A Safety and Efficacy Study of VI-0521 to Evaluate the Long Term Treatment of Obesity in Adults With Obesity-Related Co-Morbid Conditions. An Extension Study of Protocol OB-303 (NCT00553787): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Safety and Efficacy Study of VI-0521 to Evaluate the Long Term Treatment of Obesity in Adults With Obesity-Related Co-Morbid Conditions. An Extension Study of Protocol OB-303 (NCT00553787) (NCT00796367), a randomized phase 3 study. The primary outcome is Percent Weight Change at End of Treatment, Week 108..

2. Clinical Question

3. Trial Design

This was a Phase 3, randomized, double-blind, placebo-controlled, parallel-group extension study of VI-0521. It was intended for adults with obesity-related co-morbid conditions who had completed study OB-303 at selected study sites. The registry describes 676 participants enrolled and lists the study status as completed.

Participants were assigned to placebo, VI-0521 Mid, or VI-0521 Top. The registry describes the Mid group as receiving 7.5 mg phentermine and 46 mg topiramate, and the Top group as receiving 15 mg phentermine and 92 mg topiramate. The primary outcome was percent weight change at the end of treatment, week 108. The outcome population was ITT-LOCF, meaning the registry identifies an intent-to-treat population using last observation carried forward for missing endpoint values.

The study start is listed as 2008-12 and the primary completion as 2010-06. The registry frames the study as an extension of OB-303 rather than a study enrolling an entirely new group independent of that earlier trial. Results therefore describe the selected participants who continued into the extension, not all adults with obesity or all participants in the parent study. The registry record does not report the original study’s complete treatment history or the criteria used to select extension sites and participants.

The original design’s sample-size justification, planned recruitment by arm, and planned statistical analysis details are not reported in the registry record. The enrollment figure is the registry’s overall enrollment, while the outcome summaries list the analysis populations separately by arm. The record does not explain any difference between those totals.

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

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Intent-to-Treat Last observation carried forward (ITT-LOCF)

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 Percent Weight Change at End of Treatment, Week 108., measured over From baseline to end of treatment.

The primary endpoint is percent weight change from baseline to the end of treatment at week 108. It is a continuous measure: each participant’s change is represented on a numeric scale, and the registry reports least-squares mean changes for each group. Negative values in the reported outcome indicate weight loss as expressed by this endpoint.

A group mean provides a summary of the average outcome, but does not show how individual participants’ changes were distributed. The registry reports standard errors alongside the least-squares means. A standard error describes the precision of an estimated mean; it is not the same as the spread of individual participant outcomes.

Comparing changes between randomized groups helps describe the difference in average outcomes under the study conditions. The endpoint is useful for that purpose because it expresses change on a common scale and permits estimation of group contrasts. It does not, by itself, describe safety, durability beyond the recorded treatment period, or the experience of every participant.

The registry identifies the primary population as ITT-LOCF. Last observation carried forward substitutes a participant’s last recorded measurement for a missing later value. That approach can affect the estimated group means, particularly when missingness is substantial or related to participants’ outcomes. The amount and reasons for missing endpoint data are not reported in the registry record.

7. Results Posted in the Registry

• VI-0521 Mid: mean -9.32 versus -1.80 on control; difference -7.52 (95% CI -9.22 to -5.82), standardized effect d = -0.91; z-test p = <0.001.
• VI-0521 Top: mean -10.50 versus -1.80 on control; difference -8.70 (95% CI -10.16 to -7.24), standardized effect d = -1.03; z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

The registry’s primary outcome is a continuous measure, and its arm-level summaries are least-squares means with standard errors. A mean is an average on the outcome scale. A standard error describes uncertainty in the estimated mean, rather than variability among individual participants. Smaller standard errors generally indicate more precise estimates, although interpretation also depends on the analysis and data structure.

The sponsor’s reported group comparisons use ANCOVA, or analysis of covariance. In general, ANCOVA compares group outcomes while accounting for covariates included in the model. Such adjustment can change an estimated contrast and its uncertainty relative to a simple comparison of posted group means. The registry identifies ANCOVA and gives estimates, confidence intervals, and p-values, but the covariates, model specification, and other detailed analysis choices are not reported in the registry record.

The sponsor reports mean differences for final values. In general, the direction of a difference depends on the order in which groups are subtracted. The reported placebo-versus-VI-0521 comparisons are positive, while the unadjusted recalculations express VI-0521 minus placebo and are negative. This difference in signs is a matter of contrast orientation, not by itself a disagreement about which group had the more negative percent change.

The recalculations are simple unadjusted comparisons based on the posted means and standard errors. The pooled standard deviations used for these calculations are approximate, as indicated by the calculation method. These estimates are not adjusted ANCOVA results, and their confidence intervals are not the sponsor’s confidence intervals. The recalculated p-values below 0.001 indicate incompatibility with a zero mean difference under the unadjusted testing assumptions; they do not account for any analysis choices that are absent from the registry record.

A confidence interval describes a range of estimates reasonably compatible with the data and the method used. In repeated use of a method under its assumptions, intervals at a stated confidence level would cover the target parameter at that rate. It is not a probability statement that a particular fixed treatment contrast lies within one observed interval. The sponsor’s reported comparisons use 95% confidence intervals. The recalculated intervals are also presented at that level.

A p-value measures how unusual results at least as incompatible with a null hypothesis would be if that hypothesis and the test assumptions held. It does not measure the probability that the null hypothesis is true, the size of an effect, or clinical importance. A p-value below a conventional threshold does not remove concerns about bias, missing observations, analysis choices, or multiple testing.

The registry specifies ITT-LOCF for the primary outcome. In general, last observation carried forward fills a missing endpoint with the participant’s last recorded value. It can preserve participants in an analysis but imposes an assumption that the last value is an appropriate stand-in for the later outcome. If outcomes change after the last observation, or if missingness differs in meaningful ways between groups, this method can misrepresent the endpoint. The registry record does not report missingness by group, reasons for missing observations, or sensitivity analyses using other missing-data approaches.

The registry lists six analyses and two outcome measures, while no secondary outcomes are posted in the record. It reports three ANCOVA contrasts: placebo with VI-0521 Top, placebo with VI-0521 Mid, and VI-0521 Mid with VI-0521 Top. In general, making multiple comparisons increases the chance of at least one apparently unusual result when all null hypotheses are true. Methods such as adjustment of significance thresholds can address that issue, but whether such a method was planned or used is not reported in the registry record.

General interpretation also depends on dispersion and the shape of outcome distributions. A mean and standard error alone do not show whether outcomes are symmetric, contain extreme values, or differ in spread across groups. The registry’s posted summaries do not include participant-level outcomes or distribution plots that would permit those checks. The outcome is continuous, so the reported means and mean differences are the central summaries, but their adequacy depends on the underlying data and the analysis assumptions.

The estimated standardized effects used in planning illustrations are not additional observed outcomes. Standardized effect measures express a mean difference relative to an estimate of outcome variability; they can help compare effects across scales, but depend on how variability is estimated. Here the recalculated pooled standard deviations are approximate. The registry does not report that these recalculations formed part of the original trial analysis.

10. Confidence Intervals and Precision

The sponsor’s 95% confidence interval for placebo versus VI-0521 Top extends from 7.39 to 10.03 around a reported mean difference of 8.71. For placebo versus VI-0521 Mid, the interval extends from 5.95 to 9.09 around a reported mean difference of 7.52. Both intervals lie on the positive side of zero, so the reported ANCOVA estimates are not compatible with a zero contrast at this confidence level under the analysis used.

For VI-0521 Mid versus VI-0521 Top, the sponsor’s 95% confidence interval extends from -0.31 to 2.68 around a mean difference of 1.19. It includes zero and also allows for differences in either direction. The interval therefore conveys uncertainty that a point estimate alone would conceal. It does not establish equality or rule out all potentially meaningful differences.

The unadjusted recalculations have their own confidence intervals: -9.219 to -5.821 for VI-0521 Mid versus placebo, and -10.16 to -7.243 for VI-0521 Top versus placebo. These intervals use the active-minus-placebo direction and are derived from posted summary statistics, not the sponsor’s ANCOVA model. Their widths reflect sampling uncertainty under the recalculation assumptions; they do not capture uncertainty from unreported design, missing-data, or modeling decisions.

Precision is distinct from importance. A relatively narrow interval can still surround an estimate whose relevance depends on context, while an interval crossing zero can include both small and larger contrasts. The registry record does not define a threshold for a clinically meaningful difference, so the intervals alone cannot determine clinical importance.

11. Planning a Similar Trial: Sample Size

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

The planning illustrations show participants per arm for 80% power and participants per arm for 90% power. They are based on two-sided significance level 0.05, equal allocation between the two arms, and a normal approximation. Each illustration uses a stated fraction of the observed effect as the effect to detect. The trial’s own original sample-size assumptions are not reported in the registry record.

For each comparison, the planning rows consider 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. The 80% power column gives the per-arm sample size that the calculation estimates would be needed to have an 80% chance of detecting the specified effect under those assumptions, if that effect is real. The 90% power column expresses the corresponding planning target for a 90% chance. These are design calculations, not guarantees about results.

Smaller effects are harder to distinguish from random variation, so the planning illustrations require more participants per arm when the target is a smaller fraction of the observed effect. The figures are not the number enrolled in the actual trial and should not be read as evidence that the trial had a particular probability of success. They are planning illustrations, not observed power.

These calculations concern a two-arm comparison at a time. They assume equal allocation even though the posted analysis arm sizes differ, and they do not incorporate unreported features of the original study design or analysis. The registry record does not report whether the original planning used the same target effect, allocation, or other assumptions. The illustrations therefore help show how sample-size requirements vary with the assumed effect, rather than reconstructing the trial’s original design.

12. How the Sample Size Is Derived

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

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -7.520; pooled standard deviation σ = 8.287.
2. n = 2 × (1.96 + 0.842)² × 8.287² / -7.520² = 19.1, rounded up to 20 participants per arm for 80% power.

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

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -7.52                      # observed difference in means
sigma <- 8.28693007409318                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(796367)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 20 (80% power), 26 (90% power)
 75% of observed effect: n per arm = 34 (80% power), 46 (90% power)
 50% of observed effect: n per arm = 77 (80% power), 103 (90% power)
power.t.test: 20.1 
Simulated power at n = 20 per arm: 0.802 (20000 replicates)

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

13. Analyses Reported by the Sponsor

The registry reports ANCOVA comparisons of final values. The unadjusted recalculations instead use the posted least-squares means and standard errors to estimate simple mean differences. These methods are not interchangeable: ANCOVA may account for covariates, while the recalculations do not, and the detailed ANCOVA model is not reported in the registry record.

For the placebo comparisons, the sponsor’s estimates are positive because the reported contrast is expressed in the placebo-versus-VI-0521 direction. The recalculated estimates are negative because they are expressed as VI-0521 minus placebo. The sign reversal is consistent with changing the subtraction order. The estimate magnitudes are similar, but the confidence limits are not identical: the sponsor’s intervals come from ANCOVA, while the recalculated intervals come from unadjusted summary-statistic methods.

The sponsor also reports the Mid-versus-Top comparison, whose confidence interval includes zero. The recalculated summaries do not include that contrast. The sponsor’s reported analysis is the registry’s own inferential analysis; the recalculations are supplementary checks based on posted summaries and should not be mistaken for a reproduction of the ANCOVA.

14. Multiplicity, Missing Data and Analysis Populations

The registry identifies the primary outcome and reports a primary population of ITT-LOCF. The endpoint is percent weight change from baseline to the end of treatment at week 108. The use of last observation carried forward is part of the stated population definition, but the record does not give the amount of missing data, the timing of missingness, or reasons participants lacked endpoint measurements.

Last observation carried forward can be problematic when a participant’s outcome would have changed after their last recorded value. It can also introduce bias when missingness is associated with treatment group or outcome. The registry record does not report sensitivity analyses that test whether the primary findings change under alternative assumptions about missing data. Nor does it describe how many participants were included in each outcome analysis beyond the arm-level summary counts.

The record lists two outcome measures, six analyses, and no posted secondary outcomes. It reports ANCOVA results for three group contrasts. It does not report whether a multiplicity adjustment was planned or applied, how the primary comparison was designated among the contrasts, or whether the analyses were prespecified in a detailed statistical analysis plan. Without that information, the p-values should be understood as the values reported for the individual comparisons, not as evidence that the family of comparisons was controlled for multiple testing.

Other analysis details are also not reported in the registry record, including the ANCOVA covariates, handling of protocol deviations, and any interim analyses. The original sample-size justification and the original study’s design assumptions are not reported either. These omissions limit independent assessment of whether the posted analysis fully follows the intended protocol, but they do not change the values recorded for the reported outcomes.

15. 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 2% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: renal failure, acute (0.7%), appendicitis (0.7%), osteoarthritis (0.7%), transient ischemic attack (0.7%), pyelonephritis (0.7%).

The registry lists serious adverse events for 9 of 227 participants at risk in the placebo group, 4 of 153 in VI-0521 Mid, and 13 of 295 in VI-0521 Top. These are counts of participants affected as recorded in the registry, with the corresponding numbers at risk shown for each group. They should be read as descriptive safety summaries, not as evidence that an event was caused by a particular study group.

The registry also lists other adverse events affecting 182 of 227 participants in placebo, 111 of 153 in VI-0521 Mid, and 234 of 295 in VI-0521 Top. Those counts are separate from the serious adverse event counts. The record lists renal failure, acute; appendicitis; osteoarthritis; transient ischemic attack; pyelonephritis; and myocardial infarction among the top serious terms, each at 0.7%. The group-specific context for those terms is not stated in the summary list.

Deaths are not reported in the registry record. The summaries do not establish causality, event duration, severity beyond the serious-event classification, or comparative risk after adjustment. They also do not provide enough information to determine whether events differed in clinical context across groups.

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

The posted results show more negative average percent weight changes in both VI-0521 groups than in the placebo group at the week 108 endpoint. The sponsor’s ANCOVA estimates for each placebo comparison have 95% confidence intervals that exclude zero. The comparison between VI-0521 Mid and VI-0521 Top has an interval that includes zero, leaving uncertainty about the difference between those groups.

These are estimates from a particular extension study population: adults with obesity-related co-morbid conditions who had completed OB-303 at selected sites. The results cannot automatically be extended to people who did not enter the parent study, did not continue into the extension, or differ from the enrolled population. The study design and the reported analysis support comparisons within this trial, but the registry record does not provide all details needed to assess every potential source of bias or uncertainty.

A group average does not show how outcomes varied among participants, how many experienced a particular change, or whether a given participant’s result can be predicted from the group result. The registry reports an ITT-LOCF endpoint, and the handling of missing values may affect the estimates. The absence of detailed missing-data information and sensitivity analyses limits assessment of how robust the estimates are to alternative assumptions.

The results also do not determine the comparative safety of the groups. Serious and other adverse events are listed as counts, but the registry summaries do not establish causation or provide a full analysis of event timing and clinical context. Nor do the reported outcome comparisons alone establish clinical importance, because a threshold for a meaningful difference is not reported in the registry record.

Accordingly, the estimates describe the recorded endpoint and analyses in this extension study. They should not be read as a guarantee of an individual outcome, as a complete account of safety, or as a result that settles questions beyond the study population and analysis reported.

17. Why This Trial Matters Statistically

This trial is statistically informative because it combines randomization and blinding with a long-term endpoint in an extension population. Randomization supports comparisons between assigned groups by reducing systematic differences at assignment, while blinding can reduce some forms of differential behavior or assessment. These design features do not eliminate uncertainty from missing data, selection into the extension, or analysis choices that are not described in the registry record.

The three-group structure allows two active-dose groups to be compared with placebo and also permits a direct comparison between the two VI-0521 groups. The placebo contrasts have confidence intervals excluding zero, whereas the Mid-versus-Top interval includes zero. That contrast-specific uncertainty matters: evidence for differences from placebo does not automatically establish a difference between the active groups.

The trial also illustrates why estimates, intervals, and methods should be considered together. The registry’s ANCOVA results are adjusted analyses, while the recalculations based on posted summary statistics are unadjusted. Their contrast directions differ, and their uncertainty estimates are not identical. Reading only a point estimate or p-value would obscure these distinctions.

Finally, the ITT-LOCF population definition and unreported details of missingness, multiplicity, and model specification are important for interpreting precision and robustness. The record allows a reader to identify the main outcome, reported group summaries, and sponsor comparisons, while also showing where additional protocol and analysis detail would be needed for a fuller statistical assessment.

18. 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 study is an extension of OB-303 and enrolled people who had completed that earlier study at selected sites. This means the analyzed population is a selected continuation group. Its experience may not represent people who would not have entered or completed the parent study, and the registry record does not describe the selection process in detail.

The primary outcome uses ITT-LOCF. The registry record does not report missing endpoint counts by group, reasons for missingness, or results under alternative missing-data assumptions. Because carrying forward an earlier observation can misstate a later outcome when the outcome changes over time, the robustness of the estimates to missing-data handling cannot be assessed from the reported information alone.

The registry gives ANCOVA estimates but does not report model covariates, diagnostics, or the detailed analysis plan. Multiplicity adjustment is not reported despite multiple outcome and group analyses. The original sample-size justification, interim analyses, and original design assumptions are also not reported in the registry record. These gaps limit evaluation of the prespecified testing strategy and the relation between the reported analyses and the original plan.

Safety results are descriptive counts. The registry record does not provide information sufficient to assess causality, timing, or adjusted comparative risks, and deaths are not reported. The record also does not provide the participant-level outcome distribution needed to evaluate outliers or differences in variability. These limitations concern how completely the posted results can be interpreted; they do not alter the numerical summaries that are reported.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: A Safety and Efficacy Study of VI-0521 to Evaluate the Long Term Treatment of Obesity in Adults With Obesity-Related Co-Morbid Conditions. An Extension Study of Protocol OB-303 (NCT00553787) — NCT00796367.
• Registry record last updated 2012-09-10; status last verified 2012-09.
• Results first posted 2012-08-31.
• 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.

22. Record Summary

The registry lists NCT00796367 as a completed Phase 3, randomized, double-blind, parallel-group extension study sponsored by VIVUS LLC. It enrolled 676 participants and compared placebo with VI-0521 Mid and VI-0521 Top among adults with obesity-related co-morbid conditions who had completed OB-303 at selected sites.

The primary endpoint was percent weight change from baseline to the end of treatment at week 108, analyzed in an ITT-LOCF population. The registry reports least-squares mean changes and ANCOVA comparisons. Placebo comparisons with each VI-0521 group have reported confidence intervals excluding zero; the Mid-versus-Top interval includes zero. The record also lists serious and other adverse event counts. Important details about missing data, multiplicity, model specification, and the original sample-size justification are not reported in the registry record.