A Study to Evaluate Preliminary Helicobacter Pylori Eradication After Multiple Doses of TNP-2198 Capsules Combined With Rabeprazole Sodium Enteric-coated Tablets, or Rabeprazole Sodium Enteric-coated Tablets and Amoxicillin Capsules: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Evaluate Preliminary Helicobacter Pylori Eradication After Multiple Doses of TNP-2198 Capsules Combined With Rabeprazole Sodium Enteric-coated Tablets, or Rabeprazole Sodium Enteric-coated Tablets and Amoxicillin Capsules (NCT06076681), a randomized phase 1, phase 2 study. The primary outcome is The Eradication Rate of Helicobacter Pylori Infection.

2. Clinical Question

3. Trial Design

The registry describes a Phase Ⅰb/Ⅱa, single-center, randomized, parallel-group study, with enrollment of 40 and status recorded as completed. Its stated aims included safety, tolerability, pharmacokinetics, and preliminary Helicobacter pylori eradication efficacy. The posted primary outcome concerns eradication, defined as a negative urea breath test result.

Participants were assigned to four groups of 10. Each group received TNP-2198 with rabeprazole, while the TNP-2198 dose varied across Groups A, B, and C. Group D received the same listed TNP-2198 and rabeprazole doses as Group B, with amoxicillin also included. Thus, the study includes dose variation among three groups and an additional-drug combination in another group. The registry's posted comparisons use Group A as the reference for each of the other groups.

The study description calls the trial open-label, while the registry's masking classification is recorded as double. These entries do not align, and the discrepancy is not reconciled in the registry record. The record lists the start as 2021-09-27 and primary completion as 2022-01-06. More detailed information about randomization procedures, allocation concealment, treatment adherence, or operational conduct is not reported in the registry record.

The study is described as preliminary, and its small groups are consistent with reading the posted findings as early, descriptive evidence rather than as precise estimates. The registry does not report a sample-size justification or a planned primary comparison. It also does not report further outcomes as posted secondary measures. These absences limit how much can be inferred about the study's intended statistical framework and the role of the different study aims.

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

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as The Eradication Rate of Helicobacter Pylori Infection, measured over Urea breath test is assessed 4-6 weeks after the treatment (Day 44~ Day 50).

The primary endpoint is the number of participants with eradication of Helicobacter pylori infection. The registry defines eradication as a negative urea breath test result. Testing was scheduled 4-6 weeks after treatment, with the stated timeframe Day 44~ Day 50. The outcome is therefore a binary participant-level status: a participant either had a negative test or did not, according to the registry definition.

A urea breath test offers an objective way to classify the presence or absence of infection at the assessment time. Using a common assessment and definition across groups makes the outcome interpretable as a count of participants meeting the eradication criterion. The posted values are counts, not continuous measurements, and the reported comparisons express differences in the probability of meeting that criterion.

The endpoint captures test status at the specified follow-up assessment; it does not, by itself, describe how quickly a test became negative, whether the result persisted beyond that assessment, or the participant's broader clinical course. Those matters are not reported as part of the primary endpoint record. The registry also does not report how indeterminate, missing, or unevaluable test results were handled. Interpretation should therefore remain tied to the posted counts and the stated test-based definition.

7. Results Posted in the Registry

• Group B: 30.0% (3/10) versus 0.0% (0/10) on control; risk difference 30.0 percentage points (95% CI 1.6 to 58.4); unadjusted two-proportion z-test p = 0.060.
• Group C: 40.0% (4/10) versus 0.0% (0/10) on control; risk difference 40.0 percentage points (95% CI 9.6 to 70.4); unadjusted two-proportion z-test p = 0.025.
• Group D: 80.0% (8/10) versus 0.0% (0/10) on control; risk difference 80.0 percentage points (95% CI 55.2 to 104.8); unadjusted two-proportion z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

The endpoint is binary because each participant's urea breath test is classified as negative or not negative under the registry definition. For a binary outcome, the arm-level result can be described by a count and the number of participants assessed or enrolled. Here, the posted counts are 0 of 10 in Group A, 3 of 10 in Group B, 4 of 10 in Group C, and 8 of 10 in Group D. The record does not clarify whether every enrolled participant had an evaluable test or how missing tests would be treated.

What does the risk difference represent?

A risk difference subtracts the event probability in one group from that in another. In this study, the event is a negative urea breath test. Each reported risk difference compares one group with Group A, so a positive value means that the observed fraction meeting the endpoint was higher in that group than in Group A. Unlike a relative measure, a risk difference is expressed in absolute probability units. It does not describe the chance of benefit for an individual participant, and it does not by itself establish that group assignment caused the observed contrast.

The reported risk differences are 0.3 for Group B, 0.4 for Group C, and 0.8 for Group D. They align with the posted counts in each group. These estimates are based on small group sizes, so a small change in the outcome count would substantially affect the estimated group proportions and their difference. The point estimates should therefore be read alongside the confidence intervals, not as precise quantities.

How should the p values be read?

A p value describes how compatible the observed data are with a specified statistical null model, under the assumptions of the test. It is not the probability that the null hypothesis is true, nor does it measure the size or clinical importance of an effect. The registry posts unadjusted p values for comparisons of Groups B, C, and D with Group A: 0.06, 0.025, and <0.001, respectively. The particular test used, its assumptions, and any prespecified threshold are not reported in the registry record.

Because there are multiple comparisons, p values considered one at a time can give a different overall error interpretation from a method that accounts for the full set. The record does not report a multiplicity strategy. Accordingly, these values should be understood as the posted unadjusted results, not as multiplicity-controlled evidence. They also do not remove the uncertainty reflected in the confidence intervals.

What does a confidence interval add?

A confidence interval presents a range of values compatible with the data under the interval procedure and its assumptions. It makes uncertainty visible in a way that a single point estimate cannot. General interpretation: a wider interval indicates less precision, while a narrower interval indicates greater precision, all else being equal. In this study the reported intervals are wide, consistent with the limited number of participants in each group. Their lower and upper endpoints should not be read as guaranteed limits for the true effect.

The Group D interval has an upper endpoint of 1.048, beyond the possible maximum of 1 for a risk difference. Some interval procedures can produce limits outside the natural bounds of a parameter. The registry does not state the procedure used, so the reason for that endpoint cannot be determined from the record. The point estimate and interval should be reported as posted rather than silently altered.

What planning information is available?

The registry includes a planning illustration based on the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It describes effects to detect as fractions of the observed effect. Such calculations are planning illustrations, not observed power and not evidence about the probability that the study detected a true effect. The numerical planning outputs and the particular effect fractions are not reported in the registry record available for this analysis. A sample-size justification for the original trial is also not reported.

The registry records no posted statistical analyses, despite presenting outcome counts and comparisons. It does not identify the analysis population, the handling of missing observations, the statistical test or interval procedure, or whether the analysis was prespecified. General explanation: randomization can help balance participant characteristics across groups on average, but in small groups imbalances can remain, and randomization does not resolve missing-data or analysis-reporting uncertainties. These unreported details matter when judging how robustly the observed contrasts can be interpreted.

The study has four parallel groups, but the posted risk differences compare each of three groups with Group A. They do not provide a complete set of pairwise contrasts. In particular, comparing the result for Group D with the result for Group B by looking at their separate estimates against Group A is not equivalent to a direct statistical comparison between those groups. No direct Group D versus Group B result is reported. Thus, the data support describing the observed group counts and listed reference comparisons, but not attributing the difference between those groups to the addition of amoxicillin.

10. Confidence Intervals and Precision

The confidence intervals around the reported risk differences communicate substantial uncertainty. For Group B versus Group A, the risk difference is 0.3, with a 95% interval from 0.016 to 0.584. For Group C versus Group A, it is 0.4, with an interval from 0.096 to 0.704. For Group D versus Group A, it is 0.8, with an interval from 0.552 to 1.048. These ranges are compatible with a wide span of possible contrasts, rather than pinning down a narrow effect size.

General explanation: a confidence interval is a procedure-based range intended to convey the uncertainty of an estimate. Its interpretation depends on the statistical method and assumptions used. It is not a range that guarantees where the true value lies, and it is not a prediction interval for individual participants. The registry does not report the procedure used to construct these intervals, which limits assessment of their exact properties.

Group D's upper confidence limit is 1.048, exceeding the maximum possible risk difference of 1. The point estimate remains the posted 0.8, but the interval endpoint signals that the interval method may not enforce the parameter's natural bounds. The registry does not report enough methodological detail to explain this feature, and the endpoint should not be mistaken for a feasible risk difference.

All groups included 10 participants, so the precision is constrained by the small number of observations. The confidence intervals show that uncertainty remains even where the posted counts differ. The study estimates are therefore useful as preliminary descriptions, but they do not support highly precise estimates of the contrasts.

11. Multiplicity, Missing Data and Analysis Populations

The posted comparisons include Group B versus Group A, Group C versus Group A, and Group D versus Group A. Their p values are explicitly unadjusted. When several hypotheses are tested, evaluating each p value separately does not account for the chance of a false-positive result across the full set. The registry record does not report a multiplicity adjustment, a testing hierarchy, or another strategy for addressing the three comparisons. It is therefore not possible to determine whether the reported p values reflect a prespecified approach to multiple testing.

The registry records no posted statistical analyses and does not state which participants were included in the primary analysis. The primary population field is blank. Whether the endpoint was analyzed among all randomized participants, participants who completed treatment, or participants with an evaluable breath test is not reported in the registry record. The record also does not explain how withdrawals, protocol deviations, or unevaluable results were handled.

Missing outcome data can affect an estimate if the probability of having an observed test differs across participants or groups. General explanation: assumptions about missing outcomes influence what an analysis estimates, and different handling methods can yield different results. No missing-data assumptions or methods are reported. Although each arm lists 10 participants, that does not establish that all 10 had a valid outcome assessment.

The record provides no prespecified analysis plan, test specification, or detailed description of the confidence interval method. It also does not report a planned primary comparison or a statistical rationale for the sample size. These omissions do not change the posted counts, but they restrict evaluation of whether the analyses followed a planned approach and how fully the reported uncertainty accounts for the study design.

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 0% reporting threshold.

Most frequent non-serious adverse event terms

The registry's serious adverse event figures are 0 affected participants in each group. Each group lists 10 participants at risk for serious adverse events, with 0 affected in Group A, Group B, Group C, and Group D. The listed death count is also 0 in every group. These figures describe what the record reports during the recorded study period; they do not establish that no serious adverse events could occur in a larger or different population.

For other adverse events, the registry lists 6 affected participants in Group A, 8 in Group B, 5 in Group C, and 8 in Group D, with 10 at risk in each group. The record does not provide event terms in the information available for this analysis, nor does it describe severity, timing, duration, relatedness, or how multiple events within one participant were counted. The group counts should not be treated as a comparative safety test. With these small groups and without further detail, the figures are descriptive and do not establish a safety difference between regimens.

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

The results show that the number of participants with a negative urea breath test differed across the four groups at the recorded assessment: 0 in Group A, 3 in Group B, 4 in Group C, and 8 in Group D. The registry reports risk differences for Groups B, C, and D relative to Group A, with corresponding confidence intervals and unadjusted p values. These figures describe the outcomes observed in this small study and the uncertainty around the reported contrasts.

They do not establish how the regimens would perform in a broader population. Each group contained 10 participants, the intervals are broad, and the registry does not report enough analysis detail to assess the treatment of missing outcomes or the precise statistical procedures. The trial is described as preliminary, and the posted findings should be interpreted in that context.

The results also do not isolate the contribution of each component. Groups A, B, and C differ in their listed TNP-2198 doses, while Group D includes amoxicillin and shares the listed TNP-2198 and rabeprazole doses with Group B. But the posted comparisons are against Group A, not direct comparisons between those groups. A higher observed count in Group D than Group B cannot, on its own, be attributed to amoxicillin.

Nor do the primary outcome counts answer the study's other stated aims of safety, tolerability, or pharmacokinetics. The posted serious adverse event counts are zero in each group, and other adverse event counts are listed, but detail about event types and other safety analyses is not reported in the registry record. These data describe recorded observations, not a guarantee of safety or a recommendation about treatment. Follow-up beyond the stated test assessment is not described by this endpoint.

14. Why This Trial Matters Statistically

This trial illustrates how a randomized, parallel-group design can generate preliminary comparisons across different dose combinations and an additional-drug regimen. The primary outcome is concrete and defined as a negative urea breath test, and the registry provides counts for each group along with risk differences relative to a reference group. These features make the observed pattern straightforward to describe.

At the same time, the study shows why effect estimates need to be read with their precision. With 10 participants in each group, a small number of outcomes determines each estimate. The confidence intervals are broad, and the Group D upper endpoint extends beyond the possible risk-difference range. Those details matter more than a simple ranking of the observed counts.

The record also illustrates the distinction between a planned study design and a reported analysis. Randomization is recorded, but the analysis population, missing-data handling, multiplicity approach, and interval method are not reported. The study description calls the trial open-label, whereas the masking classification says double. Such inconsistencies and omissions make it harder to assess the safeguards and assumptions behind the results.

Finally, the four-group arrangement does not make every regimen contrast available. The posted analyses compare three groups with Group A, rather than directly comparing all pairs. The example underscores why an observed difference between two groups is not a tested difference unless a direct comparison is reported. These statistical features are relevant to reading small preliminary trials without overstating what their results establish.

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 principal limitation is the small number of participants in each group: 10 per group. This yields limited information for estimating an eradication fraction or a difference between groups, and the reported confidence intervals are broad. The counts are preliminary observations rather than precise estimates for a wider population.

The registry's study description identifies the trial as open-label, but its masking classification is recorded as double. The record does not clarify this inconsistency. It also does not describe the randomization method, allocation concealment, adherence, or other design procedures that could inform an assessment of bias.

The registry does not report the primary analysis population, the method used for the risk-difference intervals or p values, or how missing and unevaluable breath tests were handled. It records zero posted statistical analyses, even though comparison estimates are presented. No multiplicity plan, sample-size justification, or detailed statistical analysis plan is reported. The p values are unadjusted, and the comparisons do not include direct tests between every pair of groups.

The reported Group D confidence interval has an upper limit of 1.048, beyond the possible range of a risk difference. The record does not identify the interval method or explain the out-of-range endpoint. The safety information is also limited: adverse-event terms and detail on severity, timing, and relatedness are not reported in the record available for this analysis. Finally, the endpoint captures test status at one stated follow-up period and does not establish longer-term status.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study to Evaluate Preliminary Helicobacter Pylori Eradication After Multiple Doses of TNP-2198 Capsules Combined With Rabeprazole Sodium Enteric-coated Tablets, or Rabeprazole Sodium Enteric-coated Tablets and Amoxicillin Capsules — NCT06076681.
• Registry record last updated 2023-11-22; status last verified 2023-11.
• Results first posted 2023-11-22.
• 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 records NCT06076681 as a completed, randomized, single-center, parallel-group Phase Ⅰb/Ⅱa study with 40 participants. It evaluates TNP-2198 and rabeprazole regimens, including a group that also received amoxicillin, in asymptomatic participants with Helicobacter pylori infection.

The primary outcome is the count of participants with a negative urea breath test 4-6 weeks after treatment. Counts were 0 in Group A, 3 in Group B, 4 in Group C, and 8 in Group D, with 10 participants listed in each group. The posted risk differences compare Groups B, C, and D with Group A. The record does not report a detailed analysis plan, analysis population, missing-data approach, or multiplicity strategy. Its study description calls the trial open-label, while the masking classification records double.