This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record.
1. Trial at a Glance
E1912 was a randomized phase 3 trial evaluating ibrutinib and rituximab against a combination of rituximab, fludarabine phosphate, and cyclophosphamide in patients with untreated chronic lymphocytic leukemia or small lymphocytic lymphoma. The registered primary endpoint was the progression-free survival rate at 3 years, analyzed as a time-to-event endpoint using a log-rank test and hazard ratio.
| Feature | E1912 |
|---|---|
| Trial name | E1912 |
| ClinicalTrials.gov identifier | NCT02048813 |
| Phase | Phase 3 |
| Therapeutic area | Hematology |
| Conditions | Anemia; Chronic Lymphocytic Leukemia; Small Lymphocytic Lymphoma |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | None |
| Primary purpose | Treatment |
| Enrollment | 529 |
| Number of arms | 2 |
| Trial status | Active, not recruiting |
| Start | 2014-03-10 |
| Primary completion | 2018-10-24 |
| Lead sponsor | National Cancer Institute (NCI) |
| Sponsor type | NIH |
2. Clinical Question
The central statistical question is whether treatment assignment to ibrutinib and rituximab changes progression-free survival compared with rituximab, fludarabine phosphate, and cyclophosphamide in patients with untreated chronic lymphocytic leukemia or small lymphocytic lymphoma.
Population
Patients with untreated chronic lymphocytic leukemia or small lymphocytic lymphoma represented in the E1912 randomized trial.
Intervention
Ibrutinib and rituximab.
Comparator
Rituximab, fludarabine phosphate, and cyclophosphamide.
Primary question
Does the ibrutinib-and-rituximab treatment assignment produce a different progression-free survival experience from the comparator, with the registered hypothesis specified as superiority?
3. Trial Design
Ibrutinib + Rituximab
- Ibrutinib
- Rituximab
Rituximab + Fludarabine Phosphate + Cyclophosphamide
- Rituximab
- Fludarabine phosphate
- Cyclophosphamide
The registry also identifies laboratory biomarker analysis, pharmacogenomic study, and quality-of-life assessment among the trial interventions. These components are distinct from the registered primary time-to-event endpoint and are not used here to infer additional treatment effects.
4. Trial Timeline and Registry Status
Trial start
The registered trial start date was March 10, 2014.
Primary completion
The registered primary completion date was October 24, 2018.
Active, not recruiting
The ClinicalTrials.gov record identifies the trial as active, not recruiting.
5. Endpoints
The registered primary endpoint is a time-to-event measure expressed as a progression-free survival rate at a fixed 3-year time point. The registry describes repeated assessment during follow-up rather than treating the 3-year rate as an ordinary binary endpoint.
| Endpoint | Registry definition / time frame | Statistical role |
|---|---|---|
| Progression-free Survival (PFS) Rate at 3 Years | Assessed every 3 months until progression up to 4 years and 8 months. PFS was defined as the time from randomization to CLL progression or death, whichever occurred first. Progression is characterized by any of the following: ≥ 50% increase from nadir since start of treatment (tx) in the sum of the products of at least 2 lymph nodes on 2 consecutive examinations 2 weeks apart; ≥ 50% increase from nadir since start of tx in the size of liver and/or spleen; ≥ 50% | Primary endpoint; time-to-event; superiority |
6. Statistical Methodology
Time-to-event analysis
The primary endpoint is a time-to-event outcome. Instead of recording only whether a participant eventually experienced progression, the analysis uses the time from randomization to progression or death. This preserves information about when events occur and allows participants without an observed event by their last assessment to contribute follow-up through censoring.
Log-rank test
The registry reports a log-rank test for the primary PFS comparison. The log-rank framework compares the survival experience of the randomized groups across observed event times. Conceptually, at each event time it compares the number of events observed in each group with the number expected under the null hypothesis of no difference between the groups.
The log-rank test is therefore a global comparison of the time-to-event distributions rather than a comparison of only the proportions who have progressed at one selected time point.
Hazard ratio
The registered analysis reports the treatment effect as a hazard ratio. A hazard ratio compares the estimated instantaneous event rates between treatment groups within the time-to-event framework.
For E1912, Arm A is ibrutinib plus rituximab and Arm B is rituximab, fludarabine phosphate, and cyclophosphamide.
Confidence interval
The primary analysis reports a two-sided 95% confidence interval for the hazard ratio. The interval describes statistical uncertainty around the estimated treatment effect under the analysis framework. It is not a range containing 95% of individual patient effects.
Analysis population
The registry explicitly states that all randomized patients were included in the primary PFS analysis. This anchors the efficacy comparison to treatment assignment rather than restricting the primary comparison to patients who completed a particular treatment exposure.
Superiority hypothesis
The registered hypothesis type is superiority. Accordingly, the statistical question is whether the randomized treatment groups differ in the prespecified direction of the superiority framework, rather than whether one treatment is merely not worse than the other according to a non-inferiority margin.
7. Primary Result: Progression-Free Survival at 3 Years
The registry reports a formal statistical analysis for the primary endpoint, Progression-free Survival (PFS) Rate at 3 Years. All randomized patients were included. The groups compared were ibrutinib plus rituximab versus rituximab, fludarabine phosphate, and cyclophosphamide.
Hazard ratio for progression-free survival
95% CI: 0.22–0.56 · P < .001
Two-sided log-rank analysis; superiority hypothesis.
| Primary endpoint | Arm A | Arm B | Statistical analysis |
|---|---|---|---|
| Progression-free Survival (PFS) Rate at 3 Years | Ibrutinib + rituximab | Rituximab + fludarabine phosphate + cyclophosphamide | Log-rank; HR 0.35 (95% CI 0.22–0.56); P < .001 |
A hazard ratio of 0.35 means that the estimated instantaneous rate of progression or death in the ibrutinib-and-rituximab group was approximately 35% of the corresponding rate in the comparator group under the reported time-to-event analysis. Equivalently, this corresponds to an estimated 65% lower hazard of the event relative to the comparator.
The hazard ratio does not mean that 65% of participants avoided progression, that 65% of participants were cured, or that every participant experienced exactly the same reduction in event risk. It is a relative, model-based time-to-event measure.
The 95% confidence interval of 0.22–0.56 quantifies uncertainty around the estimated hazard ratio. Because the entire interval is below 1, the interval is consistent with a lower estimated hazard in Arm A across the range represented by the reported confidence interval.
The P < .001 value addresses the statistical evidence against the null hypothesis under the specified test. It does not measure the size of the treatment effect. Effect size is described by the hazard ratio, while the confidence interval describes precision.
The result should also be understood as a time-to-event comparison. Censoring, the timing of events, and the assumptions underlying interpretation of a hazard ratio all matter. A hazard ratio should not automatically be translated into a fixed absolute risk reduction at every point in follow-up.
8. How to Read the Primary PFS Result
The reported result contains three different pieces of statistical information: the effect estimate, its uncertainty interval, and the hypothesis-test result. Keeping these components separate is important.
Effect estimate
The HR of 0.35 summarizes the relative difference in the estimated instantaneous rate of progression or death between the randomized groups.
Precision
The 95% CI of 0.22–0.56 shows the uncertainty surrounding the reported HR estimate.
Evidence against the null
The two-sided log-rank P-value is <.001. This is evidence under the specified statistical test; it is not a measure of clinical magnitude.
What is not established
The HR alone does not provide the proportion of participants progression-free at 3 years, the median PFS, or the absolute difference in survival probability at 3 years.
9. Secondary Result: Overall Survival at 3 Years
The registry also reports a formal statistical analysis of the secondary endpoint Overall Survival (OS) Rate at 3 Years. All randomized patients were included in this analysis, and the same two randomized treatment groups were compared using a log-rank analysis with a hazard ratio.
Hazard ratio for overall survival
95% CI: 0.05–0.54 · P < .001
Two-sided log-rank analysis.
| Secondary endpoint | Arm A | Arm B | Statistical analysis |
|---|---|---|---|
| Overall Survival (OS) Rate at 3 Years | Ibrutinib + rituximab | Rituximab + fludarabine phosphate + cyclophosphamide | Log-rank; HR 0.17 (95% CI 0.05–0.54); P < .001 |
A hazard ratio of 0.17 means that the estimated instantaneous rate of death in the ibrutinib-and-rituximab group was approximately 17% of the corresponding estimated rate in the comparator group under the reported time-to-event analysis. Expressed as a relative hazard difference, this corresponds to an estimated 83% lower hazard of death.
The HR does not mean that 83% of participants survived, nor does it mean that each individual participant experienced an 83% reduction in mortality risk. It is a relative time-to-event measure.
The 95% confidence interval of 0.05–0.54 indicates uncertainty around the estimate. The interval is relatively broad on the multiplicative hazard-ratio scale, despite remaining below 1 throughout the reported interval.
The P < .001 value indicates strong statistical evidence against the null hypothesis under the reported two-sided log-rank test. It should not be interpreted as the probability that the treatment effect is real, nor as a measure of the magnitude of benefit.
As with PFS, the OS hazard ratio is subject to the structure of time-to-event analysis, including censoring and the interpretation of a single relative hazard measure across follow-up.
10. Primary and Secondary Findings Together
The two formal registry analyses provide a coherent time-to-event statistical picture: the reported hazard ratio for progression-free survival was 0.35, while the reported hazard ratio for overall survival was 0.17. Both analyses used a log-rank method, included all randomized patients, and reported two-sided P-values of <.001.
| Endpoint | Role | Analysis population | Method | HR | 95% CI | P-value |
|---|---|---|---|---|---|---|
| PFS Rate at 3 Years | Primary | All randomized patients | Log-rank | 0.35 | 0.22–0.56 | <.001 |
| OS Rate at 3 Years | Secondary | All randomized patients | Log-rank | 0.17 | 0.05–0.54 | <.001 |
These estimates should not be treated as interchangeable measures. PFS concerns progression or death, whereas OS concerns death from any cause. A lower PFS hazard and a lower OS hazard therefore answer related but distinct clinical questions.
11. Statistical Methods Explained
Why is a log-rank test appropriate for this trial?
The registered primary endpoint is a time-to-event outcome. A log-rank test compares the event-time distributions between randomized groups while incorporating the ordering and timing of observed events. This is more informative than reducing the endpoint to a simple yes/no outcome without regard to follow-up time.
What does an HR of 0.35 mean?
Within the reported time-to-event framework, an HR of 0.35 means that the estimated instantaneous rate of progression or death in the ibrutinib-and-rituximab group was approximately 35% of that in the comparator group. It is a relative hazard measure, not an absolute probability and not a statement that exactly 65% of patients benefited.
What does the 95% confidence interval of 0.22–0.56 mean?
The confidence interval describes uncertainty around the estimated hazard ratio. It provides a range of values compatible with the statistical estimation framework used for the reported analysis. It does not mean that 95% of patients have hazards within this range, and it does not describe individual treatment responses.
Why doesn't the P-value measure the treatment effect?
A P-value evaluates how compatible the observed data are with a specified null hypothesis under the statistical test. It is affected by the amount of information in the analysis and by the magnitude and variability of the observed difference. The hazard ratio describes the estimated effect, while the confidence interval provides information about its precision.
Why is PFS analyzed as a time-to-event endpoint rather than only as a 3-year proportion?
The registry names the primary endpoint as the PFS rate at 3 years, but it also defines PFS as the time from randomization to progression or death and reports a log-rank test and hazard ratio. This means that the statistical analysis uses the underlying event-time information. A fixed 3-year rate is a time-specific summary; the hazard ratio and log-rank test use the broader time-to-event structure.
Why are all randomized patients important in the primary analysis?
The registry states that all randomized patients were included. An analysis based on randomized assignment preserves the comparison created by randomization and avoids redefining the efficacy population based on treatment exposure or later events. This is especially important when interpreting treatment effects in randomized clinical trials.
Why should the hazard ratio not be interpreted as an absolute risk difference?
A hazard ratio is a relative measure of instantaneous event rates. An absolute risk difference compares probabilities at a defined time point. The same hazard ratio can correspond to different absolute differences depending on the underlying event rates and follow-up. Therefore, an HR of 0.35 does not by itself tell us how many additional participants were progression-free at exactly 3 years.
12. Censoring and Time-to-Event Interpretation
Time-to-event analysis is designed for settings in which not every participant experiences the event during observed follow-up. Participants may reach the end of available observation without progression or may otherwise have their event time censored. The registry's PFS endpoint is assessed every 3 months until progression up to 4 years and 8 months.
This is why a time-to-event analysis can use different amounts of follow-up across participants without treating participants with shorter observation as though they had the same follow-up opportunity as participants observed longer.
The statistical interpretation of a hazard ratio also depends on how the event process behaves over time. A single HR summarizes the relative hazard under the model and analysis framework; it should not automatically be interpreted as a constant absolute risk reduction throughout follow-up.
13. Randomization and Causal Interpretation
E1912 is registered as randomized, parallel, and unmasked. Randomization is central to the causal interpretation of the comparison because treatment assignment is determined by the trial rather than selected according to observed prognosis after enrollment.
What randomization contributes
It creates the treatment groups through an allocation process intended to make systematic differences between groups less likely than under non-randomized treatment selection.
What randomization does not guarantee
Randomization does not make the groups numerically identical on every baseline characteristic, nor does it eliminate uncertainty from sampling variation.
Parallel design
Participants are assigned to one of the two treatment strategies rather than sequentially receiving both randomized strategies as a crossover design.
Unmasked design
The registry identifies the masking status as none. The statistical interpretation should therefore distinguish the randomized treatment comparison from any potential effects of an unmasked treatment setting.
14. Analysis Population and the ITT Principle
The primary PFS and secondary OS analyses both state that all randomized patients were included. This is a key feature of the reported efficacy analysis.
| Endpoint | Analysis population reported |
|---|---|
| Progression-free Survival (PFS) Rate at 3 Years | All randomized patients were included in this analysis. |
| Overall Survival (OS) Rate at 3 Years | All randomized patients were included in this analysis. |
The important distinction is between the treatment assigned at randomization and what happens to a participant afterward. An efficacy analysis anchored to randomization retains the original treatment comparison rather than redefining treatment groups according to later adherence, discontinuation, or exposure.
15. Safety: Serious Adverse Events by Arm
The registry data provide serious adverse event counts by treatment arm. These figures are presented separately from the efficacy analyses because safety and efficacy answer different statistical questions.
| Safety measure | Affected participants | Participants at risk | Reported proportion |
|---|---|---|---|
| Arm A — Ibrutinib, Rituximab | 270 | 352 | 270/352 |
| Arm B — Rituximab, Fludarabine Phosphate, Cyclophosphamide | 132 | 158 | 132/158 |
The reported figures indicate that serious adverse events affected 270 of 352 participants at risk in Arm A and 132 of 158 participants at risk in Arm B. These counts should not be substituted for the efficacy hazard ratios: the denominators, event definitions, and statistical questions are different.
16. Multiplicity and Multiple Endpoints
The registry data identify one primary endpoint and also report a formal secondary analysis for overall survival. The ClinicalTrials.gov record does not describe a multiplicity-adjustment procedure, an alpha-allocation strategy, or an endpoint hierarchy beyond the distinction between primary and secondary endpoint roles.
| Endpoint | Role in the ClinicalTrials.gov record | Formal result reported |
|---|---|---|
| Progression-free Survival (PFS) Rate at 3 Years | Primary | Yes |
| Overall Survival (OS) Rate at 3 Years | Secondary | Yes |
This distinction matters because a P-value for a secondary endpoint should not automatically be interpreted as though it were the P-value for the primary endpoint. The registry labels OS as secondary, while the primary endpoint is the 3-year PFS rate.
17. Interim Analysis, Non-Inferiority, and Bayesian Methods
Non-inferiority margin
The registered hypothesis type is superiority. The ClinicalTrials.gov record does not report a non-inferiority margin, so no margin-based analysis is appropriate for this page.
Interim analysis
The ClinicalTrials.gov record does not report an interim-analysis schedule or an alpha-spending procedure.
Bayesian methods
No Bayesian analysis is reported in the ClinicalTrials.gov record.
Crossover
The ClinicalTrials.gov record does not report a crossover design or crossover analysis. The registered design model is parallel.
These distinctions are important because different trial-design features require different statistical interpretations. In particular, superiority testing should not be explained using a non-inferiority margin, and an unreported interim or Bayesian procedure should not be inferred simply because the trial is a phase 3 study.
18. Missing Data and Imputation
The ClinicalTrials.gov record does not report a missing-data or imputation method for the primary PFS or secondary OS analyses. The primary endpoints are time-to-event outcomes, for which incomplete follow-up is typically handled through censoring within the survival-analysis framework rather than by simply imputing a final outcome value for every participant.
A censored participant contributes observed follow-up information up to the censoring point. Treating an unobserved future event time as though it were a missing continuous measurement requiring a single imputed value would represent a different statistical problem.
Because the ClinicalTrials.gov record does not specify additional missing-data assumptions, this page does not attribute a particular imputation strategy to E1912.
19. Stratification
The ClinicalTrials.gov record identifies the method as a log-rank test but do not provide stratification factors. Consequently, this page does not infer stratification variables or describe the analysis as stratified.
20. Why the Hazard Ratio Is Not a Three-Year Survival Rate
The primary endpoint is named “Progression-free Survival (PFS) Rate at 3 Years,” but the formal statistical analysis reports a hazard ratio from a log-rank comparison. These are related but distinct quantities.
| Measure | Question it answers | E1912 value reported |
|---|---|---|
| 3-year PFS rate | What proportion remains progression-free at the specified time point? | Not numerically reported in the ClinicalTrials.gov record. |
| Hazard ratio | How do the estimated instantaneous event rates compare between randomized groups? | 0.35 |
| 95% confidence interval | How uncertain is the estimated hazard ratio? | 0.22–0.56 |
| P-value | How compatible are the data with the specified null hypothesis under the log-rank test? | <.001 |
This is why a hazard ratio should not be presented as though it were the percentage of patients alive or progression-free at a particular time. The registry supplies a time-specific endpoint label and a time-to-event effect measure, but the ClinicalTrials.gov record does not include the numerical 3-year PFS proportion itself.
21. What the Overall Survival Hazard Ratio Does — and Does Not — Mean
The reported OS HR of 0.17 indicates a substantially lower estimated instantaneous rate of death in the ibrutinib-and-rituximab group relative to the comparator within the reported time-to-event analysis.
It does not mean that 17% of patients died, that 83% of patients survived, or that every individual participant experienced the same relative reduction in mortality risk.
The reported 95% CI of 0.05–0.54 provides an uncertainty range for the estimated HR. The width of the interval reflects that the point estimate alone does not fully describe the precision of the treatment-effect estimate.
The P-value of <.001 indicates statistical evidence against the null hypothesis under the two-sided log-rank test. It does not tell us whether the HR is clinically large or small. That assessment begins with the effect estimate and its confidence interval and then considers the endpoint and clinical context.
22. Statistical Interpretation vs Clinical Interpretation
Statistical interpretation
The primary PFS analysis reported an HR of 0.35 with a two-sided 95% CI of 0.22–0.56 and P <.001 using a log-rank test. The secondary OS analysis reported an HR of 0.17 with a two-sided 95% CI of 0.05–0.54 and P <.001.
Clinical interpretation
The statistical results describe differences in time-to-event outcomes between randomized treatment groups. The ClinicalTrials.gov record does not provide enough numerical information to summarize absolute 3-year PFS or OS rates, median survival, or other clinical outcome measures beyond the reported hazard ratios.
This distinction prevents the statistical evidence from being stretched beyond what the ClinicalTrials.gov record can support. The reported hazard ratios provide relative time-to-event comparisons; they do not provide every clinically relevant measure of benefit.
23. Important Limitations and Interpretation Issues
- Incomplete endpoint detail: the registry definition of PFS ends after “≥ 50%”. The page does not reconstruct the remainder of the definition from external information.
- Absolute 3-year rates are not reported: although the primary endpoint is expressed as a 3-year PFS rate, the ClinicalTrials.gov record reports the HR, confidence interval, and P-value rather than the numerical 3-year PFS proportions.
- Median survival is not reported: no median PFS or median OS is included in the ClinicalTrials.gov record, so none is reported here.
- Hazard ratio interpretation: the HR is a relative time-to-event measure and should not be interpreted as an absolute risk difference or as the probability of individual benefit.
- Censoring: time-to-event analysis incorporates incomplete event observation through follow-up and censoring; the ClinicalTrials.gov record does not provide the underlying censoring pattern.
- Proportional-hazards caution: a single hazard ratio summarizes relative event rates over the analyzed follow-up. The ClinicalTrials.gov record does not provide a formal assessment of the proportional-hazards assumption.
- Multiplicity: the ClinicalTrials.gov record identifies one primary and one secondary formal analysis but do not describe a multiplicity-adjustment procedure.
- Subgroups: no subgroup estimates are reported in the ClinicalTrials.gov record, so subgroup-specific treatment effects are not presented.
- Missing-data methods: the ClinicalTrials.gov record does not specify a separate imputation strategy.
- Design-specific features: no non-inferiority margin, Bayesian method, crossover analysis, or interim-analysis procedure is reported in the ClinicalTrials.gov record.
- Safety comparison: serious adverse event counts are reported by arm, but no formal comparative safety analysis is provided in the data.
24. Why This Trial Matters Statistically
E1912 is a useful teaching example because its registry record connects a randomized phase 3 design with a classic survival-analysis framework. The primary endpoint is explicitly time-to-event, the formal method is a log-rank test, and the effect measure is a hazard ratio.
| Concept | How it appears in E1912 |
|---|---|
| Randomization | The trial uses randomized allocation across two parallel treatment arms. |
| Time-to-event endpoint | PFS is defined from randomization to progression or death, whichever occurs first. |
| Fixed-time endpoint | The primary endpoint is expressed as the PFS rate at 3 years. |
| Log-rank test | The registry reports a log-rank comparison for the primary PFS analysis and the secondary OS analysis. |
| Hazard ratio | The treatment effect is reported using HR 0.35 for PFS and HR 0.17 for OS. |
| Confidence interval | Both reported HRs include two-sided 95% confidence intervals. |
| Superiority | The registered primary hypothesis type is superiority. |
| Intention-to-treat principle | All randomized patients were included in the reported PFS and OS analyses. |
| Safety analysis | Serious adverse event counts are reported by randomized treatment arm with affected and at-risk counts. |
25. Related Tutorials
Learn more about the methods used in this trial:
26. Related Calculators
27. Sources
- ClinicalTrials.gov: E1912, NCT02048813.
- PubMed record: PMID 37801678.
- PubMed record: PMID 35427411.
- PubMed record: PMID 34865212.
- PubMed record: PMID 33439748.
- PubMed record: PMID 31365801.
Continue with the underlying statistical methods
Explore survival-analysis tutorials and statistical calculators related to confidence intervals, hazard ratios, Kaplan-Meier estimation, log-rank testing, randomization, and time-to-event endpoints.
28. Record Summary
E1912 provides a clear example of randomized clinical-trial survival analysis. The trial enrolled 529 participants in a phase 3 randomized parallel design with two treatment arms. Its registered primary endpoint was the Progression-free Survival (PFS) Rate at 3 Years, defined as the time from randomization to CLL progression or death, whichever occurred first, and assessed every 3 months until progression up to 4 years and 8 months.
The formal primary analysis included all randomized patients and used a log-rank test with a hazard ratio as the effect measure. The reported PFS HR was 0.35, with a two-sided 95% CI of 0.22–0.56 and P <.001. The secondary OS analysis used the same general time-to-event comparison and reported an HR of 0.17, with a two-sided 95% CI of 0.05–0.54 and P <.001.
The most important statistical lesson is that these hazard ratios are relative time-to-event measures. They should be interpreted together with their confidence intervals, their underlying endpoint definitions, the randomized analysis population, and the limitations of the available registry information. The ClinicalTrials.gov record does not provide the numerical 3-year survival proportions, median survival times, subgroup estimates, or detailed multiplicity and interim-analysis procedures, so those quantities are not inferred here.