Intravaginal Artesunate for the Treatment of HPV+ High Grade Cervical Intraepithelial Neoplasia (CIN2/3): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Intravaginal Artesunate for the Treatment of HPV+ High Grade Cervical Intraepithelial Neoplasia (CIN2/3) (NCT02354534), a non-randomized phase 1 study. The primary outcome is Number of Participants With Serious Adverse Events.

2. Clinical Question

The protocol’s stated primary purpose is to evaluate the safety, tolerability, and feasibility of intravaginal artesunate administration in women with CIN2/3. The listed secondary outcomes are viral clearance of HPV and histologic regression of CIN2/3. The posted results summarized in the registry record do not include analyses of those secondary outcomes.

The four cohorts differ in dose or number of cycles, but the registry does not describe random allocation. A comparison between their event counts therefore cannot isolate a treatment effect: differences could reflect the schedules, differences between the people in each cohort, or other factors not described in the record. The central statistical question is consequently descriptive: what event counts were recorded in each cohort, and how much can those counts establish about safety and tolerability?

3. Trial Design

This is a phase I study of intravaginally administered artesunate in women with high-grade cervical intraepithelial neoplasia. The registry reports a parallel, non-randomized design with no masking. Enrollment is listed as 30. The study status is completed; the recorded start is 2015-02 and primary completion is 2018-07.

The intervention is artesunate suppositories. Four cohorts are described: 50 mg for one five-day cycle; 200 mg for one five-day cycle; 200 mg for two five-day cycles; and 200 mg for three five-day cycles. Each arm description says treatment was given before therapeutic resection of the lesion if that was clinically indicated. The registry does not describe a separate untreated or placebo cohort.

The primary outcome is the number of participants with serious adverse events. Its description also includes dose-limiting toxicities related to the intervention and says assessment used Common Terminology Criteria for Adverse Events (CTCAE) 4.0. The stated timeframe is 41 weeks. The secondary outcomes are HPV viral clearance and histologic regression of CIN2/3.

The registry reports no outcome analyses. It does not report a statistical analysis plan, detailed eligibility criteria, a sample-size justification, or how participants were assigned to the cohorts. Nor does it explain whether the cohorts were enrolled concurrently or in sequence. Those details are important when interpreting comparisons across schedules, and they are not reported in the registry record.

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 Number of Participants With Serious Adverse Events, measured over 41 weeks.

The primary endpoint is the number of participants with serious adverse events or dose-limiting toxicities related to the study intervention, assessed using CTCAE 4.0 criteria. The registry labels the measure as a count of participants and gives a timeframe of 41 weeks. The clinical purpose is to describe treatment-related safety, tolerability, and feasibility in this phase I study.

A participant count is a binary outcome at the individual level: for the specified event definition and observation period, a participant either is counted as having an event or is not. Reporting the count alongside the number at risk in each cohort gives context for how many participants contributed to the observed result. These counts describe the recorded participants; without an uncertainty interval or a prespecified analysis, they do not by themselves establish the rate of events in a broader population.

The registry’s posted serious-event counts are zero in each cohort. However, the primary endpoint description combines serious adverse events with dose-limiting toxicities, and the record does not provide a separate dose-limiting toxicity result. Thus the posted counts should be read as serious-event counts, not as a complete account of every component named in the endpoint description.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a participant count mean?

The primary measure is a count of participants with the specified event, rather than a count of event episodes. In a binary-outcome analysis, each participant is classified according to whether the event definition was met during the stated observation period. This makes the denominator important: the same event count can convey a different level of information when the number observed differs. The registry gives cohort-specific numbers at risk and serious-event counts, but does not report confidence intervals or a modeled estimate.

The endpoint description refers to serious adverse events or dose-limiting toxicities related to the intervention. Those are event definitions, and their operational details affect which participants are counted. The registry names CTCAE 4.0 as the assessment criteria but does not report a separate dose-limiting toxicity count or further details about how relatedness was determined. Consequently, the reported serious-event counts cannot fill in any unreported component of the composite description.

Why does randomization matter?

Randomization is a method for assigning participants to treatment groups by chance. In general, it helps reduce systematic differences between groups, so a comparison can more credibly estimate the effect of the assigned treatments. Here, the registry identifies allocation as non-randomized and reports no masking. The four cohorts receive different dose-and-cycle schedules, but they are not randomized control groups. Their results should therefore be described separately rather than interpreted as a controlled estimate of the effect of one schedule versus another.

In a non-randomized study, differences in participant characteristics or enrollment circumstances may accompany differences in treatment. Statistical adjustment can sometimes address measured differences, but it cannot automatically remove bias from unmeasured factors. The registry lists no outcome analyses and does not report adjustment methods, baseline comparisons, or a statistical analysis plan. It is not possible to assess from the registry record whether any such methods were planned or used.

What can an uncertainty interval add?

In general, a confidence interval describes a range of values compatible with the observed data under a specified statistical method. It communicates sampling uncertainty, which is especially relevant when event counts are small. An interval around a zero observed count would not mean that every value in the interval is equally plausible or that future participants will have the same result; it would quantify uncertainty under the chosen method and assumptions. No confidence interval for the serious-event results is reported in the registry record, so no interval should be inferred from the counts alone.

Likewise, a p-value is not a measure of effect size or clinical importance. It addresses how compatible data are with a specified statistical model and null hypothesis. The registry lists zero analyses, and it reports no p-values or formal comparison of cohorts. There is therefore no posted hypothesis-test result to interpret.

How should planning calculations be read?

The record describes sample-size assumptions for planning illustrations: a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect to detect expressed as a stated fraction of the observed effect. Those calculations are planning illustrations, not the trial’s own design justification. The registry record does not report a sample-size justification for the study itself.

Such illustrations ask what sample size might be needed under specified assumptions. They do not show that the trial had a particular probability of detecting an effect, and they are not observed power. Because this study is non-randomized and its cohort sizes differ, an illustration that assumes equal allocation between two arms does not describe the actual cohort structure. The registry does not report enough information to determine how a planning calculation relates to the study’s enrollment decisions.

More generally, power calculations depend on assumptions about the outcome frequency, effect size, allocation, significance level, and statistical method. If those assumptions do not match the study question or design, the calculation may not be informative for that study. The posted assumptions should therefore be read as an illustration of a hypothetical design, not as evidence about the precision or success of the completed study.

What about over-dispersion?

Over-dispersion is a general concern when observed variability is greater than a statistical model expects. It is especially relevant to count outcomes modeled with a distribution that assumes a particular relationship between the mean and variance. The primary endpoint is a participant count for a binary event, and the registry reports no model-based analysis. There is no reported analysis in which over-dispersion was assessed or needed to be addressed.

For binary outcomes, uncertainty depends on the number of participants and the event probability, as well as on the analysis method. A model should reflect the design and outcome definition. Since no model or inferential analysis is reported, it would be inappropriate to infer a modeling approach from the posted counts.

What does the record leave unspecified?

The registry does not report an analysis population, handling of missing outcome data, detailed participant flow, or a statistical analysis plan. These matters affect who is included in an estimate and how incomplete observation is treated. The 41-week timeframe identifies the stated endpoint window, but the record does not explain the timing or completeness of assessments for each participant. No further analysis choices can be assumed from the summary results.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("50 mg Artesunate Suppositories, 1 Cycle", "200 mg Artesunate Suppositories, 1 Cycle", "200 mg Artesunate Suppositories,2 Cycles", "200 mg Artesunate Suppositories,3 Cycles")
n      <- c(3, 9, 10, 8)
events <- c(0, 0, 0, 0)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  50 mg Artesunate Suppositories, 1 Cycle  3  0  0.0  0.0  70.8  0.0  56.1  0.6  60.2
ARM  200 mg Artesunate Suppositories, 1 Cycle  9  0  0.0  0.0  33.6  -0.0  29.9  0.3  30.8
ARM  200 mg Artesunate Suppositories,2 Cycles  10  0  0.0  0.0  30.8  0.0  27.8  0.2  28.5
ARM  200 mg Artesunate Suppositories,3 Cycles  8  0  0.0  0.0  36.9  0.0  32.4  0.3  33.6

11. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary safety endpoint and two secondary outcomes: HPV viral clearance and histologic regression of CIN2/3. It reports no outcome analyses. A statistical plan describing how these outcomes would be tested, prioritized, or reported is not reported in the registry record.

Multiplicity arises when several outcomes, time points, subgroups, or comparisons are examined. In general, the more analyses performed, the greater the chance of a seemingly notable result arising by chance if each is interpreted without accounting for the full set of tests. Methods such as prespecifying a primary outcome or adjusting inference can help, but the registry does not report whether such methods were planned or used. With no listed analyses, there are no secondary-outcome test results or multiplicity-adjusted findings to interpret.

The record does not report how missing outcome data were handled, whether all enrolled participants contributed to each outcome, or how the analysis population was defined. Missingness can affect both the denominator and the interpretation of a participant count. For example, an unobserved outcome cannot automatically be treated as an event-free outcome; the appropriate handling depends on the study’s definitions and analysis plan. No imputation or other missing-data method is reported.

The registry also does not provide a detailed participant flow or explain the completeness of follow-up during the stated timeframe. These omissions limit assessment of whether the recorded counts represent complete outcome ascertainment. The registry record does not report the protocol’s statistical plan, interim analyses, or sample-size justification, so none should be inferred.

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 serious adverse event counts are reported as 0 in each cohort. The corresponding numbers at risk are 3 for 50 mg artesunate suppositories for one cycle, 9 for 200 mg for one cycle, 10 for 200 mg for two cycles, and 8 for 200 mg for three cycles. Read these as cohort-specific descriptive counts. They do not establish that serious events cannot occur, and the non-randomized design does not support a direct safety ranking among schedules.

The registry separately reports participants with other adverse events: 2 of 3, 9 of 9, 8 of 10, and 8 of 8 in the same cohort order. These figures must not be confused with serious adverse events. The record does not list the event terms in the serious-event category, and no top serious-event terms are reported. It also does not provide separate results for dose-limiting toxicities, although those are included in the primary endpoint description.

Counts are most useful when read with their denominators, definitions, and observation period. The stated primary-endpoint timeframe is 41 weeks, but the registry does not report more detail about each participant’s follow-up or event ascertainment. No confidence intervals are reported, so the precision of the counts is not quantified.

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

The posted data show that the registry recorded zero participants with serious adverse events in each of the four artesunate cohorts. They also show participants with other adverse events in every cohort. These are descriptive findings for the participants represented in the record; they are not proof that serious events are impossible or a precise estimate of how often events would occur in a broader population.

The results do not establish that one dose-and-cycle schedule is safer or more effective than another. Cohort assignment was non-randomized, and the registry does not report cohort-comparability information or an adjusted analysis. The four groups are therefore not interchangeable control arms. A contrast between their counts would not, on its own, separate the effects of schedule from differences in the people assigned to each cohort or other unreported factors.

The record does not report analyses of the secondary outcomes, HPV viral clearance and histologic regression of CIN2/3. It also does not report confidence intervals, p-values, a statistical analysis plan, or a separate dose-limiting toxicity count. As a result, it cannot establish the size or precision of an effect on those outcomes, nor provide a complete statistical evaluation of the composite safety endpoint as described.

Phase I studies commonly focus on early safety and feasibility questions as a general matter, but that general purpose should not be confused with evidence of efficacy. The registry’s stated aims include safety, tolerability, and feasibility; the posted results support only a narrow description of recorded adverse-event counts. They do not determine whether artesunate improves HPV clearance or causes histologic regression.

Planning calculations, where presented, are hypothetical sample-size illustrations based on specified assumptions. They are not observed power, do not convert the completed study into a randomized comparison, and cannot substitute for missing outcome analyses. The main statistical meaning is therefore limited: the record documents observed cohort-specific counts, while leaving substantial uncertainty about event risk, comparative safety, and the listed secondary outcomes.

14. Why This Trial Matters Statistically

This study is statistically instructive because its outcomes and design need to be read together. The primary safety endpoint is a participant count, and the registry reports cohort-specific denominators alongside zero serious adverse events. Pairing a count with its denominator is essential: a count alone can conceal how much observation contributed to it. Even with denominators, zero observed events do not establish zero underlying risk.

The non-randomized design is equally important. Different dose-and-cycle schedules may be of interest, but cohort differences cannot be interpreted as treatment effects without accounting for how participants entered each cohort and whether the groups differed in relevant ways. The registry does not report those details or an adjusted analysis. Resisting a causal comparison is therefore part of accurate statistical interpretation, not merely a technical caveat.

The record also illustrates why endpoint wording matters. The primary endpoint description includes serious adverse events or dose-limiting toxicities related to treatment, while the posted serious-event counts do not separately report dose-limiting toxicities. A numerical result should not be taken to cover components that are not separately documented.

Finally, the absence of listed outcome analyses limits what can be concluded about the secondary questions. The registry names viral clearance and histologic regression, but does not provide their analyzed results. Statistical interpretation requires distinguishing an outcome that was planned or named from one for which results are actually reported. Together, these points show why design, outcome definition, denominators, uncertainty, and reporting completeness all shape what a clinical trial record can 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 main trial-specific limitation is the non-randomized allocation. The four cohorts differ in dose or number of cycles, but the registry does not explain how cohort assignment occurred or report whether the groups were similar before treatment. Their event counts should not be used as a like-for-like comparison or as a causal estimate of differences between schedules.

The study is phase I and reports cohort sizes of 3, 9, 10, and 8 at risk for the serious-event outcome. Counts from cohorts of this size provide limited information about uncommon events. The registry does not report confidence intervals, so uncertainty around the counts is not quantified. The enrollment figure is 30, but the record does not explain participant flow or how enrollment relates to the outcome-specific at-risk numbers.

The primary endpoint description includes serious adverse events or dose-limiting toxicities related to treatment, but the posted results provide serious-event counts and no separate dose-limiting toxicity count. The record does not report serious-event terms, detailed event timing, or further information on relatedness. It also does not provide analyses for HPV viral clearance or histologic regression.

Other limitations in the registry record include the absence of a reported statistical analysis plan, sample-size justification, analysis population, missing-data approach, or detailed follow-up completeness. The endpoint timeframe is 41 weeks, but the registry does not give participant-level follow-up detail. These gaps prevent a fuller assessment of precision, outcome ascertainment, and the scope of conclusions that can be drawn.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Intravaginal Artesunate for the Treatment of HPV+ High Grade Cervical Intraepithelial Neoplasia (CIN2/3) — NCT02354534.
• Registry record last updated 2021-04-28; status last verified 2021-03.
• Results first posted 2021-04-02.
• 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

The registry record describes a completed phase I study of intravaginal artesunate suppositories in women with CIN2/3 and high-risk HPV. It lists a non-randomized, unmasked, parallel design and four dose-and-cycle cohorts: 50 mg for one cycle, 200 mg for one cycle, 200 mg for two cycles, and 200 mg for three cycles. Enrollment is listed as 30.

The primary endpoint is the number of participants with serious adverse events or dose-limiting toxicities related to treatment, assessed using CTCAE 4.0 over 41 weeks. Serious adverse event counts are 0 in each cohort. The registry also lists other adverse-event counts by cohort. HPV viral clearance and histologic regression of CIN2/3 are named as secondary outcomes, but no outcome analyses are listed.

The record does not report a statistical analysis plan, confidence intervals, missing-data methods, or a sample-size justification. Because allocation was non-randomized, cohort counts are descriptive and do not constitute a controlled comparison.