Phase II Study of Gemcitabine+Romidepsin in the Relapsed/Refractory Peripheral T-cell Lymphoma Patients: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Phase II Study of Gemcitabine+Romidepsin in the Relapsed/Refractory Peripheral T-cell Lymphoma Patients (NCT01822886), a na phase 2 study. The primary outcome is Complete Remission (CR) Rate.

2. Clinical Question

3. Trial Design

This is a Phase IIa, multicenter, single-arm, open-label study of gemcitabine plus romidepsin in relapsed or refractory PTCL. The record lists an enrollment of 20 and the study status as completed. It identifies Fondazione Italiana Linfomi - ETS as the sponsor. Allocation is listed as NA; masking is listed as none. With no concurrent control arm, the study is not a randomized comparison.

Participants are described as receiving the GEMRO regimen as salvage treatment in a selected population. The listed regimen starts with romidepsin 12 mg/m2 on days 1, 8, and 15 plus gemcitabine 800 mg/m2 on days 1 and 15 for 6 cycles of 28 days. The schedule then lists romidepsin 14 mg/m2 on days 1 and 15 to progression disease. This regimen description does not establish how many participants received each component, completed treatment, or remained under observation for a specified duration; those details are not reported in the registry record.

The primary outcome is Complete Remission (CR) Rate, with a count of participants as its unit and 18 months as its timeframe. Secondary outcomes named in the record include progression-free survival, overall survival, frequency of Grade 3 and 4 toxicities, and overall response rate. The record lists five outcome measures but zero analyses. That distinction matters: an outcome name identifies what was intended for assessment, while an analysis entry would provide results or statistical comparisons. The posted results described in the record do not supply such an analysis for the primary endpoint.

Open-label status means the record does not describe masking of treatment assignment. In a single-arm study there is also no control group against which to compare outcomes. These design features do not invalidate the study's descriptive role, but they limit what can be inferred about differences attributable to treatment. The registry record does not report a statistical analysis plan, the intended analysis population, or a sample-size justification.

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 Complete Remission (CR) Rate, measured over 18 months.

The primary endpoint is Complete Remission (CR) Rate, expressed in the registry as a count of participants, with an 18 months timeframe. The record defines complete remission as disappearance of all target lesions according to the Revised Response Criteria for Malignant Lymphoma (Cheson et al. 2007). Thus, the endpoint is intended to capture whether participants meet a defined response category, rather than a continuous measure such as change in lesion size.

General statistical context: complete remission is a categorical outcome. A count of participants meeting the definition can describe how many reached that response category in the assessed group. To interpret such a count as a rate or proportion, a denominator and a clear description of who was assessed are also needed. The registry record does not report the observed count, an endpoint-specific denominator, or a remission proportion. Enrollment is listed as 20, but the record does not say how many participants contributed to the primary endpoint assessment.

General context: a complete-response endpoint can be useful because it applies a defined response threshold, making the clinical meaning of the outcome more explicit than a loosely defined improvement. It still gives only one part of the clinical picture. By itself, a remission count does not describe how long a response lasted, whether disease later progressed, survival, or treatment-related harms. The registry lists some of these as secondary outcomes, but does not report their analyzed results in the record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

General explanation: statistical methods answer different questions. A design describes how participants were assigned and observed; an endpoint defines the outcome; and an analysis specifies how observed data are summarized or compared. The registry record describes a single-arm, open-label design and names the outcomes, but lists zero analyses. It does not report a statistical analysis plan. The methods that would have been used to analyze the outcomes are therefore not reported in the registry record.

What does a single-arm study estimate?

In a single-arm study, outcomes are observed in the group receiving the study regimen. A summary could describe the number meeting an endpoint, the time until an event, or the frequency of an adverse event, depending on the outcome and analysis. Without a concurrent comparator, such a summary does not estimate a randomized difference between treatments. Differences from outcomes in other studies would be indirect comparisons and could reflect differences in eligibility, care, measurement, or follow-up. No such comparison is reported for this study.

General explanation: random allocation helps balance measured and unmeasured factors between groups on average, while masking can reduce some forms of bias in treatment delivery or outcome assessment. This record lists allocation as NA and masking as none. It does not describe a randomized control group or masked outcome assessment. These features matter especially when outcomes involve clinical judgment, although the registry's stated remission definition provides a named criterion for classification. The record does not say how remission assessments were conducted or adjudicated.

How should a count endpoint be interpreted?

The primary endpoint is described as a count of participants with complete remission at the 18 months timeframe. A count is an absolute tally. To form a proportion, the count must be related to a defined denominator, such as participants assessed under a specified analysis population. The record does not report the count or the endpoint-specific denominator. It also does not state whether the count would include all enrolled participants, only those who completed assessment, or another population. Those analysis-population details are not reported in the registry record.

General explanation: for a binary outcome, an uncertainty interval around a proportion conveys how much imprecision remains given the available observations and the chosen statistical method. With a limited number of observations, an estimate can be imprecise even when it is calculated correctly. The appropriate interval method depends on the analysis plan and data structure. Neither a remission proportion nor an interval method or interval is reported in this record, so no precision assessment can be made for the primary outcome.

What do the planning illustrations represent?

The sample-size assumptions describe a planning illustration based on the observed effect, rather than the trial's own design. The assumptions state a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. They frame the effect to detect as a stated fraction of the observed effect. These assumptions define a hypothetical future comparison; they do not describe the single-arm structure recorded for this study.

General explanation: a planning calculation asks how much information a proposed study might need under specified assumptions. Its output is conditional on the chosen effect, significance level, allocation, outcome model, and other design choices. It should not be read as a result from the completed study. Nor is a retrospective calculation based on an observed estimate the same as observed power. The posted planning illustrations are not a substitute for an outcome estimate, an uncertainty interval, or a reported statistical analysis. The registry record does not report its own sample-size justification.

The planning assumptions distinguish between time-to-event outcomes and other outcomes: time-to-event illustrations use a total number of events, while other illustrations use participants per arm. General explanation: time-to-event precision and power often depend strongly on the number of observed events, not merely on enrollment. For other endpoint types, the relevant information depends on the outcome distribution and the analysis method. These general principles do not reveal the actual event count or analysis choices for this study; those details are not reported in the registry record.

Why do analysis choices matter?

General explanation: choices about the analysis population, missing observations, outcome timing, and treatment of events can change a result. For a remission endpoint, for example, the handling of participants without a recorded assessment must be defined before interpreting a count as a proportion. For survival outcomes, the definition of time origin, event, censoring, and follow-up matters. The registry names progression-free and overall survival but does not report these operational analysis details or results.

General explanation: statistical significance and clinical importance are different concepts. A significance test, when reported, evaluates how compatible the data are with a specified null model under its assumptions. An effect estimate describes magnitude, while an interval describes precision. Neither a test nor an interval is available for the primary remission endpoint in this record. In the absence of an estimate and analysis description, it is not possible to infer statistical significance, clinical importance, or the strength of evidence.

The safety figures also require defined methods for interpretation. General explanation: adverse-event summaries may count participants with at least one event, total events, or events classified by severity or relatedness. These are different quantities. The record gives participant counts at risk and affected for serious and other adverse events, and lists serious-event terms with percentages, but it does not explain the analysis rules, event timing, overlap among terms, or attribution process. No additional calculations can resolve those missing reporting details.

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("Romidepsin, Gemcitabine")
n      <- c(20)
events <- c(3)
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  Romidepsin, Gemcitabine  20  3  15.0  3.2  37.9  5.2  36.0  5.4  36.3

11. Multiplicity, Missing Data and Analysis Populations

The registry record names a primary outcome and several secondary outcomes: progression-free survival, overall survival, frequency of Grade 3 and 4 toxicities, and overall response rate. It lists five outcome measures but zero analyses. The result for the primary outcome, and analyzed results for the secondary outcomes, are not reported in the registry record. The record also does not report a prespecified hierarchy or testing strategy for the outcomes.

General statistical context: when several outcomes or comparisons are tested, the chance of at least one apparently notable result can increase if each is judged separately without accounting for multiplicity. A prespecified hierarchy, adjustment method, or distinction between confirmatory and exploratory outcomes can clarify how evidence should be interpreted. The registry record does not report how multiplicity was handled, so no claim about adjustment or control of false-positive findings can be made.

Missing-data handling and analysis populations are also not reported in the registry record. In particular, the record does not say how participants without a remission assessment would be treated, whether all enrolled participants would be included in the primary analysis, or how withdrawal and incomplete follow-up would be addressed. General statistical context: excluding participants with missing outcomes can bias a result if missingness is related to prognosis or treatment experience. Different assumptions or analysis strategies can produce different estimates; the relevant strategy cannot be identified.

For the time-to-event outcomes, the record does not report the event definitions, censoring rules, follow-up summaries, or analysis population. For the safety outcomes, it does not describe how events were classified or whether event terms could overlap. These reporting gaps limit evaluation of both the endpoint summaries and the assumptions behind any statistical analysis.

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: fever with shiver, hypotension, desaturation (5.0%), pneumothorax (5.0%), dyspnea (5.0%), fever, loss of consciousness and urinary infection (5.0%), fever and pulmonary insufficiency (5.0%).

The registry record lists the Romidepsin, Gemcitabine group with 20 participants at risk for serious adverse events and 4 affected. It lists 20 at risk for other adverse events and 20 affected. The deaths count is listed as 10. These are reported counts, not a comparison with another treatment group. The record does not provide further detail on the timing of deaths, their causes, or whether they were considered related to treatment.

The listed serious adverse-event terms are Fever with shiver, Hypotension, Desaturation; Pneumothorax; Dyspnea; Fever, loss of consciousness and urinary infection; Fever and pulmonary insufficiency; and Pneumonia. Each is reported as 5.0%. The registry record does not explain whether terms are mutually exclusive, whether one participant could contribute to more than one term, or whether these entries represent participants or event episodes. They should therefore be read as the record's reported serious-event entries, not as a complete account of all clinical circumstances.

General statistical context: adverse-event counts describe events observed and recorded in a particular study group. They do not establish that the regimen caused an event, and without a comparator they do not show whether event frequency differs from another treatment. Definitions, observation time, severity grading, attribution, and follow-up affect interpretation. Further information on these matters is not reported in the registry record.

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

The record establishes that a Phase IIa, multicenter, single-arm, open-label study evaluated gemcitabine plus romidepsin as salvage treatment in a selected population with relapsed or refractory PTCL. It identifies complete remission as the primary outcome and gives a definition based on disappearance of all target lesions under the named response criteria. It also lists several secondary outcomes and descriptive safety information. These are useful descriptions of the study question and the outcomes the registry tracks.

However, the record does not report the observed primary endpoint result. It gives no remission count, endpoint denominator, proportion, or interval. The number enrolled is not a substitute for the number assessed for remission. The record lists zero analyses, so no efficacy analysis or secondary-outcome result can be interpreted from the information recorded. The absence of a posted result does not mean that no participant achieved remission or that the outcome was unfavorable; it means the result is not reported in the registry record.

The single-arm design also constrains causal interpretation. A result from one treated group could describe outcomes in that study population, but it could not by itself show what would have happened to the same participants under a different treatment or no treatment. No randomized control group is recorded, and there is no like-for-like comparator in the study design. Thus, the design does not yield a randomized estimate of a treatment difference.

General statistical context: strength of inference depends on the estimate, its precision, the design, and the analysis assumptions. A point estimate without an uncertainty measure can leave substantial ambiguity; an outcome title without a result provides no estimate at all. This record does not provide the primary result or a precision estimate. It therefore cannot support a conclusion about the magnitude or certainty of an efficacy outcome.

The safety entries show that the registry records adverse events and deaths, but those counts alone cannot establish causality or comparative risk. The record does not describe timing, causes of death, exposure duration, or additional context needed to assess individual event patterns. The study's findings should not be treated as medical advice or as proof that a treatment does or does not work.

14. Why This Trial Matters Statistically

Statistically, this study is useful to examine because it makes the distinction between a registered clinical question and a reported statistical result especially clear. The record specifies a population, a combination regimen, a primary response endpoint, a timeframe, and secondary outcomes. Yet it lists zero analyses and does not report the primary remission count or its denominator. Knowing what an outcome is meant to measure is not the same as having an estimate that can be interpreted.

The single-arm design is another important teaching point. Such a study can describe the experience of a selected group receiving a regimen, and may be informative for planning further research. It cannot provide a randomized treatment comparison when there is no control arm. Keeping that distinction clear prevents descriptive outcomes from being overstated as causal effects.

The planning illustrations also show why assumptions must be kept separate from trial findings. Their stated framework uses a two-arm comparison, equal allocation, a normal approximation, and an effect defined as a fraction of an observed effect. Those are planning assumptions, not a description of the registered single-arm design or evidence that the trial had a particular power. Planning calculations can help frame later study designs, but they do not recover an unreported result.

Finally, the record provides some safety counts and event terms without the context needed for causal or comparative interpretation. Reading those data carefully means distinguishing affected participants from event descriptions, recognizing that the registry does not explain potential overlap, and avoiding assumptions about timing or attribution. Together, these features make the study a useful example of why design, endpoint definitions, result reporting, uncertainty, and safety context must be considered separately.

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 most consequential limitation for interpreting efficacy is that the registry record does not report the primary endpoint result. Although complete remission is defined and assigned an 18 months timeframe, no observed count, denominator, remission estimate, or uncertainty interval is recorded. The enrollment figure does not establish how many participants had an evaluable endpoint. As a result, the primary outcome cannot be quantified from this record.

The single-arm, open-label design has no concurrent control group and no masking. This limits comparative and causal interpretation and may leave outcome assessment more vulnerable to knowledge of treatment, depending on how assessments were conducted. The record does not report assessment procedures, independent review, or adjudication. It also does not provide participant characteristics or fuller eligibility criteria, making it difficult to judge how broadly the study population represents other patients with PTCL.

Important analysis details are not reported in the registry record. These include the statistical analysis plan, analysis populations, missing-data handling, the method for estimating uncertainty, and the approach to multiple outcomes. For the survival outcomes, event definitions, censoring rules, follow-up summaries, and estimates are not reported. The record names those outcomes but does not provide results that would permit evaluation.

Safety interpretation is limited by the absence of event timing, causes of death, attribution, and detail on whether the serious-event terms overlap. The record also does not describe exposure duration or how safety observations were collected. These omissions mean the listed counts should not be treated as complete comparative risk estimates.

Finally, the planning illustrations are explicitly distinct from the trial's own design. Their assumptions do not remedy missing trial results or establish the study's original sample-size rationale. The registry record does not report that rationale.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Phase II Study of Gemcitabine+Romidepsin in the Relapsed/Refractory Peripheral T-cell Lymphoma Patients — NCT01822886.
• Registry record last updated 2019-10-14; status last verified 2018-08.
• Results first posted 2019-10-10.
• 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

The registry record describes NCT01822886, a completed Phase IIa, multicenter, single-arm, open-label study of gemcitabine plus romidepsin as salvage treatment in a selected population with relapsed or refractory PTCL. Enrollment is listed as 20. The regimen schedule includes gemcitabine and romidepsin followed by romidepsin maintenance to progression disease.

Complete Remission (CR) Rate is the primary endpoint, defined as disappearance of all target lesions under the Revised Response Criteria for Malignant Lymphoma, with an 18 months timeframe. Secondary outcomes include progression-free survival, overall survival, Grade 3 and 4 toxicity frequency, and overall response rate. The registry lists five outcome measures and zero analyses.

The record does not report the primary endpoint result, its denominator, or statistical uncertainty. It also gives descriptive adverse-event counts and serious-event terms without sufficient information for comparative or causal interpretation. There is no control group, and the registry record does not report a statistical analysis plan, missing-data approach, or sample-size justification.