Latin American Real-world Study in Acute Leukemia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Latin American Real-world Study in Acute Leukemia (NCT05166135), a registered phase not stated study. The primary outcome is Number of Participants According to Health Insurance Type.

2. Clinical Question

3. Trial Design

The LOYAL Study is identified as a real-world study of epidemiology, treatment patterns, outcomes, and healthcare resource use among adults in Latin America with de novo AML or R/R B-cell ALL. The registry lists an enrollment of 589 and a status of COMPLETED. Pfizer is the sponsor. The listed study start is 2021-12-10 and primary completion date is 2022-11-15.

The registry describes two disease cohorts. The AML cohort includes participants at least 18 years old at diagnosis with de novo AML diagnosed between 01 January 2015 and 31 December 2019, as documented in the medical chart and according to physician notes, and with at least 1 AML treatment line within the study period. The R/R ALL cohort uses the corresponding age and diagnosis dates, requires R/R ALL documented in the chart and physician notes, and requires at least 1 line of treatment for R/R ALL within the study period.

The primary outcome timeframe describes an index period of anytime between 01-Jan-2015 and 31-Dec-2019, approximately 5 years, while data were collected and observed retrospectively over 11 months. These time descriptions refer to the registry’s outcome and data-collection wording; the record does not give further detail about how individual records were assembled or the exact dates of data collection.

The registry does not state an allocation method, and the study is not described as randomized. Its disease groups should therefore be read as observational cohorts, not as intervention and control arms in an efficacy trial. The record does not report the names of the 4 Latin American countries, participating sites, or the number enrolled in each cohort beyond the adverse-event denominators. It lists 64 outcome measures and 0 analyses.

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: A total of 628 participants were screened of which 39 did not meet eligibility criteria and were excluded and 589 participants were included in the study.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Analysis population included all eligible participants whose data were retrieved and observed in this study.

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: Analysis population included all eligible participants whose data were retrieved and observed in this study.

6. Primary Endpoint

The registry lists the primary outcome as Number of Participants According to Health Insurance Type, measured over At index (anytime between 01-Jan-2015 and 31-Dec-2019; approximately 5 years); data collected and observed retrospectively over 11 months.

The primary endpoint is the number of participants according to health insurance type. Its listed categories are public or private insurance, and its unit is participants. The intended population is all eligible participants whose data were retrieved and observed in the study. The endpoint is measured at index, defined as anytime between 01-Jan-2015 and 31-Dec-2019; the registry describes the data as collected and observed retrospectively over 11 months.

This is a descriptive measure, not a measure of treatment benefit or harm. Insurance type can characterize the healthcare context of the observed population and may be relevant when describing treatment patterns or healthcare resource use. It does not, by itself, show why a treatment was chosen, whether insurance affected access, or whether insurance caused a particular outcome. Those questions would require suitable information and analyses not reported in the registry record.

The registry describes the outcome as a count, but does not report the actual public or private insurance counts, the number with missing or unknown insurance status, or the rules used to classify insurance. Without those results, the distribution cannot be assessed. The endpoint’s stated population also depends on data retrieval and observation; details about eligibility review, record completeness, and exclusions are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry record describes a retrospective observational study with a primary outcome reported as participant counts. It does not report an inferential statistical plan for that outcome. The following concepts are general explanations of how descriptive and inferential statistics are commonly interpreted; they do not assert that a particular analysis was performed in this study.

What does a count describe?

A count states how many observations fall into a category. For this primary outcome, the categories named in the record are public and private health insurance. A count is most interpretable when the reader also knows the population from which it came, how many records had a valid category, and how missing or ambiguous information was handled. The registry defines the intended population as eligible participants whose data were retrieved and observed, but does not report the category counts, missingness, or classification rules.

Counts and percentages answer related but different descriptive questions. A count gives the number of observations in a category; a percentage expresses that count relative to a stated denominator. A percentage cannot be interpreted reliably without knowing its denominator and whether participants with missing data were included. No insurance percentages are reported in the registry record, and none can be inferred from the outcome description alone.

What would a confidence interval mean?

General statistical explanation: a confidence interval expresses uncertainty around an estimated quantity under the assumptions of the method used. For a proportion, it can convey how precisely the observed sample estimate may represent the population from which the sample arose. It does not correct for selection bias, inaccurate chart records, or an outcome definition that excludes relevant participants. The registry record does not report a confidence interval for the insurance outcome or a method for calculating one.

For a retrospective study, uncertainty has more than a sampling component. Medical records may be incomplete, and the people whose records are available may differ from those whose records cannot be retrieved. Statistical precision alone cannot resolve those concerns. Interpretation depends on transparent reporting of the eligible population, available observations, missing information, and selection process; those details are not reported in the registry record.

Why does the observational design matter?

General statistical explanation: random allocation can help balance measured and unmeasured characteristics between treatment groups, on average, when comparing interventions. In an observational study, treatment and cohort membership are not assigned by the study in that way. Differences in observed outcomes may therefore reflect characteristics associated with diagnosis, prior treatment, clinical decisions, access to care, or record availability. Adjustment can sometimes address measured differences, but it cannot guarantee removal of confounding, especially when relevant factors are not recorded or modeled.

Here, the AML and R/R B-cell ALL cohorts are defined by different diseases and eligibility criteria. They are not interchangeable treatment arms. Any numerical contrast between them would be descriptive unless a suitable analytical approach and explicit assumptions supported a stronger interpretation. The registry lists 0 analyses and does not report adjusted estimates, confounder sets, or a comparative statistical plan.

How should the planning assumptions be read?

General statistical explanation: sample-size planning asks how much information a future study may need to detect a specified effect under stated assumptions. The registry’s planning description uses equal allocation, a two-sided significance level of 0.05, and a normal approximation. It states that the effect to detect is a stated fraction of the observed effect. These are planning assumptions, not evidence that an effect was observed with a particular precision and not a measure of observed power.

The record says time-to-event planning uses the total number of events under the Schoenfeld approach, while other planning outputs are participants per arm. Those different units reflect the statistical quantity used for planning; event totals and participants per arm are not interchangeable. The record does not report numerical planning outputs, the effect estimates used, or an explanation of how the observed effect was selected. The planning illustrations therefore cannot be treated as the study’s original design justification, and the registry does not state that they were used to plan this retrospective study.

What does over-dispersion mean in general?

General statistical explanation: over-dispersion occurs when count data vary more than a chosen statistical model expects. If ignored in an inferential analysis, uncertainty may be understated and intervals or tests may be too optimistic. Whether over-dispersion is relevant depends on the outcome structure and model. The primary insurance outcome is described as category counts, but no model, dispersion estimate, or inferential analysis is reported in the registry record. It would not be appropriate to claim that an over-dispersion adjustment was or was not needed for an analysis that is not documented.

What can be said about multiplicity?

General statistical explanation: when many outcomes or tests are examined, the chance of finding an apparently unusual result by chance can increase. Statistical plans may designate a primary outcome, distinguish confirmatory from exploratory outcomes, and describe whether or how multiplicity is addressed. The record lists 64 outcome measures, but does not report a multiplicity procedure, a hierarchy among the secondary outcomes, or a plan for interpreting multiple tests. It also lists 0 analyses, so the number of outcome definitions should not be mistaken for the number of reported statistical tests.

10. Multiplicity, Missing Data and Analysis Populations

The registry names a primary outcome and lists 64 outcome measures overall, including outcomes about molecular testing, AML translocations and classifications, ALL classification, and immunophenotyping. It lists 0 analyses. The record does not report a statistical analysis plan, a hierarchy for secondary outcomes, or a multiplicity adjustment. Without those details, the purpose and inferential status of any outcome beyond the primary measure are not clear. A long outcome list by itself does not indicate that each outcome was tested or that any finding was confirmatory.

The primary population is described as all eligible participants whose data were retrieved and observed. The record does not report the number with available data for each outcome, the number with missing information, the approach to incomplete charts, or whether missing values were excluded or classified. It also does not state whether missingness was assessed by cohort or by outcome. These omissions matter because a count based on available records may describe those records rather than all eligible participants.

The registry does not describe imputation, sensitivity analyses, or methods for addressing missing data. These methods should not be presumed. Nor does it report analysis populations for secondary outcomes beyond their outcome titles. The listed enrollment of 589 is not a replacement for outcome-specific denominators, which may depend on eligibility and availability of the relevant information. Interpretation of the descriptive results therefore requires caution about both missingness and the range of outcomes listed.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: sepsis (0.4%).

The registry lists serious adverse-event figures separately for Newly AML and R/R B-cell ALL. For Newly AML, it reports 2 participants affected among 518 at risk for serious events and 2 among 518 at risk for other adverse events. For R/R B-cell ALL, it reports 0 participants affected among 71 at risk for serious events and 1 among 71 at risk for other adverse events. The record also lists deaths of 392 for Newly AML and 60 for R/R B-cell ALL. These are reported counts; they should not be combined or treated as evidence of a treatment effect.

The registry lists sepsis at 0.4% among the top serious terms. It does not explain the denominator or the coding and attribution rules behind that percentage. The record also does not report event timing, severity grading details, duration of observation by participant, causes of death, or whether events were considered related to treatment. Death counts are presented separately, and the record does not establish how they relate to adverse-event categories. Because the cohorts have different diseases and eligibility criteria, their figures do not constitute a like-for-like safety comparison.

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

The study’s central contribution is descriptive: it was designed to characterize epidemiology, treatment patterns, outcomes, and healthcare resource use among adults with de novo AML or R/R B-cell ALL in Latin America. Its primary outcome is the number of eligible participants in public or private health insurance categories at index. The registry record does not report the actual category counts, so no conclusion about the observed insurance distribution can be drawn from the outcome definition alone.

The study’s retrospective use of medical charts can provide information about care as recorded in practice. At the same time, the available record depends on what was documented and retrieved. The registry does not describe completeness of chart data, missingness, or how records were selected beyond the stated eligibility requirements. These limitations affect how broadly descriptive findings can be applied.

The AML and R/R B-cell ALL cohorts represent different clinical populations. They are not randomized treatment groups and should not be interpreted as if one were a control for the other. Differences in any observed characteristic or outcome could reflect disease and eligibility differences, clinical history, care, or data availability. The record does not report analyses designed to separate those explanations.

The listed adverse-event counts and deaths are also descriptive. The registry record does not identify causes of death, treatment attribution, or comparable observation periods. The counts cannot establish that a treatment caused an event or that one cohort had greater treatment-related risk. Likewise, the listing of outcomes related to molecular testing and classification does not reveal their results where those results are not reported.

Accordingly, the registry information supports understanding the study’s scope, population definitions, and intended descriptive questions. It does not establish treatment efficacy, comparative safety, or causal relationships. It also does not provide the numerical results needed to judge the precision of the primary insurance outcome. Any stronger interpretation would require results and methods not reported in the registry record.

13. Why This Trial Matters Statistically

Real-world data can describe populations and care patterns that are not always represented by randomized trials. The LOYAL Study focuses on adults with de novo AML and R/R B-cell ALL in Latin America, with diagnoses documented in medical charts and physician notes. Its stated objectives include epidemiology, treatment patterns, outcomes, and healthcare resource use. This scope makes clear that the study is intended to characterize care in practice rather than test an assigned intervention.

Statistically, that distinction matters. Descriptive estimates depend on how patients entered the record, what was documented, and which observations were retrievable. The primary insurance outcome depends on valid classification of public or private insurance, while the registry does not report the counts or the handling of missing information. Naming the target population and the measurement timeframe helps define the question, but does not resolve uncertainty about representativeness or completeness.

The record also illustrates why disease cohorts should not automatically be treated as comparative arms. AML and R/R B-cell ALL have separate criteria and are clinically distinct populations. A contrast between their counts would not, on its own, isolate any treatment effect. The registry lists many diagnostic and outcome measures, but no analyses, so the scope of the data should not be confused with evidence of inferential comparisons.

Finally, planning assumptions require careful labeling. Illustrations based on observed effects can help frame what a future study might need, but they are not observed power and do not prove that a retrospective study was prospectively designed around those assumptions. Keeping descriptive findings, causal claims, and planning calculations distinct is essential when interpreting real-world research.

14. 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 registry record leaves several trial-specific details unclear. It identifies the setting as 4 Latin American countries but does not name them or report the participating sites. It describes diagnoses from 01 January 2015 through 31 December 2019 and retrospective observation over 11 months, but does not give further detail about data collection procedures, record retrieval, or follow-up by participant.

The overall enrollment is 589, but the record does not report the number of participants with usable data for the primary insurance outcome. It also does not report public and private insurance counts, missing insurance data, or classification rules. For the outcome measures more broadly, outcome-specific denominators and missing-data handling are not reported in the registry record. This restricts assessment of completeness and makes it difficult to know whether available records represent all eligible participants.

The cohorts are not like-for-like comparison groups. The AML and R/R B-cell ALL definitions differ by diagnosis and disease status, and no randomized allocation is stated. The registry lists 0 analyses, with no adjusted estimates or details of confounder control reported. The record also does not provide a statistical analysis plan, multiplicity approach, or sample-size justification for the retrospective study. Numerical planning outputs are not reported in the registry record.

The safety information lacks detail on event timing, attribution, observation duration, and causes of death. The basis for the sepsis percentage is not described. As a result, the listed counts should be read as reported descriptive figures rather than comparative risks or evidence of treatment causation. These limitations are specific to the information documented in the registry record and do not establish what may or may not appear in other study materials.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Latin American Real-world Study in Acute Leukemia — NCT05166135.
• Registry record last updated 2025-03-25; status last verified 2025-03.
• Results first posted 2025-03-25.
• 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.

18. Record Summary

The LOYAL Study is a completed, retrospective real-world study sponsored by Pfizer, with listed enrollment of 589. It concerns adults with de novo AML or R/R B-cell ALL in 4 Latin American countries, diagnosed between 01 January 2015 and 31 December 2019 and meeting the stated treatment-line eligibility criteria. The listed start date is 2021-12-10 and primary completion date is 2022-11-15.

The primary outcome is the number of participants by public or private health insurance type at index. The registry record defines the outcome and intended population but does not report the category counts. It lists 64 outcome measures and 0 analyses. Serious-event counts, other adverse-event counts, deaths, and a sepsis term are also listed, without sufficient detail for causal or like-for-like comparative conclusions. The study is descriptive, not a randomized treatment comparison.