (44.1%), alkaline phosphatase increased (35.6%), anorexia (35.6%).

The registry’s adverse-event summary reports 59 participants at risk and 59 affected in the serious-event fields, 59 at risk and 56 affected in the other-event fields, and 40 deaths. The summary also lists serious-event terms with percentages: Death NOS, 67.8%; fatigue, 45.8%; hypertension, 44.1%; alkaline phosphatase increased, 35.6%; anorexia, 35.6%; and weight loss, 32.2%. These are registry adverse-event figures, not estimates from a randomized comparison.

The record does not provide enough detail to determine from these figures alone the timing, duration, severity distribution, or relationship of each event to treatment. A reported event does not by itself establish that the study drug caused it. The Death NOS percentage is listed among serious-event terms, while the record separately reports deaths; the information given does not explain the relationship between those entries or provide a cause-specific breakdown. The listed percentages should be read as reported, not recalculated or treated as comparisons between treatment regimens.

The two cohorts are not like-for-like controls, and the record says response analyses combined them. Safety outcomes are not described as a randomized comparison either. The figures therefore describe the registry’s reported adverse-event experience and do not establish comparative risks for AZD2171 alone versus AZD2171 with prednisone.

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

The record documents a phase II study of AZD2171 in metastatic androgen-independent prostate cancer and describes a six-month progression-free survival endpoint. It also reports treatment cohorts, study procedures, and adverse-event summaries. These registry entries provide context for what was studied and how the primary endpoint was intended to be assessed.

What the record does not establish is the numerical primary result. It does not report the estimated probability of progression-free survival at 6 months, its confidence interval, or the associated event count. The record therefore cannot support a quantitative statement about the magnitude or precision of that outcome. Its endpoint wording also alternates between the probability of progression-free survival and the proportion who progress or die, without resolving the distinction.

Because the study was single-group and unmasked, its observed outcomes cannot be interpreted as a randomized estimate of treatment effect against a control. The cohorts differed in prednisone use, and the registry says they were not a like-for-like comparison and were combined for response. The study does not provide a sound basis for attributing any cohort difference specifically to prednisone.

Adverse-event counts and percentages describe what the registry records but do not, by themselves, identify causation. The registry’s summary is not a comparative safety estimate and does not explain all event timing or attribution. Similarly, the sample-size illustrations are planning calculations, not evidence of observed power or a substitute for a reported treatment-effect estimate.

The most defensible reading is limited: the registry describes the trial’s population, intervention structure, endpoint definition, and selected safety figures, but the primary efficacy result and key analysis details are not reported in the registry record. No conclusion about treatment benefit or lack of benefit follows from the information recorded.

13. Why This Trial Matters Statistically

This trial illustrates why a clearly stated endpoint and an available numerical result are both necessary for statistical interpretation. A label such as six-month progression-free survival indicates the intended outcome, but a reader also needs the estimate, its uncertainty, and enough information about the analysis population to assess what the result means. Those primary result details are not reported in the registry record.

It also illustrates the limits of single-group evidence. Outcomes observed during treatment can describe the experience of enrolled participants, but without a randomized control group they cannot isolate the treatment effect from other explanations. The two cohorts do not provide a substitute control comparison: they differed in prednisone use, were not randomized against one another, and were analyzed together for response according to the registry.

The endpoint description contains a substantive wording ambiguity between progression-free probability and the proportion who progress or die. The distinction is important because these represent different event directions. Statistical reporting should make clear whether an estimate describes remaining event-free or experiencing an event, and should connect that definition to the analysis and results.

Finally, the registry lists multiple outcomes and adverse-event summaries without a detailed statistical analysis plan or multiplicity approach. This makes it harder to distinguish prespecified primary inference from exploratory reporting and to evaluate uncertainty across outcomes. The study is therefore a useful example of how design, endpoint definition, analysis transparency, and result reporting jointly determine what readers can infer.

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 main trial-specific limitation for interpretation is the absence of the primary endpoint estimate and its uncertainty in the registry record. Without those, the six-month result cannot be quantified or assessed for precision. The endpoint description itself is inconsistent: the title refers to the probability of being progression-free, but the definition refers to progression or death by 6 months. The record does not clarify which direction the reported probability represents.

The study is recorded as single-group, unmasked, and not randomized. Although the arm description gives separate AZD2171 and AZD2171-with-prednisone cohorts, the registry cautions against treating them as a like-for-like control comparison. It says the cohorts were analyzed together for response because prednisone was added to relieve toxicity. This limits comparisons between those regimens and prevents a controlled treatment-effect interpretation.

The registry reports overall enrollment and cohort sizes, but does not explain their relationship. It notes one participant was not evaluable after cord compression on day 2 and was subsequently removed from the trial; the population description is truncated. The record does not state the primary analysis denominator, full analysis-population definition, missing-data approach, censoring details, or sensitivity analyses.

Detailed statistical methods beyond the Kaplan-Meier description are not reported in the registry record. The timing and adjudication of endpoint assessments, the handling of PSA and RECIST when they differ, and any multiplicity plan are also not specified. These gaps constrain evaluation of both the endpoint estimate and the secondary outcomes. The adverse-event summary does not provide enough contextual detail to infer causation or compare safety between cohorts.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: AZD2171 to Treat Prostate Cancer — NCT00436956.
• Registry record last updated 2018-10-09; status last verified 2018-10.
• Results first posted 2014-07-01.
• 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

NCT00436956 is a completed phase II study sponsored by the National Cancer Institute. It enrolled 59 participants with metastatic androgen-independent prostate cancer. The registry describes daily AZD2171 treatment, with prednisone included in one cohort, and a primary endpoint concerning six-month progression-free survival.

The study is listed as single-group, with no masking and no randomized allocation. The registry says the cohorts were not a like-for-like comparison and were analyzed together for response. It describes Kaplan-Meier estimation for the primary endpoint but does not report the numerical estimate, confidence interval, or event count. The adverse-event summary lists serious and other events and deaths, without establishing comparative risks or causation.

Key details needed for independent statistical evaluation—including the complete analysis population, missing-data handling, full analysis plan, and multiplicity approach—are not reported in the registry record.