This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Role of Soy Supplementation in Prostate Cancer Development (NCT00255125), a randomized phase 3 study. The primary outcome is Effect of Soy Isoflavones on Serum Testosterone Levels..
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00255125 |
| Phase | Phase 3 |
| Condition | Prostate Neoplasm |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Triple |
| Enrollment | 86 |
| Arms in the primary outcome | 2 |
| Lead sponsor | VA Office of Research and Development |
| Results posted | Yes |
2. Clinical Question
Population
Patients with prostate cancer who were planning prostatectomy. The record describes evaluation of blood and prostate tissue around surgery, but does not report further eligibility details or a more specific description of the population in the primary outcome entry.
Intervention
A commercial soy supplement, described in the record as soy isoflavones. Participants were assigned to receive it before their planned surgery. The study summary describes a preoperative period of 2-4 weeks; the primary result compares measurements at two weeks.
Comparator
Placebo, assigned in the parallel comparison. The primary result contrasts the two arms at the two-week time point. The record does not provide further information about the placebo's composition or how it was matched to the supplement.
Primary question
What were serum total testosterone levels, measured in ng/ml, after two weeks on soy supplement or placebo? The registry posts the mean and an 80% confidence interval for each arm. It does not post a primary between-arm estimate, confidence interval for the contrast, or p-value, so the arm means alone do not quantify the uncertainty in the treatment contrast.
3. Trial Design
This was a randomized, triple-masked, parallel-group phase 3 study of a soy supplement and placebo in patients with prostate cancer. The registry lists an enrollment of 86 and identifies the sponsor as the VA Office of Research and Development. The study began in 2005-09, reached primary completion in 2009-04, and is recorded as completed. These record entries describe the trial's design and status; they do not establish how masking was implemented or whether it remained intact.
Participants were assigned to soy supplement or placebo before planned prostatectomy. The study summary describes treatment for 2-4 weeks before surgery. The primary outcome description says that total serum testosterone was measured at enrollment, after two weeks on assigned treatment, and again just before prostatectomy. Because the date of surgery varied, the later preoperative time point could vary as well. The reported primary comparison is specifically described as the two-week comparison, rather than a comparison at the variable preoperative time point.
The primary result gives 42 participants in the soy supplement arm and 44 in the placebo arm. It reports an arm mean and an 80% confidence interval for each. The registry lists estrogen receptor status and molecular effects of supplementation as secondary outcomes. Numerical results are posted for estrogen receptor status; the molecular-effects outcome is listed without a reported result. Although the study title refers to prostate cancer development or prevention, the primary result presented is a short-term serum biomarker, not a reported comparison of cancer development.
The registry record does not report the allocation sequence method, concealment procedures, masking procedures, eligibility criteria in detail, or the protocol's sample-size justification. It also does not report the rationale for selecting the treatment duration or the two-week primary time point. Those aspects cannot be reconstructed from the design labels and summary alone.
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
| Stage | Arm Soy Supplement | Arm Placebo |
|---|---|---|
| STARTED | 42 | 44 |
| COMPLETED | 42 | 44 |
| NOT COMPLETED | 0 | 0 |
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: Prostate cancer patient, randomized between to study groups; all patients with early stage prostate cancer.
| Characteristic | Unit | Arm Soy Supplement | Arm Placebo | Total |
|---|---|---|---|---|
| Participants analysed | 42 | 44 | 86 | |
| Age, Continuous | Years | 62 (SD 12) | 62 (SD 7) | 62 (SD 12) |
| Sex/Gender, Customized: Male | participants | 42 | 44 | 86 |
| Ethnicity (NIH/OMB): Hispanic or Latino | Participants | 1 | 2 | 3 |
| Ethnicity (NIH/OMB): Not Hispanic or Latino | Participants | 41 | 42 | 83 |
| Ethnicity (NIH/OMB): Unknown or Not Reported | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): American Indian or Alaska Native | Participants | 1 | 4 | 5 |
| Race (NIH/OMB): Asian | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): Native Hawaiian or Other Pacific Islander | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): Black or African American | Participants | 3 | 9 | 12 |
| Race (NIH/OMB): White | Participants | 38 | 31 | 69 |
| Race (NIH/OMB): More than one race | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): Unknown or Not Reported | Participants | 0 | 0 | 0 |
| Region of Enrollment: United States | Participants | 42 | 44 | 86 |
6. Primary Endpoint
The registry lists the primary outcome as Effect of Soy Isoflavones on Serum Testosterone Levels., measured over Two weeks.
| Element | Registry entry |
|---|---|
| Result type | Mean |
| Measure of spread | 80% Confidence Interval |
| Unit | ng/ml |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint is total testosterone in serum, measured in ng/ml. The outcome description identifies measurements at enrollment, after two weeks on soy supplement or placebo, and just before prostatectomy. The posted result is for the two-week time point and compares the assigned arms. The reported values are means, each accompanied by an 80% confidence interval.
A blood measure can provide a quantitative way to assess a biological response over a defined period. In this trial, the endpoint addresses whether average serum testosterone differed between the randomized groups after two weeks of assigned treatment. It is not itself a measure of prostate cancer development, cancer progression, or a clinical outcome after surgery. The study summary also mentions assessment of prostate tissue, but the primary endpoint result does not report a tissue-based treatment contrast.
The registry record does not explain why testosterone was selected as the primary endpoint, how the endpoint was chosen in relation to the study's prevention title, or whether another measurement or time point was designated as primary in the protocol. The posted arm-level means are 376.0 ng/ml for soy supplement and 405.0 ng/ml for placebo. Their 80% confidence intervals are 124.0 to 528.0 and 232.0 to 576.0, respectively. These intervals describe uncertainty around each arm's mean; they are not a confidence interval for the difference between arms.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
At two weeks, the posted mean serum testosterone level was 376.0 ng/ml in the soy supplement arm and 405.0 ng/ml in the placebo arm. The corresponding 80% confidence intervals were 124.0 to 528.0 and 232.0 to 576.0. These are arm-specific summaries. The registry record does not report a primary treatment-effect estimate, a confidence interval for the difference between arms, or a p-value for this comparison.
The two intervals overlap. That observation is not a formal test of the between-arm difference: the intervals describe uncertainty in the separate arm means, while the treatment question concerns uncertainty in a contrast between those means. A direct contrast would use the randomized comparison and its uncertainty. Without that reported estimate or a documented primary analysis, the arm-level summaries do not establish the size or precision of the treatment contrast. Nor does the absence of a reported primary p-value establish that the groups were equivalent.
The intervals are 80% confidence intervals, not 95% confidence intervals. A confidence interval is a range produced by a statistical procedure to express uncertainty in an estimate. In repeated use under the procedure's assumptions, intervals at a stated confidence level have a corresponding coverage property. An 80% interval is less inclusive than a 95% interval constructed from the same data and method, so its endpoints should not be read as especially stringent bounds. The registry does not describe the calculation method or assumptions behind these intervals.
The trial randomized 86 participants, with 42 in the soy arm and 44 in the placebo arm for this outcome. Those counts indicate the recorded arm sizes for the posted result; they do not by themselves show whether all randomized participants contributed data or how missing measurements were handled. The analysis population and missing-data procedures are not reported in the registry record.
The endpoint is a short-term biomarker measured after two weeks. The result therefore addresses the reported mean serum testosterone levels at that time point, not long-term effects or prostate cancer prevention. The study summary describes a broader interest in prostate and blood effects, but those aims do not change the scope of the primary result. The record's title should not be taken as evidence that a cancer-incidence or cancer-progression outcome is reported.
The registry reports an estrogen receptor status secondary result with an unadjusted p-value of 0.935, along with a mean difference estimate and confidence interval. That secondary result does not substitute for a missing primary treatment-effect analysis. It is also one of multiple listed secondary questions, and the registry record does not report a multiplicity adjustment. The evidence from secondary outcomes should therefore be viewed as exploratory rather than as a confirmatory answer to the primary question.
Overall, the posted primary data describe the two arm means and their 80% confidence intervals. They do not provide enough information to quantify the primary between-arm effect or its precision directly. Interpretation should remain limited to those descriptive results and to the specified short-term measurement.
9. Statistical Methods Explained
The primary outcome is presented as an arm mean with an 80% confidence interval. A mean describes the central average of the measured values in a group. A confidence interval expresses uncertainty around an estimate under a statistical procedure and its assumptions. It does not show the full distribution of participant values, and it is not a range containing a stated proportion of individual measurements. The registry does not report the method used to construct the primary intervals, so their detailed assumptions cannot be assessed from the record.
Why is the treatment contrast different from the two arm means?
The clinical comparison is between the assigned soy supplement and placebo groups. A treatment contrast expresses how the outcome estimates differ between those groups and should be accompanied by uncertainty for that contrast. Separate confidence intervals around each mean answer a different question: how precisely each arm mean has been estimated. Their overlap or non-overlap is not a substitute for a confidence interval or test of the contrast. The registry record does not report a primary contrast estimate or a p-value, so those quantities cannot be recovered as an official trial analysis from the posted primary summaries.
The primary result uses an 80% confidence level. Confidence levels describe the long-run coverage of an interval-building procedure if its assumptions hold and the procedure is repeated. They do not assign a probability to a particular fixed parameter after the interval has been calculated. The interval level matters when judging precision: an interval constructed at a lower confidence level is generally narrower than one built from the same data using a higher level. This is general statistical context, not a claim about the undocumented calculation method used for the registry's primary intervals.
What does the secondary comparison add?
For estrogen receptor status, the registry reports group means and standard deviations, as well as an unadjusted mean difference estimate of -1.0 pg/mL. Its confidence interval is -25.11 to 23.11 pg/mL, and the unadjusted p-value is 0.935. The estimate is the soy-supplement mean minus the placebo mean as represented in the posted comparison. The interval includes values on both sides of zero, indicating that the data are compatible with a range of possible contrasts under the analysis used. The p-value summarizes how unusual data at least as incompatible with a zero-difference null hypothesis would be under that test's assumptions; it is not the probability that the null hypothesis is true.
The registry labels this secondary outcome as a mean with standard deviation. Standard deviation describes variation among observations, whereas a confidence interval describes uncertainty in an estimate. They are not interchangeable. The record also gives a standardized effect estimate, d = -0.018. Standardized effects express a mean contrast relative to a measure of variability, which can aid comparison across measurement scales, but do not remove uncertainty or establish clinical importance. The reported secondary analysis is unadjusted; the registry does not identify a correction for testing multiple outcomes.
How should planning calculations be read?
The record includes sample-size planning illustrations based on fractions of the observed secondary effect. They use a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. These are planning scenarios, not the trial's original design justification and not observed power. The stated assumptions say that the effect to detect is a stated fraction of the observed effect. Such calculations can illustrate how a target effect and variability influence a future study's required enrollment, but using an effect observed in a small trial as a planning input can be unstable. The registry does not report the assumptions used for the original design, its target effect, or a prospective power calculation.
Sample-size planning is distinct from observed power. Once a study has been conducted, a post hoc power calculation based on the observed estimate largely recasts the estimate and its uncertainty; it does not add independent evidence. A planning illustration is most useful as a conditional scenario: if the assumptions and target effect are appropriate, it indicates the approximate scale of a future study. It should not be treated as a property of the completed trial or as evidence that the trial could or could not detect an effect.
What statistical details remain unknown?
The record does not report a detailed primary statistical analysis plan, the primary contrast model, covariate adjustment, handling of repeated measurements, or assumptions used for the primary confidence intervals. It does not report an analysis population definition, methods for missing observations, or an interim-analysis plan. The number of formal analyses listed is zero, even though the record contains arm-level outcome results and a secondary comparison. That distinction makes it important not to infer a complete protocol analysis from the displayed outcome entries.
General statistical considerations also include distribution shape, outliers, and departures from model assumptions. For example, if measurements are highly skewed or contain influential extreme values, a mean and a normal-approximation procedure may not capture the data well; alternative summaries or robust methods may be considered. These are general considerations, not evidence that such problems occurred in this study. The record does not report participant-level data or diagnostic checks, so the distribution and suitability of any particular model cannot be assessed.
Multiplicity is another general concern. When several outcomes or comparisons are examined, the chance of at least one apparently unusual result can increase if each is tested without adjustment. The registry lists secondary questions and an unadjusted secondary p-value, but does not report a multiplicity strategy or a prespecified hierarchy. Accordingly, the secondary result should not be treated as if it were an isolated confirmatory test.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists the primary testosterone outcome and secondary questions concerning estrogen receptor status and molecular effects. It reports an unadjusted p-value for the estrogen receptor comparison. A plan for controlling the chance of false-positive findings across multiple outcomes or analyses is not reported in the registry record. The record also does not describe a hierarchy that would distinguish confirmatory tests from exploratory ones, or explain whether particular secondary analyses were prespecified.
This matters because each additional test creates another opportunity for a result to appear unusual by chance. Without a reported multiplicity strategy, the unadjusted secondary p-value should not be interpreted as though it had been protected against the full set of possible outcomes and analyses. This does not establish that a correction was not used outside the registry entry; it means the registry record does not document one. The number of analyses listed is zero, while outcome summaries and a secondary comparison are present, so the formal analysis record is incomplete.
Missing-data methods are also not reported in the registry record. Although the primary result includes arm counts, the record does not state whether these are all randomized participants, all participants with a two-week measurement, or the analysis population after any exclusions. It does not describe missing measurements, reasons for missingness, imputation, or sensitivity analyses. Randomization supports a fair comparison at assignment, but incomplete or selectively available outcome data can affect how well the analyzed groups preserve that comparison.
The analysis population is not specified. The registry does not report whether the primary result follows an intention-to-treat approach, a per-protocol approach, or another defined population. It also does not report interim analyses, stopping rules, or the detailed statistical plan. These omissions limit assessment of prespecification and of whether the posted summaries reflect all randomized participants and intended analyses. They should be treated as reporting gaps, not as evidence that a particular method was or was not used.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Arm Soy Supplement | 2 | 42 | 4.8% |
| Arm Placebo | 6 | 44 | 13.6% |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Arm Soy Supplement | 9 of 42 |
| Arm Placebo | 4 of 44 |
Most frequent serious adverse event terms
| Event term | Organ system | Arm Soy Supplement | Arm Placebo |
|---|---|---|---|
| Mild gastrointestinal symptoms | Gastrointestinal disorders | 2 of 42 | 6 of 44 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Arm Soy Supplement | Arm Placebo |
|---|---|---|---|
| GI symptoms | Gastrointestinal disorders | 9 of 42 | 4 of 44 |
Serious events with the highest arm-level rate: mild gastrointestinal symptoms (13.6%).
The registry reports serious adverse events for 2 of 42 participants in the soy supplement arm and 6 of 44 participants in the placebo arm. It also reports other adverse events for 9 of 42 and 4 of 44 participants, respectively. These are event counts against the arm-specific numbers recorded as at risk. The record does not provide a treatment-effect estimate or uncertainty interval for these comparisons, and the counts alone do not establish a difference between groups.
The registry lists “Mild gastrointestinal symptoms” among the top serious terms, with 13.6%. The record does not identify an arm for that percentage or explain the relationship between the term label and its serious-event classification. The percentage should therefore not be assigned to either arm or treated as a complete description of serious-event patterns. Death counts are not reported in the registry record. The available entries do not describe event timing, duration, causality assessment, severity criteria, or how many events each participant experienced. These summaries are limited safety reporting, not a full assessment of comparative safety.
12. What the Result Does, and Does Not, Mean
The primary result establishes the posted arm-level average serum testosterone values after two weeks: 376.0 ng/ml for soy supplement and 405.0 ng/ml for placebo. The associated 80% confidence intervals indicate uncertainty around those separate means. The registry does not report the primary between-arm effect estimate or its uncertainty, so the magnitude and precision of the randomized treatment contrast cannot be read directly from the primary result.
These data do not establish that the arms are equivalent. Similar-looking arm summaries, or overlapping confidence intervals, are not an equivalence test. Equivalence requires a prespecified margin and an analysis designed to determine whether the plausible treatment effects fall within that margin; neither is reported in the registry record. The data also do not establish that the intervention changed testosterone, because a direct primary contrast and its inferential analysis are absent from the posted results.
The endpoint is a short-term biological measurement in patients preparing for prostatectomy. It is not a direct measure of prostate cancer prevention, cancer development, recurrence, or long-term health outcomes. The trial title and the study summary describe prevention-related aims, but the primary reported result concerns serum testosterone at two weeks. No cancer-development result is posted in the record summarized by these outcome entries.
The secondary estrogen receptor result is an unadjusted comparison with substantial uncertainty around its mean-difference estimate. It may describe the recorded data for that endpoint, but it does not answer the primary question or establish a clinical consequence. The molecular-effects outcome is listed without a reported usable comparison. The registry also does not document multiplicity handling, the analysis population, or missing-data methods, which further constrain interpretation.
Accordingly, the defensible reading is narrow: the record reports two-week arm means and confidence intervals for serum testosterone, plus partial secondary and safety summaries. It does not provide enough primary inferential detail to characterize the treatment effect precisely, and it does not establish an effect on prostate cancer development. These limits concern what the registry record reports; they are not a recommendation about treatment or care.
13. Why This Trial Matters Statistically
This trial is statistically instructive because the primary outcome is reported with separate arm means and confidence intervals, but without a direct treatment-effect estimate. A randomized comparison is most interpretable when the contrast itself and its uncertainty are reported. Group summaries can describe the observed data, but they do not supply the same information as an estimate for the randomized difference. The distinction is especially important when readers are tempted to judge a treatment contrast by comparing the endpoints of two separate intervals.
The confidence level also deserves attention. The primary summaries use 80% confidence intervals. Readers need to retain that stated level when judging the intervals' coverage and precision, rather than treating them as a conventional higher-level interval or as a test result. The registry does not document how the intervals were calculated, which leaves additional uncertainty about the assumptions used.
The study also illustrates why endpoint scope matters. Its title raises a prevention question, while the primary reported outcome is a short-term serum biomarker measured before surgery. A biological measure may be relevant to a proposed mechanism, but its result cannot automatically be extended to cancer development or other clinical outcomes. Distinguishing the stated study aim from the measured endpoint prevents overinterpretation.
Finally, the record contains secondary results and planning illustrations without reporting a full analysis plan, multiplicity strategy, or missing-data approach. These are important parts of interpreting statistical evidence, not administrative details. Planning calculations based on an observed effect are scenarios for future study design, not observed power and not a replacement for the original sample-size justification. Together, these features show why precise reporting of the estimand, analysis population, uncertainty, and prespecified methods is essential for clinical trial interpretation.
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 principal limitation for interpreting the primary finding is that only arm-level means and 80% confidence intervals are reported. The registry does not report a treatment-effect estimate, a confidence interval for the contrast, or a primary p-value. The intervals are relatively broad in relation to the scale of the means, so the arm summaries leave substantial uncertainty about the contrast; no new contrast should be inferred from them. The calculation method for the intervals is not described.
The result concerns a two-week measurement in a preoperative setting. The study summary describes treatment for 2-4 weeks, while the primary comparison is reported at two weeks and the later preoperative measurement varied with surgery timing. This limits the reported primary result to its stated time point and does not describe longer-term exposure or outcomes. The registry does not explain how the variable later measurement was analyzed.
The enrolled population consisted of patients with prostate cancer preparing for surgery. That context differs from a general prevention population, so the reported findings cannot automatically be generalized to people without diagnosed cancer or to different treatment settings. The primary endpoint is serum testosterone, not cancer development. The posted record does not include a reported cancer-incidence or progression comparison.
Additional limitations arise from incomplete statistical reporting. The analysis population, missing-data handling, primary model, covariate adjustment, interim analyses, original sample-size justification, and multiplicity plan are not reported in the registry record. Only one secondary endpoint has a numerical comparison, and that result is unadjusted; the molecular-effects outcome is listed without a usable comparison. Safety entries likewise lack detail on timing, causality, and event-level counts. These gaps make it difficult to assess robustness and completeness, without establishing that any specific unreported method was not used.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Role of Soy Supplementation in Prostate Cancer Development — NCT00255125.
- Registry record last updated 2019-06-18; status last verified 2019-06.
- Results first posted 2015-06-17.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
NCT00255125 is a completed, randomized, triple-masked, parallel-group phase 3 study sponsored by the VA Office of Research and Development. It enrolled 86 participants and compared a soy supplement with placebo before planned prostatectomy. The study began in 2005-09 and reached primary completion in 2009-04.
The primary outcome was total serum testosterone in ng/ml at two weeks. The registry reports arm means and 80% confidence intervals but no primary treatment-effect estimate or p-value. Estrogen receptor status is reported as a secondary comparison; molecular effects are listed without a usable numerical comparison. The registry record does not report the analysis population, missing-data procedures, multiplicity strategy, or original sample-size justification. The posted primary result is a short-term biomarker comparison, not a reported test of prostate cancer prevention.