Pharmacodynamics, Safety and Pharmacokinetics of BMS-663068, an HIV Attachment Inhibitor, in HIV-1: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Pharmacodynamics, Safety and Pharmacokinetics of BMS-663068, an HIV Attachment Inhibitor, in HIV-1 (NCT01009814), a randomized phase 2 study. The primary outcome is Mean Logarithm With Base 10 (Log10) Change From Baseline in Human Immunodeficiency Virus (HIV) Ribonucleic Acid (RNA) at Day 9.

2. Clinical Question

The study asked whether short-course dosing of BMS-663068, a prodrug for the HIV attachment inhibitor BMS-626529, was associated with a decrease in HIV RNA over the treatment period. Its research hypothesis proposed a mean decrease of at least 1 log10 in HIV RNA at Day 9 after 8 days of therapy in at least one safe and well-tolerated regimen among participants with Clade B HIV-1.

This is a pharmacodynamic question about a short-term laboratory measure, not a question about longer-term clinical outcomes. The record also lists safety and pharmacokinetic objectives, but the primary endpoint concerns the change in viral RNA.

3. Trial Design

The registry identifies this as a Phase 2, randomized, open-label, parallel-group study. It was completed and enrolled 50 participants. Participants were assigned among five treatment regimens; the primary results list endpoint group sizes of 9 or 10 for each regimen. The safety results list 10 participants at risk in each group. These are endpoint-specific analysis populations, and the record does not explain the difference between the posted counts.

All regimens involved BMS-663068, administered from Day 1 to Day 8. Four paired BMS-663068 with RTV: 600 mg Q12H with RTV 100 mg Q12H; 1200 mg QHS with RTV 100 mg QHS; 1200 mg Q12H with RTV 100 mg Q12H; and 1200 mg Q12H with RTV 100 mg QAM. The fifth regimen was BMS-663068 1200 mg Q12H without RTV. Thus, the comparisons concern alternative active dosing regimens, rather than treatment versus placebo.

The study was unmasked, so participants and study personnel were not blinded according to the registry record. The record does not report how randomization was generated or concealed, whether allocation was stratified, or how treatment adherence was assessed. The study dates recorded are a start of 2009-11-23 and primary completion of 2010-06-25.

The primary analysis was described as an analysis of covariance (ANCOVA) correcting for baseline HIV viral load and treatment group. Analyses were specified by antiretroviral treatment history—naive, experienced, and combined—and the combined analysis included treatment history as an additional covariate. Only Clade B participants were included in the primary endpoint population. Details such as the planned number of participants in each treatment-history subgroup, the handling of protocol deviations, and the sample-size justification 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: A total of 75 participants were screened, of which 25 were screen failure, and 50 eligible participants were enrolled into the study and were randomized.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Pharmacodynamic Population. It comprised of all participants for whom pharmacodynamic measurements were available at Baseline and at least one other time.

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 Mean Logarithm With Base 10 (Log10) Change From Baseline in Human Immunodeficiency Virus (HIV) Ribonucleic Acid (RNA) at Day 9, measured over Baseline and Day 9.

The primary endpoint was mean log10 change from baseline in HIV RNA at Day 9, measured in log10 copies per milliliter. Baseline was defined as the last non-missing observation before the first dose on Day 1. Change was calculated by subtracting that baseline value from the post-baseline visit value. A negative change therefore represents a lower measured HIV RNA value than at baseline.

HIV RNA is a direct laboratory measure of viral load, so change over a short dosing period can describe antiviral pharmacodynamic activity. The registry’s research hypothesis focused on whether at least one regimen would produce a mean decrease of at least 1 log10 by Day 9 after 8 days of therapy. That hypothesis is a stated objective, not a result: the observed means and the adjusted comparisons should be read separately from the threshold proposed in the hypothesis.

The endpoint is a mean, and the posted arm summaries include a standard error. A standard error describes the precision of an estimated mean, rather than the spread of individual participants’ measurements. The registry says ANCOVA was used to compare regimen groups while correcting for baseline viral load. This adjustment matters because participants can begin with different viral loads, and the comparison is intended to account for baseline values rather than treating the raw group means as the adjusted treatment contrast.

The measure is limited to the observed Day 9 response. It does not, by itself, describe persistence of a response after treatment, longer-term viral control, clinical outcomes, or the balance of benefits and harms over extended use. Such outcomes are not part of this primary endpoint.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary analysis used ANCOVA, a regression approach for comparing group means while adjusting for one or more covariates. In this study, the registry says the model corrected for baseline HIV viral load and treatment group. For the combined antiretroviral-treatment-history analysis, treatment history was included as an additional covariate. Adjustment can improve the fairness and precision of a comparison when baseline measurements differ across groups, but it does not remove uncertainty or compensate for limitations in the population studied.

Why adjust for baseline viral load?

Change scores depend in part on where participants started. Two groups can have different average changes partly because their baseline values differed. ANCOVA uses baseline viral load in the model so that the treatment-group contrast is estimated conditional on baseline, rather than relying only on a comparison of unadjusted mean changes. This is a general statistical rationale; the registry does not report model diagnostics or the exact fitted model details beyond the covariates described.

What do the adjusted contrasts represent?

The registry reports net mean differences for pairs of regimens, with 90% confidence intervals and p-values. A contrast compares the model-adjusted mean for one regimen with the model-adjusted mean for another, under the model’s adjustment. Its sign depends on the order of the groups in the contrast. The interval conveys uncertainty about the contrast; when it spans zero, the data are compatible with differences in either direction at that confidence level. When it does not span zero, the result is less compatible with a zero contrast under the model, but it still does not indicate whether the difference is clinically important.

The posted raw arm means and the ANCOVA contrasts answer related but different questions. The means describe observed average changes in each arm. The adjusted contrasts incorporate baseline viral load and, in the combined treatment-history analysis, treatment history. They should not be treated as interchangeable estimates. The registry’s reported pairwise analyses are the sponsor’s model-based results; they take precedence over simple unadjusted recalculations when the two differ.

What does the standard error describe?

The primary arm summaries report a standard error. In general, a standard error measures the sampling precision of an estimated mean: it reflects how much the estimate would be expected to vary across repeated samples under the statistical model. It is not the same as the standard deviation, which describes variation among individual observations. A mean and standard error alone do not reveal the full distribution of individual viral-load changes.

How should the confidence level be read?

The reported ANCOVA intervals are 90% confidence intervals. In general, a confidence interval is produced by a procedure that, under its assumptions and repeated application, covers the target quantity at the stated rate. It is not a probability statement about the particular fixed contrast after the data have been observed. A 90% interval is also not interchangeable with an interval at another confidence level. The record does not report the rationale for choosing this confidence level.

Why does multiplicity matter?

The record reports ten pairwise ANCOVA comparisons. When many hypotheses are tested, the chance of at least one apparently unusual result can increase if each test is interpreted without adjustment. Whether and how to control that issue depends on the prespecified analysis plan and the intended family of hypotheses. An adjustment method or hierarchy for these comparisons is not reported in the registry record. The reported p-values and confidence intervals should therefore be understood as the posted pairwise results, not as evidence that family-wise error was controlled.

What does the secondary-endpoint planning illustration mean?

For the physical-examination abnormality comparison between the reference regimen and BMS-663068 1200 mg Q12H with RTV 100 mg Q12H, an unadjusted risk difference and an unadjusted p-value are posted. A planning illustration uses the observed effect as a basis for estimating participants per arm under stated assumptions: a two-sided significance level of 0.05, equal allocation, and a normal approximation. Its rows represent planning for 100%, 75%, or 50% of the observed effect; the corresponding entries are participants per arm for 80% power and participants per arm for 90% power.

These figures are planning illustrations, not observed power and not the study’s actual design justification. They do not show that the study had a particular probability of detecting an effect after the results were observed. The effect used is based on a sparse binary outcome, so planning around it is sensitive to the instability of the observed estimate. A sample-size justification for the trial, including its original target and assumptions, is not reported in the registry record.

These methods explanations are general statistical context. The registry does not report model diagnostics, assumptions checks, handling of influential observations, or a detailed analysis plan beyond the ANCOVA description and the posted pairwise results.

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("BMS-663068 600 mg Q12H + RTV 100 mg Q12H", "BMS-663068 1200 mg QHS + RTV 100 mg QHS", "BMS-663068 1200 mg Q12H + RTV 100 mg Q12H", "BMS-663068 1200 mg Q12H + RTV 100 mg QAM", "BMS-663068 1200 mg Q12H")
n   <- c(9, 9, 10, 10, 10)
m   <- c(-1.3445, -1.2532, -1.2381, -1.1888, -0.876)
sp  <- c(0.07925, 0.1061, 0.10455, 0.12938, 0.22621)
sd  <- sp * sqrt(n)      # standard errors converted to SDs
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  BMS-663068 600 mg Q12H + RTV 100 mg Q12H  9  -1.345  0.238  -1.527  -1.162
CARM  BMS-663068 1200 mg QHS + RTV 100 mg QHS  9  -1.253  0.318  -1.498  -1.009
CARM  BMS-663068 1200 mg Q12H + RTV 100 mg Q12H  10  -1.238  0.331  -1.475  -1.002
CARM  BMS-663068 1200 mg Q12H + RTV 100 mg QAM  10  -1.189  0.409  -1.481  -0.896
CARM  BMS-663068 1200 mg Q12H  10  -0.876  0.715  -1.388  -0.364

11. Analyses Reported by the Sponsor

The registry reports the sponsor’s ANCOVA results for the primary endpoint. Those results adjust for baseline HIV viral load and treatment group, and the combined treatment-history analysis also adjusts for treatment history. The posted group means are descriptive summaries and should not replace the adjusted comparisons when the question concerns differences between regimens after baseline adjustment.

The adjusted reference-regimen comparisons with the other RTV-containing regimens have 90% confidence intervals that include zero. The comparison with BMS-663068 1200 mg Q12H without RTV has an interval that excludes zero and a reported p-value of 0.023. The adjusted results therefore qualify what might be inferred from visual or arithmetic comparison of arm means. They do not provide evidence of equivalence among the regimens, and the record does not report a clinically important difference threshold.

There is no stated discrepancy between the posted primary results and a recalculated primary analysis. Where an unadjusted recalculation and the sponsor’s adjusted analysis differ, the sponsor’s own analysis takes precedence because it reflects the reported ANCOVA model. The registry does not report enough detail to independently reproduce the adjusted model or verify its assumptions. The sponsor’s analysis also does not resolve the unreported multiplicity approach or limitations of the short follow-up.

12. Multiplicity, Missing Data and Analysis Populations

The registry reports ten pairwise ANCOVA comparisons for the primary endpoint, alongside a primary analysis described by antiretroviral treatment history. It does not report whether the pairwise comparisons were prespecified as a family, whether one contrast was primary among those pairs, or whether any multiplicity correction was applied. The posted p-values and 90% confidence intervals should be interpreted as the results of the individual reported comparisons. Without a stated adjustment, they do not establish control of the overall false-positive risk across all comparisons.

The record lists 32 secondary outcomes. It does not report a multiplicity strategy for those outcomes, a testing hierarchy, or which secondary analyses were intended to be confirmatory. As a general matter, many separate tests make isolated low p-values harder to interpret because chance findings become more likely across a broad set of analyses. This does not invalidate the individual estimates, but it reduces the weight that should be placed on a selected result without a prespecified framework.

Some missing-data rules are described: baseline is the last non-missing observation before the first dose, and the pharmacodynamic population requires a baseline measure and at least one other measurement. The registry does not report the amount or pattern of missing primary endpoint data, how Day 9 measurements that were unavailable were handled, or whether any imputation was used. It also does not report sensitivity analyses for missing data.

The primary population is described as the pharmacodynamic population, limited to Clade B participants. The record does not report whether an intention-to-treat analysis was conducted, the number of participants excluded from the primary population, or the reasons for exclusion. The listed primary endpoint group sizes and safety denominators are not identical in the record; an explanation is not reported. These omissions make it difficult to assess the impact of missing observations and analysis-population selection on the estimates.

Randomization is reported, but the method of sequence generation, allocation concealment, and any stratification are not. The record also does not report interim analyses, a stopping rule, or the prespecified statistical analysis plan. These details are important for judging how closely the posted analyses follow the original plan and for interpreting multiple comparisons.

13. 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 posted safety counts show zero serious adverse events in each regimen group, with 10 participants at risk in each group. No deaths are recorded in any group. These are observed counts over the reported study period; they do not demonstrate that serious events cannot occur, nor do they establish the absence of risk. With short treatment exposure and limited group sizes, uncommon events may not be observed.

The registry also reports non-serious adverse events in each group, with counts ranging from 5 to 9 among 10 participants at risk per group. These counts describe participants with an event, not the total number of event episodes. The record does not provide the serious-event terms because none are listed, and it does not report a formal comparative safety analysis in the results.

The study’s brief treatment period, from Day 1 to Day 8, is relevant to interpretation: these counts describe events captured in the reported study window and cannot characterize longer-term safety. The counts should be read as descriptive observations, not as evidence that one regimen is safer or less safe than another.

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

The results describe short-term changes in HIV RNA after five BMS-663068 regimens. Each posted group mean is a negative log10 change from baseline at Day 9, indicating a lower average measured value than at baseline. The four regimens that included RTV had means from -1.189 to -1.345, while the regimen without RTV had a mean of -0.876. Those figures describe this endpoint population and do not show how every participant responded.

The adjusted ANCOVA comparisons provide a more relevant basis for comparing regimens than the raw means alone because the model accounts for baseline viral load. The reported results distinguish the reference regimen from the regimen without RTV at the stated 90% confidence level, while the reference comparisons with the other RTV-containing regimens have intervals that include zero. The record does not describe an adjustment for the set of pairwise tests, and it does not report a clinically meaningful difference threshold.

These results do not establish that the regimens are equivalent when a comparison is not statistically distinguishable from zero. They also do not show that a statistically distinguishable contrast is clinically important. The study’s stated hypothesis refers to a mean decrease of at least 1 log10 in at least one regimen, but the posted pairwise ANCOVA tests compare regimens with one another; a formal test of each mean against that threshold is not reported.

The findings are specific to a brief dosing period and to the primary pharmacodynamic population, which included Clade B participants with baseline and at least one other measurement. They do not describe durability after treatment, long-term viral suppression, clinical outcomes, or results in other HIV-1 clades. Those outcomes are not reported in the registry record.

Safety interpretation is similarly bounded. No serious adverse events or deaths were recorded in the posted group counts, but this does not establish that serious harms are impossible or that the regimens have equivalent safety. The study offers descriptive short-term observations; it is not a basis for medical advice or a definitive judgment about treatment.

15. Why This Trial Matters Statistically

This trial is statistically informative because it compares several active dosing schedules using a continuous pharmacodynamic endpoint and reports both arm means and adjusted pairwise contrasts. The distinction between those summaries is important: the means show the observed group changes, while ANCOVA estimates differences after accounting for baseline viral load and, in the combined analysis, treatment history.

The design also illustrates why a reported hypothesis should not be mistaken for a demonstrated result. The stated objective sets a decrease threshold, whereas the principal reported inferential results are pairwise regimen comparisons. Those comparisons answer whether the groups differ from one another under the model; they do not by themselves test whether an individual regimen crosses the proposed threshold or whether a difference is clinically important.

The confidence intervals are especially useful for interpreting uncertainty. Several adjusted comparisons remain compatible with zero difference, while the reference comparison with the regimen without RTV has an interval that excludes zero. That pattern is more informative than ranking the means alone, but it must be considered alongside the unreported multiplicity approach and the absence of a stated meaningful-difference threshold.

The trial also highlights limits of short, modestly sized studies. Group counts are small, follow-up for dosing is brief, and the safety results are descriptive. Such evidence can characterize early pharmacodynamic response and generate questions for further study, but it cannot settle longer-term efficacy or uncommon safety outcomes. The registry’s incomplete reporting of missing-data procedures, analysis populations, and the sample-size rationale further constrains statistical assessment.

16. 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 endpoint is measured at Day 9 after therapy from Day 1 to Day 8. This brief observation window cannot establish whether changes persist after dosing ends or translate into longer-term outcomes. The primary analysis is restricted to Clade B participants, so the posted results do not describe other clades. Although the analysis plan describes groups by antiretroviral treatment history, the subgroup-specific results are not reported in the registry record.

The study is open-label, and the registry does not report allocation concealment or the method used to generate the randomization sequence. The primary endpoint is a laboratory measure, which may be less susceptible to subjective assessment than some outcomes, but open-label conduct can still affect other aspects of trial conduct. Details about adherence, protocol deviations, and concomitant treatment are not reported.

The registry does not provide a sample-size justification or explain the difference between primary endpoint group sizes and safety denominators. It also does not report the extent of missing primary measurements, how unavailable Day 9 values were handled, or sensitivity analyses. These gaps limit assessment of selection and missing-data bias.

The pairwise ANCOVA results include multiple comparisons, but a multiplicity approach is not reported. The record does not state a clinically meaningful difference threshold or report model diagnostics. Consequently, statistical distinguishability should not be treated as proof of clinical importance, and non-significant comparisons should not be treated as proof of equivalence.

Finally, the safety counts cover a short treatment period and limited group sizes. Zero serious events and deaths in the posted counts cannot rule out uncommon or delayed harms. Most secondary endpoint results are not detailed in the registry record, limiting the ability to evaluate how broadly the primary result aligns with other measured outcomes.

17. Related Tutorials

18. Related Statistical Calculators

19. Sources

• ClinicalTrials.gov: Pharmacodynamics, Safety and Pharmacokinetics of BMS-663068, an HIV Attachment Inhibitor, in HIV-1 — NCT01009814.
• Registry record last updated 2020-01-03; status last verified 2019-10.
• Results first posted 2020-01-03.
• 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.

20. Record Summary

NCT01009814 is a completed randomized, open-label, parallel Phase 2 study of BMS-663068 regimens in HIV-1 infection. It enrolled 50 participants and compared five dosing schedules, four with RTV and one without. The primary endpoint was mean change from baseline in log10 HIV RNA at Day 9, analyzed using ANCOVA with baseline viral load adjustment and treatment-history analyses described in the registry.

All five posted group means were negative. The adjusted comparison of the reference regimen with BMS-663068 1200 mg Q12H without RTV had a 90% confidence interval excluding zero; the reference comparisons with the other RTV-containing regimens had intervals including zero. The registry does not report a multiplicity strategy, a sample-size rationale, detailed missing-data handling, or longer-term outcomes. Posted safety counts show no serious adverse events or deaths, but these short-term observations do not rule out risk.