Immune Reconstitution of Lopinavir/Ritonavir-Based vs Efavirenz-based HAART in Advanced HIV Disease: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Immune Reconstitution of Lopinavir/Ritonavir-Based vs Efavirenz-based HAART in Advanced HIV Disease (NCT00775606), a randomized phase 4 study. The primary outcome is CD4+ (Cluster of Differentiation 4) T-cell Apoptosis.

2. Clinical Question

The broader clinical question in the registry summary is whether the regimens differ in immune enhancement for people with advanced HIV disease. The summary notes interest in immune parameters beyond the primary endpoint, including responses to tetanus-diphtheria and 23-valent pneumococcal vaccines. These are secondary outcomes, not replacements for the specified primary measure. The registry describes the trial as phase 4, randomized, and parallel, with no masking. Its status is terminated. The record does not state why the trial was terminated, nor does it describe whether or how the termination affected planned follow-up or analyses.

3. Trial Design

This was a randomized, parallel-group phase 4 study comparing two antiretroviral regimens. The registered enrollment is 15. Participants were assigned to a lopinavir/ritonavir-based regimen or an efavirenz-based regimen. The study was unmasked, meaning the record does not describe masking of participants or study personnel. No further details about the randomization procedure are reported in the registry record.

The lopinavir/ritonavir arm combined lopinavir 400 mg/ritonavir 100 mg BID with emtricitabine 200 mg/tenofovir 300 mg QD. The efavirenz arm combined efavirenz 600 mg with emtricitabine 200 mg/tenofovir 300 mg QD. The registry summary describes the target population as HIV-positive people starting antiretroviral treatment because their CD4+ T-cells were below 200. It also notes a usual treatment threshold of a CD4+ T-cell count below 350, but does not provide individual participant characteristics.

The primary outcome was change in the percentage of naive CD4 T-cells undergoing apoptosis, measured from baseline to week 24. The stated timeframe is 24 weeks from treatment initiation. Other registered outcomes concern CD4+ T-cell change and subsets, activation and proliferation of CD4+ and CD8+ T-cells, and vaccine responses. The record lists six posted secondary outcomes, although one outcome description appears twice in the registry information. It does not provide a detailed schedule for measurements beyond the primary endpoint timeframe.

The registry lists a start date of 2008-10 and a primary completion date of 2010-12. The study status is terminated. The registry record does not say when or why termination occurred, whether all intended assessments were completed, or whether termination followed a protocol-defined decision. Posted outcome summaries list n=5 in each primary outcome arm. The record does not explain how those arm summaries relate to the registered enrollment of 15, and it does not report the analysis population or participant flow.

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: Two subjects withdrew participation prior to starting study

Overall Study

Reasons for not completing this period:

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 CD4+ (Cluster of Differentiation 4) T-cell Apoptosis, measured over 24 weeks from treatment initiation (baseline and week 24).

The primary endpoint is change in the percentage of naive CD4 T-cells undergoing apoptosis. Apoptosis is cell death; the registry describes the measurement as a laboratory test using propidium iodide staining. The change is defined as the week 24 percentage minus the baseline percentage. Consequently, a negative mean change describes a lower measured percentage at week 24 than at baseline, on average within that arm. This direction describes the endpoint scale, not a causal treatment effect or proof of a difference between groups.

The endpoint focuses on one cellular feature of immune reconstitution. Measuring naive CD4 T-cell apoptosis can provide information about the survival of this cell subset over the treatment period. It is not a direct summary of every component of immune function, clinical health, or response to infection. The study also included other immune measures, including T-cell subset frequencies and vaccine responses, but these are distinct outcomes.

The primary outcome is continuous and is summarized in the registry as a mean with a standard deviation. The posted arm summaries report mean changes of -12.33 per cent for lopinavir/ritonavir and -8.01 per cent for efavirenz, with standard deviations of 12.34 and 7.97, respectively. Those summaries describe the observed changes within the two groups. The registry record does not report a confidence interval or a formal between-group analysis for this endpoint. Therefore, the precision and statistical uncertainty of the treatment contrast cannot be judged from the reported arm means and standard deviations alone.

The endpoint is relevant to the study's stated immune-reconstitution question, but it is a surrogate laboratory measure. The registry does not establish how a change in this particular measure corresponds to longer-term clinical outcomes.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary measure is continuous: it records a percentage change in a laboratory measurement. The registry summarizes each arm using a mean and standard deviation. A mean gives the arithmetic average of observed changes within a group. A standard deviation describes how spread out individual observations are around that mean. It is not a confidence interval and does not directly express how precisely the mean or the between-group contrast has been estimated.

What does the change score represent?

The registry defines the change as the percentage of naive CD4 T-cells undergoing apoptosis at week 24 minus the baseline percentage. This subtraction establishes the direction of the scale: negative values correspond to lower measurements at week 24 than at baseline. A group mean change condenses the observed participant-level changes into one average. It does not show the full distribution, whether changes were similar across participants, or whether any particular participant experienced a change in the same direction.

Change scores can be useful when a study measures participants before and after treatment, because the outcome directly reflects movement from baseline. However, an average change within a group is not itself the randomized treatment comparison. The treatment question concerns whether the change differs between the assigned groups. That requires an estimate of the between-group contrast and its uncertainty. Such an estimate is not reported in the registry record.

How should the arm summaries be read?

The reported mean changes are -12.33 per cent in the lopinavir/ritonavir arm and -8.01 per cent in the efavirenz arm. The corresponding standard deviations are 12.34 and 7.97. Each arm summary lists n=5. These are descriptive statistics, and the standard deviations show that there is participant-level variation around the reported averages. With small posted arm counts, a few observations can have substantial influence on a mean. The registry does not provide participant-level values, medians, ranges, or graphical distribution details in the record details.

A confidence interval is a range calculated by a statistical procedure to express uncertainty around an estimate. For a randomized comparison, a confidence interval around the arm contrast would help readers assess which effect sizes are compatible with the data under the analysis assumptions. No confidence interval for the primary comparison is reported. A p-value is also not reported; such a value, when calculated under a specified model, evaluates compatibility between data and a null hypothesis. It would not measure the probability that either treatment is effective or that the null hypothesis is true.

Why does the design matter?

Random allocation is intended, in general, to make assigned groups comparable on average with respect to measured and unmeasured baseline factors. A parallel design assigns participants to one regimen rather than switching them between regimens. Here, the study is listed as randomized and parallel, but the registry record does not report the allocation sequence, concealment procedures, stratification factors, or details of an analysis model. The trial was unmasked, so participants and study personnel were not described as blinded. Laboratory outcome procedures may have their own safeguards, but those are not reported in the registry record and should not be assumed.

For a continuous endpoint, a comparison could be analyzed in several ways, such as comparing change scores directly or modeling the week 24 measure while accounting for baseline. These general approaches can yield different estimates and require a prespecified analysis strategy. The registry record does not report which approach was planned or used, how missing measurements were handled, or whether the analysis adjusted for other variables. No analysis result is listed for the primary outcome, and the registry analysis count is zero.

What are the planning illustrations?

Any sample-size planning illustration based on these observed summaries is a planning exercise, not a report of what the trial was designed to detect. The stated assumptions for such illustrations are a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. The assumed effect is a stated fraction of the observed effect. The registry record does not report the trial's own sample-size justification, planned target effect, or statistical design calculation. Planning values should therefore be read only as conditional illustrations under those assumptions, not as the original protocol's requirements.

A planned sample size is not observed power. Observed power is not a substitute for reporting an estimated effect and its confidence interval; it largely restates the observed data under a particular model. No numerical sample-size planning result or observed-power analysis is reported in the registry entries. No sample-size value should be interpreted as a participant count actually analyzed or recruited beyond the registered enrollment and the stated arm summaries.

Why can variability and missingness matter?

Standard deviations reflect individual variation. When outcomes are more variable than anticipated, a study generally has less precision for a fixed amount of information. If observations are missing, the consequences depend on how much is missing, why it is missing, and how the analysis handles it. These are general statistical considerations. In this trial, the registry does not report the amount or reasons for missing primary measurements, an imputation method, or sensitivity analyses. The difference between registered enrollment and the endpoint-specific arm summaries is not explained, so it cannot be attributed to any particular missing-data mechanism.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("ARM A/Lopinavir-ritonavir", "ARM B/Efavirenz")
n   <- c(5, 5)
m   <- c(-12.33, -8.01)
sp  <- c(12.34, 7.97)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  ARM A/Lopinavir-ritonavir  5  -12.330  12.340  -27.652  2.992
CARM  ARM B/Efavirenz  5  -8.010  7.970  -17.906  1.886

11. Multiplicity, Missing Data and Analysis Populations

The registry identifies one primary endpoint: change in naive CD4 T-cell apoptosis from baseline to week 24. It also lists secondary outcomes covering CD4+ T-cell change, T-cell subset frequencies, activation and proliferation, and vaccine responses. The record lists six posted secondary outcomes, with a repeated description among the listed outcomes. Considering multiple outcomes can increase the chance of finding at least one apparently notable result by chance if many tests are conducted. This is a general statistical concern; the registry record does not report whether multiplicity adjustments were planned or used.

No detailed statistical analysis plan is reported in the registry record. In particular, it does not state the test or model for the primary endpoint, a testing hierarchy, whether secondary analyses were confirmatory or exploratory, or how multiplicity was handled. The record contains no analysis entries. Therefore, there are no posted inferential results to assess for multiplicity-adjusted or unadjusted interpretation.

The primary endpoint summaries report n=5 for each arm, while registered enrollment is 15. The record does not reconcile these figures. It does not report participant flow, reasons for non-inclusion in the outcome summaries, completeness of baseline and week 24 measurements, or the handling of missing data. It is also not reported whether the primary analysis included all randomized participants, only participants with observed measurements, or another population. No intention-to-treat or per-protocol population is specified.

These omissions matter because missingness and analysis-population choices can affect both the estimated arm changes and the validity of a comparison. Without information about why measurements are unavailable and how they were handled, readers cannot assess whether the posted arm means represent all randomized participants. No assumption about the causes of missing data or the analysis population is warranted from the registry record.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports serious adverse event figures separately by arm. In the lopinavir/ritonavir group, the serious-event safety population is listed as 8 at risk and 0 affected. In the efavirenz group, the corresponding figures are 5 at risk and 0 affected. These are the recorded group counts; they are not evidence that the treatments have identical safety profiles, particularly given the limited numbers recorded and the absence of a comparative safety analysis.

The record also gives other adverse event figures, separate from serious adverse events: 8 at risk and 1 affected in the lopinavir/ritonavir group, and 5 at risk and 0 affected in the efavirenz group. The registry does not provide the event term for the listed other event in the registry record for this summary. The serious-event term list is empty. Death counts are not reported in the registry record.

Counts should be interpreted alongside the stated safety populations and the duration of observation. The record does not explain how safety populations were defined, how events were collected, or whether the groups had the same duration of exposure and follow-up. It also does not report a formal between-group comparison of adverse events. These figures are descriptive registry counts, not a comprehensive basis for comparing safety.

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

The posted primary outcome shows negative mean changes in naive CD4 T-cell apoptosis in both treatment groups over the stated baseline-to-week-24 interval. The reported means are -12.33 per cent for lopinavir/ritonavir and -8.01 per cent for efavirenz. The standard deviations are 12.34 and 7.97. These statistics describe the observed arm summaries and the spread of changes within each arm. They do not, by themselves, quantify a causal difference between regimens.

The trial was randomized, which is a design feature intended to support a fair comparison. But a useful comparative conclusion also requires an appropriate analysis of outcomes and uncertainty. The registry record does not report a treatment contrast, confidence interval, or p-value for the primary endpoint. Without these, readers cannot assess the magnitude and precision of the between-group difference using the sponsor's own analysis. An inference should not be made by treating the separate within-arm changes as though they were a direct comparison.

The endpoint is one laboratory measure of immune reconstitution. It is not a complete measure of immune function, and it is not a clinical endpoint describing symptoms, illness, or long-term outcomes. The registered secondary measures broaden the scientific questions, including T-cell characteristics and vaccine responses, but no results for those analyses are listed. The primary endpoint alone cannot answer all the questions raised in the study summary.

The registered enrollment is 15 and the primary arm summaries each list n=5. The record does not explain the relationship between those figures or report an analysis population. It also does not give a reason for the terminated status. These gaps make it difficult to evaluate the completeness and representativeness of the outcome summaries. They do not establish that the results are invalid, but they restrict the conclusions that can be drawn.

Accordingly, the registry data support a descriptive statement about the reported mean changes, not a claim that one regimen is superior, equivalent, or ineffective. They do not establish that either regimen improves clinical outcomes or that changes in apoptosis translate into better immune protection. The result should be understood as an incomplete record of a small randomized comparison with posted descriptive primary outcome statistics and no reported inferential comparison.

14. Why This Trial Matters Statistically

This trial illustrates why a randomized label and arm-level means are not sufficient on their own for interpreting a treatment comparison. Randomization describes how treatment assignment was intended to occur. To understand the comparative result, readers also need a clearly reported estimate of the difference between groups, an uncertainty measure, and information about who contributed outcome data. The registry record supplies the arm summaries but does not report those inferential details for the primary endpoint.

The distinction between within-group and between-group change is especially important in pre-post studies. A negative mean change in both arms can describe improvement on the measured scale within each group, yet it does not reveal whether changes differ between the randomized regimens. Natural variation over time, measurement variation, and other factors can contribute to within-arm changes. The randomized comparison, analyzed according to a stated method, is the relevant estimate for the comparative question.

The enrollment and endpoint summary counts also show why participant flow and analysis populations matter. The registry records enrollment of 15, while the primary outcome summaries list n=5 in each group. Without an explanation, readers cannot determine how the analyzed observations relate to the enrolled group or assess the potential implications of missing outcomes. This is a reporting issue, not a basis for guessing what happened.

Finally, the trial's multiple immune outcomes make transparent prespecification and multiplicity handling important. The record identifies a primary measure and several secondary measures, but does not report the analysis plan or analysis results. Statistical interpretation depends not only on the numerical summaries but also on knowing which questions were designated in advance and how uncertainty was evaluated. The gaps are instructive: precise reporting of contrasts, confidence intervals, analysis populations, missingness, and outcome hierarchy is essential for readers to judge the evidence.

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 primary outcome summaries are limited to arm means, standard deviations, and counts. A between-group estimate and confidence interval are not reported in the registry record, nor is a p-value. The record therefore does not permit assessment of the precision of the comparative effect. The summaries are descriptive rather than a complete account of the randomized analysis.

The endpoint-specific counts are n=5 in each arm, whereas registered enrollment is 15. No participant flow diagram or explanation of this relationship is reported. It is not reported whether participants withdrew, lacked endpoint data, were excluded, or entered a different analysis population. No missing-data approach, sensitivity analysis, or definition of the primary analysis population is reported.

The registry lists the trial as terminated but does not state the reason. It is not reported whether the planned study duration, assessments, or analysis procedures were completed. The record also lacks details of the randomization method, concealment, laboratory masking, and the statistical model. As the study was unmasked, the record does not support an assumption that treatment assignment was concealed from participants or study staff during follow-up.

Several secondary immune outcomes are listed, but their analyses are not reported. The record does not state how multiple outcomes were handled statistically or identify an analysis hierarchy. This limits interpretation of the broader immune-reconstitution aims. The primary measure itself is a laboratory endpoint; the registry does not report its relationship to clinical outcomes or longer-term follow-up.

Safety information is also limited to counts. Serious-event terms are not listed, deaths are not reported in the registry record, and no comparative safety analysis is included. The duration and completeness of safety observation are not described. These are limitations of the registry information and should not be resolved by assumptions about the conduct or findings of the trial.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Immune Reconstitution of Lopinavir/Ritonavir-Based vs Efavirenz-based HAART in Advanced HIV Disease — NCT00775606.
• Registry record last updated 2023-05-26; status last verified 2023-05.
• Results first posted 2014-05-29.
• 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 phase 4, randomized, parallel, unmasked study of lopinavir/ritonavir-based versus efavirenz-based antiretroviral therapy in antiretroviral-naïve subjects with advanced HIV disease. The registered enrollment is 15, and the status is terminated. The registry lists a start date of 2008-10 and a primary completion date of 2010-12.

The primary outcome is change in the percentage of naive CD4 T-cells undergoing apoptosis from baseline to week 24. The posted mean change is -12.33 per cent, with standard deviation 12.34, in the lopinavir/ritonavir arm, and -8.01 per cent, with standard deviation 7.97, in the efavirenz arm. Each arm summary lists n=5. The registry does not report a primary between-group analysis, confidence interval, or p-value.

The record also lists secondary immune outcomes and descriptive adverse event counts. It does not explain the difference between registered enrollment and primary outcome counts, report an analysis population or missing-data method, or state why the trial was terminated. These omissions limit interpretation of the posted summaries and the broader study questions.