Trial Comparing Ceftriaxone Plus Azithromycin Versus Ceftriaxone for the Treatment of Gonorrhea: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Trial Comparing Ceftriaxone Plus Azithromycin Versus Ceftriaxone for the Treatment of Gonorrhea (NCT05027516), a randomized phase 4 study. The primary outcome is Macrolide Resistance Determinants.

2. Clinical Question

3. Trial Design

This was a randomized, parallel-group trial registered as Phase 4, with enrollment of 42. The planned allocation described in the summary was 1:1 between ceftriaxone alone and ceftriaxone plus azithromycin. The posted arm counts are 22 for Rocephine and 20 for Azithromycin / Rocephine®. The record marks the study completed. Its listed start date is 2022-01-17 and its primary completion date is 2022-05-09.

The trial summary describes recruitment at the Institute of Tropical Medicine’s STI and HIV clinic and a test-of-cure visit at day 14 after treatment. Urine, oropharyngeal and anorectal samples were to be collected. The primary endpoint concerns the anorectal microbiome; the summary also describes testing for cure and monitoring effects on the microbiome and resistome. The registry lists two secondary outcomes: read counts of resistance determinants for each non-macrolide antibiotic class, and acquisition of phenotypic resistance to azithromycin by N. gonorrhoeae.

The title and official title describe the study as open label, while the registry masking field lists quadruple masking. The record does not explain this difference, so the masking arrangement cannot be resolved from the posted information. Randomization and a parallel comparison are recorded. Further details of sequence generation, concealment, adherence, or deviations from assigned treatment are not reported in the registry record.

Enrollment describes the registry’s stated study size, while the posted arm counts show the group sizes used in the outcome display. The record does not give a fuller account of participant flow or explain whether every enrolled participant contributed to every analysis. That distinction matters when interpreting group summaries and precision.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Analysis population for the primary outcome, as defined in the registry: The primary analysis of assessing the ratio between the mean normalized read count of macrolide resistance determinants in anorectal microbiome between the two groups was done using permutation test with 10 000 permutations.

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 Macrolide Resistance Determinants, measured over Day 14.

The primary endpoint is the mean normalized read count of macrolide resistance determinants in anorectal samples collected at day 14, expressed in normalized determinants/million reads. Resistance determinants are genetic features classified by antibiotic class through bioinformatic analysis of sequencing reads. The endpoint therefore summarizes the abundance of identified resistance-associated genetic material in the sampled microbiome.

The registry says the primary comparison is a ratio of means: the mean in the ceftriaxone-plus-azithromycin group divided by the mean in the ceftriaxone group. A ratio is used to express the relative abundance between groups, rather than their absolute difference. A ratio of 1 would indicate equal values on this scale; values above or below 1 indicate which group has the larger estimated mean, given the stated numerator and denominator. This is a general interpretation of a ratio, not an additional result from the trial.

This endpoint is relevant to the trial’s stated aim of examining treatment effects on the resistome. It is not the same as a clinical outcome such as symptom resolution, and it should not be interpreted as a direct measure of whether gonorrhea was cured. The record describes testing for cure, but the primary result is the resistance-determinant measure. Details of the sequencing workflow beyond the stated read-based classification and normalization are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary analysis is described as a permutation test with 10 000 permutations. The outcome is continuous, and the registry identifies the reported comparison parameter as a ratio of means. Understanding these choices helps clarify what the posted estimate represents and what it cannot answer.

What does a permutation test do?

As a general statistical concept, a permutation test evaluates how unusual an observed group comparison would be if group labels could be reassigned under a null hypothesis of no treatment-group difference. It repeatedly constructs comparisons under that reassignment scheme and uses the resulting reference distribution to assess the observed result. Here, the registry specifically states that 10 000 permutations were used for the primary ratio. The record does not give further details about the permutation procedure, such as restrictions on reassignment or the exact test statistic used in each permutation.

Random assignment is central to the logic of a permutation test: if treatment allocation was conducted as intended and the null hypothesis is appropriate, group labels can provide a basis for judging the comparison. However, the registry record does not describe randomization implementation in enough detail to evaluate the procedure independently. Nor does the test, by itself, address missing outcome data, protocol deviations, or other features of study conduct.

How should the ratio be read?

The registry defines the ratio as the mean normalized read count in the ceftriaxone/azithromycin group divided by the corresponding mean in the ceftriaxone group. This makes 1 the reference value for equal group means. A ratio above 1 points toward a larger mean in the combination group, while a ratio below 1 points toward a smaller mean, using the stated order. This is general guidance on interpreting ratios. The reported estimate is 1.05, and its 95% confidence interval is 0.55 to 1.83.

A confidence interval describes the uncertainty associated with an estimated quantity under the analysis procedure. It is not a range for individual observations, and it does not guarantee that the true value lies in the interval. In this result, the interval spans the reference value of 1 and allows materially different relative values. The data therefore do not pin down the ratio precisely. A narrow interval would generally indicate more precision than a wide one, although precision alone does not establish clinical or biological importance.

What does the p-value add?

The reported p-value is 0.103. In general, a p-value summarizes how extreme the data would be under a specified null model and analysis. It does not measure the size of an effect, the probability that the result occurred by chance, or the probability that a treatment hypothesis is true. Its meaning depends on the test and assumptions. The registry does not provide a full statistical analysis plan or explain how the p-value was interpreted against a decision threshold. It should therefore be considered together with the estimate and confidence interval, not as a stand-alone verdict.

Why can the group summaries and ratio look different?

The posted summaries give a mean for each arm and a separate ratio from permutation testing. A ratio obtained through an analysis procedure is not necessarily presented as a simple arithmetic operation on the displayed summaries. The registry does not describe further adjustments or transformations, so it would be inappropriate to infer the reason for any difference between the summaries and the reported test estimate. The posted primary analysis is the relevant source for the reported inferential comparison.

What does over-dispersion mean in general?

For count-like measurements, over-dispersion means that the observed variation is greater than a simple model would expect. It can affect uncertainty estimates if an analysis assumes less variability than is present. This is a general statistical consideration, not a reported finding about this trial. The registry does not describe a model for over-dispersion, a distributional assessment, or a sensitivity analysis for the count-based read measure. The primary analysis is instead described as permutation testing.

The record lists the endpoint as a continuous measure and presents means with 95% confidence intervals. It does not report individual observations, the distribution of values, or details of how the confidence intervals for the arm means were constructed. Those omissions limit independent evaluation of the summaries. General statistical explanations cannot fill in those trial-specific details.

The posted analyses include 16 analyses and the record lists three outcome measures, but their full relationship and any analysis hierarchy are not described. Because multiple outcomes or analyses can increase the chance of at least one striking result, adjustment or prespecified prioritization may matter. Whether such procedures were used is not reported in the registry record. The primary result should be distinguished from secondary findings rather than treated as one result among an unspecified set of equivalent tests.

10. Analyses Reported by the Sponsor

The registry reports the primary comparison as a permutation-test ratio of 1.05, with a 95% confidence interval of 0.55 to 1.83 and p = 0.103. This is the trial’s posted analysis and is the principal inferential result to use when describing the primary endpoint. The registry does not flag a discrepancy between the reported analysis and any recalculation.

No separate recalculated estimate or interval is included in the record, so a numeric comparison between a sponsor analysis and an independent recalculation cannot be made. The displayed arm means and their confidence intervals provide useful descriptive context, but they are not a substitute for the reported permutation analysis. In particular, the sponsor’s ratio is defined by the registry as the primary between-group result, and the method used to obtain it is not fully specified beyond permutation testing.

The sponsor-reported analysis remains the appropriate account of what the registry says was tested. That does not make the result more conclusive than its uncertainty allows: the confidence interval includes the equality value, and the estimate should be interpreted with the limitations of the record and endpoint in mind.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists three outcome measures and 16 analyses. It names two secondary outcomes: read counts of resistance determinants for each non-macrolide antibiotic class, and acquisition of phenotypic resistance to azithromycin by N. gonorrhoeae. The primary outcome is the day-14 macrolide resistance determinant measure. The record does not explain how all listed analyses map to the outcomes, which analyses were prespecified, or whether a hierarchy was used.

When several outcomes or tests are considered, there is a general risk that some results may appear notable through chance alone. Statistical approaches can address this through prespecified prioritization or adjustment, but whether either was used is not reported in the registry record. No multiplicity adjustment is described in the information. The primary result should therefore be kept distinct from secondary or other analyses, and the number of analyses makes a complete analysis plan important for interpretation.

The primary outcome analysis is described as using the day-14 anorectal samples, but the record does not report how missing samples or unusable sequencing results were handled. It does not state the number with an observed primary endpoint, reasons for missingness, or any imputation or sensitivity analysis. The posted arm counts do not, on their own, establish that every participant contributed usable endpoint data.

The analysis population is not described in detail. The record does not say whether the primary analysis included all randomized participants, only participants with available samples, or another defined population. It also does not report how treatment adherence, crossovers, or protocol deviations were handled. These are important distinctions for assessing what population the estimate represents.

The endpoint is measured at day 14, and no later primary assessment is reported. The information available does not establish how the study handled the possibility that a day-14 sample was unavailable or difficult to classify. These gaps do not invalidate the posted result, but they limit assessment of robustness and generalizability.

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 reports zero serious adverse events among 22 participants in the Rocephine® group and zero among 20 participants in the Rocephine® + Azithromycin group. It also reports zero deaths in each group. These are the posted counts for the listed serious-event and death categories; they should not be read as proof that serious events are impossible or that the treatments have equivalent safety.

For other adverse events, the record reports 2 affected participants in the Rocephine® group and 4 in the combination group. The registry information does not provide event descriptions, severity breakdowns, timing, attribution, or the method used to collect these events. The figures count affected participants as recorded, not necessarily the total number of event episodes.

These group sizes are limited, and the absence of recorded serious events in this trial cannot rule out uncommon harms. Conversely, the difference in the posted counts of other adverse events does not, by itself, establish a treatment-related difference. The figures are descriptive safety information from this record and should be interpreted alongside the missing detail about event definitions and ascertainment.

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

The result addresses a specific question: how the mean abundance of macrolide resistance determinants in day-14 anorectal samples compared between the two randomized groups. The estimated ratio was 1.05, with a 95% confidence interval from 0.55 to 1.83. The point estimate is close to the equal-value reference, but the interval is broad enough to include values on either side of that reference. This means the posted result does not precisely determine the relative abundance.

The analysis does not establish that the groups are identical. A confidence interval that includes 1 is compatible with equality, but it also includes other values. Nor does the result establish a clear difference. The right interpretation is that this trial’s posted primary analysis leaves substantial uncertainty about the size and direction of the relative difference.

The endpoint is a read-based measure of resistance determinants in the anorectal microbiome. It does not directly measure clinical cure, symptoms, transmission, or longer-term resistance. Although the trial summary says samples were taken to test for cure and monitor treatment effects, the primary result is not a cure result. Conclusions should not be transferred from this biological measure to outcomes that are not reported.

The posted findings also do not show that either regimen caused a change from baseline: the primary comparison is between groups at day 14, and baseline values or a change-from-baseline analysis are not reported in the registry information. Details about missing samples, analysis populations, and handling of multiple analyses are also not reported. Those omissions constrain how confidently the result can be generalized beyond the observed comparison.

In short, the registry presents a randomized comparison and a primary ratio estimate, but the estimate is imprecise and addresses a narrowly defined endpoint. It should be read as evidence about the measured day-14 resistome outcome in this trial, not as a broad conclusion about treatment efficacy or safety.

14. Why This Trial Matters Statistically

This trial is statistically interesting because it compares treatment groups using a microbiome and resistome endpoint rather than a conventional clinical endpoint as its primary measure. The endpoint is a normalized read count of macrolide resistance determinants in anorectal samples, and the primary comparison is expressed as a ratio of group means. That combination makes clear definitions and analysis methods especially important: the reader needs to know what was counted, how reads were classified and normalized, and how the group comparison was estimated.

The registry identifies a permutation test with 10 000 permutations. This makes the inferential method visible, while also leaving questions about the implementation unanswered. The reported ratio and confidence interval show why a point estimate alone is insufficient. The interval conveys uncertainty and includes the equality value, so the analysis does not precisely settle the comparison.

The trial also illustrates the distinction between a mechanistic or biological outcome and a clinical outcome. Resistance determinants can be relevant to the study’s stated resistome question, but the endpoint does not itself quantify cure or patient benefit. Keeping the question tied to the measurement prevents overinterpretation.

Finally, the record lists multiple outcomes and analyses, yet does not report how multiplicity, missing samples, or analysis populations were handled. These details are central to judging the reliability and scope of any result. The trial therefore offers a useful case for reading effect estimates, uncertainty intervals, and registry methods together, while recognizing what an incomplete public analysis description cannot resolve.

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 registry record leaves several trial-specific uncertainties. Although the title describes an open-label study, the masking field records quadruple masking, and no explanation reconciles the two descriptions. The practical masking arrangements are therefore unclear. Details of allocation concealment, randomization implementation, and adherence are not reported in the registry record.

The primary endpoint is a sequencing-derived count measure. The record describes classifying resistance determinants with a read-based tool and normalizing counts, but does not give further technical details about the pipeline, quality control, or handling of samples that could not be analyzed. It also does not report the distribution of individual measurements. These omissions limit independent assessment of the endpoint and its variability.

The posted arm summaries include group sizes and confidence intervals, but the record does not explain the construction of those intervals. The primary permutation analysis is identified, yet further implementation details are not reported. The relationship between the arm summaries and the reported ratio is also not fully explained. No separate recalculated estimate is available for comparison.

The record lists multiple outcomes and analyses but does not describe a multiplicity strategy, the full analysis hierarchy, or whether analyses were prespecified. Missing data, analysis populations, participant flow, and sensitivity analyses are not reported. The record also does not provide a threshold for a meaningful change in resistance determinants, making it difficult to judge the practical importance of the estimated ratio.

Finally, the outcome is measured at day 14 in anorectal samples. The registry summary mentions other sample types and testing for cure, but the primary result does not answer those distinct questions. The modest posted group sizes and the endpoint’s narrow scope further limit the breadth of conclusions that can be drawn.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Trial Comparing Ceftriaxone Plus Azithromycin Versus Ceftriaxone for the Treatment of Gonorrhea — NCT05027516.
• Registry record last updated 2024-08-02; status last verified 2024-02.
• Results first posted 2024-08-02.
• 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.

19. Record Summary

NCT05027516 is a completed randomized parallel-group Phase 4 trial comparing ceftriaxone with ceftriaxone plus azithromycin in men with Neisseria gonorrhoeae attending the Institute of Tropical Medicine’s STI and HIV clinic. The registry lists enrollment of 42 and arm counts of 22 for Rocephine and 20 for Azithromycin / Rocephine®.

The primary endpoint was the mean normalized abundance of macrolide resistance determinants in day-14 anorectal samples. The registry reports a permutation-test ratio of 1.05, with a 95% confidence interval of 0.55 to 1.83 and p = 0.103. It also reports arm means of 110.3 and 167.5, respectively, with their displayed 95% confidence intervals.

The record reports no serious adverse events or deaths in either group, and other adverse events in 2 and 4 participants, respectively. Details about missing data, analysis populations, masking, and multiplicity are not reported or are not reconciled in the registry information. The posted primary result is an imprecise comparison of a specific microbiome measure, not a direct estimate of clinical cure.