This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for A Study of Switching From Entecavir to Tenofovir Disoproxil Fumarate in Subjects With Chronic Hepatitis B (NCT03258710), a na phase 4 study. The primary outcome is Percentage of Participants Who Achieved 0.25 Logarithm to the Base 10 (Log10) Hepatitis B Surface Antigen (HBsAg) Reduction From the Baseline at Week 48.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT03258710 |
| Phase | Phase 4 |
| Condition | Hepatitis B, Chronic |
| Allocation | Not applicable (single group) |
| Design model | Single Group |
| Masking | None (open label) |
| Enrollment | 75 |
| Arms in the primary outcome | n/a |
| Lead sponsor | GlaxoSmithKline |
| Results posted | Yes |
2. Clinical Question
The study concerns people with chronic hepatitis B who were HBeAg-positive, had undetectable HBV-DNA, and were receiving ongoing entecavir treatment. Its stated clinical question is whether HBsAg levels change after switching from entecavir to tenofovir disoproxil fumarate, and what safety events are reported during follow-up. The primary endpoint focuses on a specified HBsAg reduction from baseline at Week 48.
Population
Japanese participants with chronic hepatitis B who were HBeAg-positive and HBV-DNA undetectable while receiving entecavir. The primary endpoint population is defined in the registry as the full analysis set.
Intervention
Participants switched from ongoing entecavir treatment to tenofovir disoproxil fumarate. The registry describes one tablet of TDF 300 mg taken orally once daily for 96 weeks, starting on the day entecavir was discontinued, without overlapping treatment.
Comparator
There was no concurrent comparison group. This was a single-group study, so the registry record does not provide a like-for-like control arm receiving continued entecavir or another treatment.
Primary question
What percentage of participants in the full analysis set met the defined threshold for HBsAg reduction from baseline at Week 48, and how precisely was that percentage estimated? The observed percentage and its 95% confidence interval are not reported in the registry record.
This design can describe what was observed among participants after the switch, but it cannot by itself separate changes associated with the switch from changes that might have occurred over time or for other reasons. The absence of a concurrent comparator is central to interpreting both the efficacy endpoint and safety observations.
3. Trial Design
This was a Phase 4, prospective, multicentre, one-arm study in people with chronic hepatitis B. The official title identifies the target population as Japanese participants who were HBeAg-positive and HBV-DNA undetectable while taking entecavir. The study is recorded as completed. Its start date is 2017-10-02, and its primary completion date is 2018-12-28.
The study intervention was a switch from ongoing entecavir treatment to tenofovir disoproxil fumarate. Participants were to begin TDF on the day ETV was stopped, without a period of overlapping treatment, and receive one tablet of TDF 300 mg orally once daily for 96 weeks. The record describes no masking, and allocation is recorded as not applicable because this was not a randomized allocation between study arms.
The registry lists enrollment as 75. Separately, the brief summary says approximately 80 subjects would be screened to achieve 65 evaluable subjects. Those are distinct statements about enrollment and the screening/evaluability objective; the record does not explain their relationship or report the number ultimately included in the primary analysis.
There was no concurrent control group. Consequently, the study is descriptive of the enrolled group during and after the treatment switch. It is not a randomized comparison of TDF with continued ETV, and it cannot establish whether an observed change was caused by switching rather than by time, participant characteristics, or other influences. The registry record does not report a detailed protocol for how participants were selected, how adherence was assessed, or how deviations from the treatment schedule were handled.
The record describes virological and safety objectives, including HBsAg changes, antigen loss or seroconversion, HBeAg outcomes, and adverse events. These secondary outcome categories do not replace the primary Week 48 endpoint, and results for them are not included in the registry record for this account.
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: A total of 75 participants were enrolled in the study.
Overall Study
| Stage | Tenofovir Disoproxil Fumarate |
|---|---|
| STARTED | 75 |
| COMPLETED | 72 |
| NOT COMPLETED | 3 |
Reasons for not completing this period:
| Reason | Tenofovir Disoproxil Fumarate |
|---|---|
| Adverse Event | 1 |
| Protocol-defined stopping criteria | 1 |
| Physician Decision | 1 |
Analysis population for the primary outcome, as defined in the registry: FAS 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.
| Characteristic | Unit | Tenofovir Disoproxil Fumarate |
|---|---|---|
| Participants analysed | 75 | |
| Age, Continuous | Years | 48.4 (SD 9.35) |
| Sex: Female, Male: Female | Participants | 20 |
| Sex: Female, Male: Male | Participants | 55 |
| Race/Ethnicity, Customized: Asian-East Asian Heritage | Participants | 4 |
| Race/Ethnicity, Customized: Asian-Japanese Heritage | Participants | 71 |
6. Primary Endpoint
The registry lists the primary outcome as Percentage of Participants Who Achieved 0.25 Logarithm to the Base 10 (Log10) Hepatitis B Surface Antigen (HBsAg) Reduction From the Baseline at Week 48, measured over Baseline (Day 1, Pre-dose) and at Week 48.
| Element | Registry entry |
|---|---|
| Result type | Number |
| Measure of spread | 95% Confidence Interval |
| Unit | Percentage of Participants |
| Endpoint classification | Binary Endpoint |
The primary endpoint is the percentage of participants who achieved a 0.25 log10 HBsAg reduction from baseline at Week 48. HBsAg is hepatitis B surface antigen. The registry defines the measurement as the log10 HBsAg observation at follow-up minus the log10 baseline value; a responder value is at or below -0.25 log10, while a value above -0.25 log10 is classified as a non-responder. Baseline is the latest non-missing assessment before dosing, including an unscheduled visit when applicable.
Expressing the endpoint as a percentage makes it a binary outcome: each participant is classified as meeting or not meeting the responder threshold, and the group-level measure is the proportion classified as responders. This can provide a straightforward summary of how often a defined biomarker change occurred in the analysis population. The threshold also makes the responder rule explicit, although this classification does not show the full range of individual HBsAg changes.
The endpoint is assessed at Week 48, with baseline measurements used as the reference. The registry lists a 95% confidence interval as the measure of dispersion. However, the observed responder percentage and confidence interval are not reported in the registry record. Therefore, the endpoint definition can be described, but its observed magnitude and precision cannot be assessed from this record.
This is a biomarker endpoint, not a direct measure of symptoms or clinical events. The record does not report a clinical interpretation linking the specified HBsAg threshold to an individual outcome. The result should therefore be read as a defined virological response measure, not as proof of a broader clinical effect.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The central statistical limitation is that the primary endpoint result is absent from the registry information. Although the registry specifies the Week 48 responder definition and says that a 95% confidence interval is the dispersion measure, it does not report the observed percentage or the interval limits. Without those values, the magnitude of the response in the analysis population and the uncertainty around it cannot be evaluated. A p-value or other primary-endpoint analysis is also not reported in the registry record.
The study has a single group and no concurrent comparator. Its endpoint, if reported, would estimate the fraction meeting the responder rule among participants analyzed after the switch. It would not directly estimate a treatment difference between switching and continuing entecavir. In a one-arm study, a before-and-after contrast may describe change from baseline, but changes over time can reflect influences beyond the intervention. The design therefore supports a descriptive account, not a controlled causal comparison.
The endpoint uses a threshold to divide participants into responders and non-responders. This is easy to communicate, but it compresses potentially different degrees of biomarker change into two categories. Participants whose values fall on opposite sides of the threshold are assigned different classifications even if their measurements are close. Conversely, participants classified together may have differing amounts of change. The registry also lists change-from-baseline HBsAg values among secondary outcomes, but no numerical results for those outcomes are reported in the record.
The population definition matters. The primary outcome is specified for the full analysis set, which excludes participants meeting either of the registry's stated exclusion criteria: receiving no study treatment, or having no efficacy data from at least 15 days after treatment started. This definition is not identical to simply counting everyone enrolled. The registry record does not report the number included in the analysis or how missing Week 48 measurements were handled. Thus, even with an endpoint estimate, interpreting its denominator would require the actual analysis population and missing-data information.
The safety information reports 2 participants with serious adverse events and 36 with non-serious adverse events among 75 at risk in the TDF group, with 0 deaths recorded. The listed leading serious-event terms are osteoarthritis and appendicitis, each reported as 1.3%. These are descriptive observations in a single treated group. They do not establish whether events were caused by treatment, and the record does not provide a comparative safety estimate.
Overall, this record defines a clinically relevant biomarker question and documents the single-group treatment schedule, but it does not provide the primary efficacy estimate needed to judge the observed response or its precision. The design and reporting gaps should remain separate: the lack of a control group limits causal interpretation, while the missing endpoint estimate prevents assessment even of the response frequency within the treated group.
9. Statistical Methods Explained
The statistical methods that can be described are limited to the outcome definition, the stated confidence interval, and the full analysis set criteria. The registry record does not report a detailed statistical analysis plan. The explanations below distinguish the general meaning of these methods from trial-specific information that is absent.
What does a percentage endpoint represent?
For a binary outcome, a participant is assigned to one of two categories, such as responder or non-responder. The percentage summarizes the share of analyzed participants in the responder category. It is a group-level description, not an individual prediction. The primary endpoint uses a prespecified-looking threshold in its definition: the log10 HBsAg follow-up value minus the log10 baseline value must be at or below -0.25 log10 to count as a response. The record does not say whether this cutoff was selected through a protocol rationale or how it was justified.
A percentage depends on both the number classified as responders and the number included in the analysis. The registry identifies the full analysis set as the primary endpoint population but does not report the number analyzed or the observed responder percentage. The registry record therefore does not permit a numerical interpretation of the response rate.
What does a confidence interval communicate?
In general, a confidence interval expresses the statistical uncertainty around an estimate. For a percentage, its width depends in part on the amount of information available and on the observed response pattern. A narrower interval indicates greater precision under the method used; a wider interval indicates less precision. An interval does not eliminate uncertainty, and it is not a range that guarantees the true value lies within its limits.
The record names a 95% confidence interval for the primary endpoint, but its limits are not reported. The confidence-interval method itself is also not reported in the registry record. Without the estimate, interval, and calculation method, precision cannot be assessed, and no conclusion about compatibility with particular values can be drawn.
Why does the single-group design matter?
In a comparative trial, a treatment effect can be expressed as a contrast between groups. This study has one group, so there is no concurrent treatment contrast. An observed proportion after switching can describe participants in the study, but it cannot answer how those same participants would have fared under a different treatment strategy. That unobserved alternative is the counterfactual needed for a causal comparison.
The record reports allocation as not applicable and masking as none. Neither a randomized allocation procedure nor a blinded comparison is described. The study is therefore not a like-for-like control comparison. Any interpretation should distinguish the outcome observed in the cohort from an effect attributable to the treatment switch.
What is the role of the full analysis set?
The registry defines the full analysis set as enrolled participants excluding those who received no dose of study treatment and those with no efficacy data—HBsAg, HBV-DNA, HBcrAg, HBeAg, HBsAb, HBeAb, or ALT—from at least 15 days after treatment began. This definition identifies eligibility for the primary analysis at a broad level. It does not state how missing measurements at Week 48 were classified, whether all available measurements were used, or how participants who discontinued treatment were handled.
Analysis-population definitions affect what an estimate describes. An analysis restricted by post-treatment data availability may represent participants with evaluable follow-up rather than every person enrolled. The exact size of the full analysis set and the number with missing endpoint measurements are not reported in the registry record, so the extent and possible influence of exclusions or missingness cannot be assessed.
Why can over-dispersion matter in general?
For binary outcomes, a simple model may assume that participants have a common underlying response probability and that their outcomes are independent. In practice, response probabilities may vary across participants, and outcomes may cluster by site or other shared circumstances. This extra variability beyond a simple model's assumptions is often called over-dispersion. If it is ignored, uncertainty can be understated. This is general statistical context; the registry record does not report whether over-dispersion was assessed or whether a model was used for the primary endpoint.
What can be said about sample-size planning?
The registry's planning description refers to a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the target effect expressed as a stated fraction of the observed effect. It also distinguishes planning for participants per arm from time-to-event calculations based on total events. These are descriptions of a planning illustration, not the design of this one-arm trial. The numerical planning estimates and the observed effect on which such an illustration depends are not reported in the registry record.
The sample-size description should not be read as observed power. Planning calculations estimate what sample size might be needed under stated assumptions; they do not establish the probability that this completed study detected an effect. The record does not report the trial's sample-size justification, the assumed primary-endpoint response rate, or a power calculation for the one-arm design. No conclusion about adequacy of the enrolled sample can be drawn from the enrollment figure alone.
What is not available about the analysis?
The registry record does not report the statistical test or estimation procedure for the primary endpoint, covariate adjustment, handling of protocol deviations, sensitivity analyses, subgroup analyses, or interim analyses. It lists 36 outcome measures and no analysis entries, but that does not establish what analyses may have been conducted outside the registry record. The observed primary estimate and its confidence interval are also missing. As a result, the available description supports explanation of the endpoint and general statistical concepts, but not a reproduction or verification of the reported analysis.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists 36 outcome measures, including secondary outcomes involving HBsAg reduction, HBsAg loss and seroconversion, HBeAg loss and seroconversion, changes in HBsAg and HBcrAg log values, and adverse events. Several of these outcomes are assessed at more than one follow-up time. When many outcomes or time points are examined, the chance of finding apparently notable results by chance can increase if each is tested separately. This general issue is called multiplicity.
The registry record does not report a multiplicity strategy, such as prioritizing a testing sequence, adjusting significance thresholds, or treating secondary analyses as descriptive. It also does not report a detailed statistical analysis plan. The record's listing of 36 outcome measures describes the breadth of outcomes, but does not tell readers which were formally tested or how any multiple testing was handled. No analysis entries are reported in the registry information.
Missing data are also important for a percentage endpoint. The primary population is defined as the full analysis set, with exclusions for no treatment received or no efficacy data from at least 15 days after treatment start. The registry record does not say how a missing Week 48 HBsAg value was handled, whether missing outcomes were imputed, or whether the primary analysis used another rule. It does not report the number of participants with missing endpoint data.
The distinction between enrollment and analysis is therefore consequential. Enrollment is listed as 75, while the brief summary describes approximately 80 subjects to be screened to achieve 65 evaluable subjects. The number included in the primary endpoint analysis is not reported. Without that denominator and the missing-data approach, the extent to which the estimate might depend on exclusions or unavailable measurements cannot be judged.
The registry record does not report whether interim analyses were planned or performed. No assumptions about interim monitoring, multiplicity adjustment, or missing-data handling should be inferred from the outcome list alone.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Tenofovir Disoproxil Fumarate | 2 | 75 | 2.7% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Tenofovir Disoproxil Fumarate | 0 of 75 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Tenofovir Disoproxil Fumarate | 36 of 75 |
Most frequent serious adverse event terms
| Event term | Organ system | Tenofovir Disoproxil Fumarate |
|---|---|---|
| Osteoarthritis | Musculoskeletal and connective tissue disorders | 1 of 75 |
| Appendicitis | Infections and infestations | 1 of 75 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Tenofovir Disoproxil Fumarate |
|---|---|---|
| Nasopharyngitis | Infections and infestations | 19 of 75 |
| Beta 2 microglobulin urine increased | Investigations | 17 of 75 |
| Headache | Nervous system disorders | 4 of 75 |
| Dental caries | Gastrointestinal disorders | 4 of 75 |
| Back pain | Musculoskeletal and connective tissue disorders | 4 of 75 |
| Cough | Respiratory, thoracic and mediastinal disorders | 4 of 75 |
Serious events with the highest arm-level rate: osteoarthritis (1.3%), appendicitis (1.3%).
The registry reports safety observations for the TDF group, with 75 participants at risk. It records 2 participants with serious adverse events, 36 with non-serious adverse events, and 0 deaths. These are descriptive counts in the treated group. They do not show how event rates compare with continued entecavir or with an untreated group, because no concurrent comparator was included.
The leading serious adverse-event terms are osteoarthritis and appendicitis, each reported as 1.3%. These terms identify events recorded in the registry; the listing does not establish that either event was caused by TDF. The record does not report detailed event timing, duration, outcome, relatedness assessments, or the complete event-level context in the information.
Serious and non-serious adverse events are different reporting categories. A serious adverse event is classified by seriousness criteria, whereas non-serious events do not meet those criteria; seriousness is not a measure of how likely an event was to be caused by treatment. Counts alone also do not describe exposure-adjusted rates or account for follow-up time and differences in participants' underlying health.
Because this is a single-group study, the safety figures can describe what was reported among participants receiving TDF during the study, but they cannot isolate treatment-related risk. They should be read as registry safety summaries rather than comparative evidence.
12. What the Result Does, and Does Not, Mean
The study addresses a defined question: the frequency of a specified HBsAg reduction at Week 48 after switching from entecavir to TDF in a selected group of people with chronic hepatitis B. It also records safety outcomes during the study. The responder threshold and the treatment schedule are described, giving readers a clear outline of what the primary biomarker measure was intended to capture.
However, the primary outcome's observed percentage and its 95% confidence interval are not reported in the registry record. This means the record does not establish how many participants met the responder definition, how large the response frequency was, or how precise an estimate might be. It is not possible to characterize the magnitude of the primary result from the endpoint definition alone.
Even if the estimate were available, the one-arm design would limit interpretation. A response observed after switching would be an outcome within this cohort, not a direct estimate of the difference between switching and continuing entecavir. Without a concurrent control group, changes over time and other influences cannot be separated from changes associated with the treatment strategy. The study therefore does not provide a randomized comparison.
The safety counts are similarly descriptive. The registry reports 2 participants with serious adverse events, 36 with non-serious adverse events, and 0 deaths among 75 at risk in the TDF group. These counts do not establish causation or comparative safety. The listed serious-event terms are registry classifications, not evidence that TDF caused the events.
The record also does not report the primary analysis denominator, the handling of missing Week 48 values, or a detailed statistical plan. These omissions make it harder to assess the analysis population and uncertainty even apart from the absence of a comparator. The most defensible reading is therefore limited: the record describes a prospective one-arm switch study and defines its principal biomarker endpoint, while leaving the primary efficacy result and key analysis details unreported.
13. Why This Trial Matters Statistically
This study is statistically interesting because it makes a clear distinction between a biomarker endpoint and a comparative treatment effect. The primary endpoint classifies participants according to a specified HBsAg change from baseline at Week 48. That definition allows a response frequency to be estimated, but the registry record does not report the estimate or its confidence interval. A well-defined endpoint is not by itself enough to judge the observed evidence; the result and its uncertainty are also needed.
The one-arm structure highlights a central design issue in clinical research. Describing outcomes in a group can be useful for understanding what occurred during follow-up, but a causal treatment contrast ordinarily requires an appropriate comparison. Here, there is no concurrent control group, so a response after switching cannot be distinguished from changes that may have occurred under another strategy or over time. This is a limitation of the design, not a calculation problem.
The record also illustrates why the analysis population and missing-data rules matter. The full analysis set excludes specified participants, but the number analyzed and the handling of missing Week 48 values are not reported. Those details shape what a percentage represents and how precisely it can be interpreted. A registry entry that states an endpoint definition but omits the analyzed estimate leaves a substantial gap between the intended question and the evidence readers can evaluate.
Finally, the safety counts show the difference between reporting events and making causal claims. A single treated group can document adverse events, but comparative attribution requires more than a list of counts. Taken together, these features make the record useful for understanding the study's design and planned measurement, while demonstrating why outcome reporting, analysis transparency, and a suitable comparison are important for statistical interpretation.
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.
Several limitations specific to this study affect interpretation. First, the design is single-group, with no concurrent control arm. The study can describe HBsAg and safety observations in participants after switching, but it cannot directly estimate how those outcomes differ from continued entecavir or another comparator. The absence of masking is also recorded, although the registry does not describe any procedures intended to limit assessment or reporting influences.
Second, the primary endpoint result is not reported in the registry information. The observed percentage of responders and the 95% confidence interval are missing, so neither the result's magnitude nor its precision can be evaluated. The record also does not give the primary analysis denominator, making it impossible to verify how many members of the full analysis set contributed to the endpoint estimate.
Third, the full analysis set definition depends on treatment exposure and availability of efficacy data after treatment began. The record does not explain the handling of missing Week 48 measurements, treatment discontinuation, or deviations from the switching schedule. It also does not report a statistical analysis plan, including methods for missing data, multiplicity, interim analyses, or any adjustment for participant characteristics.
Fourth, the stated screening and evaluability objective differs from the listed enrollment figure: the brief summary describes approximately 80 subjects screened to achieve 65 evaluable subjects, while enrollment is listed as 75. The registry does not explain this relationship or state the number ultimately analyzed. These figures should not be treated as interchangeable.
Finally, the safety data are descriptive and lack a comparator. The registry record does not provide enough event-level detail to assess timing, exposure duration, or attribution. These limitations constrain interpretation of the available record, without implying a particular outcome.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: A Study of Switching From Entecavir to Tenofovir Disoproxil Fumarate in Subjects With Chronic Hepatitis B — NCT03258710.
- Registry record last updated 2020-08-04; status last verified 2020-07.
- Results first posted 2020-01-02.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
NCT03258710 is a completed Phase 4, prospective, multicentre, one-arm study of switching from entecavir to tenofovir disoproxil fumarate in Japanese participants with chronic hepatitis B who were HBeAg-positive and HBV-DNA undetectable. The registry lists enrollment as 75 and describes TDF 300 mg once daily for 96 weeks, beginning when ETV was discontinued.
The primary endpoint is the percentage meeting the defined 0.25 log10 HBsAg reduction threshold from baseline at Week 48, with a stated 95% confidence interval. The observed percentage, confidence interval, and primary analysis denominator are not reported in the registry record. The record also lacks a concurrent comparator and does not report detailed analysis or missing-data methods.
For safety, the TDF group lists 2 participants with serious adverse events, 36 with non-serious adverse events, and 0 deaths among 75 at risk. These are descriptive observations and do not establish causation or comparative risk.