Study of 2 Different Doses of Revlimid in Biochemically Relapse Prostate Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of 2 Different Doses of Revlimid in Biochemically Relapse Prostate Cancer (NCT00348595), a randomized phase 1, phase 2 study. The primary outcome is Safety, Feasibility and Tolerance of Revlimid as Assessed by Number of Participants Experiencing Grade 3 and 4 Adverse Events..

2. Clinical Question

3. Trial Design

This was a randomized, parallel-group Phase I/II study. Its official title describes it as double blinded, while the registry masking field reports triple masking. The record does not identify which people or roles were masked, so the masking arrangements cannot be specified further. The trial compared two active oral doses of Revlimid, 5mg/day and 25mg/day, using matched capsules in each arm. It did not include a placebo-only or untreated comparison group.

The study concerned prostate cancer with biochemical relapse after local treatment, in the absence of metastases (M0). The trial objectives included feasibility, safety, and tolerance during 6 months of administration, and assessment of PSA progression at 6 months. Other stated objectives concerned PSA doubling time, PSA slope, and possible relationships between serum Revlimid concentrations, toxicity, PSA measures, and disease progression.

The registry lists an enrollment of 77 and a completed status. The study start is recorded as 2006-07-20 and primary completion as 2016-06-29. The primary outcome result is reported for 26 participants in the 5mg/Day arm and 34 in the 25mg/Day arm. The registry record does not explain the difference between these outcome counts and overall enrollment, or describe an analysis population that reconciles them.

The intervention descriptions specify morning oral dosing on days 1-21 of 28-day cycles. The 5mg/day group received one active 5 mg capsule and one matched 25 mg placebo capsule; the 25 mg/day group received one active 25 mg capsule and one matched 5 mg placebo capsule. The record names plasma concentration of Revlimid at steady state as a secondary outcome, but does not post a result for it in the registry record for this analysis. The sponsor is Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins.

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.

Before assignment: 17 subjects were screen failures

Overall Study

Reasons for not completing this period:

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: Baseline characteristics were only recorded for patients who entered the study.

6. Primary Endpoint

The registry lists the primary outcome as Safety, Feasibility and Tolerance of Revlimid as Assessed by Number of Participants Experiencing Grade 3 and 4 Adverse Events., measured over 6 months post-intervention.

The primary outcome counts participants who experienced Grade 3 or 4 adverse events at 6 months post-intervention. The registry defines these events using National Cancer Institute Common Toxicity Criteria version 3.0. It is a count of affected participants in each arm, rather than a continuous measurement or a time-to-event estimate.

In general, a participant count gives a direct description of how many people in the reported group met the specified event definition during the stated assessment period. It can be useful in a safety and tolerability study because it captures whether participants experienced higher-grade events, rather than only describing average symptoms or laboratory values. The measure does not, by itself, describe event severity beyond the stated grades, the duration of events, or how many separate events each participant experienced.

The posted counts are 3 participants in the Revlimid 5mg/Day arm and 10 participants in the Revlimid 25mg/Day arm. The outcome measure is described as safety, feasibility, and tolerance, but this particular result addresses only participants with Grade 3 or 4 adverse events. Other aspects of feasibility and tolerance are not quantified by this endpoint. The registry record does not report additional details about event ascertainment, the handling of participants without a recorded assessment, or an analysis population beyond the arm counts.

7. Results Posted in the Registry

• Revlimid 25mg/Day: 29.4% (10/34) versus 11.5% (3/26) on control; risk difference 17.9 percentage points (95% CI -1.8 to 37.5), risk ratio 2.55 (0.78–8.34); unadjusted two-proportion z-test p = 0.096.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is a binary participant-level outcome: each person either experienced at least one Grade 3 or 4 adverse event within the defined assessment or did not. The registry reports the number of participants with an event and the number in each arm. In general, this setup allows the observed risk in each group to be compared, while preserving the fact that the outcome is a count of participants rather than a count of every event episode.

What does the risk difference describe?

The posted comparison gives a risk difference of 0.179 for Revlimid 25mg/Day relative to Revlimid 5mg/Day. A risk difference subtracts one group's event risk from the other group's event risk, according to the comparison direction. Here, the record identifies the 5mg/Day group as the reference and the 25mg/Day group as the comparison. A positive estimate therefore indicates a greater observed event risk in the comparison group. It is an absolute contrast, not a relative increase, and it does not itself establish a causal mechanism.

The comparison also reports arm risks of 0.115 and 0.294, corresponding to 3 of 26 and 10 of 34 participants. These estimates are based on limited group sizes, so the observed risks are subject to sampling variability. The point estimate is a single summary of the observed contrast; it should not be read without its interval, which shows a range of values compatible with the analysis.

How should the interval and p-value be read?

In general, a confidence interval represents uncertainty in an estimate under the statistical procedure used to calculate it. The posted interval for the risk difference runs from -0.018 to 0.375. Because the bounds include zero, the data are compatible with no risk difference as well as with contrasts in either direction. The registry record does not identify the interval method or its confidence level, and those details are not reported in the registry record.

The posted unadjusted p-value is 0.096. In general, a p-value evaluates how unusual data at least as incompatible with a specified null model would be if that model and the test assumptions held. It does not quantify the size or clinical importance of an effect, and it is not the probability that the null model is true. The registry record does not report the test used, whether the calculation was prespecified, or whether another analysis was intended to take priority.

The interval and p-value are related ways of expressing uncertainty, but neither substitutes for a complete analysis description. The record does not report covariate adjustment, a stratification plan, handling of repeated events, or a detailed statistical analysis plan for this endpoint. The posted comparison is identified as unadjusted; no additional adjusted estimate is available in the registry record analyzed.

Why can counts and risks differ in reliability?

In general, a participant-level binary outcome reduces each person's experience to whether the event occurred, which can simplify interpretation. It also discards information about the number, timing, and duration of events. If participants have different follow-up or if event recording is incomplete, the simple count may not represent comparable observation time. The registry record does not describe these issues or explain whether all participants had the same opportunity for outcome assessment.

The listed numbers at risk for the primary outcome are 26 and 34, while overall enrollment is 77. The record does not explain why the primary-outcome denominators are smaller than enrollment, whether exclusions or missing assessments account for the difference, or which analysis population was used. Consequently, the reported risks should be interpreted as applying to the posted outcome denominators, not automatically to everyone enrolled.

What does over-dispersion mean in general?

Over-dispersion is a general concern when a statistical model assumes less variability than the data contain. It is most often relevant to count models that analyze numbers of events, especially when event counts vary more than the model allows. The primary outcome is reported as participants with an event, rather than as a rate of repeated event counts, so the posted endpoint does not itself establish whether over-dispersion was assessed. The registry record does not report a model or an over-dispersion assessment.

More generally, model assumptions matter because interval widths and p-values depend on how uncertainty is calculated. Random allocation is part of the trial design, but it does not identify every statistical choice made after data collection. The registry record does not report interim analyses, a stopping rule, a prespecified sample-size calculation, or the original statistical analysis plan. These items should not be inferred from the posted effect estimate or from planning illustrations based on it.

How do the secondary objectives fit?

The stated secondary objectives concern PSA doubling time, PSA slope, and preliminary pharmacodynamic relationships involving serum Revlimid concentrations, toxicity, PSA measures, and disease progression. The registry also names plasma concentration at steady state as a secondary outcome. Those questions involve outcomes and analyses distinct from the primary participant count. The record does not post corresponding numerical results in the registry record analyzed, so no statistical conclusion about those objectives can be drawn from the primary safety comparison.

10. Confidence Intervals and Precision

The risk-difference estimate is 0.179, with a reported interval from -0.018 to 0.375. This interval is wide enough to include zero and positive and negative values. Thus, the point estimate alone gives an incomplete picture: plausible values under the reported interval range from a small negative difference to a considerably larger positive difference. The registry record does not state the confidence level or calculation method for the interval, so those should not be assumed.

In general, confidence intervals show the precision of an estimate under a chosen method and its assumptions. A narrower interval would indicate less uncertainty in the estimated contrast; a wider interval indicates more. The reported arm counts, 26 and 34, are modest, and the interval reflects considerable uncertainty around the difference. It does not imply that every value within the interval is equally likely, and it does not account for sources of uncertainty that may not be included in the analysis.

The unadjusted p-value is 0.096. It does not make the interval or effect size unnecessary, and it should not be treated as a measure of the probability that the groups truly differ. The record does not report adjusted analyses or a multiplicity procedure. Nor does it explain the difference between enrollment of 77 and the outcome denominators. That uncertainty about the analyzed population limits how confidently the precision can be generalized to all enrolled participants.

The reported estimate is therefore best read as an uncertain comparison of the posted outcome groups, rather than as a precise estimate of a dose effect across all eligible patients. The registry record does not report further precision measures for this endpoint.

11. 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 planning illustrations are based on the observed risk difference and give participants per arm, not total events. They show hypothetical group sizes for 80% and 90% power when planning to detect the full observed effect, 75% of the observed effect, or 50% of the observed effect. The assumptions listed for these illustrations are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is the stated share of the observed effect.

In general, power is the chance that a planned analysis will detect an effect of a specified size under the planning assumptions if that effect is present. The 80% power column gives the participants per arm for a design with an 80% chance of detecting the assumed effect; the 90% power column gives the participants per arm for a design with a 90% chance under those same assumptions. These are planning quantities, not a description of what happened in this completed trial.

For the full observed effect, the illustrations list 79 participants per arm for 80% power and 106 participants per arm for 90% power. Planning for 75% of the observed effect lists 130 participants per arm for 80% power and 173 participants per arm for 90% power. Planning for 50% of the observed effect lists 264 participants per arm for 80% power and 352 participants per arm for 90% power. The increasing requirements illustrate the general principle that smaller effects are harder to distinguish from random variation and therefore need more participants. This is not observed power.

The trial's original design assumptions and its own sample-size justification are not reported in the registry record. These illustrations are calculated from the observed effect under the listed assumptions; they should not be read as evidence that the trial was designed around those assumptions, nor as a retrospective measure of whether it succeeded in answering its question.

12. 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 = 0.115; treatment proportion p₁ = 0.294; average p̄ = 0.205.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.5707 = 1.1185.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.5565 = 0.4684.
4. n = (1.1185 + 0.4684)² / (0.1787)² = 78.8, rounded up to 79 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 <- 0.115384615384615                          # control proportion
p1 <- 0.294117647058824                          # 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(348595)
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 = 79 (80% power), 106 (90% power)
 75% of observed effect: n per arm = 130 (80% power), 173 (90% power)
 50% of observed effect: n per arm = 264 (80% power), 352 (90% power)
power.prop.test: 78.8 
Simulated power at n = 79 per arm: 0.814 (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.

13. Multiplicity, Missing Data and Analysis Populations

The study has more than one stated objective. Its primary objectives include safety, feasibility, and tolerance during 6 months of Revlimid administration, as well as PSA progression at 6 months. Secondary objectives include PSA doubling time, PSA slope, pharmacodynamic correlations, and serum drug concentration. Testing multiple outcomes can raise a multiplicity question: in general, when many hypotheses are assessed, the chance of at least one apparently unusual result can increase unless the analysis plan accounts for it.

The registry record does not report a multiplicity adjustment, a hierarchy for interpreting outcomes, or which comparisons were intended to be confirmatory versus exploratory. It does not report whether interim analyses were conducted or whether any stopping rules were used. The unadjusted p-value for the posted primary adverse-event comparison should therefore be understood as an unadjusted result, not as an estimate adjusted for the collection of study objectives.

Missing-data handling and analysis populations are also not reported in the registry record. Overall enrollment is 77, while the primary outcome denominators are 26 in the 5mg/Day arm and 34 in the 25mg/Day arm. The record does not say whether the difference reflects missing assessments, exclusions, or another reason, and it does not identify an intention-to-treat, per-protocol, or safety population. No explanation should be inferred from the counts alone.

The record also does not specify how participants without an evaluable outcome were handled, whether follow-up differed by arm, or whether any sensitivity analyses were performed. These omissions matter because participant counts and risks are most readily compared when outcome definitions, follow-up, and analysis populations are applied consistently. The posted comparison remains a description of the analyzed groups and should not be assumed to represent all enrolled participants without additional information.

14. 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: neutropenia (14.7%), venous thromboembolism (7.7%), severe rash (5.9%), fatigue (3.8%).

The serious adverse event summaries report 3 participants affected among 26 at risk in the Revlimid 5mg/Day group and 10 participants affected among 34 at risk in the Revlimid 25mg/Day group. Deaths are reported as 0 in each group. The summary also reports other adverse events affecting 26 participants in the 5mg/Day group and 34 participants in the 25mg/Day group.

These are participant counts, not counts of all event episodes. A participant may experience more than one event, but the summary does not provide event totals or detail on how events overlapped. Serious adverse events are not synonymous with the primary endpoint, which counts participants with Grade 3 or 4 adverse events. The registry record does not report whether each serious event met the primary endpoint's grade definition.

The registry lists Neutropenia at 14.7%, venous thromboembolism at 7.7%, Severe Rash at 5.9%, and Fatigue at 3.8% among top serious adverse event terms. It does not provide enough context in the record to attribute those listed percentages to a particular treatment arm or clarify their denominators. These terms should therefore be read as listed safety information, not as arm-specific rates or a complete account of adverse events.

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

The posted primary result describes the number of participants who experienced Grade 3 or 4 adverse events during the stated assessment period in two randomized dose groups. The observed counts were 3 of 26 for Revlimid 5mg/Day and 10 of 34 for Revlimid 25mg/Day. The posted risk difference is 0.179, with an interval from -0.018 to 0.375, and an unadjusted p-value of 0.096. Taken together, the point estimate suggests a higher observed event risk in the 25mg/Day arm, while the interval indicates substantial uncertainty about the size and direction of the underlying difference.

These results do not establish that the dose caused the observed difference, and they do not establish that the doses have the same safety profile. The interval includes zero and values on either side, and the registry record does not describe enough about missing outcomes, the analyzed population, or the statistical procedure to resolve those uncertainties. The reported comparison should be understood as applying to the posted group counts.

The result also does not answer every objective in the trial record. The study description refers to PSA progression at 6 months, PSA doubling time, PSA slope, and pharmacodynamic relationships, but corresponding results are not reported in the registry record for this analysis. The primary adverse-event outcome cannot be used as a substitute for results on those separate questions.

The serious adverse event summaries add descriptive safety context, including participant counts and listed terms, but do not fully characterize event timing, duration, recurrence, or arm-specific rates for the listed terms. Deaths are reported as 0 in each arm. The record does not establish broader effects in populations beyond the participants represented in the posted outcome groups. Interpretation should remain focused on what was counted and how uncertain the between-arm estimate is.

16. Why This Trial Matters Statistically

This trial illustrates a common challenge in clinical biostatistics: a randomized comparison can produce a clear direction in the observed counts while still leaving substantial uncertainty about the size of the difference. The posted counts are 3 participants with the primary adverse-event outcome in the 5mg/Day group and 10 in the 25mg/Day group. The risk-difference estimate is positive, but its interval extends across zero. Reading the point estimate, interval, and p-value together gives a more complete account than relying on any single statistic.

It also demonstrates why denominators and analysis populations matter. The registry lists enrollment of 77, but the primary outcome groups contain 26 and 34 participants. Without an explanation of how those numbers were reached, readers cannot assume that the outcome analysis includes every enrolled participant or know how missing assessments were handled. Clear reporting of participant flow and outcome availability is important for assessing how representative the analyzed groups are.

The record distinguishes a primary adverse-event count from other objectives involving PSA and pharmacodynamic measures. This is statistically important because different outcomes answer different questions and may require different models and interpretations. With multiple stated objectives, it is also important to know whether the analysis plan prespecified an outcome hierarchy or multiplicity approach; that information is not reported in the registry record.

Finally, the planning illustrations make explicit that sample-size requirements depend on the effect a future study aims to detect and on the chosen assumptions. They are not a substitute for the trial's original design rationale, which is not reported in the registry record. Taken together, these features make the trial a useful example of interpreting uncertainty, denominators, and planning calculations without treating a single posted comparison as a complete account of the study.

17. 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 primary outcome denominators are 26 for Revlimid 5mg/Day and 34 for Revlimid 25mg/Day, while total enrollment is 77. The registry record does not explain this difference, identify an analysis population, or describe how participants without a posted outcome were handled. This limits interpretation of how the reported group estimates relate to everyone enrolled.

The study record reports a comparison and an unadjusted p-value, but the statistical test, interval calculation method, and confidence level are not reported in the registry record. The record also does not provide adjusted estimates, sensitivity analyses, or a detailed analysis plan. It is therefore not possible to assess from the posted information whether other analytical choices would materially alter the estimate.

The primary endpoint is a participant count for Grade 3 or 4 adverse events at 6 months post-intervention. The record does not provide event-level detail, timing, duration, recurrence, or information about exposure and follow-up by arm. The serious adverse event terms are listed with percentages, but the registry record does not provide enough context to assign those figures to a specific arm or denominator.

The trial's stated objectives include PSA progression and several secondary PSA and pharmacodynamic measures, but their results are not reported in the registry record for this analysis. The trial's original sample-size justification, interim-analysis plan, and multiplicity approach are also not reported. These gaps do not change the posted counts, but they constrain how fully the results can be interpreted or compared with the broader set of study objectives.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Study of 2 Different Doses of Revlimid in Biochemically Relapse Prostate Cancer — NCT00348595.
• Registry record last updated 2019-04-12; status last verified 2019-04.
• Results first posted 2019-04-12.
• 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.

21. Record Summary

NCT00348595 is a completed randomized Phase I/II parallel-group study of Revlimid in people with biochemically relapsed prostate cancer after local treatment, without metastases. The registry lists enrollment of 77, with a study start of 2006-07-20 and primary completion on 2016-06-29. The official title calls the study double blinded, while the masking field reports triple masking.

The primary posted outcome counts participants with Grade 3 or 4 adverse events at 6 months post-intervention. The result records 3 of 26 participants in the Revlimid 5mg/Day arm and 10 of 34 in the Revlimid 25mg/Day arm. The reported risk difference is 0.179, its interval is -0.018 to 0.375, and the unadjusted p-value is 0.096. The registry also reports serious adverse event summaries and deaths of 0 in each arm.

The registry record does not explain the difference between total enrollment and the primary outcome denominators. Results for the stated PSA and pharmacodynamic objectives are not reported in the registry record for this analysis.