A Randomized, Double-Blind, Placebo-Controlled Study of Idelalisib in Combination With Rituximab for Previously Treated Chronic Lymphocytic Leukemia (CLL): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Randomized, Double-Blind, Placebo-Controlled Study of Idelalisib in Combination With Rituximab for Previously Treated Chronic Lymphocytic Leukemia (CLL) (NCT01539512), a randomized phase 3 study. The primary outcome is Progression-Free Survival.

2. Clinical Question

3. Trial Design

This was a Phase 3, randomized, double-blind, placebo-controlled, parallel-group study sponsored by Gilead Sciences. The registry record lists triple masking. Participants were assigned to idelalisib plus rituximab or placebo plus rituximab, with the brief summary describing a 1:1 allocation. Enrollment was 220. The study began in 2012-04, reached primary completion in 2013-10, and is listed as completed.

The registry describes the study as an evaluation of the efficacy and safety of idelalisib with rituximab in participants previously treated for CLL. The primary outcome was progression-free survival, assessed up to 17 months. Secondary outcomes listed include overall response rate, lymph node response rate, overall survival, and complete response rate. Numeric outcome results are posted for overall response, lymph node response, and overall survival; a numeric result for complete response rate is not reported in the registry record.

The brief summary also says that participants tolerating primary study therapy who experienced definitive CLL progression were eligible to receive active idelalisib therapy in an extension study. The registry record does not report how many participants entered that extension or provide further extension-study results in the posted outcomes. The original protocol’s detailed design assumptions, including its planned sample-size justification, are not reported in the registry record.

4. Participant Flow and Accounting

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

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Intent-to-Treat (ITT) Analysis Set: randomized participants with treatment group designated according to initial randomization.

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: ITT Analysis Set: randomized participants with treatment group designated according to initial randomization.

6. Primary Endpoint

The registry lists the primary outcome as Progression-Free Survival, measured over Up to 17 months.

Progression-free survival measures the interval from randomization until the earlier of definitive disease progression or death from any cause. The registry defines progression using standard CLL criteria other than lymphocytosis alone, including new lesions, specified increases in measured lymph nodes, new or enlarged liver or spleen, transformation to a more aggressive histology, or a reduction in blood cells attributable to CLL.

This endpoint combines disease progression and death into a time-to-event outcome. In general, time-to-event measures retain information about when an event occurs, rather than recording only whether it occurred by a fixed point. They can also account for participants whose event time is not observed during follow-up, if the analysis handles that censoring appropriately. The registry record does not report detailed censoring rules or the full analysis plan.

The primary outcome timeframe is listed as up to 17 months. The posted comparison uses a hazard ratio, and the arm-level results include a median and confidence interval information where reported. These statistics describe different features of the event-time data: a median gives a point on the event-time distribution, while a hazard ratio compares event rates over time under the analysis method. The record reports no median value for the idelalisib plus rituximab arm, so its median cannot be read from the posted result.

7. Results Posted in the Registry

• Idelalisib + Rituximab vs Placebo + Rituximab: hazard ratio 0.18 (95% CI 0.10–0.32), reported p < 0.001 (as reported in the registry).

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.

Overall Response Rate

• 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:
  • Idelalisib + Rituximab: 74.5% (82/110) versus 14.5% (16/110) on control; risk difference 60.0 percentage points (95% CI 49.5 to 70.5), risk ratio 5.12 (3.22–8.17); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Idelalisib + Rituximab than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Lymph Node Response Rate

• 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:
  • Idelalisib + Rituximab: 91.8% (101/110) versus 5.5% (6/110) on control; risk difference 86.4 percentage points (95% CI 79.7 to 93.0), risk ratio 16.83 (7.72–36.72); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Idelalisib + Rituximab than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Overall Survival

• Endpoint type: Time to event: compared as a hazard ratio.
• Test: Hazard ratio as reported by the sponsor; it was not recalculated.
• Result:
  • Idelalisib + Rituximab vs Placebo + Rituximab: hazard ratio 0.28 (95% CI 0.11–0.69) (as reported in the registry).
• Interpretation:
  • The 95% confidence interval excludes no difference (a ratio of 1), so the hazard of the event is statistically lower for Idelalisib + Rituximab vs Placebo + Rituximab than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Across the posted secondary outcomes, the response measures show higher reported percentages in the idelalisib plus rituximab group than in the placebo plus rituximab group. The overall-survival comparison also has a reported hazard-ratio estimate below 1, though its p-value is not reported. Together, these outcomes provide additional descriptions of response and survival alongside the primary progression-free survival result. They do not substitute for the primary endpoint, and each addresses a different aspect of outcome.

These results carry less inferential weight than the primary result for several reasons. The registry does not report a multiplicity-adjustment plan, and the posted response comparisons are unadjusted recalculations rather than sponsor-reported tests. When several outcomes are examined, the chance of at least one apparently notable result can increase unless the analysis accounts for the testing structure. The record does not report how the secondary outcomes were prioritized or whether any were part of a prespecified testing sequence.

Overall response rate, lymph node response rate, and overall survival have numeric results in the posted outcomes. Complete response rate is listed as a secondary outcome, but its numeric result is not reported in the registry record. The secondary evidence therefore has gaps in both analysis detail and outcome reporting. It can inform the broader statistical picture, but should be considered with the primary result and with the uncertainty and reporting limitations for each outcome, rather than treated as a set of independent confirmations.

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

• Overall Response Rate
• Lymph Node Response Rate
• Overall Survival
• Complete Response Rate

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is time-to-event: each participant’s outcome is represented by the time from randomization to progression or death, with the earlier event defining the endpoint. The registry reports the comparison as a hazard ratio and identifies the Log Rank method for the primary analysis. A hazard ratio compares the event rate at a given time among participants still at risk in one group with that in the other group. General interpretation often treats a value below 1 as favoring a lower event hazard in the first-named group, but the estimate is a relative measure and is not a direct measure of the difference in median survival time.

The reported primary hazard ratio is 0.18, with a 95% confidence interval of 0.10 to 0.32 and p < 0.001. The confidence interval gives a range of estimates compatible with the data under the statistical procedure and its assumptions. It is not a range containing 95% of individual outcomes, nor does it mean there is a 95% probability that the true hazard ratio lies inside the observed interval. The registry record does not report whether the proportional-hazards assumption was tested or how departures from that assumption were handled. In general, if relative hazards vary over time, a single hazard ratio may not describe that changing pattern fully.

A Log Rank test compares the event-time experience of groups across follow-up. In general, it uses information about the timing of events and the number of participants at risk as follow-up proceeds. Its interpretation depends on the analysis and assumptions, including how censoring is treated. The registry record identifies the method but does not provide the full analysis specification, detailed censoring rules, or a description of any interim analyses. Those elements are therefore not reported in the registry record.

The arm-level median is the time by which half of the relevant event-time distribution has experienced the event, as estimated from the data. The placebo plus rituximab median is reported as 5.5 months, with a 95% confidence interval of 3.8 to 7.1 months. The median for idelalisib plus rituximab is not reported, although the registry posts a lower confidence limit of 10.7 months. A missing median value should not be replaced with a derived estimate from the hazard ratio: these are different summaries and cannot be converted into one another from the posted information alone.

For binary outcomes such as response, the basic descriptive result is the proportion of participants with the outcome in each group. The registry reports overall response rates of 74.5% and 14.5%, and lymph node response rates of 92.2% and 5.9%, for idelalisib plus rituximab and placebo plus rituximab, respectively. The associated arm-level confidence intervals describe uncertainty in each proportion. The recalculated comparisons use the posted summary counts and are unadjusted. In general, an unadjusted comparison does not account for testing multiple outcomes or multiple hypotheses, and it may differ from a sponsor’s prespecified or adjusted analysis. No sponsor-reported analysis is posted for these response outcomes.

For overall survival, the registry reports a hazard ratio of 0.28, with a 95% confidence interval of 0.11 to 0.69. It does not report a p-value or analysis method for this comparison. The arm-level medians are not reported as numeric values; the placebo plus rituximab entry includes a lower confidence limit of 12.8 months, while the other median and interval limits are not reported. Consequently, interpretation should focus on the posted hazard ratio and its uncertainty rather than treating absent median values or an absent p-value as if they had been observed.

General statistical caution: estimates from clinical trials are subject to sampling variability, and analysis choices affect how that variability is represented. A confidence interval communicates precision for a particular estimate under the stated method; it does not address every source of uncertainty, such as missing data, measurement issues, or applicability beyond the study population. The registry record does not report the full protocol, the original statistical analysis plan, the handling of missing outcome data, or whether planned interim analyses were conducted. These omissions limit independent assessment of the analysis details, even though the principal reported comparison and some secondary results are available.

11. Confidence Intervals and Precision

The primary hazard-ratio estimate is 0.18, with a 95% confidence interval of 0.10 to 0.32. The interval communicates statistical uncertainty around the estimated relative event hazard. Its limits are both below 1, while the estimate itself is substantially below 1. The reported p-value is p < 0.001. These are related but distinct summaries: the confidence interval conveys a range of estimates compatible with the analysis, while the p-value measures compatibility with a null hypothesis under the test’s assumptions.

Arm-level precision for the medians is less complete. For placebo plus rituximab, the median progression-free survival is 5.5 months and its 95% confidence interval is 3.8 to 7.1 months. For idelalisib plus rituximab, the median is not reported; the record gives a lower confidence limit of 10.7 months and no upper limit. This incomplete reporting means a direct comparison of the arm-level median estimates is not available from the record.

For overall survival, the hazard ratio is 0.28 with a 95% confidence interval of 0.11 to 0.69. The interval is wider on the hazard-ratio scale than the primary interval and reflects uncertainty in this secondary estimate. The registry does not post a p-value for this comparison. For the response outcomes, arm-level 95% confidence intervals are available, but they do not account for multiplicity across outcomes. Confidence intervals quantify uncertainty under a particular analysis; they do not correct for unreported design features or missing-data choices.

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 planning figures are illustrations based on observed effects, not the trial’s original design calculations. For the time-to-event outcomes, the figures are total events needed across the comparison, not participants per arm. For the other outcomes, the figures are participants per arm. The calculations assume a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. They consider detecting the stated fraction of the observed effect. The original design assumptions and sample-size justification are not reported in the registry record.

The 80% power column describes the planning number associated with an 80% chance of rejecting the null hypothesis under the stated assumptions, if the effect used for planning were the true effect. The 90% power column uses a higher target chance under those same assumptions. These figures are hypothetical planning quantities, not a measure of the trial’s observed power. They should not be interpreted as evidence that the study had a particular probability of detecting its actual effect after the results were known.

Each planning row uses a different fraction of the observed effect. Planning for 75% of the observed effect requires more events or participants than planning for 100% of the observed effect, and planning for 50% requires more again. This illustrates a general principle: smaller effects are harder to distinguish from random variation and therefore require more information. The figures are not a recommendation for a future study; they depend on the stated assumptions and on using an observed effect as the starting point. The record does not provide the trial’s original event target or the assumptions used to choose its enrollment.

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

d is the total number of events needed (Schoenfeld's formula) with 1:1 allocation and proportional hazards.

Worked calculation for this trial

1. Hazard ratio HR = 0.180, so ln(HR) = -1.7148.
2. d = 4 × (1.96 + 0.842)² / [ln(0.180)]² = 10.7, rounded up to 11 events in total 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
hr <- 0.18                              # observed hazard ratio
events_needed <- function(power, frac = 1) ceiling(round(4 * (z_a + qnorm(power))^2 / log(hr^frac)^2, 6))   # Schoenfeld, 1:1 allocation
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: events = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(events_needed(0.80, f), big.mark = ","), format(events_needed(0.90, f), big.mark = ",")))

# Simulation: given d total events under proportional hazards, events in the treatment arm are Binomial(d, hr / (1 + hr)).
# The score (log-rank type) statistic is (D1 - d / 2) / sqrt(d / 4).
set.seed(1539512)
d    <- events_needed(0.80)
reps <- 20000
D1 <- rbinom(reps, d, hr / (1 + hr))
z  <- (D1 - d / 2) / sqrt(d / 4)
cat(sprintf("Simulated power with %s events: %.3f (%d replicates)\n", format(d, big.mark = ","), mean(abs(z) > z_a), reps))

Output from running the code

100% of observed effect: events = 11 (80% power), 15 (90% power)
 75% of observed effect: events = 19 (80% power), 26 (90% power)
 50% of observed effect: events = 43 (80% power), 58 (90% power)
Simulated power with 11 events: 0.770 (20000 replicates)

The simulation uses the fact that, with d events in total and proportional hazards, the number of events in the treatment arm is binomial. It is an approximation that conditions on the total number of events, and it should land close to the target power.

14. Analyses Reported by the Sponsor

For progression-free survival, the registry posts a sponsor-reported Log Rank analysis with a hazard ratio of 0.18, a 95% confidence interval of 0.10 to 0.32, and p < 0.001. The primary result discussed in the statistical interpretation is that reported analysis, rather than an independently reconstructed estimate. The registry does not flag a discrepancy between a recalculation and the reported result.

The response outcomes have posted arm-level results and unadjusted recalculated comparisons, but no sponsor-reported analyses are listed for them. Those recalculations are simple comparisons based on the posted summaries. They do not establish that the sponsor used the same method, adjustment, or analysis population. The record does not report a sponsor analysis for those comparisons, so no such analysis can be reconstructed from the posted results.

For overall survival, the registry reports a hazard ratio of 0.28 with a 95% confidence interval of 0.11 to 0.69. A p-value and analysis method are not reported for that comparison. There is no discrepancy flag, but the missing method and p-value still limit the detail available for interpreting the result. In general, when a sponsor-reported analysis is available, it is the primary reference for describing the trial’s own inferential result; a recalculation from summary data is supplementary and may not reproduce all prespecified analysis choices.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists multiple secondary outcomes, including overall response rate, lymph node response rate, overall survival, and complete response rate. It reports results for some, but not all, of these outcomes. The record does not report whether the statistical analysis plan controlled the overall type I error across multiple outcomes, used a hierarchical testing procedure, or applied another multiplicity strategy. Without that information, the secondary p-values from unadjusted recalculations should not be interpreted as if they were adjusted confirmatory tests.

The primary outcome population is identified as the ITT Analysis Set, with randomized participants analyzed according to initial randomization. This is a stated population definition; it does not by itself provide all details about inclusion in each endpoint analysis. The registry record does not report endpoint-specific analysis populations, missing-data handling, censoring rules, or how participants with incomplete outcome information were treated. Those details can affect estimates and their precision, especially for time-to-event outcomes.

The record also does not report the original sample-size justification, the full statistical analysis plan, whether interim analyses occurred, or whether any outcome definitions or analyses changed during the study. These aspects should not be inferred from the posted results. Their absence does not invalidate the statistics that are reported, but it limits independent assessment of prespecification, multiplicity, and potential effects of missing information. The planning illustrations based on observed effects are separate from these unreported original design and analysis details.

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: pneumonia (11.8%), pyrexia (9.1%), diarrhoea (7.3%), sepsis (6.4%), febrile neutropenia (5.6%).

The registry reports serious adverse events in 65 of 110 participants in the idelalisib plus rituximab group and 43 of 108 in the placebo plus rituximab group. Other adverse events are reported in 103 of 110 and 102 of 108, respectively. These are descriptive counts for the groups as listed in the safety record; they should not be read as proof that treatment caused or prevented an event. The record does not report death counts in these safety entries.

The listed serious adverse-event terms include pneumonia at 11.8%, pyrexia at 9.1%, diarrhoea at 7.3%, sepsis at 6.4%, febrile neutropenia at 5.6%, and cellulitis at 3.7%. The registry listing does not specify, in this summary, a group assignment alongside each term, so those percentages should not be attributed to one arm. Serious-event counts and event-term percentages describe different levels of safety information and may not represent the same unit of analysis. Detailed definitions, exposure-adjusted rates, and additional safety comparisons are not reported in the registry record.

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

The posted primary result is a statistically precise comparison of progression-free survival favoring a lower estimated hazard of progression or death in the idelalisib plus rituximab group relative to the placebo plus rituximab group. The hazard ratio is 0.18, its 95% confidence interval is 0.10 to 0.32, and the reported p-value is p < 0.001. These figures describe the trial’s statistical comparison; they do not by themselves determine the practical importance of the result for a particular person.

The registry also reports higher response percentages in the idelalisib plus rituximab arm for the posted response outcomes, and an overall-survival hazard ratio below 1. The secondary results are less conclusive as a group because the multiplicity plan is not reported, the response comparisons are unadjusted recalculations, and the complete response result is not reported. The overall-survival entry lacks a p-value and analysis method. These limits matter when weighing the secondary findings alongside the primary result.

The posted data do not establish that every participant experienced the same outcome, explain the balance of benefits and harms for an individual, or guarantee that the results apply to populations not represented in the trial. The safety entries report adverse-event counts and selected serious-event terms, but do not provide all details needed for a comprehensive safety assessment. The registry record also does not report the full statistical plan, missing-data procedures, or all analysis details. Accordingly, the results should be understood as estimates from a particular randomized comparison, with the uncertainty and reporting gaps described in the record.

18. Why This Trial Matters Statistically

This trial illustrates how a time-to-event endpoint can summarize more than a single follow-up status. Progression-free survival combines the timing of CLL progression and death, and the primary analysis reports a hazard ratio with a confidence interval and p-value. Reading these together distinguishes the estimated relative contrast from its statistical uncertainty. The separately posted median information also shows why a hazard ratio and arm-level medians should not be treated as interchangeable summaries.

The record also illustrates the importance of separating primary from secondary evidence. The primary comparison has a sponsor-reported analysis, whereas the response comparisons are unadjusted recalculations and the overall-survival report lacks a p-value and method. These distinctions affect how confidently the results can be compared and how much weight secondary findings can carry. The absence of a reported multiplicity strategy is particularly relevant when several outcomes are listed.

Finally, the sample-size illustrations demonstrate how planning needs depend on the effect size a study aims to detect and on the desired power. They are not the original design rationale and are not observed power. Keeping these planning examples distinct from the trial’s actual design assumptions is important for avoiding retrospective overinterpretation. Overall, the registry record offers a clear primary estimate while leaving some design, analysis, and secondary-outcome details unreported.

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.

The arm-level median progression-free survival for idelalisib plus rituximab is not reported, and its confidence interval is incomplete in the posted result. The placebo plus rituximab median is reported, but the absent intervention-arm median prevents a direct comparison of the two median estimates from these summaries. The primary hazard ratio remains available, but it does not fill in the missing median value.

Reporting for secondary outcomes is also incomplete. Numeric results are posted for overall response, lymph node response, and overall survival, but not for the listed complete response rate. The overall-survival comparison has no reported p-value or method. No sponsor-reported analyses are listed for the response comparisons, which are unadjusted recalculations. The registry does not report how multiplicity was handled, so the secondary comparisons should be interpreted cautiously.

The registry record does not report the original sample-size assumptions, the full statistical analysis plan, interim analyses, detailed censoring rules, or missing-data procedures. It also does not provide death counts in the posted safety entries. The serious-event term percentages are not assigned to an arm in the listed summary. These limitations restrict assessment of the full analysis and safety context. They do not change the values that are reported, but they matter when judging the completeness and scope of the evidence available in the registry record.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Randomized, Double-Blind, Placebo-Controlled Study of Idelalisib in Combination With Rituximab for Previously Treated Chronic Lymphocytic Leukemia (CLL) — NCT01539512.
• Registry record last updated 2019-05-14; status last verified 2015-04.
• Results first posted 2014-10-16.
• 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 record describes a completed, randomized, double-blind, placebo-controlled Phase 3 study of idelalisib plus rituximab versus placebo plus rituximab in previously treated CLL. Enrollment was 220, with the brief summary describing 1:1 allocation. The primary outcome was progression-free survival, defined as time from randomization to definitive disease progression or death from any cause.

The primary reported hazard ratio was 0.18, with a 95% confidence interval of 0.10 to 0.32 and p < 0.001. Secondary results include response outcomes and an overall-survival hazard ratio, with limitations in analysis details and outcome reporting. The record does not report the original sample-size justification, full analysis plan, or several other details needed for a complete independent assessment.