Imiquimod, Fluorouracil, or Observation in Treating HIV-Positive Patients With High-Grade Anal Squamous Skin Lesions: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Imiquimod, Fluorouracil, or Observation in Treating HIV-Positive Patients With High-Grade Anal Squamous Skin Lesions (NCT02059499), a randomized phase 3 study. The primary outcome is Percentage of Participants Achieving Complete Response in 5-FU Arm and Observation Arm Using ITT Population.

2. Clinical Question

3. Trial Design

This was a randomized phase III study with a crossover model and single masking. The registry record does not state who was masked. It lists three initial groups: imiquimod, fluorouracil and observation. The study enrolled 91 participants and is recorded as completed. The recorded start date is 2015-12-28, and the primary completion date is 2023-12-05.

Participants assigned to the imiquimod arm applied it intra-anally once daily for 16 weeks. Participants assigned to fluorouracil applied it intra-anally twice daily on days 1–5, repeating treatment every 2 weeks for 8 courses unless disease progression or unacceptable toxicity occurred. Participants assigned to observation received no treatment. At week 20, participants with persistent HSIL who agreed to randomization could cross over to imiquimod or fluorouracil.

The primary comparison used the intent-to-treat population and evaluated complete response at week 20 in the fluorouracil and observation arms. The posted group sizes for that comparison are 36 and 37, respectively. The registry record does not explain the difference between these outcome-group totals and overall enrollment, nor does it report the original target sample size or its justification. The primary analysis was specified as a stratified Mantel-Haenszel-Cochran test across sites, with a one-sided alpha level of 0.025. Further details of the statistical analysis plan are not reported in the registry record.

The study also registered secondary outcomes covering response, adverse events, drug consumption and HPV measures. Some response analyses were restricted to participants randomized to imiquimod or observation before closure of the imiquimod arm. The reason for that closure and additional details of its timing 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.

Before assignment: 142 patients were screened using HRA, clinical examination and laboratory tests (within 90 days) to ensure they meet the eligibility criteria. 91 patients were considered eligible and assigned to one of the three study arms.

Upto Week 20

Reasons for not completing this period:

Post-week 20

Reasons for not completing this period:

Cross-over Post-week 20 (Re-randomized)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: ITT 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.

Population: Intent-to-treat (ITT) population, which consists of all participants who were randomized

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants Achieving Complete Response in 5-FU Arm and Observation Arm Using ITT Population, measured over At week 20.

The primary endpoint is complete response at week 20 in the fluorouracil and observation arms. A participant counted as having complete response only if all biopsies lacked HSIL histology and cytology also lacked HSIL. This definition combines tissue and cytology findings, rather than relying on either alone.

The endpoint addresses whether the lesion findings meet the study’s prespecified complete-response definition at a fixed assessment time. A fixed-time binary outcome can be compared as the number of participants meeting the definition in each randomized group. The registry reports the counts and group sizes, as well as a between-group comparison.

The endpoint is distinct from complete or partial response, which is a secondary outcome. It is also distinct from longer-term persistence, new HPV infections, and adverse events. The week-20 assessment provides a time-specific result; it does not by itself describe response beyond that point. Because the observation group could cross over after week 20 if HSIL persisted and participants agreed, later follow-up may involve treatment exposure that differs from the initial assignment. Details of how that affects later outcome analyses are not reported in the registry record.

7. Results Posted in the Registry

• Arm B (Fluorouracil): 25.0% (9/36) versus 13.5% (5/37) on control; risk difference 11.5 percentage points (95% CI -6.4 to 29.4), risk ratio 1.85 (0.69–4.99); unadjusted two-proportion z-test p = 0.213.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Percentage of Participants Achieving Complete or Partial Response in 5-FU vs Observation Using ITT Population

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Arm B (Fluorouracil): 58.3% (21/36) versus 40.5% (15/37) on control; risk difference 17.8 percentage points (95% CI -4.8 to 40.4), risk ratio 1.44 (0.89–2.32); unadjusted two-proportion z-test p = 0.128.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Arm B (Fluorouracil) and the control. The interval gives the range of effects that remain compatible with the data.

Percentageof Patients Achieving Complete or Partial Response in Imiquimod vs Observation Arm, Using ITT Population Restricted to Those Randomized to Either Treatment Prior to the Closure of the Imiquimod Arm.

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Arm A (Imiquimod): 61.1% (11/18) versus 31.6% (6/19) on control; risk difference 29.5 percentage points (95% CI -1.2 to 60.3), risk ratio 1.94 (0.91–4.13); unadjusted two-proportion z-test p = 0.072.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Arm A (Imiquimod) and the control. The interval gives the range of effects that remain compatible with the data.

Amount of Drug Consumed in 5-FU and Imiquimod Arm by Week 16

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Arm B (Fluorouracil): mean 32.60 versus 39.50 on control; difference -6.90 (95% CI -13.55 to -0.25), standardized effect d = -0.47; z-test p = 0.042.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Arm B (Fluorouracil) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary outcomes cover several distinct questions: combined complete or partial response, drug consumption, adverse events, and HPV infection measures. The posted response comparisons generally point toward more complete-or-partial responses in the active-treatment groups than in observation, but the intervals are wide and the sponsor-reported p-values do not establish clear differences under a conventional threshold. The drug-consumption comparison is descriptive of the amount recorded by week 16, rather than an efficacy endpoint.

These secondary findings add context but do not replace the primary complete-response result. Complete or partial response uses a broader endpoint than complete response alone, and the imiquimod comparison uses a restricted population. Neither is a direct head-to-head comparison of the active treatments. HPV measures address infection patterns rather than the primary lesion-response definition. The registry record lists those outcomes but does not provide numerical results for all of them in the posted information.

Secondary results carry less weight when many outcomes or comparisons are examined without a reported multiplicity adjustment. In general, repeated testing increases the possibility of chance findings, and selective emphasis on a favorable result can be misleading. The registry lists 17 outcome measures and 13 analyses, but a multiplicity strategy is not reported in the registry record. Secondary p-values should therefore be interpreted as supporting, exploratory evidence rather than as confirmatory proof.

Differences in outcome definition, analysis population, and follow-up time also limit direct comparisons across secondary results. The primary question is specifically complete response at week 20 for fluorouracil versus observation. Secondary outcomes can help describe broader response and treatment experience, but they do not change what the primary endpoint tested.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Number of Participants Who Experienced an Adverse Event of Grade 1 to 5 by Week 44, Irrespective of Relatedness to the Intervention. AEs Were Stratified According to Those Reported at or Before Week 20 and After Week 20.
• Percentage of Participants Achieving Complete or Partial Response in 5-FU vs Observation Using ITT Population
• Percentageof Patients Achieving Complete or Partial Response in Imiquimod vs Observation Arm, Using ITT Population Restricted to Those Randomized to Either Treatment Prior to the Closure of the Imiquimod Arm.
• Amount of Drug Consumed in 5-FU and Imiquimod Arm by Week 16
• Percentage of Participants Achieving Complete or Partial Response in 5-FU vs Observation Using ITT Population
• Percentage of Patients Achieving Complete or Partial Response in Imiquimod vs Observation Arm, Using ITT Population Restricted to Those Randomized to Either Treatment Prior to the Closure of the Imiquimod Arm.
• Persistence and New Infections of HPV Type Specific Infections
• Comparison of the Number of hrHPV Genotypes in Each Arm Observed at Baseline vs at Week 20

9. Statistical Interpretation

10. Statistical Methods Explained

For a binary endpoint such as complete response, each participant is classified as meeting or not meeting the endpoint definition at the specified assessment. The group result can be expressed as a count alongside the number assessed. A comparison may then estimate the difference between the groups’ response proportions. The posted unadjusted risk difference is 0.115, with a lower interval limit of -0.064 and an upper limit of 0.294.

Why use a stratified comparison?

The registry specifies a stratified Mantel-Haenszel-Cochran test across sites. In general, stratification allows a comparison to account for predefined groups, such as study sites, rather than treating every participant as if site did not matter. The registry says that the comparison was to be made across sites, but it does not report the site-specific results, the number of strata, or the detailed calculation. The registry’s reported p-value for the primary comparison is 0.188.

A p-value describes how inconsistent the observed data are with a specified null hypothesis under the assumptions of the test. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of an effect. The reported primary p-value is to be read in relation to the stated one-sided alpha level of 0.025. The registry record does not report a separate confidence interval for the stratified estimate.

What does the recalculation represent?

The unadjusted recalculation uses the posted summary counts and group sizes, without adjustment for site. It gives a risk difference of 0.115 and p = 0.213. This simple calculation is useful as a descriptive check on the group-level data, but it is not the specified stratified primary method. The sponsor’s own analysis takes precedence. The two p-values need not match because the methods differ.

A risk difference is an absolute contrast between the response proportions in two groups, expressed on the proportion scale. Positive values in this comparison indicate a larger observed response proportion in the fluorouracil group; negative values indicate a larger observed response proportion in observation. The interval crosses zero, so the posted summary does not pin down the direction of the underlying difference with precision.

What can an interval communicate?

In general, a confidence interval communicates uncertainty around an estimated effect. A narrower interval indicates greater precision, while a wider interval indicates that a broader range of effects remains compatible with the data and method. The recalculated interval ranges from -0.064 to 0.294. The confidence level for this recalculated interval is not reported in the registry record, so it should not be inferred from the registry’s testing description.

An interval that includes zero does not prove there is no effect. It indicates that the data and calculation do not rule out zero at the interval’s stated level, while also allowing effects in either direction. A p-value and an interval are related summaries but answer different presentation needs: the p-value compares the result with a null hypothesis, while the interval displays uncertainty across an effect scale.

Why does the analysis population matter?

The primary outcome is described as using the intent-to-treat population. In general, an intent-to-treat analysis compares participants according to their randomized groups, regardless of treatment adherence or later treatment changes, when outcome data are available. The registry identifies the population but does not provide details about missing outcome handling, deviations, or the exact operational rules used. The posted primary group sizes are 37 for observation and 36 for fluorouracil, while overall enrollment is 91. The registry record does not explain that difference.

The study’s crossover design matters especially after week 20. Before crossover, the primary comparison is between initial assignment groups at the stated time point. For later outcomes, participants in observation could receive an active treatment after week 20 if HSIL persisted and they agreed to randomization. Analyses after that point may therefore answer a different question from the primary comparison of initial assignments. Specific methods for handling crossover in later analyses are not reported in the registry record.

Why does over-dispersion matter?

Over-dispersion is a general concern when observed variability exceeds what a statistical model assumes. It can make standard errors too small and intervals too narrow if the model does not accommodate the extra variability. For a binary comparison, the main posted uncertainty comes from the number of participants and the observed counts; the registry record does not report an over-dispersion assessment or a model that requires one. This concept is therefore general context, not a reported feature of this trial’s primary analysis.

The registry also reports several secondary outcomes and analyses. Multiple statistical tests raise a general multiplicity concern: when many comparisons are examined, the chance of at least one apparently unusual result can increase. No multiplicity adjustment for the secondary analyses is reported in the registry record. Secondary p-values should consequently be treated as exploratory unless a prespecified strategy is documented.

11. Confidence Intervals and Precision

The primary recalculated risk difference is 0.115, and its interval runs from -0.064 to 0.294. This span is broad relative to the point estimate and crosses zero. The observed result therefore does not determine whether the underlying response difference favors fluorouracil, favors observation, or is close to no difference. Both the magnitude and direction remain uncertain.

The interval is informative in a way that the p-value alone is not: it displays a range of effect estimates rather than a single test statistic. Its lower limit permits a negative risk difference, while its upper limit permits a positive one. These are not separate observed outcomes; they describe uncertainty around the recalculated estimate under its method. The registry does not state a confidence level for this unadjusted interval, so no confidence percentage should be attached to it.

The primary outcome counts are 9 of 36 and 5 of 37. Although these groups are close in size, the number of observed complete responses is limited. Binary endpoints can be imprecise when the number of participants and events is limited, because a small change in counts can have a noticeable effect on the estimated proportion. The registry record does not report a precision target or a sample-size justification for the original design.

The stratified sponsor analysis reports p = 0.188, but no interval for its adjusted effect estimate is reported. The interval is an unadjusted recalculation and should not be mistaken for an interval around the stratified estimate. More detailed site-level information would be needed to assess how stratification affects the estimated effect and its precision.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size illustrations are planning calculations based on the observed effect, not the trial’s original design assumptions. For these binary outcomes, the figures represent participants per arm, not a combined total. The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; each calculation uses the stated fraction of the observed effect as the effect to detect. The original trial’s sample-size assumptions and justification are not reported in the registry record.

The 80% power column describes the planned number of participants per arm under which the calculation would have an 80% chance of detecting the specified effect if that effect were present and the assumptions held. The 90% power column represents a higher chance of detection under the same conditions, and therefore generally requires more participants. These are planning illustrations, not observed power.

The rows consider planning for 100%, 75% and 50% of the observed effect. A smaller target effect is harder to distinguish from random variation, so more participants are needed. The figures depend on the observed effect in this trial, which is uncertain; they should not be treated as a recommendation or as evidence that the original study was designed to detect those effects. They also do not account for details not included in the stated planning assumptions, such as potential changes in allocation or missing outcomes.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.135; treatment proportion p₁ = 0.250; average p̄ = 0.193.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.5576 = 1.0930.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.5517 = 0.4643.
4. n = (1.0930 + 0.4643)² / (0.1149)² = 183.8, rounded up to 184 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
p0 <- 0.135135135135135                          # control proportion
p1 <- 0.25                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(2059499)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 184 (80% power), 246 (90% power)
 75% of observed effect: n per arm = 309 (80% power), 413 (90% power)
 50% of observed effect: n per arm = 651 (80% power), 871 (90% power)
power.prop.test: 183.8 
Simulated power at n = 184 per arm: 0.807 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

The registry’s primary analysis is a stratified Cochran-Mantel-Haenszel comparison across sites, with p = 0.188. The posted recalculation is an unadjusted comparison based on the summary counts and group sizes, with p = 0.213 and a risk difference of 0.115. The recalculation does not reproduce the site-stratified analysis, so the difference between the p-values is not evidence of a data inconsistency.

The sponsor’s own analysis takes precedence because the registry specifies a stratified primary method. It is the appropriate result to cite when describing the trial’s primary analysis. The unadjusted recalculation provides a transparent summary of the absolute difference and its interval, but it should be labeled as unadjusted and not substituted for the registry analysis.

The registry reports the sponsor’s p-value but does not report the adjusted effect estimate, its interval, or site-level results. Consequently, the size and precision of the stratified estimate cannot be reconstructed from the posted summary. The result should be described with both the primary method and its reported p-value, while recognizing that the simple recalculation is a different analysis.

15. Multiplicity, Missing Data and Analysis Populations

The registry identifies a primary endpoint and specifies a one-sided alpha level of 0.025 for the stratified comparison across sites. It also lists multiple secondary outcomes and analyses. When multiple hypotheses are tested, the chance of at least one nominally unusual result can rise even if none of the underlying effects is real. This is the general multiplicity problem.

A plan to control multiplicity might specify which comparisons are confirmatory, how testing order is set, or how the threshold is adjusted. Such a strategy for the secondary outcomes is not reported in the registry record. Without it, secondary p-values should not be interpreted as if each were the sole prespecified test. This does not make the results useless; it means they are less conclusive and should be viewed as exploratory evidence.

The registry names the intent-to-treat population for the primary endpoint and selected secondary comparisons. It does not report how missing outcome data were handled, how many outcomes were missing, whether any imputation was used, or how participants with protocol deviations were treated. These details matter because missingness can affect the estimated group results, particularly if missing outcomes differ systematically between groups.

The crossover design adds a further distinction between the week-20 primary comparison and later outcomes. Observation participants with persistent HSIL could cross over after week 20 if they agreed to randomization. The registry record does not report the analysis approach for later outcomes in relation to crossover, nor does it state whether any interim analyses were conducted. The original sample-size justification and full statistical analysis plan are also not reported in the registry record.

16. 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 serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: enterocolitis infectious (5.6%), hypoglycemia (5.6%), lung infection (5.6%), infections and infestations (bacteremia) (2.8%), arthralgia (2.8%).

The registry reports serious adverse events by treatment group and follow-up period. In the initial imiquimod group, 2 of 18 participants were reported with a serious adverse event through up to 44 weeks. In the fluorouracil group, 4 of 36 were reported with a serious adverse event through up to 44 weeks. In the observation group through week 20, none of 37 were reported with a serious adverse event. After week 20, the crossover and non-crossover observation subgroups each had no serious adverse events reported in their listed follow-up.

These groups have different exposure periods and, after week 20, different treatment histories. They should not be combined as though they were parallel groups observed over identical periods. The registry also lists serious adverse-event terms, including infectious enterocolitis, hypoglycemia, lung infection, bacteremia, arthralgia and thromboembolic event, with the recorded percentages. It does not establish from these entries whether a particular event was caused by study treatment.

Serious adverse events are distinct from all adverse events. The registry also reports other adverse events, but those counts should be interpreted separately. These data describe events recorded in the trial; they are not, on their own, a complete comparison of safety risks across treatments.

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

The primary result describes complete response at week 20 among participants assigned to fluorouracil or observation. The observed counts were 9 of 36 and 5 of 37, respectively. The registry’s stratified primary test reported p = 0.188, while the unadjusted recalculation estimated a risk difference of 0.115 with an interval from -0.064 to 0.294. The data are compatible with a range of differences, including values in either direction.

This result does not establish that fluorouracil and observation are equivalent. A non-significant test is not proof of no effect, especially when the estimate is imprecise. Conversely, the observed difference does not by itself establish a treatment benefit. The registry’s prespecified primary analysis and threshold should guide interpretation, and the sponsor-reported test does not meet the stated one-sided alpha level of 0.025.

The result also does not compare fluorouracil directly with imiquimod. The trial included an imiquimod arm, but its response analysis was separate and restricted to participants randomized before closure of that arm. Nor does the week-20 endpoint describe persistence, recurrence, longer-term response, HPV outcomes, or outcomes after crossover.

The registry reports adverse events and other secondary outcomes, but they answer different questions and have their own limitations, including multiple testing and differences in follow-up. The primary finding should therefore be understood narrowly: it is an uncertain comparison of a specific complete-response definition at a specific assessment time in the posted analysis population. It cannot by itself support a broad conclusion about comparative effectiveness or safety.

18. Why This Trial Matters Statistically

This trial illustrates why the analysis method and the outcome definition need to be read together. Complete response was defined using both biopsy histology and cytology, and the primary comparison was specified as a stratified test across sites. A simple comparison of group counts gives a useful descriptive estimate, but it is not interchangeable with the sponsor’s stratified analysis.

The result also shows why effect estimates and intervals matter alongside p-values. The sponsor-reported p-value alone does not communicate the range of possible differences. The unadjusted risk difference and its interval make the uncertainty visible: the estimate is positive, but the interval crosses zero and includes effects in either direction. This distinction helps prevent a binary significant/not-significant label from being mistaken for certainty.

The crossover design makes timing important. The week-20 primary outcome concerns the initial comparison, while later observation participants could receive active treatment. Follow-up beyond the primary assessment therefore has a different treatment context. The registry also reports several secondary outcomes, raising the general need to distinguish confirmatory analyses from exploratory comparisons when multiplicity control is not described.

Finally, overall enrollment and the primary outcome group sizes are not the same totals, and the registry record does not explain why. That is a reminder that the analysis population and outcome availability should be clearly documented. These features make the study a useful example of interpreting a modest binary endpoint with uncertainty, stratified analysis, multiple outcomes, and crossover, without treating any single summary statistic as the whole result.

19. 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 primary comparison has limited precision: the unadjusted interval ranges from -0.064 to 0.294 and crosses zero. The sponsor reports a stratified p-value, but an adjusted effect estimate and interval are not reported in the registry record. Site-level results are also not reported, so the contribution of stratification cannot be assessed from the posted summary.

The registry reports 91 enrolled participants, while the primary comparison includes 37 in observation and 36 in fluorouracil. The reason for the difference is not reported in the registry record. Missing outcome counts and the handling of missing data are also not reported, limiting assessment of whether the posted analysis reflects all randomized participants as intended.

The study included a crossover option after week 20 for observation participants with persistent HSIL who agreed to randomization. Later comparisons may therefore have different treatment exposure from the initial randomized groups. The registry does not describe the statistical handling of crossover for later outcomes. The reason and timing of imiquimod-arm closure are also not reported in the registry record.

Multiple secondary outcomes are listed, but a multiplicity strategy is not reported. Some secondary comparisons use restricted populations, and the registry record does not include numerical results for every listed outcome. The original sample-size rationale, interim analyses, and full statistical plan are not reported in the registry record. These omissions limit evaluation of the intended precision, analysis choices, and the confirmatory status of secondary findings.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Imiquimod, Fluorouracil, or Observation in Treating HIV-Positive Patients With High-Grade Anal Squamous Skin Lesions — NCT02059499.
• Registry record last updated 2025-07-23; status last verified 2025-07.
• Results first posted 2025-05-20.
• 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.

23. Record Summary

NCT02059499 is a completed randomized phase III study sponsored by the AIDS Malignancy Consortium. It enrolled 91 HIV-positive men and women with high-grade anal squamous intraepithelial lesions and compared imiquimod, fluorouracil and observation, with crossover possible after week 20 for eligible observation participants.

The primary endpoint was complete response at week 20, defined by absence of HSIL histology in all biopsies and absence of HSIL cytology. In the fluorouracil and observation groups, the recorded counts were 9 of 36 and 5 of 37. The registry’s stratified primary analysis reported p = 0.188; the unadjusted recalculation is imprecise, with a risk difference of 0.115 and an interval from -0.064 to 0.294.

Secondary outcomes include combined response, adverse events, drug consumption and HPV measures. The registry record does not report the original sample-size justification, missing-data handling, or a multiplicity strategy for secondary analyses.