A Study Being Conducted at Multiple Locations to Compare Safety and Efficacy of Three Different Regimens; (1) High-Dose Lenalidomide; (2) Lenalidomide + Azacitidine; or (3) Azacitidine in Subjects ≥ 65 Years With Newly-Diagnosed Acute Myeloid Leukemia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study Being Conducted at Multiple Locations to Compare Safety and Efficacy of Three Different Regimens; (1) High-Dose Lenalidomide; (2) Lenalidomide + Azacitidine; or (3) Azacitidine in Subjects ≥ 65 Years With Newly-Diagnosed Acute Myeloid Leukemia (NCT01358734), a randomized phase 2 study. The primary outcome is Kaplan Meier Estimates for One Year Survival.

2. Clinical Question

3. Trial Design

This was a phase 2, multicenter, randomized, open-label, parallel-group study sponsored by Celgene. It enrolled people aged ≥65 years with newly diagnosed acute myeloid leukemia and compared three treatment strategies: lenalidomide alone, azacitidine followed sequentially by lenalidomide, and azacitidine alone. The registry lists enrollment of 88 and marks the study completed.

The treatment schedules differed in both drugs and timing. The combination arm used azacitidine 75 mg/m^2/day subcutaneously on Days 1-7, then lenalidomide 50 mg/day by mouth on Days 8-28, followed by a 14-day break. The lenalidomide-alone schedule began with 50 mg daily for the first 2 cycles, then 25 mg daily for the next 2 cycles, followed by continuous 28-day cycles of 10 mg daily. The azacitidine-alone schedule used azacitidine 75mg/m^2/day subcutaneously on Days 1-7, followed by a 21-day break. Each arm description included best supportive care.

Randomization assigns treatment groups by chance, which can reduce systematic differences between groups at the start of a trial. The record does not report the randomization procedure, allocation ratio, or stratification factors. Open-label means that masking was not used; the record does not describe whether this affected assessment or treatment decisions.

The primary analysis population was the intention-to-treat population, described as all randomized participants. The registry record does not report the number randomized to each group as a separate design statement, nor does it describe adherence, crossovers, or protocol deviations. Its reported group sizes for the primary outcome are Lenalidomide 15.0, Azacitidine Plus Lenalidomide 39.0, and Azacitidine 34.0. The original sample-size justification and the trial's planned statistical analysis document 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.

Before assignment: Participants were centrally randomized 1:1:1 and stratified by the Eastern Cooperative Oncology Group (ECOG) performance score (0 to 1 versus 2) and peripheral blood blast count (\<1 versus ≥ 1 x 10\^9/L).

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: ITT included all randomized participants

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: Intent To Treat (ITT) includes all randomized participants

6. Primary Endpoint

The registry lists the primary outcome as Kaplan Meier Estimates for One Year Survival, measured over Up to 24 months.

The primary outcome is titled “Kaplan Meier Estimates for One Year Survival.” The registry describes the measure as median survival, in months, with a 95% confidence interval, over a timeframe of up to 24 months. Its definition says that deaths within one year from randomization counted as events, while other participants were censored at year 1 or at discontinuation. The coexistence of a one-year survival label, this event definition, and a median reported over a timeframe of up to 24 months is not fully explained in the registry record.

In general, survival outcomes are useful when the timing of an event matters, rather than only whether it occurred. Kaplan-Meier methods estimate the event-free proportion over time while accounting for participants whose follow-up ends before an event is observed. Censoring means that the participant contributes information up to the stated censoring point, without being treated as having experienced the event afterward.

A median is the estimated time at which the Kaplan-Meier survival curve reaches the halfway point. It gives a time-based summary, but does not describe the entire curve or the timing of every event. The hazard ratio comparisons provide a separate summary of relative event rates over follow-up. Neither summary alone reveals all features of the survival experience.

7. Results Posted in the Registry

• Lenalidomide vs Azacitidine Plus Lenalidomide: hazard ratio 1.79 (95% CI 0.86–3.72), reported p = 0.119 (as reported in the registry).
• Lenalidomide vs Azacitidine: hazard ratio 2.66 (95% CI 1.21–5.82), reported p = 0.014 (as reported in the registry).
• Azacitidine Plus Lenalidomide vs Azacitidine: hazard ratio 1.37 (95% CI 0.70–2.65), reported p = 0.356 (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.

Percentage of Participants With 30-Day Treatment-Related Mortality

• 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:
  • Azacitidine Plus Lenalidomide: 17.9% (7/39) versus 13.3% (2/15) on control; risk difference 4.6 percentage points (95% CI -16.4 to 25.6), risk ratio 1.35 (0.31–5.76); unadjusted two-proportion z-test p = 0.684.
  • Azacitidine: 5.9% (2/34) versus 13.3% (2/15) on control; risk difference -7.5 percentage points (95% CI -26.4 to 11.5), risk ratio 0.44 (0.07–2.84); unadjusted two-proportion z-test p = 0.380.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Azacitidine Plus Lenalidomide and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Azacitidine and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists secondary outcomes covering response, remission duration, cytogenetic response, progression-free survival, event-free survival, relapse-free survival, and 30-day treatment-related mortality. These outcomes examine different aspects of treatment experience. Response measures concern disease status, while time-to-event outcomes concern the timing of later events. They are related to the primary survival question but are not interchangeable with it.

The numerical secondary result available for interpretation in the record is 30-day treatment-related mortality: 13.3% for Lenalidomide, 17.9% for Azacitidine Plus Lenalidomide, and 5.9% for Azacitidine. The record does not list sponsor-reported inferential analyses for these percentages. Numerical results for the other listed secondary outcomes are not reported in the registry record, so an overall pattern across response and other time-to-event outcomes cannot be assessed from the available record.

Secondary outcomes can help describe whether findings across different measures point in a similar direction, but they do not automatically confirm the primary result. Some outcomes may be based on fewer events, different definitions, or different follow-up, and the registry does not report enough detail to assess those features for every endpoint. A numerical pattern in one secondary measure should not be treated as a substitute for the primary survival analysis.

Unadjusted secondary results carry less weight when many outcomes or comparisons are examined because repeated testing can increase the chance of finding an apparently notable result by chance. The registry lists 13 outcome measures and 3 primary pairwise analyses, but does not report a multiplicity plan for the secondary outcomes. That absence does not invalidate descriptive results; it does mean that isolated secondary findings should be considered exploratory unless stronger prespecified analysis details are available.

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

• Percentage of Participants With a Complete Response or Morphologic Incomplete Response.
• Duration of Remission (DoR)
• Cytogenetic Complete Remission Rate (CRc)
• Percentage of Participants With an Overall Response Rate (CR +CRi+ PR)
• Progression-Free Survival (PFS)
• Event-Free Survival (EFS)
• Relapse-Free Survival (RFS)
• Percentage of Participants With 30-Day Treatment-Related Mortality

9. Statistical Interpretation

10. Statistical Methods Explained

Time-to-event analysis is used when the outcome includes both whether an event occurred and when it occurred. In this trial, the primary outcome concerns survival, with death as the event described in the registry definition. Participants may have different lengths of follow-up, so analyses need to account for observations that end before an event is recorded. The record says that other participants were censored at year 1 or at discontinuation. Further detail about the handling of censoring is not reported in the registry record.

What does Kaplan-Meier estimation summarize?

In general, the Kaplan-Meier method estimates the probability of remaining event-free over time. At each event time, the estimate reflects the number of participants still under observation and at risk just before that time. Participants who are censored contribute information up to their censoring point. They are not counted as having experienced the event after follow-up ends. This approach can use incomplete follow-up without assuming that every participant has the same observation period.

The reported group summaries are medians in months, accompanied by 95% confidence intervals. A median is a point on the estimated survival curve, not a statement that every participant experienced the event at that time. The reported medians were 3.0 months for Lenalidomide, 9.61 months for Azacitidine Plus Lenalidomide, and 13.67 months for Azacitidine. The upper confidence limit for Azacitidine is not reported. The registry outcome title refers to one-year survival, while its stated summary statistic is the median and its timeframe is up to 24 months; the relationship among these descriptions is not further clarified.

What does the log-rank test compare?

In general, the log-rank test compares the event experience of groups across follow-up. It uses the timing of events and the numbers still at risk, rather than comparing only the medians. The registry reports log-rank analyses for all three pairwise group comparisons. It does not report whether the analyses were adjusted for prognostic characteristics, or whether checks of model assumptions were performed.

The reported effect measure is a hazard ratio. In general, a hazard is the event rate among participants who have not yet experienced the event and remain under observation. A hazard ratio compares these rates between groups over time. A ratio of 1 represents equal hazards under the model; values on either side represent different estimated hazards. The comparison order matters: the registry names the first group and then the second. A hazard ratio does not directly express a difference in median survival, an absolute probability of survival, or the number of events avoided.

The sponsor-reported hazard ratios were 1.789 for Lenalidomide versus Azacitidine Plus Lenalidomide, 2.659 for Lenalidomide versus Azacitidine, and 1.367 for Azacitidine Plus Lenalidomide versus Azacitidine. Their 95% confidence intervals and p-values are reported with the results. The record does not specify the hazard-ratio model in detail. In particular, it does not report diagnostics for proportional hazards, which is the assumption that the relative hazards are adequately represented by a constant ratio over the relevant follow-up.

How should confidence intervals and p-values be read?

In general, a confidence interval expresses the precision of an estimate under the statistical method used. A wider interval indicates greater uncertainty about the value compatible with the data and method; it does not assign a probability to the true effect after the interval has been calculated. For a hazard ratio, an interval including 1 is compatible with equal hazards as well as with effects in either direction. An interval excluding 1 is evidence against that particular null value under the stated analysis, but does not eliminate uncertainty about the size of the effect.

A p-value describes how unusual results at least as extreme as those observed would be under a specified null model and its assumptions. It is not the probability that the null hypothesis is true, nor a measure of clinical importance. The reported p-values are 0.119, 0.014, and 0.356 for the respective pairwise comparisons. The registry does not report a multiplicity adjustment, so the p-values should be interpreted as reported rather than as familywise-adjusted evidence across the full set of comparisons.

The primary analysis population is described as intention-to-treat and includes all randomized participants. In general, intention-to-treat analysis retains participants in their assigned groups, preserving the basis of randomized comparison. The registry does not report how missing outcome information, treatment discontinuation, or deviations from assigned treatment were handled beyond the stated censoring description.

Finally, statistical significance and estimation answer different questions. A test summarizes evidence against a null hypothesis, while the point estimate and interval show the estimated magnitude and its uncertainty. In this trial, the pairwise comparisons vary in precision; interpretation should account for the interval widths, the reported p-values, the number of comparisons, and the endpoint definition.

11. Confidence Intervals and Precision

The confidence intervals around the primary medians show that the estimates are not precise enough to identify a narrow range of plausible survival times. For Lenalidomide, the median is 3.0 months and the 95% confidence interval is 1.18 to 11.93. For Azacitidine Plus Lenalidomide, the median is 9.61 months and the interval is 3.18 to 19.31. For Azacitidine, the median is 13.67 months, with a lower confidence limit of 6.75 and no upper limit reported. The missing upper limit should not be read as certainty that the median is below a particular time.

The hazard-ratio intervals are also important for judging precision. Lenalidomide versus Azacitidine Plus Lenalidomide has an estimate of 1.789 and a 95% confidence interval of 0.861 to 3.718. Lenalidomide versus Azacitidine has an estimate of 2.659 and an interval of 1.214 to 5.822. Azacitidine Plus Lenalidomide versus Azacitidine has an estimate of 1.367 and an interval of 0.704 to 2.653. All three intervals span a range of possible effect sizes; the first and third also include 1.

In general, confidence intervals reflect both the observed information and the analysis method. Small or uneven outcome groups, limited event information, and censoring can contribute to uncertainty in time-to-event estimates. The registry does not report event counts for the primary endpoint, the number censored in each group, or the detailed basis for the confidence intervals, so the contribution of each factor cannot be assessed from the record.

Precision is distinct from whether a result crosses a conventional significance threshold. An estimate may be compatible with a range of effects, even when its interval excludes the null value. Conversely, an interval that includes the null does not prove equivalence. The estimates are best read together with their intervals and the reported analysis, rather than as exact values or definitive rankings.

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 the observed effect, not the trial's original design calculations. The registry record does not report the trial's original sample-size justification or its planning assumptions. For the time-to-event comparisons, the figures refer to the total number of events needed, not participants per group. For the 30-day treatment-related mortality comparisons, the figures refer to participants per arm.

The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set at the stated fraction of the observed effect. For example, planning for 75% of the observed effect means evaluating a smaller effect than the full observed estimate. The event calculations use the Schoenfeld approach. These assumptions make the displayed figures conditional planning examples, rather than a reconstruction of the original trial design.

In plain terms, the 80% power column gives the event or participant count that the planning calculation associates with an 80% chance of detecting the specified effect under its assumptions, if that effect were real. The 90% power column applies the same idea at a higher target chance of detection. Higher target power requires more information. The figures should be read within each comparison and outcome type, taking care not to confuse total events with participants per arm.

Smaller effects are harder to distinguish from random variation, so the planning figures increase when the target is reduced to 75% or 50% of the observed effect. These counts are not observed power. They are planning illustrations, and they do not show that the completed trial had a particular probability of finding a result.

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 = 1.789, so ln(HR) = 0.5817.
2. d = 4 × (1.96 + 0.842)² / [ln(1.789)]² = 92.8, rounded up to 93 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 <- 1.789                              # 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(1358734)
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 = 93 (80% power), 125 (90% power)
 75% of observed effect: events = 165 (80% power), 221 (90% power)
 50% of observed effect: events = 372 (80% power), 497 (90% power)
Simulated power with 93 events: 0.815 (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

The primary hazard ratios, confidence intervals, and p-values are the sponsor-reported analyses in the registry record. The comparisons used the log-rank method. The reported figures are 1.789, 2.659, and 1.367, with p-values of 0.119, 0.014, and 0.356, respectively. The statistical interpretation of those results should give priority to the sponsor's own analysis rather than to an independently reconstructed estimate.

No discrepancy is flagged between the posted sponsor analyses and the statistical summaries used in this explanation. The primary analysis is not replaced by a recalculation from group-level summaries: the registry reports the hazard ratios directly, and the interpretation uses those reported values. This distinction matters because time-to-event results depend on event timing, censoring, and the analysis method, details that cannot be recreated from medians alone.

The secondary 30-day treatment-related mortality comparisons do not have sponsor-reported inferential analyses listed in the record. Any simple recalculation based on posted group counts is unadjusted and is not equivalent to a prespecified sponsor analysis. The descriptive percentages are therefore the appropriate registry-level summary for that outcome; no independently calculated p-value should be mistaken for the sponsor's conclusion.

The record does not report a full statistical analysis plan, including any planned adjustments, model diagnostics, or hierarchy among endpoints. Consequently, the posted results can be described and interpreted in their recorded form, but unavailable analysis details cannot be assumed.

15. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when statistical tests are conducted across several comparisons or outcomes. Each test has its own chance of a false-positive result under its null hypothesis. When many tests are interpreted together, the chance of at least one apparently positive finding can increase unless the analysis plan addresses that issue. The registry reports three pairwise primary comparisons and lists multiple secondary outcomes. A multiplicity adjustment or a prespecified hierarchy for interpreting these results is not reported in the registry record.

The reported p-value of 0.014 for Lenalidomide versus Azacitidine is therefore a result for that individual comparison as reported. The record does not state whether the p-value was adjusted for the other primary comparisons. It should not be silently treated as a familywise-adjusted value. The other reported primary p-values are 0.119 and 0.356. Considering the full set of comparisons is important when drawing an overall conclusion, rather than focusing only on the smallest p-value.

Missing-data handling is only partly described. The primary analysis population is identified as intention-to-treat, and the endpoint definition says other participants were censored at year 1 or at discontinuation. The registry record does not report the amount or pattern of missing outcome information, how missingness was handled beyond that censoring statement, or whether sensitivity analyses were conducted. It also does not explain how discontinuation relates to the stated timeframe of up to 24 months.

The record does not report the randomization ratio, allocation procedure, stratification factors, detailed model specification, proportional-hazards checks, or the original sample-size assumptions. These omissions limit assessment of whether the analysis followed a prespecified plan and whether key assumptions were satisfied. The posted results remain interpretable as reported, but the unreported details prevent a complete audit of the analysis choices.

These limitations are especially relevant when comparing several endpoints with different definitions and follow-up. Without a reported multiplicity strategy, a missing-data account, and a fuller analysis plan, the strongest reading is descriptive and cautious: report the estimates, intervals, and sponsor analyses as they stand, while avoiding claims that depend on undocumented procedures.

16. 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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: febrile neutropenia (42.9%), pneumonia (21.9%), anaemia (14.3%), thrombocytopenia (14.3%), sepsis (14.3%).

The adverse-event record lists serious adverse events among participants at risk and those recorded as affected. For Lenalidomide, 14 were at risk and 13 were affected; for Azacitidine Plus Lenalidomide, 38 were at risk and 29 were affected; for Azacitidine, 32 were at risk and 25 were affected. For other adverse events, the corresponding at-risk and affected counts were 14 and 14, 38 and 38, and 32 and 32. These are descriptive counts, not adjusted comparisons between randomized groups.

The serious adverse-event terms listed most often include febrile neutropenia at 42.9%, pneumonia at 21.9%, and anaemia, thrombocytopenia, sepsis, and septic shock each at 14.3%. The registry record does not report the denominator or additional detail needed to interpret these term percentages fully. The terms may also describe events rather than unique participants, and the record does not clarify that distinction.

Deaths are recorded as 7 for Lenalidomide, 34 for Azacitidine Plus Lenalidomide, and 29 for Azacitidine in the adverse-event data. These counts should not be read as treatment-related death counts: the record does not make that equivalence. The separate secondary outcome concerns 30-day treatment-related mortality and reports percentages. The registry does not provide enough detail to reconcile those two presentations.

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

The posted results describe survival estimates and pairwise hazard-ratio analyses for three randomized treatment groups in older adults with newly diagnosed acute myeloid leukemia. The primary analysis reports different median estimates across groups. One pairwise comparison, Lenalidomide versus Azacitidine, has a reported confidence interval that excludes 1 and p = 0.014. The other two comparisons have confidence intervals that include 1 and larger reported p-values. The estimates are uncertain, and the registry's endpoint description has unresolved details about its one-year label, censoring, median statistic, and timeframe.

These findings do not establish that one regimen works or does not work, and they do not determine an individual treatment decision. A statistical comparison summarizes the observed data under an analysis method; it is not proof of a biological mechanism or a guarantee that a similar result will occur in another population. The record does not report all information needed to assess the assumptions behind the hazard-ratio analysis or the handling of missing follow-up.

The result also does not demonstrate equivalence when an interval includes 1. Such an interval means that the analysis has not clearly distinguished the observed comparison from equal hazards; it can still be compatible with effects in either direction. Conversely, an interval excluding 1 does not eliminate uncertainty about the magnitude or practical importance of the difference.

The record includes secondary outcomes and safety summaries, but results for several listed secondary endpoints are not reported in the registry record. The available treatment-related mortality percentages are descriptive and lack sponsor-reported inferential analyses. These data add context but do not settle the primary question or establish a consistent pattern across all outcomes.

The appropriate reading is therefore bounded: the registry records a randomized comparison, reports a particular set of survival estimates and sponsor analyses, and leaves important design and analysis details undocumented. Conclusions should remain tied to those posted results and their precision.

18. Why This Trial Matters Statistically

This trial illustrates why a time-to-event outcome cannot be reduced to a single group percentage or median. Kaplan-Meier estimation accommodates varying follow-up through censoring, while the log-rank analysis compares event experience across the observed period. The two summaries answer related but different questions. Median estimates describe a point on the survival curve, and hazard ratios summarize relative hazards under an analysis model.

The results also show the importance of reading estimates with their confidence intervals. The point estimates alone can suggest a clear ordering, but the intervals are broad and overlap substantially in their plausible values. A comparison's p-value does not convey the size or precision of the estimated effect. Looking at the interval and the point estimate together gives a more informative account of uncertainty.

Multiple pairwise comparisons create an additional interpretive issue. The registry reports three primary comparisons, but does not report a multiplicity adjustment. A single smaller p-value should therefore be read in the context of the full comparison set and the missing analysis-plan details. The same caution applies more strongly to secondary outcomes when inferential procedures are not reported.

Finally, an intention-to-treat population is identified, but the registry provides only a limited account of censoring and missing data. These details matter because time-to-event methods rely on follow-up information and assumptions about participants whose observation ends before an event. The trial is a useful example of how registry results can support careful statistical description while still leaving unanswered questions about precision, assumptions, and analysis planning.

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 endpoint description is not fully aligned within the registry record: the title refers to one-year survival, the definition describes deaths within one year and censoring at year 1 or discontinuation, the summary statistic is median survival, and the timeframe is up to 24 months. The record does not explain how these descriptions fit together. This makes the exact interpretation of the reported medians less straightforward than the label alone suggests.

The primary group sizes are uneven, with 15.0 for Lenalidomide, 39.0 for Azacitidine Plus Lenalidomide, and 34.0 for Azacitidine. Uneven group sizes can affect the amount of information available for each comparison. The registry does not report the allocation procedure or why the observed group sizes differ. The original sample-size justification is also not reported in the registry record.

For the primary endpoint, the record does not report event counts, censoring counts by group, detailed handling of discontinuation, or sensitivity analyses. It does not provide model diagnostics or state whether the proportional-hazards assumption was assessed. These omissions limit evaluation of the hazard ratios and the precision of the survival estimates.

The trial was open-label, and the registry does not report how knowledge of assigned treatment may have affected outcome assessment or care. The record also does not give a multiplicity plan for the pairwise comparisons and secondary outcomes. Numerical results for most listed secondary outcomes are not reported in the registry record, limiting assessment of consistency across endpoints.

The safety information is also incomplete for comparative interpretation. The serious-event percentages are not accompanied by their denominators in the record, and the relation between the death counts in the adverse-event data and the separate treatment-related mortality outcome is not clarified. These constraints do not erase the reported results, but they limit what can be concluded from them.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study Being Conducted at Multiple Locations to Compare Safety and Efficacy of Three Different Regimens; (1) High-Dose Lenalidomide; (2) Lenalidomide + Azacitidine; or (3) Azacitidine in Subjects ≥ 65 Years With Newly-Diagnosed Acute Myeloid Leukemia — NCT01358734.
• Registry record last updated 2019-06-25; status last verified 2019-06.
• Results first posted 2016-07-04.
• 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

NCT01358734 is a completed phase 2, multicenter, randomized, open-label, parallel-group study sponsored by Celgene. It enrolled 88 participants aged ≥65 years with newly diagnosed acute myeloid leukemia and compared lenalidomide, azacitidine plus lenalidomide, and azacitidine.

The primary outcome was described as Kaplan-Meier estimates for one-year survival, with median estimates and 95% confidence intervals reported. Sponsor-reported log-rank hazard ratios were 1.789 for Lenalidomide versus Azacitidine Plus Lenalidomide, 2.659 for Lenalidomide versus Azacitidine, and 1.367 for Azacitidine Plus Lenalidomide versus Azacitidine. The registry also lists multiple secondary outcomes and adverse-event summaries.

Interpretation is limited by uncertainty in the primary estimates, missing detail about censoring and analysis assumptions, and the absence of a reported multiplicity plan. Numerical results for several listed secondary outcomes are not reported in the registry record.