Bosutinib in Treating Patients With Chronic Myeloid Leukemia in Chronic Phase After Frontline TKI Failure: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Bosutinib in Treating Patients With Chronic Myeloid Leukemia in Chronic Phase After Frontline TKI Failure (NCT02906696), a na phase 2 study. The primary outcome is Response Rate.

2. Clinical Question

This is a question about response within a treated group, not a direct comparison of bosutinib with another treatment or with no treatment. The primary endpoint counts participants meeting a response definition. For those not currently in partial cytogenetic response, achieving major cytogenetic response qualifies; for those already in partial cytogenetic response, achieving complete cytogenetic response qualifies. Thus, the endpoint uses different qualifying milestones according to starting status.

The registry record lists an enrollment of 8 and a terminated status. It does not report the reason for termination or an observed primary response result. Consequently, the record does not establish how many participants met the endpoint or the precision of any response estimate. The question remains clinically and statistically specific, but the available registry information does not provide the result needed to answer it.

3. Trial Design

This was an open-label phase II study sponsored by M.D. Anderson Cancer Center. The registry describes a single-group model with no masking and allocation recorded as not applicable. Participants were to receive bosutinib; there is no randomized or concurrent control arm described. As a result, the study design can describe outcomes among treated participants but cannot directly estimate a treatment contrast against a control group.

The listed starting dose was 300 mg daily. The treatment description specifies bosutinib by mouth daily on days 1–28, with cycles repeated every 28 days for up to 2 years unless disease progression or unacceptable toxicity occurred. Laboratory biomarker analysis is listed as another intervention. The primary outcome timeframe is up to 6 months, which is shorter than the maximum treatment period stated in the treatment description.

The registry gives a start date of 2016-10-28 and a primary completion date of 2019-08-08. The study status is terminated. The reason for termination, the timing of termination relative to the planned study stages, and whether the intended follow-up was completed are not reported in the registry record. Enrollment is listed as 8.

The primary endpoint is response rate, defined by cytogenetic response milestones and planned for estimation with a 95% credible interval. The description also states that a Simon’s optimal two-stage design will be used for interim futility monitoring. This is a description of the planned design; the registry does not report whether the stages were completed, whether a futility decision was made, or the outcome data used in any such decision. The available record therefore documents important design intentions but not their implementation or the study’s final response analysis.

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

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 Response Rate, measured over Up to 6 months.

The primary endpoint is the number of participants with a response up to 6 months. Response is defined according to cytogenetic status at the outset: for a participant not currently in partial cytogenetic response, achieving major cytogenetic response counts as response; for a participant currently in partial cytogenetic response, achieving complete cytogenetic response counts.

This endpoint is used to represent whether participants reach a specified disease-response milestone, while allowing the qualifying milestone to reflect their starting cytogenetic status. Because it is a participant count, interpretation would ordinarily depend on both the number meeting the definition and the population assessed. The registry describes estimating the response rate with a 95% credible interval, which would communicate uncertainty around the estimated rate if the analysis results were available.

The record does not report the observed number of responders, the denominator used in the endpoint analysis, or the resulting estimate and interval. Enrollment alone does not supply those quantities: participants may have differing follow-up or evaluability, and the registry record does not specify how these issues were handled. Thus, the endpoint is defined, but the posted information does not permit a numerical conclusion about the response rate.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a single-group design estimate?

The study is recorded as single-group, with allocation not applicable and no masking. Its primary question concerns response among participants receiving bosutinib, rather than a difference between treatment groups. In general, a single-group study estimates an outcome within the treated population. It does not create a concurrent counterfactual group, so the observed outcome cannot isolate the treatment effect from other explanations such as the natural course of disease, participant selection, or changes in care over time.

The registry does not report an observed response count or estimate. Therefore, no estimate of response or its uncertainty can be discussed from the posted primary outcome information. The listed enrollment of 8 describes enrollment, not necessarily the denominator for the primary endpoint; the analysis population and handling of participants without endpoint data are not reported in the registry record.

How should the primary endpoint definition be read?

The endpoint counts participants meeting a cytogenetic response milestone within up to 6 months. The qualifying milestone differs by the participant’s current status: major cytogenetic response for those not currently in partial cytogenetic response, and complete cytogenetic response for those already in partial cytogenetic response. This structure makes baseline status relevant to classification. The registry does not report the number of participants in either baseline category or how many in each category met the corresponding criterion.

In general, a binary endpoint reduces each participant’s outcome to whether a prespecified event or response occurred. A count of responders is not, by itself, a response rate: a rate also requires a defined denominator. If follow-up is incomplete or some participants cannot be classified, the analysis rule for those cases can affect the estimate. The rules for missing endpoint data and the endpoint analysis population are not reported in the registry record.

What does a 95% credible interval convey?

The primary outcome description says the response rate will be estimated with a 95% credible interval. In general, a credible interval is an interval from a Bayesian analysis that expresses uncertainty about a parameter under the model and prior assumptions used. It should be interpreted as an uncertainty range conditional on that analysis, rather than as a guarantee that the true value lies within the interval. The registry record does not identify the prior, the model details, or the calculated interval. As a result, the stated interval level alone is not enough to reproduce or assess the planned estimate.

What is the role of the two-stage design?

The record states that a Simon’s optimal two-stage design will be used for interim futility monitoring. In general, a two-stage design evaluates early study data against prespecified criteria. If the early results meet a futility boundary, the study may stop rather than proceed; otherwise, it may continue to another stage. Such designs can limit exposure of participants to a study that is unlikely to meet its target and can reduce resources spent on an unpromising result. Their operating characteristics depend on the chosen response thresholds, error rates, and stage sizes.

The registry record does not report those design parameters, the number of participants planned at each stage, an interim result, or whether a stopping boundary was met. Although termination is recorded, it cannot be attributed to the two-stage futility procedure from the information available. The reason for termination is not reported in the registry record.

How should sample-size planning illustrations be understood?

The registry record’s planning assumptions describe illustrations based on an observed effect, a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is described as a stated fraction of the observed effect. For time-to-event outcomes, the planning quantity is total events; for other outcomes, it is participants per arm. These are planning conventions, not evidence of the trial’s actual design.

Those assumptions describe a two-arm planning framework, whereas this study is recorded as single-group and has no comparator arm. The record explicitly characterizes the calculations as a planning illustration rather than the trial’s own design. They should not be interpreted as observed power, a reconstruction of the study’s enrollment target, or a result of the Simon two-stage method. No numerical planning estimates are stated in the registry record.

What can event counts tell readers?

For adverse events, counts describe the number of participants recorded as affected within a group. In general, an affected-participant count differs from an event count: one participant may experience more than one event, and the same participant may contribute to more than one event category. Event percentages for named terms do not establish that those categories are mutually exclusive. The record does not provide enough detail to calculate event rates over time or to determine the total number of events.

Overall, the registry documents intended methods—response estimation with a credible interval and two-stage futility monitoring—but does not report the analyses needed to evaluate their results. The method descriptions should therefore be treated as protocol-level information, not as evidence that a particular analysis was completed or produced a particular conclusion.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Treatment (Bosutinib)")
n      <- c(8)
events <- c(6)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Treatment (Bosutinib)  8  6  75.0  34.9  96.8  40.9  92.9  40.0  92.5

11. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary response-rate outcome and secondary outcomes that include treatment interruptions and dose reductions; major molecular response, MR4, MR4.5, and complete molecular response; BCR-ABL/ABL thresholds; overall, event-free, and transformation-free survival; and change in ABL kinase domain mutation status. Multiple endpoints create the possibility of multiple statistical questions. In general, testing many outcomes can increase the chance of at least one apparently notable result arising by chance, so an analysis plan may specify a hierarchy, adjustment, or descriptive approach. No multiplicity strategy is reported in the registry record.

The record describes a 95% credible interval for the primary response rate and planned two-stage futility monitoring. It does not report the prior or full model for the interval, the detailed futility boundaries, or how secondary outcomes were to be analyzed. The number of analyses listed is 0, so there are no reported analysis entries from which to assess endpoint-specific methods or multiplicity handling.

Missing-data rules are not reported in the registry record. In general, interpretation of a binary response endpoint can depend on whether participants without an assessment are excluded, treated as nonresponders, or handled using another approach. The available record does not say which rule applied, how incomplete follow-up was addressed, or whether participants who discontinued treatment remained in the endpoint analysis.

The record also does not report the analysis populations—for example, whether the primary result was to include everyone enrolled or a subset meeting particular criteria. Enrollment of 8 is not enough to infer the number assessed for response. The absence of these details limits reproducibility and prevents readers from determining how missing observations, eligibility, or analysis-population choices might affect an estimate.

12. 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: chest pain (12.5%), diarrhea (12.5%), gastrointestinal bleed (12.5%), arterial rupture (12.5%), viral infection (12.5%).

The serious adverse-event information is recorded for the bosutinib treatment group. It lists 4 of 8 participants as affected by serious adverse events, 8 of 8 as affected by other adverse events, and 0 deaths. These are participant-level figures in the record; they are not a comparison between treatment arms, because the study has a single group.

The listed serious-event terms are chest pain, diarrhea, gastrointestinal bleed, arterial rupture, and viral infection, each shown with 12.5%. These term percentages should not be added: the record does not say that event categories are mutually exclusive, and a participant may have experienced more than one event. The term-specific figures also do not reveal how many events occurred over time or whether events were recurrent.

In general, adverse-event counts describe what was recorded during the study, not necessarily what was caused by the treatment. The registry record does not report event timing, duration, severity grades, attribution, exposure time, or the criteria used to classify events as serious. These limitations mean the figures are descriptive and should not be read as comparative estimates or as a complete account of safety.

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

The registry record establishes that a phase II, open-label, single-group study evaluated bosutinib in adults with chronic-phase chronic myeloid leukemia after frontline tyrosine kinase inhibitor failure. It defines a primary cytogenetic response endpoint with a timeframe of up to 6 months, describes a planned credible interval, and states an intention to use a Simon’s optimal two-stage design for interim futility monitoring. It also lists molecular, survival, treatment-modification, and mutation-status outcomes.

Those design details do not answer whether participants met the primary endpoint. The observed response count, denominator, rate, and interval are not reported in the registry record. The study status is terminated, but the reason is not reported. Neither fact supports a conclusion about efficacy, futility, toxicity, or the cause of study closure.

Because the design has no concurrent comparator, any outcome data would describe the treated group rather than establish a comparative treatment effect. General statistical interpretation: outcomes in a single group may be influenced by factors other than the intervention, and the design cannot separate those explanations in the way a suitable controlled comparison can. The record does not provide outcome analyses for the listed secondary measures either.

The adverse-event counts provide limited descriptive information: serious adverse events affected 4 of 8 participants, other adverse events affected 8 of 8, and there were 0 deaths. The listed serious-event terms each carry a percentage, but details on timing, attribution, and overlap are not reported. These figures cannot on their own determine causality or comparative risk.

Accordingly, the record is informative about the study’s intended question and broad design, but not about its primary response result. It supports no claim that bosutinib did or did not produce a particular response rate. The distinction between a defined endpoint and a reported endpoint result is essential: a plan specifies what would be measured, while an analysis is needed to show what was observed and how uncertain that estimate is.

14. Why This Trial Matters Statistically

This study raises a clear statistical question: how often participants with chronic-phase chronic myeloid leukemia after frontline tyrosine kinase inhibitor failure meet a response milestone during treatment with bosutinib? The endpoint definition is clinically structured around cytogenetic status, and the planned follow-up window for the primary measure is up to 6 months. Recording the endpoint and its intended estimation method helps identify what the investigators sought to learn.

The design also illustrates the limits of interpreting a single-group phase II study. A treated-group response estimate can describe what happened among participants, but without a concurrent comparator it cannot directly quantify how outcomes differ from an alternative approach. This is particularly important when the record does not report an actual endpoint estimate or its uncertainty.

The planned two-stage futility procedure is statistically relevant because such designs can incorporate an early decision point rather than waiting for all planned observations. Yet a design description and a study status do not reveal whether an interim boundary was reached. The registry does not give the stage-specific rules or report a decision, so the procedure’s actual role in this study remains unknown.

The record also highlights why transparent reporting extends beyond the endpoint name. Interpretation depends on the response denominator, analysis population, treatment of missing assessments, and interval method. Those details are not reported. Likewise, the listed secondary outcomes show a broad measurement agenda, but no analysis results or multiplicity approach are described.

For readers assessing evidence, the key contribution is therefore methodological context rather than a numerical efficacy conclusion. The registry records a defined clinical question, a single-group design, planned monitoring, and limited safety counts. It does not report the response analysis needed to quantify the outcome or assess its precision.

15. 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.

Important information needed to interpret the primary endpoint is not reported in the registry record. There is no observed responder count, response denominator, estimated response rate, or 95% credible interval. The registry does not state the analysis population or how participants without an evaluable response assessment were handled. Enrollment of 8 cannot be assumed to equal the endpoint denominator.

The trial is recorded as single-group, open-label, and without masking. No control comparison is described, so the design does not support a direct estimate of the difference between bosutinib and another treatment. The record also does not report whether the planned treatment and follow-up were completed as intended. It lists a primary completion date of 2019-08-08 and a terminated status, but the reason for termination is not reported.

The statistical descriptions are incomplete for independent assessment. The registry says a 95% credible interval was planned but does not report its model or prior assumptions. It names Simon’s optimal two-stage monitoring but does not report the response thresholds, stage sizes, interim result, or whether a stopping decision occurred. A listed design intention cannot be treated as evidence of implementation.

Secondary outcome results and a multiplicity strategy are not reported. The safety record gives participant counts and term percentages, but lacks details on timing, severity, attribution, exposure duration, and overlap across event categories. These limitations prevent more detailed interpretation of either efficacy or safety from the registry information.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Bosutinib in Treating Patients With Chronic Myeloid Leukemia in Chronic Phase After Frontline TKI Failure — NCT02906696.
• Registry record last updated 2020-05-11; status last verified 2020-04.
• Results first posted 2020-04-22.
• 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.

19. Record Summary

NCT02906696 is a phase II, open-label, single-group study of bosutinib in adults with chronic-phase chronic myeloid leukemia after frontline tyrosine kinase inhibitor failure. The registry lists a starting dose of 300 mg daily, enrollment of 8, a start date of 2016-10-28, a primary completion date of 2019-08-08, and terminated status. The reason for termination is not reported.

The primary outcome is response rate up to 6 months, with cytogenetic response criteria that depend on current status. A 95% credible interval and Simon’s optimal two-stage interim futility monitoring are described as planned methods. No primary or secondary outcome analyses are reported in the registry record. Safety listings include serious adverse events affecting 4 of 8 participants, other adverse events affecting 8 of 8, and 0 deaths.