A Study of CS1001 in Subjects With Relapsed or Refractory Extranodal Natural Killer/ T Cell Lymphoma(ENKTL): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of CS1001 in Subjects With Relapsed or Refractory Extranodal Natural Killer/ T Cell Lymphoma(ENKTL) (NCT03595657), a na phase 2 study. The primary outcome is Objective Response Rate (ORR) Assessed by IRRC.

2. Clinical Question

This is a descriptive question about outcomes observed in a group receiving CS1001, not a randomized estimate of the difference between treatments. Without a concurrent comparator, the trial cannot directly distinguish an effect of treatment from other influences on observed responses, such as characteristics of the enrolled participants or the natural course of illness. The registry entry describes the trial as evaluating efficacy and safety, but the information recorded for the primary endpoint gives its definition rather than a numerical result.

The primary endpoint is an objective response rate assessed by an independent review committee. The registry says the assessment used the Criteria for Response Assessment of Lymphoma: Lugano 2014 Classification. It does not give the number or percentage of participants classified as responders, or the confidence interval around that percentage. The reported median follow-up was 29 months, with the endpoint timeframe described as enrollment to the end of follow-up. These details define the context for interpreting the endpoint, but do not substitute for the missing response result.

3. Trial Design

This was a multicenter, single-arm, phase II clinical trial of CS1001 in participants with relapsed or refractory ENKTL. The registry lists CStone Pharmaceuticals as the sponsor and gives the study status as completed. Enrollment was 80. The study start date is 2018-06-12, and primary completion is recorded as 2023-03-20.

All participants were assigned to the CS1001 group; there was no separate control arm. The intervention is described as CS1001 monotherapy, given as 1200 mg by intravenous infusion every 3 weeks. Allocation is listed as not applicable, consistent with the single-group model, and masking is recorded as none. A randomized comparison or blinded treatment assignment is therefore not described in the registry record.

The primary outcome was objective response rate assessed by the IRRC. The endpoint description says response was evaluated according to the Criteria for Response Assessment of Lymphoma: Lugano 2014 Classification. Its assessment timeframe is described as enrollment to the end of follow-up, with a median follow-up of 29 months. Two patients were excluded from the primary endpoint analysis: one was not confirmed as having ENKTL by central pathology, and one was considered by the IRRC to have no measurable or evaluable disease at baseline.

The registry lists 14 outcome measures, including investigator-assessed objective response rate, complete and partial response rates, duration of response, and time to response. It does not list analysis entries. The primary endpoint's numerical result, the statistical analysis plan, and further details of endpoint handling are not reported in the registry record. Enrollment and completion dates describe the study timeline; they do not establish when individual participants were assessed or how long each participant was followed.

The single-arm design can describe outcomes among enrolled participants, but does not provide a like-for-like control comparison. The registry record does not specify a protocol-based sample-size justification, interim analyses, or other design features beyond those noted above. Any conclusions should remain tied to the reported outcome definitions and the absence of a comparator.

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

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Two patients were excluded from analysis: one patient was not confirmed as ENKTL by central pathology, and the other patient was identified as having no measurable or evaluable disease at baseline by IRRC.

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) Assessed by IRRC, measured over From enrollment to end of follow-up, a median of 29 months.

The primary endpoint is objective response rate (ORR) assessed by the independent radiological review committee. ORR is a proportion: it represents the percentage of participants classified as having an objective response under the stated response criteria. The registry identifies the Criteria for Response Assessment of Lymphoma: Lugano 2014 Classification as the basis for the IRRC assessment.

Response rates are used to describe how often participants meet prespecified categories of tumor response during a study. An independent review committee can provide an assessment distinct from the treating investigators' evaluations. In general, independent assessment may help limit the influence of expectations or knowledge of clinical care on the interpretation of imaging; that is a general methodological rationale, not a finding about this trial's conduct. The registry record identifies IRRC assessment but does not report details of committee procedures, blinding, or adjudication.

The endpoint timeframe is described as running from enrollment to the end of follow-up, with a median follow-up of 29 months. This provides context for when responses were assessed, but it does not indicate that every participant had the same follow-up duration. The registry record does not state the number of responders, the estimated ORR, or the 95% confidence interval. It also does not report the time-specific response status of individual participants. Two patients were excluded from the primary endpoint analysis for the reasons specified in the endpoint description, so the analysis population was not all 80 enrolled participants.

A response rate is useful as a descriptive outcome, but in a single-arm study it cannot by itself establish how outcomes would have differed under another treatment or no treatment. Its interpretation also depends on the response criteria, assessment process, analysis population, and completeness of follow-up. The registry identifies some of these elements, while other implementation and analysis details are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Several statistical ideas help clarify what the registry record does and does not establish. The discussion below distinguishes general statistical principles from trial-specific information. Trial-specific statements are limited to the design and outcome details recorded for this study.

What does a response rate describe?

As a general concept, a response rate is a proportion of participants meeting a defined outcome. In this trial, the primary outcome is objective response rate assessed by the IRRC under the Lugano 2014 lymphoma response criteria. The registry describes the unit as a percentage of participants. A response rate summarizes how many participants meet the response definition in the analysis population; it does not describe how long a response lasts or how quickly it begins. Duration of response and time to response are listed separately as secondary outcomes.

For a proportion to be interpreted, the numerator, denominator, outcome definition, and analysis population must be clear. The registry defines the endpoint and reports that two patients were excluded from its analysis, with reasons. However, it does not give the number of responders or the analysis denominator. As a result, the response percentage cannot be recovered from the registry entry. Enrollment of 80 is not interchangeable with the endpoint analysis denominator because the record explicitly describes exclusions.

What does a confidence interval add?

In general, a confidence interval communicates uncertainty around an estimate. A point estimate alone can give an impression of certainty that the available data may not support. The interval's width depends on features such as the amount of information in the data and the method used to calculate it. For a response proportion, the number of participants contributing to the analysis and the number with a response are important to the uncertainty estimate.

The primary outcome lists a 95% confidence interval as its dispersion measure, but the interval limits and point estimate are not reported in the registry record. Therefore, no claim about the estimated response rate's precision can be made from this record. It is also not possible to check whether an interval was calculated using a particular method, because the method is not stated.

Why does the single-arm design matter?

In general, a comparison group helps estimate how outcomes differ between treatment strategies under the study conditions. Random allocation can help balance measured and unmeasured participant characteristics between groups, although it does not guarantee perfect balance in every trial. This study is recorded as single-group, with allocation not applicable and masking none. It does not include a concurrent control arm, so it cannot produce a randomized between-group treatment effect.

A single-arm outcome can still describe what happened among participants who received the intervention. However, interpretation is more limited because the outcome cannot be separated from other influences by a randomized comparison. Historical or external comparisons are not described in the registry record and should not be assumed. The design does not by itself determine whether the treatment is beneficial or ineffective.

What is the role of independent review?

In general, independent outcome review can offer an assessment separate from investigators involved in participant care. This may be relevant when outcomes require interpretation of radiological findings. The registry identifies the IRRC as the assessor of the primary endpoint and cites the Lugano 2014 classification. It does not describe whether committee members were blinded to treatment details, the procedures for resolving disagreements, or how assessments were handled when information was incomplete. Those details are not reported in the registry record.

The registry also lists outcomes assessed by investigators and outcomes assessed by the IRRC. Having more than one assessor or response endpoint can provide different descriptive perspectives, but those outcomes should not be treated as interchangeable. Their definitions and assessment processes may differ. The record does not provide the numerical results needed to compare those assessments.

How should the follow-up timeframe be read?

Time-to-event measures and response outcomes depend on observation over time. The primary endpoint timeframe is described as from enrollment to the end of follow-up, and the median follow-up is 29 months. In general, a median follow-up describes the midpoint of the observed follow-up distribution; it does not mean every participant was observed for that duration. The registry does not report individual follow-up durations or the response assessment schedule. It lists duration of response and time to response as secondary outcomes, but gives no results for them in this record.

What do the planning assumptions mean?

General sample-size calculations can be used to illustrate how many participants might be needed under specified assumptions. The assumptions recorded for planning illustrations are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect expressed as a stated fraction of the observed effect. Time-to-event illustrations are described as total event counts, while other illustrations are described as participants per arm.

These are planning illustrations, not observed power and not a report of this trial's actual design justification. The trial itself is single-arm, whereas the stated illustration assumptions use equal allocation between two arms. The registry record does not report the protocol's sample-size justification or the planning values used by the study team. The observed primary response effect is also not reported in this registry record, so the planning assumptions cannot be interpreted as evidence about the primary endpoint result.

More broadly, statistical interpretation requires alignment among the trial design, endpoint definition, analysis population, estimate, uncertainty, and analysis methods. The registry gives some design and endpoint details, including the planned response criteria and exclusions, but does not give the primary numerical result or analysis entries. Accordingly, the statistical methods used to produce a primary response estimate cannot be fully assessed from the registry record.

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("R/R ENKTL")
n      <- c(78)
events <- c(36)
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  R/R ENKTL  78  36  46.2  34.8  57.8  35.5  57.1  35.5  57.2

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 14 outcome measures, including the primary IRRC-assessed ORR and multiple secondary measures. The secondary outcomes include investigator-assessed ORR, complete response rate, partial response rate, duration of response, and time to response, with some measures also assessed by the IRRC. When a study evaluates multiple outcomes, multiplicity can arise if many statistical tests are conducted. In general, testing more outcomes can increase the chance of at least one apparently notable result arising by chance, unless the analysis strategy accounts for the family of comparisons.

The registry record does not report an adjustment for multiplicity, a hierarchy of testing, or a statistical analysis plan specifying how multiple endpoints were to be interpreted. It also lists no analysis entries. Therefore, it is not possible to determine from this record whether multiplicity adjustments were planned or used, or whether secondary outcomes were intended as confirmatory or descriptive.

Missing-data handling is also not reported in the registry record. The endpoint description does identify two exclusions from the primary analysis and gives the reasons: lack of central pathology confirmation of ENKTL for one patient, and no measurable or evaluable disease at baseline according to the IRRC for another. The record does not state how other unavailable assessments, withdrawals, incomplete follow-up, or unevaluable outcomes were handled. Nor does it provide the number of participants with missing endpoint assessments.

The primary analysis population is not named beyond the description of those exclusions. The registry does not state whether an intention-to-treat, modified intention-to-treat, per-protocol, or other population was used. The 80 enrolled participants therefore should not automatically be treated as the denominator for the primary response analysis. Without the analysis denominator and missing-data rules, the potential influence of exclusions and incomplete observations cannot be evaluated.

These omissions do not prove that the study lacked a statistical plan or that missing data were mishandled. They mean only that these methods are not reported in the registry record. Interpretation of the endpoint and secondary outcomes should account for that reporting gap.

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: pneumonia (6.2%), pyrexia (3.8%), death (2.5%), haemophagocytic lymphohistiocytosis (2.5%), septic shock (2.5%).

The safety record lists 80 participants at risk for serious adverse events, with 22 affected, and 80 at risk for other adverse events, with 76 affected. It also records 37 deaths. The listed serious-event terms and percentages are pneumonia, 6.2%; pyrexia, 3.8%; death, 2.5%; haemophagocytic lymphohistiocytosis, 2.5%; septic shock, 2.5%; and febrile neutropenia, 1.2%.

These figures describe reported events among participants in the CS1001 group. A serious adverse event is not the same category as an event of any severity, and the registry separates serious events from other adverse events. The listed terms are selected serious-event categories, not a complete accounting of every event reported in the study. The registry record does not provide further detail about event definitions or the period of risk used for each count.

Counts and percentages do not establish causality. In general, an adverse event occurring during a trial may have multiple possible causes, and determining whether it is related to treatment requires clinical and study-specific assessment. The registry figures alone do not show whether each event was attributed to CS1001. The listed term “Death” as a serious event is not a statement that every recorded death was caused by treatment. The record does not provide the causes of the 37 deaths or their relationship to the intervention.

Because this study has no comparison group, these safety figures cannot show how event rates compare with those under another treatment. They should be read as descriptive reports from the enrolled group, not as comparative or causal estimates.

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

The registry record establishes that CS1001 was studied in a completed, multicenter, single-arm phase II trial involving 80 participants with relapsed or refractory ENKTL. It describes the intervention as 1200 mg by intravenous infusion every 3 weeks, identifies IRRC-assessed ORR as the primary endpoint, and cites the Lugano 2014 response criteria. It also records a median follow-up of 29 months and gives reasons for two exclusions from the primary endpoint analysis.

What the record does not establish is the observed primary response rate. The numerical estimate and 95% confidence interval are not reported in the registry record. Without those values, the magnitude of the response outcome and the precision of its estimate cannot be assessed. The registry lists secondary response and time-to-event measures but does not report their numerical results in the record summarized for this analysis.

The single-arm design also limits what can be inferred. The trial describes outcomes among participants who received CS1001, but does not include a randomized control group. It therefore does not provide a direct estimate of how outcomes would differ under another treatment or under no treatment. Any comparison with outcomes outside the trial would require information and assumptions not reported in the registry entry.

The safety figures provide descriptive counts for serious adverse events, other adverse events, and deaths, as well as percentages for selected serious-event terms. They do not by themselves identify the cause of those events or show a comparative safety difference. The record does not report detailed attribution or the full context needed to interpret event timing and exposure.

These reporting limits should not be mistaken for evidence that no response occurred, that the intervention was ineffective, or that an adverse event was caused by treatment. The appropriate conclusion is narrower: the registry record contains design, endpoint, follow-up, and selected safety information, but does not contain the primary response estimate or enough statistical detail to evaluate the efficacy result's magnitude and precision.

14. Why This Trial Matters Statistically

This trial illustrates the distinction between describing outcomes in a treatment group and estimating a comparative treatment effect. A single-arm study can record response and safety outcomes among enrolled participants. However, without a concurrent comparator, it cannot provide a randomized estimate of the difference between treatment strategies. That distinction is central when interpreting clinical trial evidence, particularly when the primary endpoint is a response proportion.

The endpoint is clearly identified as IRRC-assessed ORR under the Lugano 2014 lymphoma response criteria. The registry also identifies investigator-assessed and other response outcomes, along with duration and time to response measures. This makes the outcome framework visible, but the numerical primary result and confidence interval are not reported in the registry record. The case therefore shows why endpoint definitions alone are insufficient for assessing the size and precision of an observed result.

Analysis populations and exclusions matter for a proportion. The record notes two exclusions from the primary analysis and explains why they occurred, but does not report the response denominator or how other missing assessments were handled. Statistical interpretation depends on knowing which participants contributed to the estimate and how unavailable data were treated. Enrollment alone does not answer those questions.

The study also demonstrates why registry outcome lists, analysis details, and numerical results should be distinguished. The record lists 14 outcomes but no analysis entries. That does not establish that analyses were not performed; it means their results and methods are not reported in this registry record. Similarly, the planning assumptions listed for illustrative sample-size calculations are not a substitute for a protocol-based design justification or observed trial findings.

For biostatisticians and clinical researchers, the key lesson is methodological: judge what a study can support from its design, endpoint definition, analysis population, estimate, uncertainty, and reporting completeness together. Here, design and endpoint details are available, but the primary estimate is not. The limitation is about what can be concluded from the registry record, not a judgment about the treatment's clinical value.

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.

Several limitations specific to this trial constrain interpretation of its registry record. The most important is that the numerical primary endpoint result is not reported. The IRRC-assessed ORR and its 95% confidence interval are identified, but neither the response estimate nor the interval limits are available. The number of responders and the primary analysis denominator are also not reported.

The record says two patients were excluded from the primary endpoint analysis and provides the reasons for those exclusions. It does not describe how any other missing, unevaluable, or incomplete assessments were handled. It also does not name the analysis population beyond the exclusions. These omissions prevent assessment of how the exclusions or other missing information may affect the response estimate.

The study is single-arm, with no concurrent comparison group. This limits the ability to attribute observed outcomes to the intervention or to make a direct comparison with another treatment. The registry also records masking as none. Details about independent review procedures, including whether the IRRC was blinded and how disagreements were handled, are not reported in the registry record.

The endpoint timeframe is described as enrollment to end of follow-up, with median follow-up of 29 months. Individual follow-up times, assessment schedules, and outcome-specific follow-up completeness are not given. Although multiple secondary outcomes are listed, their numerical results are not reported in the registry record. The statistical analysis plan, multiplicity handling, missing-data methods, and protocol-based sample-size justification are likewise not reported.

Safety interpretation is limited by the absence of detailed event timing, treatment attribution, and fuller event definitions in the registry information. The listed counts and terms remain descriptive and cannot establish causal relationships. These limitations identify reporting gaps and design constraints; they do not establish that unreported analyses or procedures were absent from the study itself.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study of CS1001 in Subjects With Relapsed or Refractory Extranodal Natural Killer/ T Cell Lymphoma(ENKTL) — NCT03595657.
• Registry record last updated 2024-06-18; status last verified 2024-05.
• Results first posted 2024-06-18.
• 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

NCT03595657 is a completed, multicenter, single-arm phase II trial sponsored by CStone Pharmaceuticals. It enrolled 80 participants with relapsed or refractory ENKTL and evaluated CS1001 monotherapy, described as 1200 mg by intravenous infusion every 3 weeks. The registry lists no comparator and records masking as none.

The primary endpoint was IRRC-assessed objective response rate using the Lugano 2014 lymphoma response criteria, from enrollment to end of follow-up; median follow-up was 29 months. Two patients were excluded from the primary analysis for the reasons recorded in the endpoint description. The primary response estimate and its 95% confidence interval are not reported in the registry record.

The registry lists 14 outcome measures and records selected serious adverse event counts and terms. It does not report analysis entries or the statistical plan, missing-data methods, or multiplicity procedures. The absence of a control arm limits comparative interpretation, and the safety figures are descriptive rather than causal.