This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Combination Chemotherapy, Radiation Therapy, and an Autologous Peripheral Blood Stem Cell Transplant in Treating Young Patients With Atypical Teratoid/Rhabdoid Tumor of the Central Nervous System (NCT00653068), a non-randomized phase 3 study. The primary outcome is Event-free Survival.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00653068 |
| Phase | Phase 3 |
| Condition | Childhood Atypical Teratoid/Rhabdoid Tumor |
| Allocation | Non-Randomized |
| Design model | Parallel |
| Masking | None (open label) |
| Enrollment | 70 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Children's Oncology Group |
| Results posted | Yes |
2. Clinical Question
Population
Young patients with childhood atypical teratoid/rhabdoid tumor of the central nervous system. The primary endpoint description notes ineligible patients in Strata I and III and no enrollment in Strata 2 or 4.
Intervention
Intensive chemotherapy, 3-D conformal radiation therapy, and autologous peripheral blood stem cell transplantation, with treatment sequence differing in the registry’s two arm descriptions.
Comparator
The registry lists two non-randomized treatment arms. Their labels describe different sequences of chemotherapy, radiation, and transplantation; they are not a randomized, like-for-like control comparison.
Primary question
What is the estimated probability of event-free survival up to 4 years after study enrollment, with events defined as disease progression, relapse, a second malignant neoplasm, or death from any cause?
The study concerns a rare childhood tumor and an intensive, combined treatment approach. Its primary question is framed around event-free survival (EFS), a time-to-event outcome that captures several kinds of disease or health events rather than a single event type. The registry describes EFS as the time from enrollment until progression, relapse, a second malignant neoplasm, death from any cause, or last follow-up, whichever occurs first. Patients without an event are censored at last contact.
The record identifies two treatment arms and labels them by treatment sequence. Arm I is described as chemotherapy, autologous peripheral blood stem cell transplantation, then 3-D conformal radiation therapy; Arm II is described as chemotherapy, radiation, then transplantation. Because allocation is non-randomized, the contrast between these arms cannot be interpreted as if treatment sequence had been assigned by randomization. The registry record does not report an estimated EFS probability or confidence interval, so the stated clinical question has no posted numerical answer in the record.
3. Trial Design
This completed phase III study, NCT00653068, was sponsored by the Children’s Oncology Group. The registry lists an enrollment of 70 and a start date of 2009-02-10; primary completion is dated 2016-09-27. The condition is childhood atypical teratoid/rhabdoid tumor of the central nervous system. The official title describes surgery, intensive chemotherapy, and 3-D conformal radiation, while the brief description focuses on chemotherapy, radiation, and autologous peripheral blood stem cell transplantation.
The design is listed as parallel, non-randomized, and unmasked. The record names Arm I as chemotherapy, autologous peripheral blood stem cell transplantation, and 3-D conformal radiation therapy; Arm II is labeled chemotherapy, 3-D conformal radiation therapy, and autologous peripheral blood stem cell transplantation. Thus, the registry presents treatment pathways with different sequences, not a randomized comparison. The reason participants entered one arm rather than the other is not reported in the registry record.
The intervention list includes 3-D conformal radiation therapy, autologous hematopoietic stem cell transplantation, carboplatin, cisplatin, cyclophosphamide, etoposide, filgrastim, and laboratory biomarker analysis. The brief description explains that high-dose chemotherapy is used before autologous transplantation and that stem cells are collected, stored, and returned after treatment. Arm descriptions also mention vincristine, high-dose methotrexate, leucovorin calcium, and G-CSF. The available arm descriptions are incomplete, so the full protocol details are not reported in the registry record.
Two patients were ineligible, one in Stratum I and one in Stratum III, according to the primary endpoint population description. The record says there were no patients enrolled in Stratum 2 or 4. It does not explain the basis for stratum assignment, the rationale for non-randomized allocation, or the sample-size justification. Those design details 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.
Overall Study
| Stage | Treatment |
|---|---|
| STARTED | 70 |
| COMPLETED | 20 |
| NOT COMPLETED | 50 |
Reasons for not completing this period:
| Reason | Treatment |
|---|---|
| Death | 9 |
| Lack of Efficacy | 34 |
| Physician Decision | 4 |
| Withdrawal by Subject | 1 |
| Ineligible | 2 |
Analysis population for the primary outcome, as defined in the registry: One patient was ineligible in stratum I and one patient was ineligible in stratum III. There were no patients enrolled in Stratum 2 or 4.
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.
| Characteristic | Unit | Stratum I | Stratum II | Stratum III | Stratum IV | Total |
|---|---|---|---|---|---|---|
| Participants analysed | 58 | 0 | 12 | 0 | 70 | |
| Age, Categorical: <=18 years | Participants | 58 | — | 12 | — | 70 |
| Age, Categorical: Between 18 and 65 years | Participants | 0 | — | 0 | — | 0 |
| Age, Categorical: >=65 years | Participants | 0 | — | 0 | — | 0 |
| Age, Continuous | years | 1 (0 to 2) | — | 4 (3 to 14) | — | 1 (0 to 14) |
| Sex: Female, Male: Female | Participants | 28 | — | 8 | — | 36 |
| Sex: Female, Male: Male | Participants | 30 | — | 4 | — | 34 |
| Ethnicity (NIH/OMB): Hispanic or Latino | Participants | 15 | — | 5 | — | 20 |
| Ethnicity (NIH/OMB): Not Hispanic or Latino | Participants | 39 | — | 7 | — | 46 |
| Ethnicity (NIH/OMB): Unknown or Not Reported | Participants | 4 | — | 0 | — | 4 |
| Race (NIH/OMB): American Indian or Alaska Native | Participants | 1 | — | 0 | — | 1 |
| Race (NIH/OMB): Asian | Participants | 1 | — | 2 | — | 3 |
| Race (NIH/OMB): Native Hawaiian or Other Pacific Islander | Participants | 0 | — | 0 | — | 0 |
| Race (NIH/OMB): Black or African American | Participants | 7 | — | 1 | — | 8 |
| Race (NIH/OMB): White | Participants | 40 | — | 7 | — | 47 |
| Race (NIH/OMB): More than one race | Participants | 0 | — | 0 | — | 0 |
| Race (NIH/OMB): Unknown or Not Reported | Participants | 9 | — | 2 | — | 11 |
| Region of Enrollment: Canada | Participants | 6 | — | 2 | — | 8 |
| Region of Enrollment: United States | Participants | 51 | — | 10 | — | 61 |
| Region of Enrollment: Australia | Participants | 1 | — | 0 | — | 1 |
6. Primary Endpoint
The registry lists the primary outcome as Event-free Survival, measured over Up to 4 years after study enrollment.
| Element | Registry entry |
|---|---|
| Result type | Number |
| Measure of spread | 95% Confidence Interval |
| Unit | Estimated probability |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint is event-free survival (EFS), reported in the registry as an estimated probability over a timeframe of up to 4 years after study enrollment. EFS is measured from enrollment until the first listed event: disease progression, disease relapse, occurrence of a second malignant neoplasm, or death from any cause. If none of these events occurs before last follow-up, the patient is censored at last contact.
Time-to-event endpoints are useful when both whether an event occurs and when it occurs matter. A Kaplan–Meier estimate can use follow-up information from patients who have not yet had an event by accounting for their time under observation and censoring them at their last contact. This method does not treat a censored patient as though an event occurred; it records that the event-free time was at least as long as the observed follow-up.
The registry identifies Kaplan–Meier estimation and a 95% confidence interval as the endpoint’s statistical summary. The interval would convey uncertainty around the estimated EFS probability, rather than guarantee that the true probability lies within a particular range. However, the numerical EFS estimate and its confidence interval are not reported in the registry record. The event count and the underlying analysis results are also not reported, so the endpoint’s observed magnitude and precision cannot be assessed from the record.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry defines a clinically relevant time-to-event outcome and specifies how it is estimated, but it does not post the numerical primary result. There is no reported 4-year EFS probability, 95% confidence interval, event count, or analysis result. Consequently, the record does not establish from the posted primary endpoint whether EFS was high or low, nor how precisely it was estimated. A definition and planned summary method describe what would be measured; they are not substitutes for the result.
The study is listed as phase III and completed, with enrollment of 70. Those details describe the trial record, not the outcome. They do not by themselves indicate the treatment effect, the precision of the effect estimate, or the success of the study. The registry does not report a numerical comparison of the two treatment arms on EFS. Its allocation is non-randomized, and the record explicitly cautions that the arms are not a like-for-like control comparison. Differences in observed outcomes between the arms, if available elsewhere, could reflect differences in participant characteristics or other factors as well as treatment sequence.
The posted safety figures are separate from the primary EFS result. For Stratum I, the serious adverse-event summary lists 57 at risk and 7 affected; for Stratum III, it lists 11 at risk and 0 affected. The other-adverse-event summary lists 57 at risk and 43 affected in Stratum I, and 11 at risk and 8 affected in Stratum III. These are group-level counts and should be read as safety summaries, not as estimates of EFS or proof of a difference between strata. The registry does not provide a randomized safety comparison.
The record also lists serious adverse-event terms with percentages, including sepsis, hypokalemia, intracranial hemorrhage, and respiratory failure at 3.5% each, and disseminated intravascular coagulation and left ventricular systolic dysfunction at 1.8% each. Those entries describe the listed terms; the registry information does not establish that these percentages can be attributed to a particular treatment arm or used to compare arms.
Overall, the record permits interpretation of the endpoint definition, the non-randomized design, and the posted safety counts. It does not permit a numerical interpretation of the primary EFS outcome or a treatment-effect estimate. The sponsor’s analysis results are not reported in the registry record, and there are 0 analyses listed.
9. Statistical Methods Explained
The statistical methods described in the registry are limited. The primary endpoint is estimated with the Kaplan–Meier method, with a 95% confidence interval. The record does not report a numerical estimate, event count, model-based comparison, or analysis results. General explanations below describe what these methods mean; they do not imply that unreported analyses were performed.
What does a Kaplan–Meier estimate represent?
Kaplan–Meier estimation is a way to describe the probability of remaining free of an event over time when participants have different lengths of follow-up. At each observed event time, the estimate updates using the participants still under observation and event-free immediately before that time. Participants who have not had an event when follow-up ends contribute information up to their last contact, then are censored. The registry says patients without an event are censored at last contact.
Censoring is not the same as confirming that no event will ever occur. It indicates that the event was not observed before the recorded follow-up ended. Interpretation of a Kaplan–Meier estimate depends on follow-up and on assumptions about censoring. The registry record does not report follow-up distributions, the number censored, or details that would allow the censoring assumptions to be assessed.
What does the confidence interval add?
A confidence interval communicates uncertainty around an estimate. A narrower interval generally indicates greater precision, while a wider interval indicates less precision. The stated 95% confidence interval is the planned form of uncertainty summary for EFS, but its endpoints are not reported in the registry record. Without the estimate and interval, readers cannot assess either the EFS level or the precision of that estimate.
Why combine several events in EFS?
EFS includes progression, relapse, a second malignant neoplasm, and death from any cause. A composite endpoint can capture several outcomes in one time-to-event measure. Its meaning depends on the components: a participant can reach the endpoint through any listed event, and the overall result does not by itself show which component was most common or most influential. The registry record does not report component-specific event counts or estimates.
What does non-randomized allocation imply?
Randomization is a general method for balancing, on average, measured and unmeasured baseline factors between treatment groups. This study is listed as non-randomized, and the registry cautions that its arms are not a like-for-like control comparison. Therefore, an arm contrast should not be read as a causal treatment effect without appropriate design and analysis information. The reason for allocation, baseline comparability, and any adjustment strategy are not reported in the registry record.
What do the planning illustrations mean?
The planning illustrations are not observed trial results or observed power. Their stated assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is expressed as a fraction of the observed effect. For time-to-event calculations, the planning quantity is the total number of events using the Schoenfeld approach; other calculations are expressed as participants per arm. These illustrations describe hypothetical planning under stated assumptions. They do not establish that the trial was designed that way, and the registry record does not report its own sample-size justification.
General sample-size calculations depend on assumptions about the target effect, variability, event occurrence, follow-up, and analysis. For a time-to-event endpoint, information is especially connected to the number of observed events, not enrollment alone. Since the registry does not report the EFS estimate or event count, the planning illustrations should not be mistaken for evidence about the observed precision or findings of this trial.
What analysis details are unavailable?
The registry lists 0 analyses. It does not report an inferential comparison for EFS, covariate adjustment, a statistical model beyond Kaplan–Meier estimation, an interim-analysis plan, or a multiplicity strategy. These details are not reported in the registry record. No conclusions about which analyses were conducted, or how their results were handled, can be drawn from the endpoint definition alone.
10. Multiplicity, Missing Data and Analysis Populations
The registry identifies EFS as the primary endpoint and lists a secondary outcome concerning non-hematological toxicity associated with chemotherapy, defined as Grade 3 or higher during protocol therapy. It reports 4 outcome measures overall, but does not describe all of them in the information available for this analysis. The registry record does not report a multiplicity strategy, such as an adjustment across outcomes or a hierarchy for interpreting tests. Without such information and without posted analysis results, the extent to which multiple outcomes or comparisons influenced interpretation cannot be assessed.
The primary endpoint description gives a population note: one patient was ineligible in Stratum I and one patient was ineligible in Stratum III; no patients were enrolled in Stratum 2 or 4. The record does not provide a full analysis-population definition, identify whether any other exclusions occurred, or state how participants were handled in every analysis. It would therefore be inappropriate to label the EFS population as intention-to-treat or per-protocol based on the available record.
For EFS, patients without an event are censored at last contact. This tells readers how the endpoint definition handles event-free follow-up, but it does not report the amount or pattern of missing follow-up, the number censored, or any sensitivity analysis. The registry record does not report a missing-data strategy beyond the stated censoring rule, nor whether missing outcome information was assessed for its possible effect on results.
Finally, the registry lists 0 analyses and does not report endpoint estimates or statistical tests. Information about interim analyses, analysis timing, and any adjustments for multiple comparisons is not reported in the registry record. These omissions limit assessment of the analysis plan and the strength of any inferential claims.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Stratum I | 7 | 57 | 12.3% |
| Stratum III | 0 | 11 | 0.0% |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Stratum I | 43 of 57 |
| Stratum III | 8 of 11 |
Most frequent serious adverse event terms
| Event term | Organ system | Stratum I | Stratum III |
|---|---|---|---|
| Sepsis | Infections and infestations | 2 of 57 | 0 of 11 |
| Hypokalemia | Metabolism and nutrition disorders | 2 of 57 | 0 of 11 |
| Intracranial hemorrhage | Nervous system disorders | 2 of 57 | 0 of 11 |
| Respiratory failure | Respiratory, thoracic and mediastinal disorders | 2 of 57 | 0 of 11 |
| Disseminated intravascular coagulation | Blood and lymphatic system disorders | 1 of 57 | 0 of 11 |
| Left ventricular systolic dysfunction | Cardiac disorders | 1 of 57 | 0 of 11 |
| Multi-organ failure | General disorders | 1 of 57 | 0 of 11 |
| Enterocolitis infectious | Infections and infestations | 1 of 57 | 0 of 11 |
| Upper respiratory infection | Infections and infestations | 1 of 57 | 0 of 11 |
| Lymphocyte count decreased | Investigations | 1 of 57 | 0 of 11 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Stratum I | Stratum III |
|---|---|---|---|
| Neutrophil count decreased | Investigations | 37 of 57 | 5 of 11 |
| Platelet count decreased | Investigations | 31 of 57 | 5 of 11 |
| White blood cell decreased | Investigations | 30 of 57 | 3 of 11 |
| Lymphocyte count decreased | Investigations | 18 of 57 | 1 of 11 |
| Infections and infestations - Other, specify | Infections and infestations | 7 of 57 | 0 of 11 |
| Hypokalemia | Metabolism and nutrition disorders | 6 of 57 | 1 of 11 |
| Anemia | Blood and lymphatic system disorders | 6 of 57 | 0 of 11 |
| Febrile neutropenia | Blood and lymphatic system disorders | 6 of 57 | 0 of 11 |
| Alanine aminotransferase increased | Investigations | 5 of 57 | 0 of 11 |
| Hypoxia | Respiratory, thoracic and mediastinal disorders | 4 of 57 | 1 of 11 |
Serious events with the highest arm-level rate: sepsis (3.5%), hypokalemia (3.5%), intracranial hemorrhage (3.5%), respiratory failure (3.5%), disseminated intravascular coagulation (1.8%).
The registry’s serious adverse-event summary is organized by stratum. In Stratum I, it lists 57 at risk and 7 affected; in Stratum III, it lists 11 at risk and 0 affected. The other-adverse-event summary lists 57 at risk and 43 affected in Stratum I, and 11 at risk and 8 affected in Stratum III. These are the counts as recorded, not comparative estimates. The record does not state that the strata are randomized treatment groups, and the study itself is non-randomized.
The serious adverse-event terms include sepsis, hypokalemia, intracranial hemorrhage, and respiratory failure, each listed at 3.5%. Disseminated intravascular coagulation and left ventricular systolic dysfunction are each listed at 1.8%. These percentages should be interpreted as the registry’s reported figures for those terms; the record does not give enough detail to attribute them to a specific arm or to compare risks across arms.
Safety counts depend on the recorded definitions, observation period, and population included. The registry record does not report additional detail needed to establish how the listed terms relate to the stratum-level affected counts, or to make a causal comparison between treatment pathways. The figures describe reported safety events; they do not by themselves determine the balance of risks and benefits.
12. What the Result Does, and Does Not, Mean
The registry record documents a completed phase III study of an intensive treatment approach for childhood central nervous system atypical teratoid/rhabdoid tumor. It specifies a primary EFS endpoint, a Kaplan–Meier estimation method, a 95% confidence interval, and a follow-up timeframe of up to 4 years after enrollment. It also reports stratum-level safety counts and selected serious adverse-event terms. These details allow readers to understand what the study set out to measure and some of the posted safety information.
What the record does not provide is the numerical primary endpoint result. No estimated EFS probability, confidence interval endpoints, or EFS event count are reported. As a result, the registry record cannot support a conclusion about the observed EFS level or its precision. Nor does it provide a reported statistical comparison between the treatment arms.
The study’s non-randomized allocation is central to interpretation. Its arm labels describe different sequences of chemotherapy, radiation, and transplantation, but the registry cautions that the arms are not a like-for-like control comparison. A difference between arm outcomes, even if one were reported, would not automatically establish that treatment sequence caused the difference. Information about why a participant followed one pathway, baseline characteristics, and any adjustment for such differences is not reported in the registry record.
The planning illustrations do not fill this evidence gap. They use stated assumptions to express hypothetical sample-size requirements for planning; they are not observed power and do not summarize what happened in the trial. Likewise, a completed status and an enrollment of 70 do not reveal whether the primary endpoint met any target or whether one arm performed better.
The most defensible reading is therefore bounded: the registry describes the trial design, endpoint definition, and selected safety information, but does not post the numerical EFS outcome or analyses needed for an efficacy interpretation. Those missing results and analysis details are not reported in the registry record.
13. Why This Trial Matters Statistically
This trial illustrates why a registry entry must be read at more than the level of its title or completion status. The study addresses a childhood tumor and evaluates a complex combination of chemotherapy, radiation, and autologous stem cell transplantation. Its primary endpoint is a time-to-event measure, which requires attention to event definitions, follow-up, censoring, and uncertainty—not merely a count of enrolled participants.
It also highlights the importance of allocation. The record lists a parallel, non-randomized design and explicitly cautions that the arms are not a like-for-like control comparison. In such a design, treatment pathways may differ alongside participant characteristics or other factors. The design description should therefore guide how any arm-level contrast is interpreted; the registry does not report a numerical EFS contrast in any case.
The endpoint definition shows both the usefulness and limits of a composite outcome. EFS combines progression, relapse, second malignant neoplasm, and death from any cause into a time-to-event measure. That can provide a broad summary, while leaving unanswered which components occurred and how they contributed. Component-specific results are not reported in the registry record.
Finally, the record shows that method transparency and result reporting are separate needs. It names Kaplan–Meier estimation and a confidence interval, but does not provide the estimate or interval. The listed safety counts add information, but they do not substitute for the primary endpoint analysis. This distinction is statistically important: a clear endpoint definition is valuable, yet interpretation of the trial’s observed outcome requires the outcome estimate, its precision, and sufficient analysis context.
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 trial-specific limitation is that the registry record does not report the numerical primary EFS estimate, its 95% confidence interval, the event count, or an analysis result. The stated endpoint method cannot reveal the observed outcome without those results. The registry lists 0 analyses, leaving no reported inferential comparison to interpret.
The non-randomized design limits causal interpretation of any comparison between the listed treatment pathways. The record does not explain why participants entered a particular arm, provide baseline comparisons, or describe an adjustment strategy. It also does not report the reason for the different treatment sequences. The record explicitly cautions that the arms are not a like-for-like control comparison.
The population information is incomplete. The endpoint note identifies one ineligible patient in Stratum I and one in Stratum III, and no enrollment in Stratum 2 or 4. It does not report a complete analysis-population definition, the number censored, details of follow-up, or a fuller account of missing outcome information. Those gaps matter for interpreting a Kaplan–Meier estimate and its precision.
The registry gives a broad composite EFS definition but no component-specific results, so the contribution of progression, relapse, second malignant neoplasm, and death cannot be separated. It reports selected safety counts and serious adverse-event terms, but does not provide enough context to make arm-level causal comparisons from them.
Details of the statistical plan, including sample-size justification, interim analyses, multiplicity handling, and analysis populations, are not reported in the registry record. The planning illustrations are explicitly hypothetical and cannot be treated as the trial’s own design justification or as observed power.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Combination Chemotherapy, Radiation Therapy, and an Autologous Peripheral Blood Stem Cell Transplant in Treating Young Patients With Atypical Teratoid/Rhabdoid Tumor of the Central Nervous System — NCT00653068.
- Registry record last updated 2024-04-23; status last verified 2024-04.
- Results first posted 2017-02-13.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
NCT00653068 is a completed, non-randomized phase III study sponsored by the Children’s Oncology Group. It concerns childhood atypical teratoid/rhabdoid tumor of the central nervous system and lists enrollment of 70. The registry dates the study start to 2009-02-10 and primary completion to 2016-09-27.
The parallel treatment arms are labeled with different sequences of chemotherapy, 3-D conformal radiation therapy, and autologous peripheral blood stem cell transplantation. The primary outcome is estimated EFS up to 4 years after enrollment, calculated with Kaplan–Meier estimation and reported with a 95% confidence interval. The record does not report the numerical EFS result or analysis results.
Safety information includes stratum-level serious and other adverse-event counts and selected serious adverse-event terms. The registry identifies non-hematological toxicity of Grade 3 or higher during protocol therapy as a secondary outcome. It cautions that the non-randomized arms are not a like-for-like control comparison.