Tipifarnib in Subjects With Myelodysplastic Syndromes: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Tipifarnib in Subjects With Myelodysplastic Syndromes (NCT02779777), a randomized phase 2 study. The primary outcome is Objective Response Rate (ORR).

2. Clinical Question

The study was intended to investigate tipifarnib activity in myelodysplastic syndromes among subjects with no known curative treatment. The registry describes a randomized study, while the posted primary outcome is presented as counts by cohort or regimen rather than as a comparative estimate. For each regimen, the outcome table shows one participant and zero participants with an objective response. The Original Cohort row shows zero participants, so it does not contribute an interpretable response proportion.

The central statistical question is therefore descriptive: what do these very small posted groups show, and how much can they establish about response? The record does not post a comparative analysis. The regimen-level counts are too sparse to support a precise comparison, and the posted information does not explain how these outcome rows relate to the overall enrollment. The results should be read as limited registry summaries, not as a definitive estimate of the relative effects of the schedules.

3. Trial Design

This is an adaptive Phase 2 study of tipifarnib in subjects with myelodysplastic syndromes. The registry lists randomized allocation and a single masking designation. Its brief summary describes an open-label, randomized, two-stage study with two tipifarnib regimens. The design model field, however, is listed as single group. These entries do not give a fully consistent account of the design, and the registry record does not explain the discrepancy.

The two schedules differed in dose and duration within each 28-day cycle. Regimen 1 was 600 mg twice daily for 7 days; Regimen 2 was 300 mg twice daily for 21 days. The primary endpoint was objective response rate at 1 year, assessed by the investigator. The record lists additional outcomes covering transfusion independence, hematologic improvement, duration of response, survival, and treatment-emergent adverse events.

Enrollment is listed as 16, while the study summary describes approximately 36 eligible subjects as the intended study size. The record lists the study status as terminated. It gives a start date of 2016-06-01 and a primary completion date of 2018-08-28. The reason for termination is not reported in the registry record.

The posted primary outcome shows one participant in each regimen and zero responses in each. It also shows zero participants in the Original Cohort row. Those counts do not account for the enrollment figure on their own, and the record does not explain the relationship between the rows, the enrollment total, and the endpoint population. No formal comparative analysis is posted. The adaptive or two-stage features, including any decision rules and the timing of stages, are not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: This study included a screening period, treatment period (28-day cycles until disease progression), end of treatment visit, follow-up visit (for subjects who terminated treatment for reasons other than death or disease progression), and follow-up contact (upon disease progression, all subjects were followed approximately every 12 weeks for survival and use of subsequent therapy until either death or accrual in the subject's study cohort was completed, whichever occurred first).

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full analysis set (FAS) includes subjects enrolled in Protocol Amendment 3 (version 09 Jul 2017) and received at least one dose of tipifarnib.

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 Objective Response Rate (ORR), measured over 1 year.

The primary endpoint is objective response rate at 1 year. It counts participants whose response met one of the listed categories: complete response, partial response, marrow complete remission, or hematologic improvement. The registry says investigators assessed response according to the MDS International Working Group criteria. The endpoint is reported as a count of participants, rather than as an estimate with a confidence interval.

Response endpoints are used in clinical studies to summarize whether participants meet prespecified disease-response criteria during a specified assessment period. As a general statistical concept, a response count can be converted into a proportion only when the relevant denominator is clear. That proportion is more informative when accompanied by an interval showing its uncertainty. Here, each regimen row contains one participant and zero responses, while the Original Cohort row contains zero participants. The posted information does not include an interval or a comparison between regimens.

The population definition is also important for interpreting the endpoint: the full analysis set includes subjects enrolled in Protocol Amendment 3 who received at least one dose of tipifarnib. The registry does not explain how that defined population maps onto the displayed counts. Consequently, the endpoint describes the listed observations, but the posted record does not support a precise estimate of response frequency in the intended population.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry records the primary outcome as a count of participants with objective response. The response definition includes complete response, partial response, marrow complete remission, and hematologic improvement, assessed by the investigator according to MDS International Working Group criteria. The endpoint timeframe is 1 year. These details define what was counted, but they do not specify a statistical model or an inferential analysis.

What can a count tell us?

A count reports how many observed participants met the endpoint. To interpret it as a rate or proportion, the count needs a clear denominator: the number of participants assessed under the stated endpoint definition. The posted regimen rows each show one participant and zero responses. The Original Cohort row shows zero participants. A zero denominator cannot provide a response proportion, and the registry does not give a separate estimate that resolves this issue.

In general, a confidence interval expresses the uncertainty around an estimate. For a response proportion, an interval would communicate how imprecisely the observed group may represent a wider population. The posted primary outcome does not include confidence intervals. This absence matters especially when the groups are as small as those shown, because a point count by itself does not show the range of plausible response frequencies.

Why is there no regimen comparison?

The study description refers to randomization between two schedules, but the posted results contain no comparative analysis. The record lists zero analyses. It gives neither a test of a difference nor an effect estimate with uncertainty. In general, a comparison requires interpretable outcome data for the groups being compared and a defined analysis that respects the study design. The available outcome counts do not support a stable estimate of the difference between schedules.

The record includes a sample-size section based on planning assumptions: two-sided significance level 0.05, equal allocation, a normal approximation, and an effect defined as a stated fraction of the observed effect. These are described as planning illustrations, not as the study’s own design justification. The record does not report the planned group allocation or the analysis used to derive those illustrations. Any sample-size figures displayed elsewhere on the page should be understood as planning scenarios, not as observed power or as evidence that the study had adequate power.

What does over-dispersion mean in general?

For count outcomes, over-dispersion means that observed variation exceeds what a simple model assumes. It can affect standard errors and confidence intervals when the model is appropriate to the outcome and data. This is a general statistical consideration; the registry does not report a count model, an over-dispersion assessment, or a fitted model for this endpoint. Given that the primary outcome is a participant count and no analysis is posted, there is no basis to infer how such issues were handled.

What remains unknown?

The registry record does not report the statistical analysis plan, the method for handling missing or unevaluable response assessments, or a justification for the target enrollment. Although the study is described as adaptive and two-stage, decision rules, stage-specific analyses, and interim analyses are not reported in the registry record. It also does not explain why the design model is listed as single group while allocation is listed as randomized and the study summary describes two regimens. These omissions limit independent assessment of how the study was intended to be analyzed and how the displayed results were produced.

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("Regimen 1: Tipifarnib", "Regimen 2: Tipifarnib")
n      <- c(1, 1)
events <- c(0, 0)
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  Regimen 1: Tipifarnib  1  0  0.0  0.0  97.5  0.0  79.3  1.3  84.2
ARM  Regimen 2: Tipifarnib  1  0  0.0  0.0  97.5  0.0  79.3  1.3  84.2

11. Multiplicity, Missing Data and Analysis Populations

The record lists the objective response rate as the primary outcome and lists secondary outcomes involving transfusion independence, hematologic improvement, response duration, progression-free survival, overall survival, and treatment-emergent adverse events. It reports eight outcome measures, while seven secondary outcomes are listed. The registry record does not explain this difference or report a strategy for controlling multiplicity across outcomes, timeframes, or treatment comparisons. No adjustment for multiple testing is reported, and no inferential analyses are posted.

Multiplicity matters in general when researchers test or estimate many outcomes: the chance of finding an apparently notable result can change as the number of analyses increases. Without a stated analysis plan, readers cannot determine whether an outcome was designated in advance, how comparisons were prioritized, or whether any adjustment was planned. The posted record does not provide enough detail to answer those questions.

The primary endpoint population is described as subjects enrolled under Protocol Amendment 3 who received at least one dose of tipifarnib. The record does not report how missing, unevaluable, or incomplete response assessments were handled, nor does it explain how this population relates to the one participant shown in each regimen row and zero in the Original Cohort row. The handling of missing data is therefore not reported in the registry record.

In general, an analysis population definition helps clarify who contributes to each endpoint and whether participants are included according to their assigned treatment or their treatment received. Here, the listed definition and the displayed outcome counts cannot be reconciled from the registry information. The record also does not report a protocol-level analysis plan, interim-analysis methods, or the adaptive design decision rules. These gaps limit interpretation of both the outcome summaries and the intended statistical control of the study.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: anaemia (14.3%), pneumonia (14.3%), bradycardia (7.1%), palpitations (7.1%), pyrexia (7.1%).

The serious adverse event summary lists 7 of 14 participants at risk in the Original Cohort as affected, with 14 of 14 listed for other adverse events. For Regimen 1 and Regimen 2, the summary lists zero serious adverse events among one participant at risk in each group, and one participant with other adverse events in each group. Deaths are listed as zero in all three rows.

The registry also lists serious adverse event terms with percentages: Anaemia and Pneumonia at 14.3% each, and Bradycardia, Palpitations, Pyrexia, and Hip fracture at 7.1% each. These percentages are reproduced as posted; the record does not provide enough detail in this summary to establish their denominators or to connect each term to a treatment regimen.

These are descriptive safety counts, not estimates of comparative risk. The regimen groups each contain only one participant in this summary, and the record does not show an analysis comparing adverse event rates. An adverse event reported during a study is not, by itself, proof that treatment caused it. The registry record does not provide a causal assessment in these figures. The reason the study was terminated is also not reported in the registry record.

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

The posted primary outcome shows no objective responses among the one participant listed for each tipifarnib regimen. This is the result that can be stated directly from the table. The Original Cohort row has zero participants, so its zero response count should not be interpreted as an observed outcome in a group of participants. The registry does not report a comparative analysis between regimens.

These counts do not establish that tipifarnib cannot produce an objective response, nor do they establish that the two schedules have equivalent or different effects. The groups shown are too small to characterize response frequency, and the registry does not report uncertainty intervals. The outcome rows also do not account for the enrollment total of 16, and the record does not clarify the relationship between enrollment, the full analysis set, and the displayed regimen counts.

The safety summaries describe events among the listed participants, with more participants represented in the Original Cohort than in either regimen row. They do not support a reliable comparison of harms between schedules. The adverse event percentages are descriptive entries in the record and should not be treated as proof of treatment causation.

In general, a registry result is one component of a study’s evidence base. Interpretation also depends on the design, prespecified analysis, completeness of follow-up, and clarity of the analysis population. Here, several of those details are not reported in the registry record, and no inferential analysis is posted. The study’s terminated status is recorded, but the reason is not reported. The appropriate conclusion is therefore narrow: the posted endpoint counts are sparse, and they do not provide a precise or comparative estimate of response for the intended study population.

14. Why This Trial Matters Statistically

This study illustrates why the structure of a registry result matters as much as the endpoint label. Objective response rate is a familiar clinical-trial endpoint, but a response count needs a clear denominator and an interpretable analysis population. The posted regimen rows each contain one participant, while the Original Cohort row contains none. The overall enrollment is listed as 16, so the displayed counts do not by themselves show how the endpoint results cover the enrolled study population.

The record also shows why design fields should be read together rather than in isolation. Allocation is listed as randomized, the summary calls the study open-label and randomized, and the design model is listed as single group. These descriptions are not reconciled in the record. The study is described as adaptive and two-stage, yet the decision rules and analysis details are not reported. A label alone cannot tell readers how the study was conducted or how conclusions were meant to be drawn.

Statistically, the example underscores the distinction between a descriptive observation and an estimate with uncertainty. Zero responses in a very small displayed group is a factual count, but it does not provide a precise estimate of response frequency. Without a comparative analysis or clear denominators, readers should not turn these rows into claims about relative efficacy.

The same caution applies to safety: event counts and percentages describe what the registry lists, but small groups and absent comparative analyses constrain interpretation. Clear reporting of populations, missing data, analysis methods, and design decisions is essential for readers to assess what a trial’s results can support.

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.

The primary outcome rows are sparse: each regimen has one participant and zero responses, while the Original Cohort has zero participants. The record does not explain how these displayed counts relate to the enrollment of 16 or to the defined full analysis set. This limits the ability to assess whether the posted rows represent the intended primary endpoint population.

Several design descriptions are not reconciled. The record lists randomized allocation, a single-group model, and a study summary that describes an open-label, randomized, two-stage design. The registry does not explain how these entries fit together. It also does not describe the adaptive decision rules or report interim analyses.

No inferential analyses are posted, and there are no confidence intervals or comparative estimates for objective response. The statistical analysis plan, missing-data handling, and sample-size justification are not reported in the registry record. As a result, the displayed counts cannot be placed in the context of a prespecified analysis or evaluated for precision using posted inferential results.

The study is marked terminated, but the reason is not reported in the registry record. The record lists eight outcome measures and seven secondary outcomes without explaining the difference or describing multiplicity control. The serious adverse event summary gives counts and term percentages, but does not provide enough information to establish a regimen-specific comparison or causal relationship. Together, these limitations make the record useful as a description of posted observations, but insufficient for strong comparative or population-level conclusions.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Tipifarnib in Subjects With Myelodysplastic Syndromes — NCT02779777.
• Registry record last updated 2024-07-31; status last verified 2024-07.
• Results first posted 2024-07-31.
• 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

This Phase 2 registry record concerns tipifarnib in subjects with myelodysplastic syndromes. It describes two dosing regimens and lists objective response rate at 1 year as the primary outcome. The record reports enrollment of 16 and a terminated study status.

The primary outcome rows show one participant and zero responses for each regimen, and zero participants in the Original Cohort. No inferential analyses are posted. The record also lists serious and other adverse event counts, including serious adverse events in 7 of 14 participants in the Original Cohort and none among one participant in each regimen group.

Interpretation is limited by sparse outcome rows, unresolved differences among design descriptions, and absent information on the analysis plan, missing data, adaptive decision rules, and reason for termination. The posted results are descriptive and do not provide a precise comparison of the two regimens.