Efficacy and Safety Study of GPX-150 to Treat Soft Tissue Sarcoma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy and Safety Study of GPX-150 to Treat Soft Tissue Sarcoma (NCT02267083), a na phase 2 study. The primary outcome is Number of Subjects Progression-free at 12 Months Per RECIST 1.1.

2. Clinical Question

3. Trial Design

This was a phase 2, single-group study of GPX-150 in patients with soft tissue sarcoma. The registry record lists Gem Pharmaceuticals as the sponsor and enrollment of 22. The study start date is 2015-01-07, the primary completion date is 2015-11-11, and the status is completed. These registry entries describe the study record, not its clinical outcome.

GPX-150 for injection was administered intravenously at 265 mg/m2 every 21 days for 16 cycles or until death, disease progression, unacceptable toxicity, or subject withdrawal. The brief summary describes treatment every 3 weeks. There was no randomized allocation, masking was none, and no concurrent comparison group is described. Consequently, the study is not a like-for-like treatment comparison: its outcome cannot be read as a contrast against a concurrently observed control group.

The primary endpoint was the number of subjects progression-free at 12 months from the beginning of study treatment, per RECIST 1.1. The record describes the endpoint as obtained by inversion of the Kaplan-Meier progression-free survival curve at 12 months. Secondary outcomes include the number of subjects progression-free at six months per RECIST 1.1 and the number experiencing adverse events. The registry record lists four outcome measures, but does not give numerical results for the primary endpoint in the information recorded for this analysis.

The record notes that a historical-data comparison described in the protocol was not performed because enrollment was less than the planned sample size of 30 subjects. No further detail about the planned historical comparison, its assumptions, or its statistical test is reported in the registry record. Without that analysis, the reported study design supports description of outcomes in the treated group, not a controlled estimate of a treatment effect.

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

Analysis population for the primary outcome, as defined in the registry: Consists of all subjects in the safety population (22 subjects) and who had at least one on-study tumor assessment using RECIST 1.1 (19 subjects). One subject withdrew consist before first on-study tumor assessment and two subjects died before first on-study tumor assessment.

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.

Population: The Safety Population consists of all subjects who received at least 1 dose of GPX-150 and the Intent-to-Treat (ITT) Population consists of all subjects in the Safety Population who had at least one tumor assessment using RECIST 1.1 or who died prior to any tumor assessment.

6. Primary Endpoint

The registry lists the primary outcome as Number of Subjects Progression-free at 12 Months Per RECIST 1.1, measured over 12 months from the beginning of study treatment.

The primary endpoint is the number of subjects progression-free at 12 months from the beginning of study treatment. The registry record defines progression using RECIST 1.1: at least a 20% increase in the sum of diameters of target lesions, taking as reference the smallest sum recorded on study, including the baseline sum if it is the smallest.

The endpoint is described as being obtained by inversion of the Kaplan-Meier curve for progression-free survival at 12 months. In general, a Kaplan-Meier curve estimates the proportion of a group remaining free of an event over time while accommodating participants whose event status is not observed for the entire follow-up period. Inverting that curve at a specified time expresses the estimated progression-free proportion as a participant count. This differs conceptually from simply counting only people whose status is fully observed at that time point.

Progression-free survival is used to describe time until progression or death, and the 12-month landmark gives the endpoint a specific time horizon. The record identifies the population for this endpoint as subjects in the safety population who had at least one on-study tumor assessment. It reports 22 subjects in the safety population and 19 with at least one assessment. The numerical 12-month endpoint result and its uncertainty are not reported in the registry record, so the magnitude and precision of the estimate cannot be assessed from the recorded result.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The statistical approach most directly described in the registry record is the use of a Kaplan-Meier curve for progression-free survival, with the primary endpoint defined by inversion of that curve at 12 months. The record also defines progression by RECIST 1.1 and specifies a population with at least one on-study tumor assessment. It does not report the numerical curve estimate, a confidence interval, a hypothesis test, or the historical comparison described in the protocol.

What does a Kaplan-Meier curve represent?

As a general statistical explanation, a Kaplan-Meier curve estimates the probability of remaining free of an event over time. For a progression-free survival outcome, the event definition includes progression or death. Participants may have different lengths of observation. When a participant's event status is not observed through the end of follow-up, that information is commonly treated as censored at the last time their status is known, rather than as proof that an event did or did not occur later. This method uses the observed follow-up while recognizing that not everyone has complete follow-up.

The curve is an estimate, not a guarantee about each participant's later status. Its precision generally depends on the amount and timing of follow-up and on the number of observed events. A confidence interval, when reported, communicates uncertainty around an estimate under the statistical method used. No confidence interval or other precision measure for this endpoint is reported in the registry record, so the uncertainty around the 12-month estimate cannot be described.

What does inversion at a fixed time mean?

The registry record says that the primary endpoint is obtained by inversion of the progression-free survival curve at 12 months. In general, a curve-based estimate of the proportion remaining event-free can be expressed as a participant count for the endpoint population. This is an estimated count based on the survival curve, rather than necessarily a direct tally of participants whose complete 12-month status is observed. The registry record does not explain the computational details of the inversion or provide the resulting value.

How is progression defined?

The record uses RECIST 1.1 and defines progression as at least a 20% increase in the sum of the diameters of target lesions, with the smallest sum on study as the reference. The baseline sum is included if it is the smallest sum on study. This rule supplies an operational definition for classifying progression; it does not itself report how many participants met that definition or when events occurred. The record does not provide lesion-level measurements or individual event times.

Why does the analysis population matter?

The primary endpoint population is described as subjects in the safety population who had at least one on-study tumor assessment using RECIST 1.1. The record gives 22 subjects in the safety population and 19 with at least one such assessment. It also states that one subject withdrew before the first on-study tumor assessment and two died before that assessment. In general, analysis populations define which participants contribute to a particular estimate. Different population definitions can affect how an outcome is interpreted, so the stated population should accompany any reported endpoint estimate.

What can a single-group design estimate?

In general, a single-group study can describe outcomes observed or estimated among participants receiving the study intervention. Without a concurrent comparison group, it cannot directly estimate a between-group contrast under the same study conditions. Historical comparisons can sometimes be planned, but their interpretation depends on the selected historical data and the comparability of populations, outcome definitions, follow-up, and other factors. The registry record says the protocol’s historical comparison was not performed because enrollment was less than the planned sample size of 30 subjects. Its methods and results are therefore not available for interpretation.

What do the sample-size assumptions mean?

In general, a sample-size calculation is a planning illustration based on stated assumptions about the effect to detect, significance level, allocation, and statistical approximation. It is not a measure of the probability that a completed study has found a real effect, and it is not observed power. The record describes planning assumptions using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect set as a stated fraction of the observed effect. Such assumptions do not convert this single-group study into a randomized comparison. No sample-size justification or additional statistical plan is reported in 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("GPX-150")
n      <- c(19)
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  GPX-150  19  3  15.8  3.4  39.6  5.5  37.6  5.7  37.9

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists four outcome measures and identifies progression-free status at 12 months as the primary endpoint. Secondary outcomes include progression-free status at six months per RECIST 1.1 and the number of subjects experiencing adverse events. The record reports zero analyses. No statistical plan for handling multiple outcomes, repeated testing, or multiplicity is reported in the registry record. Without such a plan and analysis results, it is not possible to determine whether any testing adjustment was planned or applied.

The registry record describes the primary efficacy population as subjects in the safety population who had at least one on-study tumor assessment using RECIST 1.1. It reports 22 subjects in the safety population and 19 with at least one assessment. It also notes that one subject withdrew before the first on-study tumor assessment and two died before that assessment. These are relevant distinctions between the safety population and the endpoint population, but the record does not provide a fuller account of participant disposition or missingness.

No method for handling missing tumor assessments, incomplete follow-up, or missing endpoint data is reported in the registry record. The Kaplan-Meier description indicates a time-to-event approach, but the record does not explain censoring rules, follow-up details, or sensitivity analyses. No imputation method is reported. The available information therefore does not establish how any missing data beyond the stated population definition affected the endpoint estimate.

The registry record also does not report whether analyses were adjusted, whether interim analyses were conducted, or what statistical testing procedure was intended. Those details should not be inferred from the study phase or endpoint type. The absence of a reported analysis result prevents assessment of how the stated population and any missing information were handled in the final estimate.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: vomiting (9.1%), febrile neutropenia (4.5%), anaemia (4.5%), myocardial infarction (4.5%), abdominal pain (4.5%).

The safety record lists a GPX-150 group and records 22 subjects in the serious adverse event at-risk count and 22 in the affected count. It also records 2 deaths. The registry record does not provide enough detail to interpret these figures as a comparison with another group or to attribute an event to GPX-150. The study has no concurrent control arm.

The serious adverse event terms with the highest listed percentages are vomiting, 9.1%; febrile neutropenia, 4.5%; anaemia, 4.5%; myocardial infarction, 4.5%; abdominal pain, 4.5%; and diarrhoea, 4.5%. These are reported as serious adverse event terms, not as a complete account of every adverse event. The record does not provide event timing, severity details beyond the serious-event classification, or participant-level context for these terms.

Percentages describe the reported event information, but they do not establish whether an event was caused by treatment. In general, safety interpretation also depends on how events were defined, collected, and followed, as well as the population under observation. No comparative safety estimate is available from this single-group study. The recorded figures should therefore be read as descriptive registry safety information, not as evidence of a relative risk or causal effect.

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

The registry record establishes the design and endpoint definition of a phase 2 single-group study of GPX-150 in soft tissue sarcoma. It describes a primary endpoint based on progression-free status at 12 months, with progression defined using RECIST 1.1 and the endpoint derived from a Kaplan-Meier curve. It also identifies the primary endpoint population and provides some information about participants who did not have an on-study tumor assessment.

What the record does not establish is the numerical primary efficacy result. The 12-month progression-free count, its precision, and an inferential test are not reported. The protocol’s planned statistical comparison with historical sarcoma data was not performed because enrollment was less than the planned sample size of 30 subjects. Therefore, no result from that comparison is available to interpret, and the single-group design offers no concurrent control contrast.

These limitations do not mean that the study had no observations; they mean that the available registry information does not support a numerical conclusion about the primary efficacy endpoint. A study being recorded as completed is not equivalent to a reported efficacy result. Likewise, an endpoint definition explains what was intended to be measured but does not show the observed value.

The serious adverse event terms and percentages provide descriptive safety information for the GPX-150 group. They do not establish that treatment caused an event, and the lack of a concurrent control group prevents a like-for-like estimate of comparative event frequency. The listed deaths are also not enough on their own to establish cause or treatment relationship.

In general, interpreting a trial requires both an estimate and an understanding of its uncertainty, population, design, and analysis. Here, the record gives important design and endpoint details, but does not report the primary numerical result or its uncertainty. The appropriate statistical reading is therefore limited to describing what the study planned and how outcomes were defined, while recognizing that the posted results do not permit a conclusion about the magnitude or precision of the primary outcome.

14. Why This Trial Matters Statistically

This trial is statistically informative as an example of the distinction between an endpoint definition and an endpoint result. The registry record specifies a 12-month progression-free endpoint, its RECIST 1.1 progression rule, and its derivation from a Kaplan-Meier curve. Those details clarify what the study intended to measure. They do not, by themselves, reveal the estimated outcome or how precise it was.

The design also matters for interpretation. A single-group study can document outcomes among participants receiving an intervention, but it does not provide a concurrent treatment contrast. The protocol described a historical comparison, yet the record says that comparison was not performed because enrollment was less than the planned sample size of 30 subjects. The result is an important distinction between a planned analysis and an analysis actually reported.

The record also distinguishes the safety population from the population with at least one on-study tumor assessment. Such population definitions are essential context for an endpoint estimate. The stated withdrawals and deaths before the first assessment further show why participant flow and missing assessment information matter when interpreting efficacy outcomes. The registry record does not report a fuller missing-data approach, so the effect of missing information cannot be evaluated.

Finally, the safety information illustrates the limits of descriptive event reporting. Serious adverse event terms and percentages can characterize reported events, but cannot establish causation or a comparative safety profile in the absence of a concurrent control group. Overall, the study record is useful for understanding design choices, outcome definitions, and the boundary between descriptive information and comparative inference, even though the primary numerical outcome is not reported.

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 direct limitation for interpreting efficacy is that the numerical primary endpoint result is not reported in the registry record. No confidence interval, other uncertainty measure, or statistical test result for the primary endpoint is reported either. Thus, neither the size nor precision of the 12-month estimate can be assessed from the recorded information.

The study is single-group, with no concurrent comparator. The historical-data comparison described in the protocol was not performed because enrollment was less than the planned sample size of 30 subjects. The registry record does not describe the historical data or the planned comparison in sufficient detail to evaluate its assumptions, and it supplies no result from that analysis.

The primary endpoint population is defined by membership in the safety population and having at least one on-study tumor assessment. The record reports 22 subjects in the safety population and 19 with such an assessment, and notes one withdrawal and two deaths before the first assessment. It does not report further details about follow-up, missing tumor assessments, censoring, or sensitivity analyses. The consequences of these features for the endpoint estimate therefore cannot be judged.

The record lists four outcome measures but does not report an approach to multiplicity or a fuller statistical analysis plan. It also reports zero analyses. The serious adverse event terms and percentages are descriptive and do not establish treatment causality or comparative safety. In general, these limitations mean that the registry information supports a description of the study and its recorded safety terms, but not a complete assessment of efficacy or comparative benefit and risk.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Efficacy and Safety Study of GPX-150 to Treat Soft Tissue Sarcoma — NCT02267083.
• Registry record last updated 2018-01-10; status last verified 2017-12.
• Results first posted 2018-01-10.
• 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 NCT02267083, a completed phase 2 single-group study of intravenous GPX-150 in patients with soft tissue sarcoma, sponsored by Gem Pharmaceuticals. Enrollment was 22. Treatment was described as 265 mg/m2 every 21 days for 16 cycles or until specified stopping conditions.

The primary endpoint was the number of subjects progression-free at 12 months from the beginning of treatment, based on inversion of a Kaplan-Meier progression-free survival curve and progression defined by RECIST 1.1. The endpoint population included subjects in the safety population with at least one on-study tumor assessment; the record reports 22 in the safety population and 19 with an assessment. One subject withdrew and two died before the first assessment.

The record says the planned historical comparison was not performed because enrollment was less than the planned sample size of 30 subjects. It does not report the numerical primary endpoint result or its precision. Serious adverse event terms and percentages are recorded, but there is no concurrent control group for comparative interpretation.