Analytical Treatment Interruption in HIV Positive Patients: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Analytical Treatment Interruption in HIV Positive Patients (NCT02590354), a na na study. The primary outcome is Assessment of the Number of Participants With a HIV Plasma Viral Load Below the Lower Limit of Detection 48 Weeks Following Interruption of Antiretroviral Treatment.

2. Clinical Question

3. Trial Design

This was a completed, single-group study of analytical treatment interruption in HIV-1 infection. The registry lists allocation as not applicable and masking as none; its summary says randomization was not foreseen. There is no randomized control group, so the study is not a like-for-like comparison of treatment interruption against continued therapy.

The study planned a two-step screening process. People with normal peripheral blood CD4+ T-cell counts and an undetectable viral load were to be recruited in four Belgian HIV reference centers. Viral reservoir measurement was part of screening, and treatment interruption was for selected patients with a very low viral reservoir. The intervention is described as interrupting longstanding antiretroviral therapy.

The registry lists enrollment as 114. That is the study enrollment figure; it should not be read as the number who entered the treatment interruption phase. The primary analysis population is described as all subjects enrolled in phase 2, but the number enrolled in that phase is not reported in the registry record.

After interruption, the planned follow-up included clinical and laboratory assessments during 48 weeks and a further 12 weeks after the intervention. The primary endpoint is assessed at 48 weeks. Secondary outcomes include intervention-related adverse events, viral reservoir measurements at baseline and viral rebound, and the kinetics of viral-load rebound. The registry lists the study as completed, with start and primary completion dates of 2016-01-25 and 2018-06-22. The statistical analysis plan and the reasons for the selected design 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.

Phase 1

Reasons for not completing this period:

Phase 2 - Treatment Interruption

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Analysis population contains all subjects enrolled in phase 2 (treatment interruption phase)

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: 114 participants were included in phase 1 of the study. 16 of those continued to phase 2 (treatment interruption).

6. Primary Endpoint

The registry lists the primary outcome as Assessment of the Number of Participants With a HIV Plasma Viral Load Below the Lower Limit of Detection 48 Weeks Following Interruption of Antiretroviral Treatment, measured over 48 weeks after treatment interruption.

The primary endpoint is the number of participants whose HIV plasma viral load is below the lower limit of detection 48 weeks after antiretroviral treatment interruption. The record defines the threshold as less than 50 HIV RNA copies/ml of plasma. Participants below that threshold at the assessment are classified as post-treatment controllers, under the record’s description of people receiving antiretroviral therapy at baseline, with low peripheral blood proviral DNA, who maintain viral suppression at 48 weeks after interruption.

This endpoint uses a clinically interpretable laboratory measure at a specified follow-up time. A count describes how many participants met the stated criterion; it is not, by itself, a comparison of interruption with continued therapy. The repeated viral-load measurements are also intended to support assessment of rebound over follow-up, but the primary endpoint is the status at 48 weeks.

The registry describes measurements every two weeks, or every four weeks after W12, until W48. It does not report the primary endpoint count or the denominator used in the posted result. Accordingly, the proportion meeting the criterion, and the precision of any estimated proportion, cannot be described from the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Several statistical ideas help clarify what this study was designed to measure and what the registry record can support. The distinctions below are general explanations, not claims about analyses that were carried out unless the registry record says so.

What does a count endpoint describe?

A count endpoint reports how many participants meet a defined condition. Here, the condition is a plasma viral load below the lower limit of detection at 48 weeks after treatment interruption. A count is different from a rate or proportion: interpreting a count in relation to the number assessed requires a denominator. The primary outcome description names the phase 2 analysis population, but the result and denominator are not reported in the registry record.

A threshold can make an outcome easier to interpret, but it also reduces a range of laboratory values to categories: below the limit or not below it. Participants near the threshold may be classified differently from participants with much higher viral loads, even though the difference in their measurements could be small. The registry’s definition uses less than 50 HIV RNA copies/ml of plasma as the threshold.

What would a confidence interval add?

In general, a confidence interval expresses the uncertainty around an estimate under a specified statistical model. For a count-based outcome, an interval around a proportion would help readers judge how precisely the observed proportion estimates the underlying proportion in a population like the one studied. An interval does not remove bias from selection or replace a control group. No confidence interval for the primary endpoint is reported in the registry record, and the missing result prevents describing its precision.

Why do repeated measurements matter?

The protocol description calls for repeated plasma viral-load measurements during follow-up, with measurements every two weeks or every four weeks after W12 until W48. Repeated measurements can describe when viral rebound occurs and how the measured viral load changes over time. Such trajectories answer a different question from whether a participant is below the detection threshold at the specific 48-week assessment.

Repeated observations from the same participant are generally related rather than independent. Statistical methods for longitudinal data account for that correlation when estimating patterns over time. If measurements are unevenly spaced, missed, or unavailable after a participant leaves follow-up, methods also need to address how those data contribute to the analysis. The registry record does not report the longitudinal model or the handling of missing measurements.

What does a single-group design allow?

In general, a single-group study can describe what happened in the group receiving the intervention. Random assignment to a comparator can help balance measured and unmeasured characteristics between groups, supporting a causal comparison. No randomization or control group is described. Any result following interruption is therefore a description of outcomes in the selected study group, not a randomized estimate of the effect of interruption relative to another treatment strategy.

Selection is particularly relevant because participants were screened and only those with a very low reservoir were intended to undergo interruption. This design may be appropriate for studying a selected population, but the results would apply most directly to people who resemble those selected. The registry does not report enough information to characterize the screening yield or compare those who entered phase 2 with those who did not.

How should planning calculations be read?

In general, a sample-size calculation is a planning tool based on assumptions about the outcome, effect to detect, allocation, and statistical test. It is not an observed result and does not establish the precision of the completed study. The registry’s planning description refers to a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It also describes the effect to detect as a stated fraction of the observed effect. These assumptions do not match the described single-group, nonrandomized design, so any such illustration should not be treated as a study-specific design justification or as observed power. The statistical rationale for the study is not reported in the registry record.

Power is a design-stage probability under assumed conditions, not a measure of whether a completed study has demonstrated a finding. A post hoc calculation based on an observed effect would not resolve uncertainty or substitute for a confidence interval. No numerical planning values or primary outcome estimate are reported in the registry record.

What is over-dispersion?

For count outcomes, some statistical models assume that variability follows a particular relationship to the expected count. Over-dispersion means the observed variability is greater than that model assumes; ignoring it can make standard errors too small and intervals too narrow. Whether over-dispersion is relevant depends on the outcome and model. The primary outcome is described as a number of participants meeting a binary threshold, and the registry does not report a model or dispersion assessment. This general concept should not be mistaken for a reported finding about this study.

The registry lists five outcome measures and zero analyses. It does not report an analysis plan, model specification, treatment of repeated measures, or inferential procedure for the primary endpoint. Statistical methods beyond the outcome definitions and planning assumptions are therefore not reported in the registry record.

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("Treatment Interruption")
n      <- c(16)
events <- c(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  Treatment Interruption  16  0  0.0  0.0  20.6  0.0  19.4  0.1  19.5

11. Multiplicity, Missing Data and Analysis Populations

The registry lists five outcome measures, including the primary 48-week viral-load threshold and secondary outcomes concerning adverse events, viral reservoir measurements, and viral-load rebound kinetics. When a study examines multiple outcomes or multiple analyses, the chance of at least one apparently unusual finding can increase. Statistical approaches may address this through a prespecified hierarchy, adjusted significance thresholds, or clear emphasis on estimation rather than multiple hypothesis tests. No multiplicity strategy is reported in the registry record.

The record lists zero analyses. It does not provide outcome-specific estimates, confidence intervals, p-values, or an analysis plan. It is therefore not possible to determine whether the listed outcomes were analyzed as planned, whether any adjustment for multiple comparisons was used, or how exploratory analyses were distinguished from confirmatory ones.

The primary analysis population is defined as all subjects enrolled in phase 2, the treatment interruption phase. The number in that population is not reported. The registry does not report how participants with missing viral-load measurements, withdrawal, or other unavailable outcome data were handled. It also does not state whether the primary analysis used all phase 2 participants as classified, a subset with an observed W48 measurement, or another approach.

In general, missing data can affect an estimate when the likelihood of missingness is related to a participant’s outcome or clinical course. Complete-case analyses, imputation, and sensitivity analyses make different assumptions; none can be attributed to this study based on the registry record. The reasons for missing data and any methods to address them are not reported in the registry record.

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 safety figures for the Treatment Interruption group. For serious adverse events, 16 participants are listed as at risk and 0 as affected; deaths are listed as 0. For other adverse events, 16 participants are listed as at risk and 13 as affected. These figures are counts of participants, not a total number of event episodes. The registry does not list top serious adverse-event terms.

The outcome description says adverse events related to the study intervention were graded according to NCI CTCAE Version 4.0. The registry figures should be read as reported counts within the treatment interruption group, rather than as a comparison against a control group. Because there is no randomized comparator, they cannot establish how these counts compare with what would have occurred under continued therapy.

The record does not provide event timing, duration, severity details for the other adverse events, or participant-level descriptions. The zero serious-event and death counts do not provide a complete account of all safety outcomes, and they should not be interpreted as proof that no safety concerns occurred. The safety figures are limited to the reported categories and the participants listed as at risk.

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

The study addresses a specific question: whether selected participants with a very low viral reservoir remain below the stated plasma viral-load detection threshold 48 weeks after antiretroviral treatment interruption. The endpoint is a defined status at a follow-up time, and the study also planned repeated viral-load and viral-reservoir assessments.

The registry record does not report the primary endpoint count, its denominator, or an uncertainty interval. It therefore cannot show how many participants met the post-treatment controller definition or how precisely that outcome was estimated. The record’s completed status does not itself provide the missing outcome result.

Even if a primary result were available, the single-group, nonrandomized design would describe outcomes among a selected group rather than provide a randomized comparison with continued antiretroviral therapy. Selection based on screening for a very low viral reservoir also limits how broadly findings could be generalized. The registry does not report the screening yield or how many participants entered phase 2.

The endpoint classification at W48 would not, by itself, explain the full course of viral load during follow-up. Repeated measurements could add information about rebound timing and patterns, but those outcome results are not reported in the registry record. Likewise, the listed safety counts summarize reported categories, but cannot be used as comparative safety estimates without a control group.

Thus, the registry record describes a focused study design and a threshold-based primary question, but it does not contain the primary result needed to characterize the observed outcome. It cannot support a conclusion about treatment efficacy, a causal effect, or the likelihood of maintaining suppression in people outside the selected study population.

14. Why This Trial Matters Statistically

This study is statistically notable for asking a tightly defined outcome question in a selected group while using a single-group design. The primary endpoint is a count of participants below a laboratory threshold at 48 weeks after interruption. That format is straightforward to communicate, but its interpretation depends on knowing both the count and the population assessed. The registry record does not report those primary result details.

The distinction between a treatment-interruption cohort and a randomized comparison is central. Outcomes observed after interruption can describe what happened in the enrolled group, but cannot by themselves estimate what would have happened to comparable participants continuing therapy. This is a design limitation for causal interpretation, not a flaw in the descriptive question.

The study also separates a specific time-point endpoint from longitudinal questions. Repeated viral-load measurements and reservoir quantification could address rebound timing and changes in the reservoir, which are different statistical targets from a single W48 classification. Keeping those questions distinct helps avoid treating one endpoint as a complete description of viral dynamics.

Finally, the record illustrates why a completed study entry and an outcome definition are not equivalent to a reportable estimate. Without the result, denominator, analysis approach, and uncertainty, readers cannot assess the magnitude or precision of the primary outcome. Transparent reporting of those elements is necessary for interpreting even a well-defined endpoint.

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 does not report the primary endpoint result, its denominator, or a confidence interval. It also does not report the number who entered phase 2, the number screened, or how many met the very-low-reservoir criterion. The overall enrollment figure of 114 should not be taken as the treatment-interruption analysis population.

The study has no randomized comparator, and the record describes no randomization. As a result, outcome descriptions after interruption cannot provide a controlled estimate of the effect of interruption relative to continued antiretroviral therapy. The selected screening population further limits generalizability beyond people meeting the study’s criteria.

The record describes repeated viral-load measurement but does not report the statistical methods for analyzing longitudinal outcomes, handling irregular or missing measurements, or estimating rebound kinetics. It does not describe missing-data patterns, participant withdrawals, or sensitivity analyses. The statistical analysis plan, model choices, and multiplicity approach are not reported in the registry record.

The listed planning assumptions refer to equal allocation between two arms, whereas the study is described as single-group and nonrandomized. No numerical planning results or study-specific sample-size justification are reported. The safety information is also limited: serious-event terms are not listed, and the registry does not provide timing or detailed descriptions for the other adverse events. These omissions constrain interpretation of both efficacy-related outcomes and safety.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Analytical Treatment Interruption in HIV Positive Patients — NCT02590354.
• Registry record last updated 2019-09-24; status last verified 2019-09.
• Results first posted 2019-09-24.
• 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 describes a completed, single-group study of analytical antiretroviral treatment interruption in people with HIV-1 infection and a very low viral reservoir after screening. Allocation is listed as not applicable, masking as none, and randomization was not foreseen. Overall enrollment is listed as 114.

The primary endpoint is the number of participants with plasma viral load below the lower limit of detection, defined as less than 50 HIV RNA copies/ml, 48 weeks after interruption. The primary result and its denominator are not reported in the registry record. The record lists 0 serious adverse events among 16 at risk, 13 participants with other adverse events among 16 at risk, and 0 deaths. The analysis plan, missing-data methods, and primary endpoint precision are not reported.