A Safety Study of GSK3039294 in Healthy Volunteers and Patients With Systemic Amyloidosis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Safety Study of GSK3039294 in Healthy Volunteers and Patients With Systemic Amyloidosis (NCT02603172), a non-randomized phase 1 study. The primary outcome is Part A:Number of Participants With Adverse Events (AEs) and Serious Adverse Events (SAEs).

2. Clinical Question

The registry describes GSK3039294 as a candidate oral alternative to parenteral CPHPC for plasma serum amyloid P component depletion before anti-SAP monoclonal antibody use. That rationale supplies context for studying the compound, but the listed primary outcome does not measure SAP depletion, antibody effects or patient benefit. No efficacy outcome is identified as the primary question. The results should therefore be understood as safety and pharmacokinetic observations from an early-phase study, not as a comparison of clinical outcomes.

3. Trial Design

This was a phase 1, open-label, non-randomised, dose-escalation study with three parts. The registry lists 23 participants enrolled and describes the study as terminated. The reason for termination is not reported in the registry record.

Part A evaluated single doses in healthy volunteers. Two cohorts were described, with participants potentially taking part in two dosing periods. Cohort 1 received Dose level 1 on Day 1 and Dose level 2 on Day 15; Cohort 2 received Dose level 3 on Day 1 and Dose level 4 on Day 15. The registry describes a wash-out period between these dosing periods. The posted safety groups include GSK3039294 200 mg, 600 mg and 1200 mg.

Part B evaluated repeat dosing in healthy volunteers for a total of 21 days. The plan described dose escalation over the study, with a further cohort to be conducted if required. The posted safety results identify a Cohort 3 group receiving GSK3039294 600 mg QD; the other listed Part B cohorts have zero participants at risk in the safety results.

Part C was intended to evaluate repeat dosing for 21 days in participants with systemic amyloidosis, at a dose level selected from Part B. No participants at risk are listed for Part C in the posted safety results. The registry record does not report a concurrent control group, allocation concealment or masking; it explicitly identifies the study as non-randomised and open-label. Consequently, the design primarily supports descriptive assessment within dose cohorts. It does not provide a randomised estimate comparing GSK3039294 with a control treatment.

The planned population and dosing differed across parts, so cohort-level counts have to be interpreted in their specific context. The registry record does not report why enrollment ended at 23 or provide a detailed account of which planned activities were completed.

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 was 3 parts study, participants received GSK3039294: single ascending dose in Part A; repeated dose in Part B; repeat dose in systemic amyloidosis in Part C. Participants in Part A also participated in Part B of the study.

Only the first 6 of 8 groups are shown.

Part A (Approximately 16 Weeks)

Reasons for not completing this period:

Part B (Approximately 7 Weeks)

Part C (Approximately 13 Weeks)

Analysis population for the primary outcome, as defined in the registry: Safety population.

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 study was terminated due to safety reasons during conduct of Part B (end of Cohort 3) whereas Cohort 4A and 4B were not recruited in Part B and Part C was not initiated.

Only the first 6 of 9 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Part A:Number of Participants With Adverse Events (AEs) and Serious Adverse Events (SAEs), measured over Up to Day 14.

The primary outcome counted participants with adverse events (AEs) and serious adverse events (SAEs) in Part A, up to Day 14. The registry defines an AE as an untoward medical occurrence temporally associated with use of a medicinal product, whether or not considered related to it. An SAE is an untoward medical occurrence meeting seriousness criteria such as death, a life-threatening event, inpatient hospitalisation or its prolongation, persistent disability, a congenital anomaly or another important medical event judged by the investigator.

Counting affected participants is a direct descriptive way to report how many people in the safety population experienced an event. The safety population is defined in the registry as participants who received at least one dose of study medication. For an early-phase dose-escalation study, this kind of outcome helps describe observed safety events during the specified follow-up. It does not, by itself, establish whether an event was caused by the drug, and it does not quantify the severity of every event beyond the separate serious-event classification.

The primary outcome is limited to Part A and its stated timeframe. It is not a measure of pharmacokinetics, SAP depletion or clinical benefit in systemic amyloidosis. The registry lists pharmacokinetic measures among secondary outcomes, including concentration-time measures for GSK3039294 and GSK2315698, but numerical results for those measures are not included in the registry record for this analysis. A prespecified method for handling repeated events or attributing events to treatment is not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The posted results are counts of participants with events, not continuous measurements or time-to-event estimates. For the listed safety groups, the record gives the number at risk and the number affected. These quantities allow a reader to see how many participants were represented in each group and how many had an event. They are descriptive summaries: they report what was observed in those groups, without automatically estimating a causal effect.

General: For a binary outcome such as whether a participant had an event, a proportion is often used to describe the share of participants with the event. The count and the denominator should be read together, because the same event count can convey different amounts of information in groups of different sizes. The posted results include counts and numbers at risk, but no confidence intervals are listed in the registry record. A confidence interval, in general, describes the uncertainty around an estimate and helps distinguish a relatively precise estimate from one compatible with a wide range of values. It does not remove limitations in the study design.

The study was non-randomised and open-label. As a result, cohort membership was not assigned by randomisation, and participants and investigators were not masked to treatment. The registry record does not report a control arm. General statistical principle: randomisation can help balance known and unknown characteristics between comparison groups, while masking can reduce some forms of influence on care or assessment. Without those protections, apparent differences between cohorts may reflect differences in who entered each cohort, when they entered, the dose schedules or other factors, as well as chance. This is why the registry's non-randomised dose groups should not be treated as a controlled comparison.

The design used dose escalation and included different study parts. Part A involved single doses in healthy volunteers; Part B involved repeat dosing in healthy volunteers; Part C was intended to involve repeat dosing in people with systemic amyloidosis. These parts address related but distinct questions. A count from a single-dose cohort is not automatically comparable with a count from a repeat-dose cohort, because exposure duration and population differ. The registry record does not report a model that adjusts for these differences, nor does it report an inferential analysis of dose trends.

General: In studies with event counts, a statistical model may be used when the goal is to compare rates or estimate associations while accounting for design features. Such a model requires a clear outcome definition, an analysis population and assumptions about the data. When observations vary more than a simple model expects, over-dispersion can make standard errors too small and uncertainty appear narrower than it is. This is a general consideration, not a claim that over-dispersion occurred in this study. No statistical model, test or adjustment strategy is reported in the registry record.

The registry's primary outcome is the number of participants with AEs and SAEs in Part A up to Day 14. The separate classifications matter: an SAE is defined by seriousness criteria, while an AE need not meet those criteria and need not be considered related to the drug. A count of people affected does not describe every event's timing, duration, recurrence or relationship to dosing. The registry record does not report a statistical method for repeated events or a detailed event-level analysis in the registry record used for this interpretation.

The sample-size assumptions associated with planning illustrations describe a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation and an effect expressed as a stated fraction of the observed effect. These are planning assumptions, not evidence of the study's original design or sample-size justification. Equal allocation between two arms is also unlike the non-randomised, multi-cohort structure described in the registry. Any sample-size values based on those assumptions are illustrations for planning, not observed power and not a retrospective measure of whether the study was adequately powered. The registry record does not report its own sample-size justification.

General: Power is a design concept concerning the chance that a planned analysis will detect an effect of a specified size under stated assumptions. It is not a measure of the probability that a result is true, and it should not be calculated from the observed result as if it were a finding. A planning calculation based on an observed effect can be especially unstable when the original groups are small or not comparable. It should be read as a hypothetical design exercise, not as a property of the completed trial.

The registry lists 51 outcome measures and no statistical analyses. It does not report p-values, confidence intervals, covariate adjustment, a statistical analysis plan or interim-analysis methods. Their absence from the registry record means the posted counts cannot be expanded into a claim of statistical significance or a formal dose-response relationship. The appropriate interpretation remains descriptive: event counts are reported for the represented safety groups, with limitations arising from the design and the amount of information available.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 51 outcome measures, including the Part A safety outcome and pharmacokinetic outcomes. When many outcomes, dose levels or comparisons are examined, the chance of finding an apparently notable result by chance can increase. General statistical practice may address this issue through a prespecified hierarchy, adjustment or careful emphasis on descriptive estimates. The registry record does not report a multiplicity strategy, a hierarchy among outcomes or a statistical analysis plan. It also lists no statistical analyses, so there are no posted inferential results to assess for adjustment.

Missing-data methods are not reported in the registry record. The registry entries do not describe how participants who withdrew, missed assessments or had incomplete event follow-up were handled. This matters because event counts are interpretable only in relation to who was observed and for how long. The registry gives numbers at risk for the listed safety groups, but the information does not establish whether those counts reflect complete follow-up for every planned assessment.

The primary endpoint identifies the safety population as all participants who received at least one dose of study medication. That definition clarifies who is included in the stated safety analysis population, but the registry record does not provide further detail on analysis populations for pharmacokinetic outcomes or other endpoints. Nor does it report a method for handling participants who received a dose but lacked later observations.

Because the study was non-randomised and open-label, and because the record describes different parts and dosing schedules, analysis choices would be important for any formal comparison. Such methods are not reported in the registry record. The posted counts can therefore be read as descriptive safety information for participants represented in each listed group, but not as a complete account of missing-data handling or multiplicity control.

11. 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: supraventricular tachycardia (12.5%).

The serious adverse event counts are zero for each of the three listed Part A dose groups: GSK3039294 200 mg, 600 mg and 1200 mg. The corresponding numbers at risk are 8, 16 and 8. In Part B Cohort 3, the 600 mg QD group, one participant had a serious adverse event among eight at risk. No participants at risk are listed for the other Part B cohorts or Part C. The registry reports no deaths in the listed groups.

The serious-event term reported as supraventricular tachycardia is accompanied by 12.5%. The registry entries do not establish which listed group that term belongs to, whether it was considered related to the study drug or what its clinical course was. An SAE classification reflects seriousness criteria, not proof that the drug caused the event. Likewise, zero serious events in a cohort means none were recorded among its listed participants; it does not establish that serious events are impossible or exclude events outside the recorded follow-up.

These figures are descriptive and come from a non-randomised study. They should be interpreted alongside the cohort sizes, the different dosing schedules and the stated Part A follow-up through Day 14. The registry record does not report confidence intervals or a controlled safety comparison.

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

The posted data show that adverse events, including serious adverse events, were recorded in some of the listed safety groups. Part A had no serious adverse events reported in the three dose groups, while Part B Cohort 3 had one participant with a serious adverse event. Other adverse events were also recorded, with counts varying across the listed groups. These observations describe what the registry reports for the participants at risk; they do not establish a causal relationship between GSK3039294 and any event.

The results do not show that one dose is safer than another. The study did not use randomised allocation or a control group, and the cohorts differed by part and dosing schedule. The group counts cannot isolate the effect of dose from differences in participants, study conditions or duration of exposure. No inferential analysis or uncertainty intervals are reported, so the precision of any between-group contrast cannot be assessed from the registry record.

The study also does not establish clinical benefit for systemic amyloidosis. Its stated goals were safety, tolerability and pharmacokinetics, and the primary outcome concerned Part A safety events. The rationale mentions plasma SAP depletion before anti-SAP monoclonal antibody use, but no result about SAP depletion or patient outcomes is included in the registry record for this analysis. The pharmacokinetic measures listed as secondary outcomes do not, by themselves, demonstrate a clinical effect.

Nor do the listed counts describe every aspect of safety. They do not reveal event timing, duration, recurrence, severity beyond the serious classification, or investigator attribution. The record is marked terminated, but the reason for termination and the extent to which every planned study part was completed are not reported in the registry record. The defensible conclusion is limited: the record documents selected event counts among the participants represented in the safety results, with substantial limits on comparisons and precision.

13. Why This Trial Matters Statistically

This study illustrates why early-phase safety findings should be kept distinct from evidence of treatment benefit. The stated focus was safety, tolerability and pharmacokinetics of GSK3039294, an oral drug candidate being studied in healthy volunteers and participants with systemic amyloidosis. The primary outcome counted adverse events and serious adverse events in Part A. Those outcomes are relevant to characterising observed events during exposure, but they do not answer whether a treatment changes disease outcomes.

Its structure also highlights the limits of dose-cohort comparisons. Dose escalation can support sequential investigation of different doses, yet a non-randomised design does not ensure that cohorts are comparable. Open-label conduct may also influence event reporting or assessment. The posted group counts are useful descriptions, but they should not be treated as if they came from a randomised trial with a control group.

The small, unevenly sized listed groups make the distinction between an observed count and a precise estimate especially important. A zero count is a record of no observed events in the represented participants, not proof that the event cannot occur. A count in another cohort does not establish a dose effect. Without posted uncertainty intervals or inferential analyses, readers should avoid claims that exceed the descriptive evidence.

Finally, the study was intended to connect an oral candidate with a treatment-development rationale involving SAP depletion, while its primary outcome remained safety-focused. The record makes clear why outcome selection matters: biological rationale, pharmacokinetic measurement and clinical benefit are different questions. The registry record does not report results that answer all of them. This trial is statistically instructive because it calls for careful separation of design intent, safety observations and conclusions that the available data can support.

14. 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 main limitation is the non-randomised, open-label design without a control group. Differences in event counts across dose groups cannot be interpreted as controlled estimates of dose effects. Cohort membership may be related to participant characteristics, study part or dosing schedule, and the registry record does not report an adjustment strategy for those factors.

The listed safety groups are limited in size, and no confidence intervals or formal statistical analyses are reported. As a result, the precision of the event estimates and any differences between groups cannot be evaluated from the posted results. Zero serious adverse events in the Part A groups should not be read as proof that such events cannot occur.

The endpoint's stated timeframe is up to Day 14 for Part A, while the study also included planned repeat-dose parts. A direct safety comparison across parts would be difficult because their populations and exposure schedules differed. The registry lists zero participants at risk for the further Part B cohorts and Part C, so their safety experience is not represented in the posted results. The reason for these entries and the extent of completion are not reported in the registry record.

Important details are also absent. The registry record does not report missing-data methods, event attribution procedures, multiplicity handling, the original sample-size justification or the reason the study was terminated. The registry entries also do not include detailed event-level information or numerical pharmacokinetic results. These gaps limit assessment of follow-up completeness, statistical uncertainty, dose-response patterns and the relationship between the study's rationale and clinical outcomes.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Safety Study of GSK3039294 in Healthy Volunteers and Patients With Systemic Amyloidosis — NCT02603172.
• Registry record last updated 2019-09-23; status last verified 2019-08.
• Results first posted 2019-09-23.
• 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.

18. Record Summary

NCT02603172 is a phase 1 study of GSK3039294 in healthy volunteers and participants with systemic amyloidosis. It was open-label, non-randomised and designed with three parts: single-dose assessment in healthy volunteers, repeat dosing in healthy volunteers and repeat dosing intended for participants with systemic amyloidosis. The registry reports enrollment of 23 and status as terminated; the reason for termination is not reported in the registry record.

The primary outcome counted participants with adverse events and serious adverse events in Part A up to Day 14. Posted safety results include Part A dose groups and a Part B Cohort 3 group; the other listed Part B cohorts and Part C have zero participants at risk. The registry lists 51 outcome measures and no statistical analyses. The reported counts therefore support a descriptive account of events in the represented groups, not a controlled comparison or a conclusion about clinical benefit.