Combination Chemotherapy in Treating Young Patients With Newly Diagnosed T-Cell Acute Lymphoblastic Leukemia or T-cell Lymphoblastic Lymphoma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Combination Chemotherapy in Treating Young Patients With Newly Diagnosed T-Cell Acute Lymphoblastic Leukemia or T-cell Lymphoblastic Lymphoma (NCT00408005), a randomized phase 3 study. The primary outcome is Disease-free Survival (DFS) for Randomized Nelarabine T-ALL Cohort (Arm I vs. Arm II vs. Arm III vs. Arm IV).

2. Clinical Question

3. Trial Design

This was a phase III, randomized, open-label, parallel-group trial sponsored by the National Cancer Institute. The registry reports enrollment of 1895 participants. Its stated population was young patients newly diagnosed with T-ALL or T-LLy. Masking is listed as none, so the record does not describe blinded treatment assignment or outcome assessment.

Patients received common induction therapy before risk assignment and possible post-induction randomization. The registry describes two randomization questions: escalating-dose methotrexate versus high-dose methotrexate during interim maintenance, and no nelarabine versus nelarabine during consolidation. Eligibility was not uniform. Low-risk T-ALL patients were not eligible for the nelarabine randomization; patients with CNS disease at diagnosis were assigned to high-dose methotrexate, and induction failures were assigned to nelarabine and high-dose methotrexate. T-LLy patients were assigned to escalating-dose methotrexate, with risk-based differences in nelarabine assignment.

The primary outcome concerns only randomized T-ALL patients in the nelarabine comparison cohort. T-LLy patients are excluded, as are induction failures assigned non-randomly to Arm IV. The posted primary results give disease-free probability for four arms, with the number of participants and a 95% confidence interval for each estimate. The registry record does not report a statistical test comparing those arms. The published outcome counts should therefore be read within this defined analysis population, not as estimates for every enrolled participant or for both disease groups together.

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: Patients were stratified between the 2 disease groups (T-ALL and T-LLy) for Induction and then risk assigned treatment based upon the patient's response.

Only the first 6 of 8 groups are shown.

Induction

Reasons for not completing this period:

Post Induction Therapy by Tx Assignment

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Includes only T-cell Acute Lymphoblastic Leukemia (T-ALL) patients randomized to Nelarabine versus No Nelarabine. T-LLY patients are excluded from this analysis. Induction failures non-randomly assigned to Arm IV were excluded from this analysis.

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.

6. Primary Endpoint

The registry lists the primary outcome as Disease-free Survival (DFS) for Randomized Nelarabine T-ALL Cohort (Arm I vs. Arm II vs. Arm III vs. Arm IV), measured over 4 years from randomization at the end of induction.

The primary endpoint is disease-free survival (DFS), reported as a percent probability at 4 years from randomization at the end of induction. The registry defines the DFS time origin as that randomization. The first event is relapse, a second malignant neoplasm, or death; for participants without an event, the endpoint uses the date of last contact. This definition combines clinically different events into one time-to-event measure, so the posted DFS probability reflects remaining free of any of the listed events rather than the probability of avoiding relapse alone.

Time-to-event endpoints are useful when participants may have different lengths of follow-up or when the timing of an event matters. They can use information about event timing and follow-up status rather than treating every participant as if observed for the same duration. The registry reports each arm’s estimated disease-free probability with a 95% confidence interval, which communicates uncertainty around the estimate. It does not report the estimation method, handling of censoring, or the number of events contributing to the primary outcome, so those details are not reported in the registry record.

The analysis population is specifically randomized T-ALL patients in the nelarabine versus no-nelarabine cohort. T-LLy patients and induction failures assigned non-randomly to Arm IV are excluded. This definition is essential to interpreting the endpoint: the reported percentages do not represent all 1895 enrolled participants.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Statistical interpretation begins with the outcome definition and the population to which it applies. DFS is a time-to-event endpoint: participants are followed from randomization until the first qualifying event or, if no event is recorded, their last contact. Relapse, second malignant neoplasm, and death all count as events under the registry definition. The endpoint therefore captures a composite of event types. It does not distinguish their separate contributions in the reported primary estimates.

What does a confidence interval communicate?

A confidence interval describes the precision of an estimate under a statistical procedure. The registry gives a 95% confidence interval for each arm’s four-year disease-free probability. In general, greater uncertainty about an estimate tends to be reflected in a wider interval. The interval is not a probability statement that the true value has a 95% chance of lying inside the particular reported limits. Its interpretation depends on the method and assumptions used to calculate it. Those details are not reported in the registry record.

The four intervals accompany four separate arm estimates. They do not substitute for a direct comparison. Comparing estimates by eye, or noting whether intervals overlap, is not the same as a formal test of a treatment contrast. A comparative analysis would need to define the contrast and report its estimate and uncertainty, or a test result. The registry record lists no analysis results for the primary outcome. No comparative p-value or arm-level adjusted estimate is reported.

Why does the analysis population matter?

Randomization can support an unbiased comparison when the randomized groups are analyzed in a way that preserves the randomized comparison and follows an appropriate analysis plan. The primary outcome population is described as randomized T-ALL patients in the nelarabine versus no-nelarabine cohort. T-LLy patients are excluded, as are induction failures non-randomly assigned to Arm IV. These exclusions define the scope of the reported endpoint; they mean the four estimates should not be generalized to every enrolled participant without additional evidence.

Eligibility also varied by disease and risk. The record states that some patients were assigned to particular methotrexate strategies or nelarabine treatment rather than randomized to them, while other patients were eligible for one or both post-induction randomizations. An analysis of the randomized cohort is not automatically an analysis of all treatment assignments in the trial. The registry does not describe whether its primary analysis followed an intention-to-treat approach, used another population definition beyond the stated exclusions, or applied adjustments for baseline characteristics.

How should planning calculations be read?

General sample-size calculations ask how many participants might be needed to detect a specified effect under stated assumptions. They are planning illustrations, not observed power and not evidence that a study did or did not detect an effect. The assumptions described for the planning illustrations are a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect set as a stated fraction of the observed effect. The registry record does not report the trial’s own sample-size justification. No pairwise primary analysis is posted, and no planning result should be treated as a test of the observed arm outcomes.

What about event rates and over-dispersion?

For a time-to-event endpoint, information depends on event occurrence and follow-up, not just the number enrolled. General statistical methods for event-time outcomes use the timing of events and censoring, where a participant’s event-free follow-up ends without a qualifying event being observed. The registry defines last contact as the end of observation for participants who remain event-free, but does not report the censoring method, completeness of follow-up, or number of primary events. Over-dispersion is a concern in some count-data models when variability exceeds what a simple model assumes; it is not a concept that can be assessed from these arm-level DFS probabilities alone.

There are no reported primary-outcome analysis results beyond the arm-specific estimates and confidence intervals. The statistical analysis plan, model specification, handling of missing outcomes, interim analyses, and multiplicity procedures are not reported in the registry record. Those omissions limit what can be inferred about how the endpoint was estimated and how any treatment contrasts were intended to be judged.

10. Multiplicity, Missing Data and Analysis Populations

The trial includes two post-induction randomization questions: methotrexate strategy during interim maintenance and nelarabine during consolidation. The primary outcome is titled as a comparison across four arms in the randomized nelarabine T-ALL cohort, but the registry record provides arm-specific DFS probabilities rather than a reported comparative analysis. It does not report a multiplicity procedure, such as how statistical error across multiple arms, endpoints, or treatment questions was controlled. The statistical plan is not reported in the registry record.

The registry lists one secondary outcome: cumulative incidence of CNS relapse for T-ALL by risk group. Its results are not reported in the record details available for interpretation. Whether there were additional prespecified outcomes, analyses, or hypotheses is not reported in the registry record. The number of posted analyses is zero, so the arm estimates should not be mistaken for a full set of inferential results or for evidence that all planned comparisons were performed.

The primary analysis population is explicitly limited to randomized T-ALL patients in the nelarabine versus no-nelarabine cohort. T-LLy patients and induction failures assigned non-randomly to Arm IV are excluded. The registry does not report further details about analysis populations, missing outcome data, loss to follow-up, imputation, censoring procedures, or sensitivity analyses. It also does not report an interim-analysis plan. These omissions mean the published percentages cannot be evaluated for sensitivity to missingness or for the statistical consequences of repeated testing from this registry record alone.

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: seizure (6.5%), anaphylaxis (4.8%), sepsis (4.1%), alanine aminotransferase increased (3.7%), allergic reaction (3.2%).

The registry reports safety counts separately for induction and post-induction groups, including T-ALL and T-LLy groups and named methotrexate/nelarabine treatment groups. For each group it lists the number at risk, the number affected by serious events, the number affected by other events, and deaths. These are separate reported counts; they should not be added together or converted into rates without a defined analysis and denominator interpretation. The registry record does not provide comparative statistical tests for these safety figures.

The listed serious-event terms are seizure (6.5%), anaphylaxis (4.8%), sepsis (4.1%), alanine aminotransferase increased (3.7%), allergic reaction (3.2%), and leukoencephalopathy (3.2%). These are reported as the top serious terms. The record details do not establish that these percentages are directly comparable across treatment groups, nor do they provide the context needed to attribute a particular event to a specific regimen. Counts from induction, post-induction, and individual treatment groups concern different sets of participants and exposure periods. Safety figures describe reported events in their recorded groups; they do not, on their own, establish a causal treatment effect.

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

The posted primary results describe four-year DFS probabilities for four arms in a selected randomized T-ALL cohort. They show the magnitude and precision of each arm’s estimate as reported by the registry. The confidence intervals communicate uncertainty around those estimates, and the participant counts show the size of each analyzed arm. This is useful descriptive information about the specified cohort and endpoint.

The estimates do not answer every comparison question. The registry does not post a test or estimated contrast between arms, so the results do not establish whether any observed difference is statistically supported. They do not establish superiority, equivalence, or non-inferiority. Those conclusions require a prespecified contrast and suitable inferential analysis; none is reported in the registry record. The arm percentages also do not show whether the observed outcomes arose from treatment assignment, chance, or other features of the analysis.

The results are not an estimate for all trial participants. The primary population excludes T-LLy patients and induction failures assigned non-randomly to Arm IV, and is limited to randomized T-ALL patients in the nelarabine comparison cohort. The registry record separately describes risk-based eligibility and assignments, so the overall enrollment cannot be treated as the denominator for these estimates.

DFS is a composite time-to-event outcome including relapse, second malignant neoplasm, or death. The reported probabilities do not separate those events or explain their individual frequency. The secondary CNS-relapse outcome is listed without results in the record details. Finally, arm-specific safety counts and listed serious-event terms provide descriptive reporting, not proof that a regimen caused or prevented an adverse event. The registry record leaves the statistical analysis plan, missing-data methods, and multiplicity approach unreported.

13. Why This Trial Matters Statistically

This trial illustrates why the label of a randomized study is not enough to interpret its posted results. Although the record describes randomization and gives four arm-specific DFS estimates, it does not report the comparative analysis that would quantify differences between arms. Separate outcome estimates can communicate what was observed in each group, but they do not replace a direct treatment contrast with uncertainty and an inferential result.

It also demonstrates the importance of matching a result to its analysis population. The trial enrolled 1895 participants across T-ALL and T-LLy, while the primary outcome is restricted to randomized T-ALL patients in a particular post-induction cohort. Recognizing that distinction prevents a reader from treating a subgroup result as a trial-wide estimate. The risk-based eligibility and non-random assignments described in the registry further underscore that the treatment comparisons do not necessarily involve every participant.

Finally, the endpoint definition and uncertainty both matter. DFS combines relapse, second malignant neoplasm, and death into a time-to-event outcome, and the confidence intervals indicate that each arm estimate has uncertainty. Without event counts, analysis details, missing-data methods, and a stated multiplicity strategy, the precision and comparative interpretation cannot be fully assessed from the registry record. These are central statistical questions, not merely technical details: they determine what the posted numbers can support and where interpretation must remain cautious.

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 main limitation for interpreting the primary outcome is that the registry gives four arm-specific estimates but no reported arm-to-arm analysis. There is no posted comparative effect estimate, p-value, or adjusted analysis. The record therefore supports description of the four probabilities and their confidence intervals, but not a statistical conclusion about differences among arms.

The analysis population is narrower than overall enrollment. It includes randomized T-ALL patients in the nelarabine versus no-nelarabine cohort, while excluding T-LLy patients and induction failures assigned non-randomly to Arm IV. The record also describes risk-dependent eligibility and assignments after induction. These features mean the primary outcome should not be interpreted as covering all participants or all treatment pathways.

Several details needed to assess robustness are not reported in the registry record: the statistical analysis plan, the method used to estimate DFS and its confidence intervals, the number and timing of primary events, the handling of censoring and missing data, and any sensitivity analyses. The record does not report a multiplicity adjustment, interim-analysis plan, or the trial’s sample-size justification. The stated planning assumptions are not a substitute for the trial’s design rationale or a comparative outcome analysis.

The endpoint is composite, so the arm-level DFS estimates do not reveal separate results for relapse, second malignant neoplasm, or death. Safety counts are listed across groups and phases but are not accompanied by comparative tests. The secondary CNS-relapse outcome is named without results in the record details available for interpretation. These limitations restrict causal and comparative conclusions based on the registry record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Combination Chemotherapy in Treating Young Patients With Newly Diagnosed T-Cell Acute Lymphoblastic Leukemia or T-cell Lymphoblastic Lymphoma — NCT00408005.
• Registry record last updated 2026-06-23; status last verified 2026-05.
• Results first posted 2019-06-20.
• 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

NCT00408005 is a completed phase III randomized, open-label, parallel-group trial sponsored by the National Cancer Institute. It enrolled 1895 young patients with newly diagnosed T-ALL or T-LLy and studied post-induction chemotherapy strategies involving methotrexate and nelarabine. Participants received common induction therapy before risk assignment and eligibility for one or both post-induction randomizations.

The primary outcome is four-year DFS in the randomized T-ALL nelarabine cohort, defined by time to relapse, second malignant neoplasm, or death. The registry reports probabilities and 95% confidence intervals for Arms I through IV, but no comparative analysis results. The primary analysis excludes T-LLy and induction failures assigned non-randomly to Arm IV. A secondary CNS-relapse outcome is listed without results in the registry record. Methods for missing data, multiplicity, interim analyses, and the primary statistical model are not reported in the registry record.