Study of GSK3965193 in Healthy Participants and Participants Living With Chronic Hepatitis B Infection: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of GSK3965193 in Healthy Participants and Participants Living With Chronic Hepatitis B Infection (NCT05330455), a randomized phase 1, phase 2 study. The primary outcome is Part 1: Number of Participants With Treatment Emergent Adverse Events (TEAEs), Serious Treatment Emergent Adverse Events (STEAEs), and Treatment Withdrawals Due to TEAEs.

2. Clinical Question

The study addressed several related questions rather than a single efficacy comparison. In healthy participants, Part 1 evaluated single-dose safety and pharmacokinetics, Part 2A evaluated repeat dosing, and Part 2B evaluated pharmacokinetics under fed and fasted conditions. In participants with chronic hepatitis B, Part 3 assessed GSK3965193 monotherapy, including change in hepatitis B virus surface antigen, with an option for subsequent open-label bepirovirsen. Part 4 was planned to study GSK3965193 with bepirovirsen and the potential for sustained virologic response.

The registered primary outcome is a safety measure in Part 1, not a measure of hepatitis B antigen reduction or virologic response. The record lists 47 outcome measures, including safety, pharmacokinetic, and pharmacodynamic outcomes. The registry record does not report one common treatment-effect question spanning all parts; the populations, treatment schedules, and aims differ across the study.

3. Trial Design

The registry describes NCT05330455 as a randomized, triple-masked, placebo-controlled crossover study with Phase 1 and Phase 2 designations. The study title describes four parts, with Parts 1, 2A, 3, and 4 randomized and double-blind, and includes the additional Part 2B food-condition assessment. The record lists enrollment of 74 and a terminated status. It lists a start date of 2022-04-14 and a primary completion date of 2025-05-19. The reason for termination is not reported in the registry record.

Parts 1 and 2 enrolled healthy participants. Part 1 used treatment periods in which participants received placebo and single doses of GSK3965193 in different sequences. Part 2A studied repeat doses of placebo or GSK3965193 twice daily for 14 days. Part 2B used a crossover sequence to evaluate Dose 10 under fasted and fed conditions. These within-part treatment sequences mean that the treatment groups should not automatically be interpreted as independent parallel arms.

Parts 3 and 4 concerned participants living with chronic hepatitis B infection on stable nucleos(t)ide analog therapy. Part 3 compared placebo with GSK3965193 Dose 9 for 28 days; participants who completed monotherapy could opt for subsequent open-label bepirovirsen. Part 4 planned 28 days of either placebo plus bepirovirsen or GSK3965193 plus bepirovirsen, followed by continued bepirovirsen injections. The adverse-event record lists no participants at risk in either Part 4 group, so the planned combination comparison has no recorded safety exposure in these results.

The registry describes the study's broad design and treatment sequences, but detailed allocation procedures and masking arrangements by part are not reported in the registry record. The study's status is terminated, but the reason and the circumstances around termination are not reported.

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: The study enrolled 74 participants (20 in Part 1, 23 in Part 2A, 13 in Part 2B, and 18 in Part 3). Part 2B was planned to have 3 treatment periods (TPs). However, TP3 was not conducted following safety and tolerability findings in TP2. The study was terminated after completion of Part 3 as a strategic decision. Hence, no participants were enrolled in Part 4. Dose 1 (D1) was the lowest dose. Doses in ascending order are as follows: D1, D2, D3, D4, D9, D5, D10, D6, D7, and D8.

Only the first 6 of 18 groups are shown.

Part 1, Cohort 1: Treatment Period 1

Part 1, Cohort 1: Treatment Period 2

Part 1, Cohort 1: Treatment Period 3

Part 1, Cohort 1: Treatment Period 4

Analysis population for the primary outcome, as defined in the registry: Safety population included all participants who received at least one dose of study treatment.

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 after completion of Part 3 as a strategic decision. Hence, no participants were enrolled in Part 4.

Only the first 6 of 19 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Part 1: Number of Participants With Treatment Emergent Adverse Events (TEAEs), Serious Treatment Emergent Adverse Events (STEAEs), and Treatment Withdrawals Due to TEAEs, measured over From the start of study intervention (Day 1) up to 12 weeks.

The registered primary outcome counts participants with treatment-emergent adverse events (TEAEs), serious treatment-emergent adverse events (STEAEs), and treatment withdrawals due to TEAEs in Part 1. The timeframe is from the start of study intervention on Day 1 up to 12 weeks. The registry defines an adverse event as an untoward medical occurrence temporally associated with study intervention, whether or not it is considered related to that intervention. An event is treatment-emergent when its onset is on or after the intervention start date through Day 5 after administration; an event starting after Day 5 is not considered treatment-emergent under the stated definition.

The record defines serious events by outcomes such as death, life-threatening illness, inpatient hospitalization or its prolongation, persistent or significant disability or incapacity, congenital anomaly or birth defect, abnormal pregnancy outcomes, or other situations judged serious by the investigator. Recording both TEAEs and STEAEs helps describe the observed safety experience while distinguishing serious events from the broader adverse-event category. Counting withdrawals due to TEAEs captures whether an adverse event led to stopping treatment.

The registry defines the safety population as all participants who received at least one dose of study treatment. The primary outcome is a participant count rather than a continuous measure such as a laboratory value. Counts need to be interpreted alongside the number at risk, exposure schedule, and crossover structure; they do not by themselves establish whether an event was caused by treatment.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a participant count describe?

The primary outcome is expressed as the number of participants with events, with the number at risk recorded for each treatment group. Such counts describe how many people met the event definition during the specified observation period. They are not the same as the total number of event episodes: one participant may experience more than one event, while the participant count records whether the person had an event. The posted endpoint also separates serious events and withdrawals due to TEAEs as outcome categories.

General: A count is easiest to interpret when read with its denominator, the event definition, and the period in which events were collected. When group sizes are small, each participant represents a substantial part of the group, so counts can change noticeably with one person's outcome. This is one reason descriptive counts alone should not be treated as precise estimates of comparative safety.

Why does the crossover design matter?

Part 1 assigned participants to treatment sequences that included placebo and single doses of GSK3965193 across treatment periods. Part 2B also used sequences for fed and fasted administration. In a crossover design, the same participant can contribute observations under more than one treatment condition. These observations are linked within a participant and are not independent in the way observations from separate parallel groups would be.

General: An analysis of crossover data may account for the within-person structure and for period or sequence effects, depending on the design and analysis plan. The registry record does not report the statistical analysis plan or how such features were handled. The posted participant counts should therefore be read as descriptive group summaries, not as an adjustment for crossover structure.

What does the record say about comparative analyses?

The registry record lists no analyses. The primary outcome contains several categories and the record does not reduce it to one comparison. No test, model, adjusted estimate, confidence interval, or comparative p-value is reported in the registry record. The adverse-event counts can be described, but they do not supply a formal estimate of the difference between treatment conditions.

General: A confidence interval expresses uncertainty around an estimated effect and indicates a range of values compatible with the data under the method used. A hypothesis test assesses compatibility with a specified null hypothesis under its assumptions. Neither a confidence interval nor a test can be reconstructed responsibly from these posted counts alone without choosing a comparison, accounting for the design, and specifying an analysis method. No such method is reported.

How should planning illustrations be understood?

The stated sample-size assumptions describe a planning illustration based on an observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The assumptions are not the trial's own design justification. The record does not report the trial's sample-size justification. Sample-size planning figures, where displayed, are illustrations under their stated assumptions; they are not observed power and do not describe the probability that the trial detected an effect.

General: Planning depends on a defined outcome, a target effect, variability or event frequency, allocation, and a chosen statistical method. A planning illustration can change when those assumptions change. In a multi-part study with crossover sequences and distinct populations, one generic planning comparison would not necessarily represent every part. The registry record does not report a part-specific calculation or the assumptions used by the investigators to choose enrollment.

Why does over-dispersion matter for some outcomes?

The registered primary outcome is a count of participants with an event, not a repeated event-count measure modeled as a rate. Over-dispersion is therefore not a reported analysis issue for this endpoint in the registry record.

General: In count models, over-dispersion means that observed variability is greater than the model expects. If ignored, estimated uncertainty can be too small and inferences can be misleading. It matters when selecting or checking a model for count data; it should not be invoked as if it were an analysis performed in this trial.

Overall, the registry supports straightforward descriptive reading of the recorded counts and treatment exposure groups. It does not document inferential procedures, adjustment for repeated treatment periods, or uncertainty estimates. Those absences constrain what can be concluded statistically from the posted registry results.

10. Multiplicity, Missing Data and Analysis Populations

The study included multiple parts and 47 outcome measures spanning safety, pharmacokinetics, and pharmacodynamics. The registry record does not report a multiplicity strategy, such as a hierarchy of outcomes or an adjustment for testing multiple outcomes. It also does not report a statistical analysis plan that would establish which outcomes or comparisons were designated for formal inference. Because the registered primary outcome contains several categories and is not reduced to one comparison, no single confirmatory treatment contrast is available in the posted record.

Missing-data methods are not reported in the registry record. The record does not explain how participants with incomplete follow-up, missing measurements, or early discontinuation were handled in analyses. The reason for termination is also not reported, and the available information does not establish whether termination affected the completeness of any particular outcome.

The primary endpoint identifies the safety population as participants who received at least one dose. Other analysis populations are not reported in the registry record. The adverse-event entries give group-specific numbers at risk, but they do not describe participant flow by part or explain whether every participant enrolled contributed to each outcome. Part 4 has zero participants recorded at risk in both planned groups; consequently, the registry offers no observed safety comparison for that planned part.

These omissions do not imply that investigators used no procedures outside the registry entry. They mean those procedures cannot be described from the registry record. Results should therefore be read as the reported descriptive information, without assuming a multiplicity adjustment, missing-data approach, or analysis population beyond the stated safety definition.

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 0% reporting threshold.

Most frequent non-serious adverse event terms

The serious adverse-event figures are participant counts in the serious-event category. The registry records zero serious events and zero deaths for every group with participants listed at risk. This is a description of the recorded experience during the study, not proof that serious events are impossible or that a treatment has no serious-event risk. Small group sizes make the absence of a recorded event especially limited as evidence about less common outcomes.

The other adverse-event counts are a separate category in the posted results. They should not be treated as counts of serious events. The primary outcome also includes treatment withdrawals due to TEAEs, but group-specific withdrawal figures are not reported in these entries. Specific serious-event terms are not listed.

Part 4 requires a different reading: both planned combination groups have zero participants at risk and zero events recorded. These zeros do not indicate that participants received the planned combinations without events; no participants are recorded at risk in those groups. For the other parts, comparisons across counts should account for differing group sizes, treatment periods, and crossover sequences. The counts are descriptive and do not provide an adjusted treatment comparison.

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

The posted results describe adverse-event counts across a complex early-phase study of GSK3965193. In the groups with participants at risk, no serious adverse events or deaths are recorded. Other adverse events are recorded in several groups, with the number of participants affected varying by treatment group. The results also document the distinction between single-dose exposure, repeat dosing, fed and fasted conditions, and treatment in participants with chronic hepatitis B.

These observations do not establish that GSK3965193 is safe or unsafe, and they do not show whether it caused any particular event. The registry definition includes events whether or not considered related to the intervention. No causal assessment or formal comparative analysis is reported. The available counts also do not answer the study's pharmacokinetic or hepatitis B antigen questions, for which corresponding numerical results are not reported in the registry record described by this record.

The study's registered aims extend beyond the safety endpoint. Part 3 was intended to assess hepatitis B surface antigen changes, and Part 4 was planned to assess combination therapy and the potential for sustained virologic response. Those aims should not be confused with the Part 1 primary safety outcome. The Part 4 group entries record no participants at risk, so they do not provide an observed combination-treatment result.

The registry status is terminated, but the reason is not reported. The record does not establish why the study ended or whether termination was related to safety, efficacy, enrollment, or another consideration. The most defensible reading is therefore descriptive and bounded: these are the adverse-event counts and exposure entries recorded by group, without a reported inferential analysis or a basis for claims beyond those observations.

13. Why This Trial Matters Statistically

This study matters statistically because it combines several distinct questions within one multi-part record. Healthy-participant dose escalation, repeat dosing, a food-condition crossover, hepatitis B monotherapy, and planned combination treatment do not share a single population or treatment contrast. A reader must keep each part's design and purpose in view rather than treating all groups as arms in one conventional parallel trial.

The crossover sequences are especially important for interpreting the safety counts. Participants in Part 1 received treatment conditions in sequence, and Part 2B compared fed and fasted periods. The group counts therefore do not necessarily represent independent sets of people. A formal comparison would need to reflect the treatment sequence and within-participant structure, but the registry record does not report such an analysis.

The registry also illustrates the difference between a protocol aim and an observed result. Part 4 planned a combination comparison, yet no participants are recorded at risk in either group. A planned treatment question is not an observed comparison when there is no recorded exposure. Likewise, a count of zero serious events is a useful description of the recorded experience but is not proof of zero risk.

Finally, a multi-outcome study raises questions about multiplicity, missing data, and analysis populations. Those methods are not reported in the registry record. Their absence limits inferential interpretation, while the detailed group counts still allow a transparent descriptive account of what is recorded. Keeping that distinction clear is central to reading early-phase, multi-part clinical research responsibly.

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 registry record is terminated and does not state the reason. It does not report whether termination affected planned enrollment, follow-up, or completion of specific outcomes. Although enrollment is listed as 74, the record does not provide a full participant flow by part in the information.

The primary outcome is a multi-category safety measure, but no single comparison or inferential analysis is reported. Group-level counts do not include confidence intervals, adjusted estimates, or p-values. The registry does not report a statistical analysis plan, a multiplicity procedure, or methods for missing data. It also does not report group-specific withdrawal counts for the primary endpoint in the adverse-event entries.

Interpretation is further limited by variation in treatment schedules and population. Parts 1 and 2 enrolled healthy participants, while Parts 3 and 4 involved participants living with chronic hepatitis B on stable therapy. Part 1 and Part 2B include crossover sequences, and Part 3 included an option for subsequent open-label bepirovirsen. The treatment groups are therefore not interchangeable for a single across-study comparison.

The Part 4 entries list zero participants at risk in both groups, leaving no recorded exposure for the planned combination comparison. In the other parts, several groups have small numbers at risk. The registry does not list specific serious-event terms, and the absence of recorded serious events cannot establish that such events would not occur in a larger or different population. These limitations frame the results as descriptive registry evidence rather than a complete evaluation of comparative safety or efficacy.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Study of GSK3965193 in Healthy Participants and Participants Living With Chronic Hepatitis B Infection — NCT05330455.
• Registry record last updated 2026-10-14; status last verified 2026-09.
• Results first posted 2026-06-16.
• 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

NCT05330455 is a randomized, triple-masked, placebo-controlled, crossover Phase 1/2 study of GSK3965193 in healthy participants and participants living with chronic hepatitis B. It enrolled 74 participants and is listed as terminated. The primary outcome concerns Part 1 counts of TEAEs, STEAEs, and treatment withdrawals due to TEAEs, with a timeframe from Day 1 up to 12 weeks.

The registry records no serious adverse events or deaths in groups with participants at risk and lists other adverse-event counts across Parts 1 through 3. The two planned Part 4 combination groups have no participants recorded at risk. No inferential analyses are listed, and the reason for termination, multiplicity methods, missing-data methods, and statistical analysis plan are not reported in the registry record.