Study of GSK3901961 In Previously Treated Advanced (Metastatic OR Unresectable) Synovial Sarcoma/ Myxoid/Round Cell Liposarcoma, and Previously Treated Metastatic Non-Small Cell Lung Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of GSK3901961 In Previously Treated Advanced (Metastatic OR Unresectable) Synovial Sarcoma/ Myxoid/Round Cell Liposarcoma, and Previously Treated Metastatic Non-Small Cell Lung Cancer (NCT06048705), a na phase 1 study. The primary outcome is Number of Participants With Dose Limiting Toxicities (DLTs).

2. Clinical Question

The study was designed to assess safety and tolerability and to determine a recommended Phase 2 dose. It concerns a selected population defined by cancer type, prior treatment, antigen status, and HLA status; the record does not describe a broad population of people with these cancers. The primary outcome focuses on treatment-related toxicities meeting prespecified severity, persistence, or recovery criteria, rather than on tumor response.

Several secondary outcomes are listed, including investigator-assessed overall response rate according to RECIST v1.1, duration of response, and measures of transgene expansion over time. Those outcomes could address different questions from the primary safety endpoint, but their results are not reported in the record information available for interpretation. The central clinical-trial question is therefore a safety and dose-finding question, not a comparative test of treatment benefit.

3. Trial Design

This is a Phase 1 study sponsored by Adaptimmune. The registry record gives a start date of 2021-03-09, a primary completion date of 2023-06-08, an enrollment of 7, and a status of terminated. The reason for termination is not reported in the registry record. The record describes the model as single group, allocation as not applicable, and masking as none.

Participants were intended to have previously treated advanced synovial sarcoma or myxoid/round cell liposarcoma, or previously treated metastatic non-small cell lung cancer, with NYESO1 and/or LAGE1a positivity. The study description additionally identifies HLA-A*02:01, HLA-A*02:05 and/or HLA-A*02:06 positivity. These eligibility details define a selected study population; they should not be assumed to apply to all patients with the listed cancers.

The intervention list includes GSK3901961, cyclophosphamide, and fludarabine. The arm description says eligible participants would undergo leukapheresis to manufacture engineered T cells and receive GSK3901961 after completing lymphodepleting chemotherapy. The full treatment schedule and the detailed roles and schedules for the chemotherapy drugs are not reported in the registry record.

The primary endpoint is the number of participants with dose-limiting toxicities up to 28 days. The record also lists overall response rate, duration of response, and pharmacokinetic or cell-expansion measures among the secondary outcomes. It lists eight outcome measures in total, but only five secondary outcomes are named in the information; the other outcome details are not reported in the registry record.

Because the record describes no random allocation and no masking, it does not establish a blinded, randomized comparison. The safety listing includes dose-range groups and a no-treatment category, but the record does not establish that these categories form comparable study arms. The protocol, detailed dose-selection rules, statistical analysis plan, and sample-size justification 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 is a sub study of the master protocol (NCT04526509). The study was terminated due to a change in GSK's R\&D priorities.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: DLT Evaluable included participants in the mITT population (all ITT participants (All participants who started leukapheresis procedure) who received any dose of New York esophageal antigen-1 \[NY ESO 1\] specific T cells) who are part of the dose confirmation phase that either had a DLT or have completed the DLT assessment period of 28 days since last T cell infusion.

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 Number of Participants With Dose Limiting Toxicities (DLTs), measured over Up to 28 days.

The primary endpoint counts participants with dose-limiting toxicities (DLTs) during a timeframe of up to 28 days. DLTs are severe or persistent adverse events defined using grading criteria, including NCI-CTCAE v5.0. The record describes criteria involving serious toxicity related to GSK3901961, certain severe toxicities that do not improve within specified periods, pneumonitis, cytokine release syndrome, neurotoxicity, and organ toxicity. These definitions aim to identify events of sufficient severity or duration to matter for dose evaluation.

Counting participants with at least one qualifying event provides a direct safety summary for a dose-finding study. It differs from counting every event: a participant with multiple qualifying events may still contribute as one participant to a participant-count endpoint. The endpoint is intended to inform whether a dose is tolerable enough to consider for further study, alongside the rest of the safety evidence. It does not, by itself, describe the full range of adverse events, the timing of every event, or the balance of risks and benefits.

The record specifies a DLT-evaluable population associated with the modified intention-to-treat population and a dose-confirmation phase, but its population description is incomplete in the registry information available. No observed DLT count or denominator for this primary endpoint is reported in the record information used. Consequently, the primary safety result and its precision cannot be assessed from these details.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What a participant-count endpoint means. The primary outcome is a count of participants who meet the DLT definition, rather than a measure of how many separate DLT events occurred. In general, this kind of binary outcome records whether each eligible participant did or did not meet the endpoint definition during the stated observation period. The resulting count is interpretable only alongside the population denominator and the rules for determining who is evaluable. Neither the primary count nor its denominator is reported in the registry information.

Why define DLTs in advance? A prespecified definition sets a consistent threshold for which toxicities count toward a dose-limiting endpoint. The record ties its definition to NCI-CTCAE v5.0 grading and includes severity, attribution, persistence, and recovery conditions. Such conditions matter because an event’s grade alone may not capture whether it resolves promptly or persists. The specified observation period is up to 28 days; the record does not provide the endpoint result within that period.

Counts and uncertainty. A count without a denominator cannot communicate an event frequency. Even when both are reported, estimates based on few participants are generally imprecise: a small change in the observed count can meaningfully alter the estimated frequency. A confidence interval is a statistical range expressing uncertainty around an estimate under a stated method. It is not a guarantee that the true value lies in the interval, and it does not address design limitations such as selection or lack of a comparator. No confidence interval for the primary outcome is reported in the registry information.

Why a comparator matters. A randomized comparison can help separate differences associated with assigned treatment from differences due to participant characteristics or other factors, when the design and analysis are appropriate. This record describes allocation as not applicable, no masking, and a single-group model. The safety listing also includes a no-treatment category, but there is no reported randomized assignment or basis for treating that category as a control group. Comparisons across the listed categories are therefore descriptive, not estimates of treatment effect.

What the planning assumptions mean. The record describes sample-size illustrations based on a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with an effect to detect expressed as a fraction of an observed effect. It specifies that time-to-event illustrations use a total event count based on the Schoenfeld method, while other illustrations are expressed as participants per arm. These are planning assumptions, not evidence that the trial had those design features, and not observed power. The record does not report the underlying calculated sample-size values in the information. The study's single-group description and non-randomized safety categories also mean those generic planning assumptions should not be read as a description of the actual allocation.

Interpreting a significance level. In general, a two-sided significance level sets a threshold for how unusual data would need to be under a specified null hypothesis to count as statistically significant. It does not measure the probability that a treatment works, and it does not establish the size or practical importance of an effect. A significance level is only one component of an analysis plan. The registry information does not report a statistical analysis plan for the primary endpoint, including a hypothesis test, interval method, or planned covariate adjustment.

Over-dispersion and model choice. In general, over-dispersion means that observed variability is greater than a statistical model assumes. It can make standard errors too small if a model fails to account for that extra variability. This issue may arise in some count or rate analyses, but the primary endpoint is described as a participant count, and no model or dispersion estimate is reported. It would be inappropriate to claim that an over-dispersion adjustment was used or was needed for this study.

Time-to-event and cell-expansion outcomes. Duration of response is listed as a secondary outcome, and the record also names maximum transgene expansion, time to that maximum, and area under the curve from zero to 28 days. In general, time-to-event methods can incorporate the time until an event and account for participants whose event time is not observed during follow-up, provided the analysis rules and censoring information are available. The record details no such analysis. The transgene measures describe different aspects of cell expansion over time, but their names alone do not provide observed values, variability, or clinical interpretation.

Planning is not observed power. A sample-size calculation can illustrate the number of participants per arm needed under selected assumptions for a planned comparison. It is not an estimate of the probability that the completed study detected an effect, and a calculation based on an observed effect can be particularly dependent on that effect estimate. The record explicitly characterizes these calculations as planning illustrations. They should not be treated as observed power or as evidence that the completed single-group study was adequately powered for a treatment comparison.

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("GSK3901961 1 × 10^9 - 8 × 10^9 Transduced Cells", "GSK3901961 0.1 × 10^9 - 0.8 × 10^9 Transduced Cell")
n      <- c(1, 4)
events <- c(1, 1)
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  GSK3901961 1 × 10^9 - 8 × 10^9 Transduced Cells  1  1  100.0  2.5  100.0  20.7  100.0  15.8  98.7
ARM  GSK3901961 0.1 × 10^9 - 0.8 × 10^9 Transduced Cell  4  1  25.0  0.6  80.6  4.6  69.9  5.3  71.6

11. Multiplicity, Missing Data and Analysis Populations

The registry information lists a primary safety outcome and several secondary outcomes, including response, duration of response, and cell-expansion measures. When a study assesses multiple outcomes, multiplicity can arise because repeated statistical tests increase the chance of at least one apparently notable result under a set of null hypotheses. In general, an analysis plan may address this through a testing hierarchy or an adjustment procedure, but the approach depends on the study’s objectives and is not inferable from the endpoint list alone. Multiplicity handling is not reported in the registry record.

Missing-data rules are also not reported in the registry record. For a participant-count safety endpoint, the treatment of participants without complete observation through the stated timeframe can affect the denominator and interpretation. The available information does not say how incomplete follow-up, missing toxicity assessments, or withdrawal was handled. It does not report imputation, sensitivity analyses, or the number of participants with missing endpoint data.

The population description says that the DLT-evaluable population included participants in the modified intention-to-treat population who were part of the dose-confirmation phase and either had a DLT or had a further condition that is cut off in the registry text. The record describes the modified intention-to-treat population as participants who started leukapheresis and received any dose of NYESO1-specific T cells. Because the DLT-evaluable definition is incomplete, its full eligibility rules cannot be reconstructed. The relationship between the total enrollment of 7 and the endpoint denominator is not reported.

No statistical analyses are listed in the information. The record does not specify an analysis population for the secondary outcomes, an approach to multiplicity, or a plan for missing outcome data. These omissions limit the ability to assess whether all outcomes were analyzed as intended or how uncertainty and incomplete observations were handled.

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: neutropenia (100.0%), non-cardiac chest pain (100.0%), pyrexia (100.0%), graft versus host disease in skin (100.0%), rash maculo-papular (100.0%).

The safety listing gives the number at risk and the number affected within each displayed category. In the higher listed GSK3901961 dose range, 1 participant was at risk and 1 had a serious adverse event; 1 had another adverse event, and 0 deaths were listed. In the lower listed dose range, 4 were at risk, 2 had serious adverse events, 4 had other adverse events, and 2 deaths were listed. In the No Treatment category, 2 were at risk, 2 had serious adverse events, 1 had another adverse event, and 1 death was listed.

The registry also names serious adverse event terms: neutropenia, non-cardiac chest pain, pyrexia, graft versus host disease in skin, and rash maculo-papular are each shown with 100.0%; atrial fibrillation is shown with 50.0%. The denominator and calculation basis for these term-specific percentages are not reported in the registry record. They should not be read automatically as percentages of all enrolled participants.

The categories are descriptive and not a randomized comparison. The record does not report event-level timing, severity details beyond the serious-event classification, causality, or circumstances of deaths. Counts alone cannot establish that a listed event was caused by the study treatment.

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

The record establishes that the study was intended to assess safety, tolerability, and a recommended Phase 2 dose for GSK3901961 in selected participants with previously treated, antigen-positive cancers. It defines a DLT outcome and a timeframe of up to 28 days. Those design details clarify the question the study set out to address, but they are not a substitute for the observed primary outcome.

The primary DLT count and its evaluable denominator are not reported in the registry information. As a result, the principal safety endpoint cannot be quantified or assessed for precision from these details. The adverse-event listing supplies group-specific counts, but the categories are not described as randomized, comparable treatment groups. They therefore cannot establish a causal contrast between GSK3901961 and no treatment or between dose ranges.

Secondary outcomes are named, including tumor response and duration of response, but no corresponding result values are reported. Their inclusion in the record does not demonstrate that a response occurred, how long it lasted, or how the cell-expansion measures behaved. No conclusion about efficacy follows from the outcome names or from the safety counts.

The status is terminated, and the reason is not reported in the registry record. That status does not by itself explain whether the study stopped for safety, operational, enrollment, or other reasons. Nor does it establish whether all planned follow-up and analyses were completed. The enrollment of 7 indicates the scale of the registered study, but it does not specify how many participants were eligible for the primary endpoint.

In general, a Phase 1 safety and dose-finding study can generate evidence to guide further investigation, but the strength of any inference depends on the observed outcomes, the evaluable population, follow-up, and analysis. Those elements are not fully. The defensible reading is limited to the stated design and the descriptive adverse-event counts, with the primary result and several important analysis details unavailable in the registry record.

14. Why This Trial Matters Statistically

This study is statistically important as an example of how the objective of a trial shapes the meaning of its outcomes. The stated primary purpose is safety, tolerability, and dose selection. Its primary endpoint is a participant count based on defined DLT criteria, not a direct comparative estimate of tumor response. Readers should distinguish that safety question from the secondary outcomes of response, duration, and cell expansion.

The design information also matters. The record describes a single-group model, no masking, and allocation as not applicable, while the safety listing contains dose-range categories and a no-treatment category. Those labels do not establish randomized allocation. Without a like-for-like comparison, simple differences between category counts cannot be interpreted as treatment effects. This is a central statistical distinction between describing outcomes and estimating what would have happened under a different treatment condition.

The small registered enrollment of 7 makes the endpoint denominator and uncertainty especially consequential. Counts from a limited number of participants can be difficult to generalize, and a missing primary count prevents even a descriptive assessment of the main endpoint. The record’s DLT-evaluable population definition is also incomplete in the available text, reinforcing the need to distinguish total enrollment from the population actually analyzed.

Finally, the record illustrates why planning calculations, protocol details, and completed results are separate kinds of information. Generic sample-size assumptions do not show that the study had a randomized comparison or that it achieved a particular level of power. The absence of a reported analysis plan, missing-data methods, and multiplicity approach leaves uncertainty about how outcomes were intended to be analyzed. Clear reporting of these elements would help readers judge what the study can support and what remains unresolved.

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 constrain interpretation of this trial record. The primary outcome is defined, but its observed DLT count and evaluable denominator are not reported in the registry information. Without both, the main safety endpoint cannot be summarized as an observed event frequency, and its precision cannot be judged.

The total registered enrollment is 7, but the record does not establish how many participants contributed to each outcome. The DLT-evaluable population description is incomplete, and the relationship between the enrolled population, the modified intention-to-treat population, and the endpoint population cannot be fully determined from the available text.

The record describes a single-group model with no masking and allocation as not applicable. It does not establish randomized assignment to the dose-range categories or to the no-treatment category. Accordingly, the safety counts do not support a causal comparison. The listed deaths have no reported circumstances, timing, cause, or treatment relationship in the information.

The study is marked terminated, but the reason is not reported in the registry record. The record also does not report the statistical analysis plan, missing-data rules, multiplicity approach, or whether the planned outcome analyses were completed. Although secondary endpoints are named, their results are not reported. These gaps limit assessment of both the completeness of reporting and the conclusions that can be drawn from the study.

The population is selected by cancer type and antigen and HLA status, and includes more than one cancer category. Findings from such a study, even if fully reported, would need to be interpreted within that defined population. The available record does not provide enough outcome information to assess the magnitude or precision of any treatment effect.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Study of GSK3901961 In Previously Treated Advanced (Metastatic OR Unresectable) Synovial Sarcoma/ Myxoid/Round Cell Liposarcoma, and Previously Treated Metastatic Non-Small Cell Lung Cancer — NCT06048705.
• Registry record last updated 2024-11-13; status last verified 2024-11.
• Results first posted 2024-05-10.
• 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

NCT06048705 is a terminated Phase 1 study sponsored by Adaptimmune. The record gives a start date of 2021-03-09, a primary completion date of 2023-06-08, and enrollment of 7. It describes a single-group model, allocation as not applicable, and no masking.

The study concerns GSK3901961, engineered autologous T cells targeting NYESO1 and LAGE1a, in previously treated participants with selected antigen-positive advanced synovial sarcoma or myxoid/round cell liposarcoma, or metastatic non-small cell lung cancer. Cyclophosphamide and fludarabine are also listed as interventions.

The primary endpoint is the number of participants with DLTs up to 28 days. The registry information does not report the primary result, its denominator, or the reason for termination. Safety-group counts are listed, but no randomized comparison is established. Secondary outcomes are named without results in the information.