Safety and Immunotherapeutic Activity of an Anti-PD-1 Antibody (Cemiplimab) in Participants With HIV-1 on Suppressive cART: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Safety and Immunotherapeutic Activity of an Anti-PD-1 Antibody (Cemiplimab) in Participants With HIV-1 on Suppressive cART (NCT03787095), a randomized phase 1, phase 2 study. The primary outcome is Count of Participants With a Grade >=3 Adverse Event (AE) or Grade >=1 Immune-related AE (irAE) Related to Study Treatment.

2. Clinical Question

The study was designed to evaluate safety and immunotherapeutic activity of cemiplimab in this population. Its primary endpoint is a safety count, not a measure of HIV control or a direct measure of immunotherapeutic activity. The record also lists secondary outcomes involving changes in HIV-1 Gag-specific CD8+ T cells, including responses assessed using intracellular staining for IFNg, CD107a, IL-2, and TNFa.

The registered comparison is cemiplimab versus placebo in Cohort 1. The record describes random allocation and double masking, but the small enrollment means that the observed counts are descriptive and provide limited information about how outcomes might vary in a broader population. The results answer a narrow question about the recorded qualifying events during the specified follow-up, not a general question about all possible benefits or harms.

3. Trial Design

NCT03787095 is a phase I/II, randomized, double-blind, placebo-controlled study with a sequential model. The official title describes an ascending multiple-dose study. The registered arms are Cohort 1: Cemiplimab and Cohort 1: Placebo. Four participants were assigned to the cemiplimab cohort and one to placebo, for an enrollment of five.

In Cohort 1, cemiplimab was administered at 0.3 mg/kg on Day 0 and Week 6, for two infusions. Placebo was administered on the same schedule. Participants in both groups continued their current non-study provided ART regimen. The record identifies the allocation as randomized and masking as double; details of the randomization procedure and of who was masked are not reported in the registry record.

The study started on 2019-08-13, and primary completion was recorded on 2020-08-18. Its status is terminated. A reason for termination is not reported in the registry record. The record names the National Institute of Allergy and Infectious Diseases (NIAID) as sponsor.

The primary outcome was assessed from study entry through Week 48 or premature discontinuation. The trial also registered secondary cellular immune-response outcomes. Four secondary outcomes are listed as posted, covering changes in HIV-1 Gag-specific CD8+ T-cell responses from baseline to post-baseline and after each infusion. The record contains six outcome measures in total. The protocol details for any later cohorts or dose escalation, including how escalation decisions were made, are not reported in the registry record.

With five participants overall and only one in the placebo cohort, the arm counts should be read as a very small clinical-trial description. Randomization and masking describe design features; they do not make estimates precise when the number of participants is this limited.

4. Participant Flow and Accounting

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

Overall Study

Analysis population for the primary outcome, as defined in the registry: All participants

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: All participants

6. Primary Endpoint

The registry lists the primary outcome as Count of Participants With a Grade >=3 Adverse Event (AE) or Grade >=1 Immune-related AE (irAE) Related to Study Treatment, measured over Study Entry through Week 48 or premature discontinuation.

The primary endpoint counts participants who experienced a Grade >=3 adverse event or a Grade >=1 immune-related adverse event (irAE) judged related to study treatment. The outcome applies to all participants and covers study entry through Week 48 or premature discontinuation. Treatment relationship was judged by the Clinical Management Committee (CMC), which was blinded to treatment arm.

This endpoint combines two event criteria. One concerns an adverse event at Grade >=3; the other concerns an immune-related adverse event at Grade >=1. The record says that irAEs include, but are not limited to, pneumonitis, colitis, adrenal insufficiency, and hypothyroidism. Because the outcome is a participant count, it records how many participants met the qualifying definition rather than the total number of event episodes or the duration of events.

A count endpoint is useful for describing how many people experienced a specified safety outcome during the observation period. The grading thresholds and the requirement that the event be judged related to study treatment define which events contribute to the primary result. The record says adverse events were graded according to the DAIDS AE Grading Table, corrected Version 2.1, July 2017; infusion reactions, adrenal insufficiency, and pneumonitis were graded using NCI CTCAE Version 5.0 guidelines.

The endpoint does not by itself describe every adverse event, establish causation, or measure immunotherapeutic activity. Its interpretation depends on the recorded event definitions, grading, treatment-relatedness assessment, and follow-up window.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is binary at the participant level: each participant either met the qualifying event definition during follow-up or did not. The reported outcome is a count of participants, rather than a continuous measurement such as a laboratory value. The endpoint combines a Grade >=3 adverse event with a Grade >=1 immune-related adverse event, and requires a judgment that the event was related to study treatment. The interpretation therefore begins with the definition, not simply with the word “adverse event.”

What does a participant count describe?

A count gives the number of people meeting an endpoint definition. For this trial, the reported counts are 2 in the cemiplimab cohort and 0 in the placebo cohort. The corresponding arm enrollments were 4 and 1. A count is straightforward to read, but by itself it does not express precision. With very few participants, a change in the experience of a single participant can substantially alter the descriptive picture. General statistical principle: estimates based on small groups tend to be uncertain, and the uncertainty should be considered when interpreting an observed count.

Why is a comparison not available?

The registry record marks the two-arm comparison for this binary endpoint as missing or not estimable. It does not report a comparative effect estimate, confidence interval, or p-value. A confidence interval is a range of values compatible with the data and a specified statistical method; it conveys precision as well as the direction and scale of an estimated effect. Without a reported interval or another inferential result, the arm counts cannot be turned into a quantified estimate of the difference between treatment groups. No recalculated comparison is reported, and the record does not state why a two-arm estimate was not estimable.

Randomization is intended, as a general design principle, to support a fair comparison by allocating participants without systematically favoring one arm. Double masking can reduce some forms of bias in treatment delivery and assessment. Neither feature removes uncertainty from a study with very few participants, and the registry record does not describe the allocation sequence, masking procedures, or any departures from them.

How should grading and treatment relationship be understood?

The endpoint includes a grading threshold and a treatment-relatedness judgment. The record states that the CMC judged relatedness while blinded to treatment arm. It also names the DAIDS AE Grading Table, corrected Version 2.1, July 2017, as the grading framework, with NCI CTCAE Version 5.0 guidelines used for infusion reactions, adrenal insufficiency, and pneumonitis. These definitions standardize how qualifying events are identified. They do not mean that every event in the record belongs to the primary endpoint: an event must meet the specified criteria.

The primary result is a participant count, not an event-rate analysis. The record does not report event timing, repeated-event handling, or an analysis of time to first event. These methods are not necessary to state the recorded count, but their absence limits interpretation of the event pattern and follow-up experience.

What do the sample-size assumptions mean?

The record describes a planning illustration based on the observed effect, with a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation; the effect to detect is described as a stated fraction of the observed effect. Such an illustration is a planning calculation, not observed power and not evidence that the trial was designed with those assumptions. The record does not provide the numerical planning results in the information reported for this analysis, and the actual allocation was not equal: four participants were in the cemiplimab cohort and one was in placebo. The trial’s own sample-size justification is not reported in the registry record.

General statistical principle: a sample-size calculation depends on assumptions about the effect to detect, variability or event frequency, allocation, and the chosen error threshold. A planning illustration based on an observed effect is especially dependent on that observed effect and should not be treated as a prospective design justification. It also cannot establish that the study had adequate power after the fact. No observed-power claim follows from the reported counts.

What can be concluded from the available analysis?

The primary safety outcome is reported descriptively by arm, and the registry record does not provide an estimable two-arm comparison. There are three analyses recorded, but the methods for those analyses are not reported in the registry record. The statistical model, any adjustment for baseline characteristics, and any sensitivity analyses are also not reported in the registry record. The data support describing the observed counts and their definitions; they do not provide a precise comparative estimate or a basis for broader claims.

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("Cohort 1: Cemiplimab", "Cohort 1: Placebo")
n      <- c(4, 1)
events <- c(2, 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  Cohort 1: Cemiplimab  4  2  50.0  6.8  93.2  15.0  85.0  14.7  85.3
ARM  Cohort 1: Placebo  1  0  0.0  0.0  97.5  0.0  79.3  1.3  84.2

11. Multiplicity, Missing Data and Analysis Populations

The record lists six outcome measures, including four secondary outcomes concerning HIV-1 Gag-specific CD8+ T-cell responses. It reports three analyses. A multiplicity strategy—such as an adjustment to account for testing several outcomes—is not reported in the registry record. Without that information, it is not possible to determine whether the listed analyses were part of a prespecified testing hierarchy, whether any adjustment was used, or how findings across outcomes were intended to be interpreted together.

The numerical results for the secondary immune-response outcomes are not reported in the registry record. The record also does not describe methods for handling missing outcome measurements, missing visits, or participants who discontinued early. The primary endpoint’s timeframe ends at Week 48 or premature discontinuation, but that wording alone does not specify how incomplete follow-up was handled in analysis.

The primary population is identified as all participants. Separate analysis populations, such as a per-protocol or safety population, are not reported in the registry record. Details about exclusions, adherence requirements, or which participants contributed to each analysis are also not reported in the registry record.

These omissions matter because missing observations and multiple outcomes can affect how results are interpreted. General statistical principle: incomplete data can introduce uncertainty, and analyzing several outcomes without a clear plan can make it harder to distinguish prespecified evidence from chance patterns. The registry record does not provide enough information to assess those issues for the secondary outcomes or to reconstruct the analysis strategy.

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 serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: thyroiditis (25.0%), blood thyroid stimulating hormone decreased (25.0%), thyroxine increased (25.0%).

The serious adverse event entries report 1 affected participant among 4 at risk in the cemiplimab cohort and 0 affected among 1 at risk in the placebo cohort. The entries for other adverse events report 4 affected among 4 at risk in the cemiplimab cohort and 1 affected among 1 at risk in the placebo cohort. Deaths are recorded as 0 in each cohort.

These serious-event figures are distinct from the primary endpoint. The primary outcome counts participants with a Grade >=3 adverse event or a Grade >=1 immune-related adverse event judged related to study treatment. The serious-event entries do not, on their own, state whether each event met the primary endpoint’s grading and relatedness criteria. A serious adverse event and a qualifying primary-endpoint event should therefore not be assumed to be interchangeable categories.

The record lists Thyroiditis, Blood thyroid stimulating hormone decreased, and Thyroxine increased among the top serious-event terms, each reported as 25.0%. These terms and percentages are registry entries, not comparative estimates. With only five participants overall, the safety counts describe this study’s recorded experience and cannot reliably characterize the frequency of uncommon events in a broader population.

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

The primary result establishes a limited descriptive fact: 2 participants in the cemiplimab cohort and 0 in the placebo cohort met the recorded primary safety endpoint. The registry record does not provide an estimable two-arm comparison for that binary outcome. The figures are therefore not a precise estimate of a treatment-group difference, and the placebo count from one participant is particularly limited as a basis for inference.

The result does not establish that the groups are equivalent or that a difference is certain. No confidence interval or p-value for the comparison is reported. The absence of an estimable comparison should be understood as a limitation of the posted analysis, not as a finding of no difference.

The endpoint is a safety measure defined by event grade, immune-related-event status, treatment-relatedness judgment, and a specified follow-up period. It does not summarize every possible safety outcome. The separate serious adverse event figures give additional descriptive information, but they do not replace the primary endpoint or show that every serious event met its criteria.

The study’s stated purpose also included evaluating immunotherapeutic activity. The registered secondary outcomes concern HIV-1 Gag-specific CD8+ T-cell responses, but their numerical results are not reported in the registry record. Consequently, these primary safety counts do not establish whether those immune responses changed. They also do not establish an effect on HIV-1 or on any clinical outcome not reported in the record.

General statistical principle: a small randomized study can be informative about feasibility and observed events while still leaving substantial uncertainty about comparative effects. The counts should be interpreted as observations from the five enrolled participants, not as a dependable forecast for a larger group or as a complete account of cemiplimab’s risks and effects.

14. Why This Trial Matters Statistically

This trial illustrates why the design label and the amount of information in the data need to be considered together. It was randomized and double-blind, but only five participants were enrolled, with four assigned to cemiplimab and one to placebo. The primary safety outcome is easy to state as a count, yet the record does not provide an estimable two-arm comparison. Design features that can support a comparison cannot compensate for the limited information carried by very small groups.

It also shows the importance of keeping related safety summaries distinct. The primary outcome requires specified severity or immune-related-event criteria and a judgment of treatment relationship. Serious adverse events and other adverse events are separately recorded. Treating these categories as if they were identical would obscure the definitions that govern what each result means.

The outcome list distinguishes safety from immunotherapeutic activity. The primary result concerns qualifying adverse events, while the secondary outcomes concern HIV-1 Gag-specific CD8+ T-cell responses. Without numerical results for the secondary outcomes, the safety counts cannot answer the immunologic question. Clear interpretation depends on matching each conclusion to the outcome actually reported.

Finally, the record’s missing inferential comparison and unreported analysis details are themselves relevant to statistical reading. They limit what can be concluded about precision, missing data, and multiplicity. A disciplined interpretation reports the observed counts, explains their scope, and leaves unanswered questions unresolved rather than treating a small descriptive result as definitive.

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 principal limitation is the very small enrollment: five participants overall, with four in the cemiplimab cohort and one in the placebo cohort. In particular, the placebo cohort provides very little information for a comparative safety assessment. The record marks the two-arm binary comparison as missing or not estimable, and it gives no confidence interval or p-value for the primary outcome.

The study status is terminated, but the reason for termination is not reported in the registry record. The recorded primary endpoint covers study entry through Week 48 or premature discontinuation; the number of participants completing follow-up and the handling of incomplete follow-up are not reported in the registry record. The analysis populations beyond the primary outcome population of all participants are not reported in the registry record.

The record does not describe the randomization procedure, the specific masking arrangements, or the statistical methods used for the three analyses. Multiplicity procedures, missing-data methods, sensitivity analyses, and the trial’s own sample-size justification are not reported in the registry record. This prevents a detailed assessment of the prespecified analysis strategy.

The secondary outcomes are listed, but their numerical findings are not reported in the registry record. As a result, the immunotherapeutic-activity component cannot be interpreted from the outcome information. The small number of participants and limited comparative information also mean that the observed safety entries cannot characterize uncommon events or support broad conclusions about outcomes outside the study.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Safety and Immunotherapeutic Activity of an Anti-PD-1 Antibody (Cemiplimab) in Participants With HIV-1 on Suppressive cART — NCT03787095.
• Registry record last updated 2022-03-02; status last verified 2022-02.
• Results first posted 2021-12-09.
• 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

NCT03787095 was a phase I/II, randomized, double-blind, placebo-controlled study with a sequential model, evaluating cemiplimab in participants with HIV-1 on suppressive cART. The record lists five participants: four in Cohort 1: Cemiplimab and one in Cohort 1: Placebo. The cemiplimab dose was 0.3 mg/kg at Day 0 and Week 6; both cohorts continued their current non-study provided ART regimen.

The primary endpoint counted participants with a Grade >=3 adverse event or a Grade >=1 immune-related adverse event judged related to study treatment, from study entry through Week 48 or premature discontinuation. The recorded counts were 2 of 4 in the cemiplimab cohort and 0 of 1 in placebo. A two-arm comparison was marked missing or not estimable.

The study is terminated. Its reason for termination, the statistical analysis details, and numerical secondary-outcome results are not reported in the registry record.