The ClinicalTrials.gov record lists MAJIC as Active, not recruiting. The study began on September 12, 2022, and the primary completion date is listed as July 15, 2027. The record reports enrollment of 607 participants.
1. Trial at a Glance
MAJIC is a randomized, open-label, parallel phase 3 treatment trial evaluating whether minimal residual disease (MRD)-driven finite treatment with acalabrutinib plus venetoclax is non-inferior to MRD-driven finite treatment with venetoclax plus obinutuzumab with respect to progression-free survival in previously untreated chronic lymphocytic leukemia or small lymphocytic lymphoma.
| Feature | MAJIC |
|---|---|
| Study acronym | MAJIC |
| ClinicalTrials.gov identifier | NCT05057494 |
| Phase | Phase 3 |
| Status | Active, not recruiting |
| Condition | Chronic Lymphocytic Leukemia or Small Lymphocytic Lymphoma |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | None |
| Primary purpose | Treatment |
| Enrollment | 607 |
| Number of arms | 2 |
| Lead sponsor | AstraZeneca |
| Sponsor type | Industry |
2. Clinical Question
The primary statistical question is whether MRD-driven finite AV treatment is non-inferior to MRD-driven finite VO treatment with respect to progression-free survival in previously untreated patients with chronic lymphocytic leukemia or small lymphocytic lymphoma.
Population
Patients with previously untreated chronic lymphocytic leukemia or small lymphocytic lymphoma.
Intervention
Acalabrutinib plus venetoclax, with treatment described in the registry as MRD-driven finite AV treatment.
Comparator
Venetoclax plus obinutuzumab, with treatment described in the registry as MRD-driven finite VO treatment.
Primary question
Is MRD-driven finite AV treatment non-inferior to MRD-driven finite VO treatment for progression-free survival?
3. Trial Design
Acalabrutinib + Venetoclax
- Acalabrutinib
- Venetoclax
- MRD-driven finite AV treatment
Venetoclax + Obinutuzumab
- Venetoclax
- Obinutuzumab
- MRD-driven finite VO treatment
The registry identifies the study as randomized, parallel, and unmasked. Its primary purpose is treatment. Two treatment arms are listed, and the interventions are acalabrutinib, venetoclax, and obinutuzumab.
4. Study Timeline and Registry Status
Study start
The ClinicalTrials.gov record lists September 12, 2022 as the study start date.
Active, not recruiting
The registry currently identifies the study as Active, not Recruiting.
Primary completion date
The registry lists July 15, 2027 as the primary completion date.
The timeline is important for interpreting the absence of efficacy estimates. The registry describes an ongoing study whose primary completion date is in 2027. The primary endpoint has a follow-up horizon extending to up to 6.6 years.
5. Primary Endpoint
| Endpoint | Registry definition | Time frame |
|---|---|---|
| Progression-free Survival (PFS) | To assess whether minimal residual disease (MRD)-driven finite AV treatment is NI to MRD-driven finite VO treatment with respect to PFS. PFS is defined as the time from the date of randomization until date of objective progressive disease per International Workshop on Chronic Lymphocytic Leukemia (iwCLL) 2018 criteria as assessed by the investigator or death from any cause in the absence of progression. | Until progressive disease (PD) [assessed Up to 6.6 Years]. |
What PFS measures
PFS is a time-to-event endpoint. The event is the first occurrence of either objective progressive disease according to iwCLL 2018 criteria, as assessed by the investigator, or death from any cause in the absence of progression.
This definition has an important statistical consequence: a patient does not need to experience both progression and death to have a PFS event. Either qualifying progression or death can establish the event. Patients who have not experienced a qualifying event by the relevant follow-up point contribute information up to their censoring time under the applicable analysis rules.
6. Planned Analysis
Primary statistical objective
Primary endpoint: progression-free survival
Registry assessment window: Until progressive disease, assessed up to 6.6 years
The ClinicalTrials.gov record identifies the primary endpoint as PFS and explicitly frames the primary comparison as a non-inferiority assessment of MRD-driven finite AV treatment versus MRD-driven finite VO treatment.
For a randomized time-to-event endpoint such as PFS, a typical analysis estimates the distribution of time to progression or death using the Kaplan-Meier method and compares treatment groups using a time-to-event treatment-effect measure such as a hazard ratio. A Cox proportional-hazards model is commonly used to estimate such a hazard ratio and its confidence interval. The registry information presented here does not report the specific statistical test, model specification, confidence level, or non-inferiority margin.
Progression is defined using iwCLL 2018 criteria and assessed by the investigator. Death without progression is also a PFS event.
Why non-inferiority changes the statistical question
A non-inferiority trial does not ask whether the experimental treatment must produce a numerically better outcome than the comparator. Instead, it asks whether the experimental treatment can be shown to be not unacceptably worse than the comparator according to a prespecified non-inferiority margin.
For a hazard-ratio formulation, the precise non-inferiority logic depends on how the endpoint and margin are defined. The important principle is that the confidence interval for the treatment effect must be evaluated against the prespecified non-inferiority boundary. A conventional two-sided test of equality is not, by itself, the defining criterion for a non-inferiority conclusion.
7. Statistical Methodology
Randomization
MAJIC is described as a randomized study with two parallel treatment arms. Randomization is the design feature that creates the principal basis for a causal comparison between the two treatment strategies, because treatment assignment is determined independently of subsequent outcome development under the randomized allocation.
The key estimand associated with a randomized comparison is therefore a treatment effect attributable to assignment to one strategy versus assignment to the other, rather than simply an association between treatment received and outcome.
Kaplan-Meier estimation
Because PFS records the time until progression or death, Kaplan-Meier estimation is a standard method for describing the event-time distribution. It allows participants who have not experienced an event by the end of their observed follow-up to contribute information without treating their follow-up as if an event had occurred.
Here, di represents events at a particular event time and ni represents participants at risk immediately before that time.
Hazard ratios
A hazard ratio is a relative comparison of the instantaneous event rates between treatment groups under a specified time-to-event model. For PFS, the event is progression or death as defined by the endpoint.
The direction of the hazard ratio depends on which treatment is placed in the numerator. A hazard ratio is not a probability, a median, or a direct measure of how many patients benefit.
Cox proportional-hazards modeling
A Cox model is a common method for estimating a hazard ratio for a time-to-event endpoint. Its proportional-hazards interpretation assumes that the relative hazard is adequately summarized by a treatment effect that does not vary in a way that materially violates the model over time.
That assumption should be examined when interpreting a single hazard ratio. If the treatment effect changes substantially over time, a single HR may compress a more complicated pattern of separation between the treatment groups.
Censoring
PFS analysis commonly involves right censoring. A participant who has not experienced progression or death during observed follow-up is not treated as though an event occurred at the end of follow-up. Instead, the available event-free follow-up contributes information to the estimated PFS distribution.
The validity of survival-analysis estimates depends on appropriate handling of censoring according to the prespecified analysis rules. The registry information presented here does not specify the detailed censoring or missing-data rules.
Investigator-assessed progression
The registry definition specifies objective progressive disease per iwCLL 2018 criteria as assessed by the investigator. Thus, the PFS endpoint incorporates an investigator assessment rather than defining progression solely through an independently described central-review process.
8. Statistical Methods Explained
Why is PFS a time-to-event endpoint?
PFS records not only whether progression or death occurs, but when the first qualifying event occurs after randomization. Two patients can both experience progression while having substantially different event times. A time-to-event framework preserves this timing information.
Why use Kaplan-Meier estimation?
Not every participant will necessarily have experienced progression or death at a particular analysis time. Kaplan-Meier estimation handles this incomplete follow-up through censoring rather than requiring every participant to have an observed event.
What does a hazard ratio compare?
A hazard ratio compares estimated instantaneous event rates between the treatment groups within a time-to-event model. For this trial, the event is PFS progression or death. An HR below 1, if observed, would indicate a lower estimated event hazard in the AV group relative to the VO group under the model's direction of comparison.
Why is non-inferiority judged against a margin?
Non-inferiority requires a clinically or scientifically prespecified boundary defining how much worse the experimental strategy could be while still satisfying the trial's non-inferiority criterion. The question is therefore not simply whether the observed treatment effect differs statistically from zero.
Why does the confidence interval matter?
A confidence interval describes statistical uncertainty around an estimated treatment effect. In a non-inferiority trial, the interval is particularly important because the entire interval is evaluated in relation to the prespecified non-inferiority margin. A point estimate alone cannot establish non-inferiority.
Why does the PFS definition include death?
Including death prevents death without documented progression from being treated as though it were a permanently event-free observation. The registry explicitly defines death from any cause in the absence of progression as a PFS event.
9. Non-Inferiority Interpretation
The central statistical feature of MAJIC is that the primary question is framed as non-inferiority. This makes the interpretation different from a conventional superiority trial in which the principal question is whether one treatment has a statistically detectable advantage over another.
Superiority question
Is the treatment effect sufficiently different from the null value to support evidence that one treatment improves the endpoint relative to the comparator?
Non-inferiority question
Is the experimental treatment sufficiently close to the comparator that any disadvantage does not cross the prespecified acceptable margin?
For MAJIC, the registry explicitly states that the objective is to assess whether MRD-driven finite AV treatment is NI to MRD-driven finite VO treatment with respect to PFS. The numerical non-inferiority margin is not reported in the registry information available here.
A future PFS estimate must be interpreted in relation to the prespecified non-inferiority margin, not simply by asking whether the confidence interval crosses the conventional null value. The margin determines what degree of potential loss of efficacy remains compatible with the non-inferiority objective.
Demonstrating non-inferiority would not mean that the two treatment strategies are mathematically identical. It would mean that the observed uncertainty around their relative PFS effect is compatible with a difference no worse than the prespecified acceptable boundary.
10. MRD-Driven Finite Treatment as a Statistical Design Feature
The primary endpoint description specifically refers to minimal residual disease (MRD)-driven finite treatment in both strategies. This wording is important because the treatment strategies are not described simply as indefinite exposure to the assigned drug combinations.
| Feature | AV strategy | VO strategy |
|---|---|---|
| Drug combination | Acalabrutinib + venetoclax | Venetoclax + obinutuzumab |
| Treatment framework | MRD-driven finite treatment | MRD-driven finite treatment |
| Primary comparison | Progression-free survival | |
| Registry objective | Assess whether AV is non-inferior to VO with respect to PFS | |
From a statistical perspective, this means that PFS is evaluating the consequences of two treatment strategies rather than merely comparing exposure to one drug with exposure to another. Treatment duration and the MRD-driven treatment framework are therefore part of what the randomized comparison represents.
11. What the Primary Endpoint Can Tell Us
PFS is particularly suited to questions in which the timing of disease progression is clinically important. It can distinguish treatment groups even when the ultimate survival experience is not yet mature, although the interpretation of PFS remains dependent on how progression is defined and assessed.
What PFS captures
Time from randomization until objective progressive disease or death from any cause in the absence of progression.
What PFS does not directly capture
Overall survival, quality of life, treatment burden, or the complete safety profile are not represented by a PFS hazard ratio alone.
Why timing matters
A participant remaining progression-free for a longer period contributes a different amount of time-to-event information from a participant progressing shortly after randomization.
Why the endpoint is composite
PFS combines objective progression and death without progression into one time-to-event endpoint.
12. Limitations
- No posted statistical analyses: The ClinicalTrials.gov record does not currently list posted statistical analyses for the primary endpoint.
- No posted outcome estimates: The registry information available here does not provide a PFS hazard ratio, confidence interval, p-value, median PFS, event count, or treatment-specific PFS estimate.
The registry describes the primary objective as non-inferiority but does not provide a numerical non-inferiority margin in the available record. - Detailed analysis specification not reported: The available registry information does not specify the primary statistical test, model details, confidence level, or detailed censoring rules.
- Single primary endpoint: The registry identifies PFS as the registered primary endpoint. Other efficacy or safety outcomes should not be substituted for the primary statistical question.
- Investigator assessment: Progression is defined using iwCLL 2018 criteria as assessed by the investigator. The registry information available here does not describe an independent central review process for this endpoint.
- Ongoing study timeline: The registry lists a primary completion date of July 15, 2027, with PFS assessed up to 6.6 years.
- Open-label design: The study is unmasked. Knowledge of treatment assignment can be relevant to the interpretation of outcomes that depend on assessment or clinical management, although the specific magnitude of any such effect cannot be determined from the registry information alone.
13. Why This Trial Matters Statistically
MAJIC is a useful teaching case because its primary question illustrates the difference between a conventional superiority framework and a non-inferiority time-to-event analysis. It also shows how treatment strategy, endpoint definition, randomization, censoring, and follow-up duration fit together in a phase 3 clinical trial.
| Statistical concept | How it appears in MAJIC |
|---|---|
| Randomization | The study uses randomized allocation to two treatment arms. |
| Parallel design | The two treatment strategies are evaluated in parallel arms. |
| Non-inferiority | The primary objective is to assess whether MRD-driven finite AV treatment is non-inferior to MRD-driven finite VO treatment with respect to PFS. |
| Time-to-event analysis | PFS measures time from randomization to objective progression or death. |
| Composite endpoint | PFS incorporates objective progressive disease and death from any cause in the absence of progression. |
| Kaplan-Meier estimation | A standard framework for estimating PFS over time while accommodating censored observations. |
| Hazard ratio | A standard relative treatment-effect measure for comparing event rates over time. |
| Confidence interval | Used to quantify uncertainty around an estimated treatment effect and, in a non-inferiority setting, to assess the relationship with the prespecified margin. |
| Analysis population | The registry reports total enrollment of 607 participants; detailed analysis-population definitions are not reported in the available information. |
| Long follow-up | The PFS time frame extends until progressive disease, assessed up to 6.6 years. |
14. Reading a Future MAJIC PFS Result
If a future registry update reports a PFS hazard ratio and confidence interval, the first task should be to identify the direction of the comparison. For example, if AV is the numerator, an HR below 1 would correspond to a lower estimated PFS event hazard for AV relative to VO. The next task in a non-inferiority analysis is to compare the confidence interval with the prespecified non-inferiority margin.
The numerical boundary is essential. A point estimate alone does not establish non-inferiority, and the conventional null value is not a substitute for the prespecified non-inferiority margin.
The key distinction is between estimating the relative PFS effect and making a non-inferiority conclusion. The first requires an appropriate time-to-event analysis. The second additionally requires the prespecified non-inferiority margin and the trial's formal inferential framework. Those elements should be read together rather than treating a hazard ratio as a standalone conclusion.
15. PFS, Progression, and Censoring
The registry definition begins the PFS clock at randomization. This establishes a common time origin for the randomized treatment comparison. The event occurs at objective progressive disease according to iwCLL 2018 criteria or at death from any cause in the absence of progression.
| Situation | Relationship to the registered PFS definition |
|---|---|
| Randomization | Starting point for PFS time |
| Objective progressive disease | PFS event when assessed according to iwCLL 2018 criteria by the investigator |
| Death without progression | PFS event |
| No progression or death during observed follow-up | Typically contributes censored event-time information in a time-to-event analysis |
| Assessment horizon | Until progressive disease, assessed up to 6.6 years |
This structure prevents a common statistical mistake: treating participants without an observed event as if their follow-up were equivalent to an observed progression-free outcome at the end of the study. Survival methods instead distinguish observed events from incomplete event-free follow-up.
16. Trial Design vs Statistical Evidence
The design establishes what comparison the trial is capable of addressing; the analysis establishes how evidence for that comparison is quantified. In MAJIC, the design is randomized and parallel, while the primary endpoint is a time-to-event measure and the primary question is non-inferiority.
Design layer
Two randomized treatment strategies are compared in a phase 3 parallel trial with 607 enrolled participants.
Endpoint layer
PFS is measured from randomization to objective progression or death.
Inference layer
The primary question is whether AV is non-inferior to VO with respect to PFS.
Interpretation layer
A future treatment-effect estimate must be interpreted together with its uncertainty and the prespecified non-inferiority margin.
17. What the Registry Currently Reports
The ClinicalTrials.gov record provides the principal design information for MAJIC: it is a phase 3 randomized, parallel, unmasked treatment study with 607 participants and two arms. Its primary endpoint is PFS, with the explicit objective of assessing non-inferiority of MRD-driven finite AV treatment to MRD-driven finite VO treatment.
| Information category | Registry information available here |
|---|---|
| Study identity | MAJIC, NCT05057494 |
| Phase | Phase 3 |
| Status | Active, not recruiting |
| Enrollment | 607 |
| Allocation | Randomized |
| Design | Parallel |
| Masking | None |
| Primary endpoint | Progression-free Survival (PFS) |
| Primary objective | Assess whether MRD-driven finite AV treatment is NI to MRD-driven finite VO treatment with respect to PFS |
| PFS definition | Time from randomization to objective progressive disease per iwCLL 2018 criteria as assessed by the investigator or death from any cause in the absence of progression |
| PFS timeframe | Until progressive disease, assessed up to 6.6 years |
| Posted statistical analyses | None |
18. Current Statistical Interpretation
There is currently no numerical efficacy result in the registry information available here from which to estimate a treatment effect for MAJIC's primary endpoint. The appropriate statistical focus is therefore the structure of the planned PFS comparison rather than an outcome estimate.
The most important feature is the non-inferiority framework. A valid interpretation will require at least three linked pieces of information: the estimated relative PFS effect, its confidence interval, and the prespecified non-inferiority margin. The interpretation also depends on the analysis population, censoring rules, and statistical model specified for the primary analysis.
19. Important Questions for Interpreting the Final Analysis
What is the PFS hazard ratio?
The registry information available here does not report a PFS hazard ratio.
What is the non-inferiority margin?
The registry information available here does not report a numerical non-inferiority margin.
How many PFS events occurred?
The registry information available here does not report a PFS event count.
What was median PFS?
The registry information available here does not report a median PFS for either treatment strategy.
These distinctions matter because a trial's enrollment number and design characteristics do not provide enough information to reconstruct its eventual PFS treatment effect. Event timing, censoring, and the prespecified inferential framework are integral parts of the analysis.
20. Why Non-Inferiority Requires More Than a P-Value
A common misconception is that a non-inferiority trial is successful whenever a conventional hypothesis test produces a non-significant p-value. That is not the correct general logic. Failure to demonstrate superiority or a statistically significant difference does not establish non-inferiority.
Instead, non-inferiority is based on whether the estimated treatment effect and its confidence interval remain within a prespecified acceptable loss of efficacy. The direction of the comparison and the definition of the margin must be established before interpreting the result.
A non-inferiority conclusion requires a prespecified margin and an inferential procedure designed to evaluate that margin.
For MAJIC, this principle is central because the registry explicitly defines the primary objective as a non-inferiority comparison of the two MRD-driven finite treatment strategies.
21. Statistical Perspective on the 6.6-Year Assessment Window
The PFS time frame is specified as until progressive disease, assessed up to 6.6 years. This means the primary endpoint is inherently longitudinal rather than a single fixed-time binary outcome.
A time-to-event framework can use information from events occurring at different times throughout follow-up. It also allows participants who remain event-free at a particular assessment point to contribute their observed follow-up through censoring, subject to the prespecified analysis rules.
The long assessment horizon also reinforces why a single number should not be interpreted without knowing the underlying event-time distribution. A hazard ratio summarizes a relative treatment effect under a model, while Kaplan-Meier estimates can show how the estimated event-free probability evolves over time.
22. Clinical Interpretation vs Statistical Interpretation
Statistical interpretation
MAJIC is designed to compare PFS between two randomized treatment strategies using a non-inferiority framework.
Clinical interpretation
The treatment strategies differ in their drug combinations: acalabrutinib plus venetoclax versus venetoclax plus obinutuzumab, both described in the registry as MRD-driven finite approaches.
These two perspectives should remain distinct. Statistical non-inferiority concerns the prespecified PFS comparison and its uncertainty. Broader clinical interpretation would require consideration of additional outcomes, including benefits, harms, treatment burden, and patient-relevant consequences, which are not represented by the registered primary endpoint alone.
23. Why This Trial Is a Useful Non-Inferiority Teaching Example
MAJIC illustrates several important principles of clinical-trial statistics without requiring a numerical result to understand the underlying framework.
| Lesson | Statistical significance |
|---|---|
| Randomization | Creates the foundation for a comparative treatment-effect estimate. |
| Time-to-event endpoint | Preserves information about when progression or death occurs. |
| Composite PFS event | Counts objective progression and death without progression as qualifying events. |
| Non-inferiority | Tests whether potential loss of efficacy stays within an acceptable prespecified boundary. |
| Confidence interval | Quantifies uncertainty and is central to non-inferiority interpretation. |
| Censoring | Allows incomplete event-time observations to contribute information without being treated as events. |
| Follow-up | Provides the time over which progression and death can be observed. |
24. Limitations of Interpreting an Unreported Primary Result
The absence of posted statistical analyses means that the numerical evidence needed to evaluate the primary PFS comparison is not present in the registry information represented here. This is particularly important in a non-inferiority trial, where the treatment-effect estimate, confidence interval, and non-inferiority margin work together.
The registry does establish the clinical question, treatment strategies, primary endpoint definition, and follow-up timeframe. It does not provide the numerical result needed to determine where the eventual PFS estimate lies relative to a non-inferiority boundary.
The distinction is methodological rather than merely documentary: a non-inferiority conclusion cannot be established from enrollment, treatment assignment, or endpoint definition alone.
25. Sources
- ClinicalTrials.gov: MAJIC — NCT05057494.
- PubMed: PubMed record associated with MAJIC.
Continue through the Clinical Biostats clinical trial library
Explore additional clinical trial statistical analyses and educational material covering study design, endpoints, survival analysis, and statistical inference.
26. Record Summary
MAJIC is a phase 3 randomized, parallel, unmasked treatment trial enrolling 607 participants with previously untreated chronic lymphocytic leukemia or small lymphocytic lymphoma. It compares acalabrutinib plus venetoclax with venetoclax plus obinutuzumab, with both strategies described in the primary endpoint as MRD-driven finite treatment.
The registered primary endpoint is progression-free survival, defined from randomization until objective progressive disease according to iwCLL 2018 criteria as assessed by the investigator or death from any cause in the absence of progression. The assessment timeframe extends until progressive disease, assessed up to 6.6 years.
The central statistical question is explicitly one of non-inferiority: whether MRD-driven finite AV treatment is non-inferior to MRD-driven finite VO treatment with respect to PFS. The registry information represented here does not report a numerical non-inferiority margin or posted statistical analyses, so the current statistical interpretation appropriately focuses on the design and inferential framework rather than an outcome estimate.