A Phase 2 Study With Enzastaurin Plus Chemotherapy or Placebo Plus Chemotherapy for Prostate Cancer Patients: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Phase 2 Study With Enzastaurin Plus Chemotherapy or Placebo Plus Chemotherapy for Prostate Cancer Patients (NCT00466440), a randomized phase 2 study. The primary outcome is Part 2: Percentage of Participants With Objective Tumor Response (Response Rate).

2. Clinical Question

This question is about a difference between two treatment groups on a defined tumor-response measure. It is not a comparison of all outcomes collected in the trial: the record also lists outcomes involving prostate-specific antigen, progression-free survival, overall survival, duration of response, adverse events, and pharmacokinetics. Those outcomes address different aspects of the study and should not be treated as interchangeable with the primary response-rate question.

The registry reports 108 participants enrolled overall. That total is distinct from the group sizes used for the primary Part 2 response results. The primary analysis population is defined by Part 2 participation and receipt of at least one study-drug dose; the registry states that Part 1 was not included in that population.

3. Trial Design

The registry describes a randomized, double-blinded, placebo-controlled phase 2 study with a parallel-group design and quadruple masking. The active comparison was docetaxel plus prednisone with enzastaurin versus docetaxel plus prednisone with placebo. The trial record lists 108 enrolled participants and a completed status. The recorded study start was 2007-06, and primary completion was 2009-08.

The official title describes first-line treatment in hormone-refractory metastatic prostate cancer and includes an enzastaurin maintenance phase. The study also had a Part 1 safety lead-in with pharmacokinetic characterization. The registry states that Part 1 was open-label and was not included in the primary efficacy analysis; the primary response population was drawn from Part 2 and included participants who received at least one dose of study drug.

The regimen descriptions identify docetaxel and prednisone in both Part 2 groups, with enzastaurin or placebo distinguishing the randomized comparison. The record also describes a modified enzastaurin regimen for pharmacokinetic characterization. The full schedule information in the arm descriptions is incomplete, so a complete account of the dosing schedule is not reported in the registry record.

Random allocation and masking are features of the design intended to reduce systematic differences in group assignment and assessment. Those design features do not remove uncertainty from a modest-sized efficacy population, nor do they establish that all outcomes were assessed or analyzed in the same way. Details of the randomization procedure, allocation concealment, and the trial's original statistical design assumptions are not reported in the registry record.

The enrollment total and the analysis-specific group sizes need to be read separately. The primary outcome reports Part 2 groups of 46 and 40 participants, while other outcomes and the safety summaries use their own reported denominators. Differences between those denominators should not be treated as errors or resolved by combining populations unless the record explains how they relate.

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: All enrolled participants from Part 2 group, who received at least one dose of study drug. Per protocol, Part 1 was not included as it is an open-label safety lead-in portion with PK characterization.

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 Part 2: Percentage of Participants With Objective Tumor Response (Response Rate), measured over Baseline up to 3 years.

The primary endpoint is the percentage of participants with an objective tumor response. The registry defines objective responders as participants who met the criteria for complete response or partial response under RECIST. Complete response means disappearance of all target lesions; partial response means a 30% decrease in the sum of the longest diameters of target lesions. The definition also describes progressive disease as a 20% increase in that sum and stable disease as changes that do not meet the other listed criteria.

A response rate reduces these categories to a binary outcome for the primary comparison: a participant either qualified as an objective responder or did not. The resulting group percentages can be compared directly as proportions. As a general statistical point, a binary endpoint can be clinically and analytically straightforward to interpret, but it does not preserve all the detail about the amount or timing of tumor change. It also does not by itself describe how long a response lasted.

The registry gives the assessment timeframe as baseline up to 3 years. The primary analysis population is Part 2 participants who received at least one dose of study drug; Part 1 was excluded because it was an open-label safety lead-in with pharmacokinetic characterization. The registry reports a 90% confidence interval for each group’s response percentage. Those intervals communicate uncertainty in the estimated response percentages, rather than showing the range of individual participant outcomes.

7. Results Posted in the Registry

• Part 2: Docetaxel + Prednisone + Enzastaurin: 15.2% (7/46) versus 15.0% (6/40) on control; risk difference 0.2 percentage points (95% CI -15.0 to 15.4), risk ratio 1.01 (0.37–2.77); unadjusted two-proportion z-test p = 0.978.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Percentage of Participants Exhibiting a Decline in Prostate-Specific Androgen (PSA) From Baseline ≥30% Within First 3 Months of Treatment

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Part 2: Docetaxel + Prednisone + Enzastaurin: 60.7% (17/28) versus 72.4% (21/29) on control; risk difference -11.7 percentage points (95% CI -36.0 to 12.6), risk ratio 0.84 (0.58–1.22); unadjusted two-proportion z-test p = 0.349.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Part 2: Docetaxel + Prednisone + Enzastaurin and the control. The interval gives the range of effects that remain compatible with the data.

Part 2: Progression Free Survival (PFS)

• Endpoint type: Time to event: compared as a hazard ratio.
• Test: Hazard ratio as reported by the sponsor (Log Rank); it was not recalculated.
• Result:
  • Part 2: Docetaxel + Prednisone + Enzastaurin vs Part 2: Docetaxel + Prednisone + Placebo: hazard ratio 0.86 (95% CI 0.50–1.40), reported p = 0.524 (as reported in the registry).
• Interpretation:
  • The 95% confidence interval includes no difference (a ratio of 1), so these data do not show a statistically significant difference between Part 2: Docetaxel + Prednisone + Enzastaurin vs Part 2: Docetaxel + Prednisone + Placebo and the control. The interval gives the range of effects that remain compatible with the data.

Part 2: Overall Survival (OS)

• Endpoint type: Time to event: compared as a hazard ratio.
• Test: Hazard ratio as reported by the sponsor (Log Rank); it was not recalculated.
• Result:
  • Part 2: Docetaxel + Prednisone + Enzastaurin vs Part 2: Docetaxel + Prednisone + Placebo: hazard ratio 0.74 (95% CI 0.30–1.60), reported p = 0.441 (as reported in the registry).
• Interpretation:
  • The 95% confidence interval includes no difference (a ratio of 1), so these data do not show a statistically significant difference between Part 2: Docetaxel + Prednisone + Enzastaurin vs Part 2: Docetaxel + Prednisone + Placebo and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes cover several different kinds of evidence: PSA change, progression-free survival, overall survival, duration of response, adverse events, and pharmacokinetic measures. These outcomes do not all answer the same question as the primary tumor-response endpoint. PSA change concerns a biomarker response, time-to-event outcomes incorporate follow-up timing, and safety outcomes describe harms rather than tumor response.

Among the secondary results with numerical group estimates, the PSA decline outcome and the time-to-event results show uncertainty rather than a uniform, precise separation between regimens. The posted progression-free and overall-survival hazard-ratio intervals are broad, while the group medians also have intervals. Those results are compatible with a range of possible group differences and should be considered alongside the primary endpoint’s imprecision, not as a substitute for it.

The registry lists 12 posted secondary outcomes. When many outcomes are considered, the chance of observing apparently notable results by chance can increase. General statistical practice therefore gives less weight to isolated unadjusted secondary p-values unless the analysis plan specifies how multiplicity is handled. The registry record does not report a multiplicity adjustment or a hierarchy for interpreting the secondary endpoints.

Secondary endpoints can still provide useful context, particularly when they address outcomes different from the primary endpoint. Their interpretation depends on their definitions, analysis populations, follow-up, and uncertainty. The available registry results do not establish that every secondary outcome had the same denominator or was analyzed under one common model. Those details should not be assumed from the primary response analysis.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Percentage of Participants Exhibiting a Decline in Prostate-Specific Androgen (PSA) From Baseline ≥30% Within First 3 Months of Treatment
• Prostate-Specific Androgen (PSA) Velocity at 2 Months
• Prostate-Specific Androgen (PSA) Velocity at 3 Months
• Part 2: Progression Free Survival (PFS)
• Part 2: Overall Survival (OS)
• Part 2: Duration of Response
• Number of Participants With Adverse Events (AEs)
• Part I: Pharmacokinetic (PK) Parameter: Maximum Observed Drug Concentration During a Dosing Interval at Steady State (Cmax,ss) of Enzastaurin, LSN326020, and Total Analyte (Enzastaurin + LSN326020)

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is binary: each participant is classified according to whether they met the complete- or partial-response criteria. For a binary outcome, the response percentage is the number of participants who meet the event definition divided by the number in the analysis population. The registry’s response-rate definition specifies that the denominator is the number of participants qualified for efficacy analysis. It identifies that population as Part 2 participants who received at least one study-drug dose.

What does the risk difference describe?

The recalculated comparison uses an unadjusted risk difference. In general, a risk difference expresses the difference between the event proportions in two groups, on the probability scale. A value of zero corresponds to equal proportions; a positive or negative value indicates which group has the larger observed proportion according to the ordering used for the contrast. The posted estimate is 0.002, and its interval extends from -0.149 to 0.154. Because that interval includes zero and spans both directions, it does not provide a precise estimate of the direction or size of the group difference.

The recalculated result is explicitly an unadjusted comparison based on the posted summary statistics. It is not an adjusted estimate accounting for baseline characteristics or other covariates. No adjusted primary response analysis is reported in the registry record. The recalculated p-value is 0.978; it should be read alongside the effect estimate and its interval, not as a stand-alone verdict.

How do confidence intervals help?

A confidence interval represents uncertainty in an estimated quantity under the statistical method used. In general, a wider interval indicates less precision, while a narrower interval indicates greater precision. An interval that includes the no-difference value does not prove that there is no difference. It indicates that the data and method do not isolate a direction of difference at the interval’s stated level. The registry gives 90% confidence intervals for the group-specific response percentages. The recalculated risk-difference interval is reported without a confidence level in the comparison details, so its level is not reported in the registry record.

How should the sponsor-reported analysis be read?

The registry lists a chi-squared analysis with a parameter described as median difference in final values, an estimate of 0.2, a 90% confidence interval from -12.52 to 12.95, and p=1.000. This parameter label does not match the primary endpoint’s definition as a percentage with objective response. The registry record does not explain how this reported analysis maps to the response-rate endpoint. It is therefore important to report it as a distinct registry analysis and avoid presenting it as the same estimand as the recalculated binary risk difference.

For progression-free survival and overall survival, the registry reports log-rank analyses and hazard ratios. As a general concept, a hazard ratio compares event rates over follow-up under the assumptions of the time-to-event analysis; it is not the same quantity as a difference between medians. A median survival time describes the time by which half of the relevant group has experienced the event, as estimated by the analysis. The registry reports both medians and hazard ratios for these outcomes, but the estimates are imprecise and their intervals are broad.

For a binary endpoint, over-dispersion is not the central issue in the same way it can be for count data. In general, over-dispersion matters when observed count variability exceeds what a simple model assumes, because standard errors can then be too small if the extra variability is ignored. The registry’s primary endpoint is a response proportion, and no over-dispersion adjustment is reported. Other modeling choices, including covariate adjustment, missing-outcome handling, and any sensitivity analyses, are not reported in the registry record.

The trial is randomized and parallel-group, but the registry excerpt does not give the exact statistical analysis plan, the rationale for the selected primary interval level, the original sample-size justification, or whether any interim analyses were planned or conducted. These details cannot be inferred from the posted endpoint results or from the recalculated comparison.

11. Confidence Intervals and Precision

The primary group estimates are 15.2% with a 90% confidence interval from 7.37 to 26.69 for the enzastaurin regimen, and 15.0% with a 90% confidence interval from 6.74 to 27.47 for the placebo regimen. These intervals are broad relative to their point estimates. They show that the response percentages are not estimated with high precision in either group.

The recalculated risk difference is 0.002, with an interval from -0.149 to 0.154. This interval spans values below and above zero. In general, that means the posted data do not precisely locate the response-rate contrast: the estimate is compatible with a difference favoring either group, as well as a value close to no difference. The interval should not be reduced to a binary declaration based on whether it crosses zero.

Confidence intervals apply to the parameter being estimated. The two 90% intervals for separate group percentages do not themselves provide the interval for the difference between those percentages. The comparison interval is the relevant uncertainty measure for the risk difference. No confidence level for that recalculated interval is stated in its comparison details, and it should not be assigned one by assumption.

The registry also reports uncertainty for secondary estimates, including intervals around median times and hazard ratios. These estimates address different parameters and use different analysis methods. Their intervals should be interpreted within those respective outcomes, rather than combined with the primary response interval as if they measured the same effect.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size illustrations are planning calculations based on the observed effect, not the trial’s original design assumptions. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation; the effect to detect is a stated fraction of the observed effect. For time-to-event outcomes, the calculations use the Schoenfeld approach and report total events. For the other outcomes, the figures are participants per arm.

The 80% power column describes the planned group size or event total that would give an 80% chance of detecting the specified effect under those assumptions, if the effect and model assumptions held. The 90% power column describes the corresponding target for a 90% chance. These are planning concepts, not guarantees about what a future study will find. The binary-outcome entries are participants per arm, while the progression-free and overall-survival entries are total events across the comparison, not participants per arm.

The rows use different fractions of the observed effect. Planning for a smaller effect requires more information to distinguish it from no difference, so the listed participant or event requirements increase as the target effect becomes smaller. The figures for the primary response comparison are very large relative to the efficacy group sizes in the posted results, illustrating how uncertain observed effects can translate into demanding planning targets.

The plotted relationship should be read as the same planning principle: lower target effects require larger sample sizes or more events for a given power target. These calculations are not observed power. They are illustrations based on the observed effect and the listed assumptions, and do not reveal the trial’s own original sample-size justification, which is not reported in the registry record.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.150; treatment proportion p₁ = 0.152; average p̄ = 0.151.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.5065 = 0.9927.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.5065 = 0.4263.
4. n = (0.9927 + 0.4263)² / (0.0022)² = 426,032.1, rounded up to 426,033 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
p0 <- 0.15                          # control proportion
p1 <- 0.152173913043478                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(466440)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 426,033 (80% power), 570,336 (90% power)
 75% of observed effect: n per arm = 756,272 (80% power), 1,012,432 (90% power)
 50% of observed effect: n per arm = 1,699,089 (80% power), 2,274,598 (90% power)
power.prop.test: 426032.1 
Simulated power at n = 426,033 per arm: 0.805 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

The registry reports a chi-squared analysis with a parameter labeled median difference in final values. Its estimate is 0.2, its 90% confidence interval runs from -12.52 to 12.95, and its p-value is 1.000. The primary endpoint, by contrast, is defined as the percentage of participants with an objective tumor response. The registry does not explain how the reported median-difference analysis corresponds to that binary endpoint.

The recalculated unadjusted response-rate comparison is expressed as a risk difference of 0.002, with an interval from -0.149 to 0.154 and p=0.978. These figures describe a different parameterization from the registry’s reported analysis. They should not be combined into a single estimate or treated as a direct replication of the sponsor-reported result.

The posted data do not flag a discrepancy between the reported analysis and the recalculation. Even so, their distinct labels and methods warrant transparency: the registry analysis should be quoted in its own terms, while the recalculated result should be identified as an unadjusted calculation from the posted group summaries. A reason for the difference in parameter or analysis is not reported in the registry record.

15. Multiplicity, Missing Data and Analysis Populations

The registry describes a primary response endpoint and lists multiple secondary outcomes, including PSA measures, progression-free survival, overall survival, duration of response, adverse events, and pharmacokinetic measures. Testing or estimating many outcomes raises a general multiplicity concern: when numerous hypotheses are examined, some results may appear notable by chance. A planned hierarchy, adjustment procedure, or distinction between confirmatory and exploratory outcomes can help frame that issue.

A multiplicity strategy is not reported in the registry record. It is also not reported whether the secondary analyses were prespecified in a detailed statistical analysis plan, whether any outcomes were designated exploratory, or how the investigators intended to interpret multiple comparisons. Consequently, unadjusted secondary findings carry less evidential weight than a single prespecified primary comparison and should not be treated as independent confirmation.

The primary population is described as all enrolled participants from Part 2 who received at least one dose of study drug. Part 1 was excluded from that efficacy analysis because it was an open-label safety lead-in with pharmacokinetic characterization. The registry does not report a broader intention-to-treat analysis for the primary endpoint, nor does it explain how participants with missing response assessments were handled.

Missing-data methods, imputation procedures, sensitivity analyses, and reasons for exclusion from outcome-specific analyses are not reported in the registry record. The denominators differ across some efficacy and safety summaries, but the record does not provide enough detail to determine the reasons for every difference. For time-to-event outcomes, details about censoring rules and the handling of incomplete follow-up are not reported in the registry record.

The original analysis plan, including any interim analyses, is not reported in the registry record. These omissions limit the ability to assess how fully the posted analyses reflect the prespecified plan and how robust the findings are to alternative assumptions.

16. 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: febrile neutropenia (14.3%), extravasation (14.3%), thrombosis (14.3%), infection (7.1%), sepsis (7.1%).

The safety summaries report serious adverse events among participants at risk in each group. In Part 1, 8 of 14 participants were listed as affected. In Part 2, 16 of 47 participants in the enzastaurin group and 14 of 41 participants in the placebo group were listed as affected. The record separately lists other adverse events as affecting all participants at risk in each of these groups. These are registry counts, not a randomized comparison adjusted for follow-up or other factors.

The serious adverse event terms with the largest listed percentages are febrile neutropenia, extravasation, and thrombosis, each at 14.3%; infection, sepsis, and oesophageal carcinoma are each listed at 7.1%. The registry excerpt does not provide enough detail to attribute an event to a regimen or to assess its timing, duration, or clinical context. These percentages should be read as descriptive safety reporting, not as proof of causation or a direct estimate of comparative risk.

Death counts are not reported in the registry record. The differing safety denominators should be kept distinct from the primary efficacy populations, and the available summaries alone do not support a full comparative safety assessment.

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

The primary result means that the posted response percentages were similar in the two Part 2 groups, while the estimated difference remains uncertain. The group estimates and the recalculated comparison do not establish a precise direction or size of difference. The interval around the risk difference spans both sides of zero, and the p-value does not provide evidence of a clearly separated response rate under the unadjusted comparison.

This result does not prove that the regimens are equivalent. Equivalence requires a prespecified margin and an analysis designed to assess whether the difference lies within that margin; such a margin and an equivalence analysis are not reported in the registry record. Nor does a non-small p-value show that the treatments have no effect. It indicates that the observed data do not provide a clear distinction under the stated comparison.

The endpoint is objective response under RECIST, not overall clinical benefit, symptom change, quality of life, or long-term survival. The record lists other outcomes, but their definitions and estimates address separate questions. No single endpoint should be taken as a complete account of all possible benefits and harms.

The findings also apply to the registry’s stated Part 2 efficacy population, not automatically to all enrolled participants or to other populations. The record identifies a specific disease setting and regimen comparison. Generalizing beyond that study context requires evidence that is not contained in these results.

The statistical reading is therefore limited to what the estimates support: observed response percentages were close, but uncertainty remains substantial. This is not medical advice and does not determine what treatment is appropriate for any individual. Treatment decisions require clinical context and evidence beyond this statistical comparison.

18. Why This Trial Matters Statistically

This trial illustrates why an effect estimate and its precision should be read together. The response percentages are close, but the uncertainty around the comparison is wide enough to allow differences in either direction. A point estimate alone could suggest a stable conclusion that the available data do not support.

It also illustrates the importance of keeping analysis populations distinct. The trial enrolled 108 participants overall, while the primary response analysis uses the Part 2 population who received at least one dose. Part 1 had a different purpose as an open-label safety lead-in with pharmacokinetic characterization. Safety and secondary outcomes use their own reported denominators, so interpreting every result as if it came from one common analysis set would be misleading.

The registry’s reported chi-squared result is labeled as a median difference in final values, while the primary endpoint is a binary response rate. Comparing the registry analysis with a recalculated risk difference shows why the estimand and method matter: two analyses should not be treated as interchangeable merely because they appear in the same outcome record.

Finally, the planning illustrations make clear that estimates from a small efficacy population can produce demanding future sample-size targets, particularly when planning to detect a smaller fraction of an observed effect. Those targets are not the trial’s own design justification and are not observed power. They demonstrate how uncertainty and the size of the effect one hopes to detect shape the information required for a future comparison.

19. 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.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

The primary efficacy groups include 46 and 40 participants, and the response estimates have broad 90% confidence intervals. That limits precision and leaves considerable uncertainty about the comparison. The registry reports an overall enrollment of 108, but the primary outcome is restricted to the defined Part 2 efficacy population.

The registry’s own primary analysis is labeled as a median difference in final values, whereas the endpoint definition concerns the percentage of participants with objective response. The record does not explain the relationship between that reported analysis and the endpoint definition. The recalculated response-rate contrast is therefore a separate unadjusted analysis rather than a reconstruction of the sponsor’s stated method.

Several details needed to evaluate the analysis fully are not reported in the registry record. These include missing-response handling, outcome-specific exclusion reasons, sensitivity analyses, the original sample-size justification, and the statistical plan for multiplicity. Details of the randomization procedure and any interim analyses are also not reported in the registry record.

The registry lists different denominators across efficacy and safety results. It does not explain all of those differences in the available record. The safety summaries are descriptive, and death counts are not reported in the registry record. These limitations restrict comparisons across outcomes and prevent a complete appraisal of the balance of benefits and harms.

These results are specific to the recorded phase 2 study, its regimens, and its Part 2 population. The record does not support assuming the same estimates for other populations or treatment settings.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Phase 2 Study With Enzastaurin Plus Chemotherapy or Placebo Plus Chemotherapy for Prostate Cancer Patients — NCT00466440.
• Registry record last updated 2020-11-06; status last verified 2020-10.
• Results first posted 2020-11-06.
• 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.

23. Record Summary

The registry describes a completed, randomized, quadruple-masked, placebo-controlled phase 2 study of docetaxel and prednisone with either enzastaurin or placebo for prostate cancer. It reports 108 enrolled participants. The primary Part 2 endpoint was objective tumor response under RECIST among participants who received at least one dose of study drug; Part 1 was an open-label safety and pharmacokinetic lead-in.

The posted primary response percentages are 15.2% and 15.0%. The recalculated unadjusted risk difference is 0.002, with an interval from -0.149 to 0.154 and p=0.978. The results show close observed percentages but substantial uncertainty about the group difference. Secondary outcomes and safety summaries provide additional descriptive information, while important details about missing data, multiplicity, and the original statistical plan are not reported in the registry record.