ALK21-018: Effects of Medisorb® Naltrexone (VIVITROL®) on Alcohol Craving in Treatment-seeking, Alcohol-dependent Adults: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for ALK21-018: Effects of Medisorb® Naltrexone (VIVITROL®) on Alcohol Craving in Treatment-seeking, Alcohol-dependent Adults (NCT00511836), a randomized phase 4 study. The primary outcome is Change From Baseline in Blood Oxygen-level-dependent (BOLD) Signal Activation Values Detected in the Reward Circuitry of the Brain in Alcohol-dependent Subjects After Presentation of Alcohol-related Cues..

2. Clinical Question

3. Trial Design

This was a randomized, quadruple-masked, parallel-group phase 4 study sponsored by Alkermes, Inc. The registry lists 31 participants enrolled and describes the condition as alcohol dependence. The study began in 2007-07, reached primary completion in 2009-06, and is listed as completed. The randomized double-blind comparison assigned participants to a single administration of VIVITROL 380 mg or placebo. The registry does not report the allocation ratio or detailed arm descriptions beyond the intervention names.

The primary outcome used data from randomized, dosed participants who had fMRI on Day 14. The posted counts for this outcome are 15 in the VIVITROL 380 mg group and 13 in the placebo group. Those are outcome-specific analysis counts; the registry does not report why they differ from the overall enrollment count. It states that no data were imputed for the primary analysis. The original sample-size justification, recruitment targets by arm, and reasons for missing or unavailable imaging data are not reported in the registry record.

Participants who completed the double-blind portion could choose to continue in an optional open-label period, receiving 2 additional months of VIVITROL 380 mg. That continuation is not part of the randomized placebo comparison, and the registry does not describe it as randomized. Its safety group includes 22 participants at risk. The registry also lists two secondary outcomes measured from baseline to Day 28: change in Obsessive-Compulsive Drinking Scale score and change in daily craving score based on Actiwatch data. The timing and populations for these outcomes differ from the primary fMRI analysis.

The record reports one sponsor analysis for the primary comparison, identified as a one-sided t-test. It does not provide an estimate, confidence interval, or other test details for that analysis. The record’s full statistical plan, including any prespecified adjustments or interim analyses, is 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: Screening evaluations were conducted during a 1-week period. All screening evaluations were to be completed at least 2 days before randomization and dosing.

Double-blind Period

Reasons for not completing this period:

Optional Open-label VIVITROL (2 Months)

Analysis population for the primary outcome, as defined in the registry: Analyses include all randomized, dosed subjects who had functional magnetic resolution imaging (fMRI) on Day 14. No data were imputed.

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 Change From Baseline in Blood Oxygen-level-dependent (BOLD) Signal Activation Values Detected in the Reward Circuitry of the Brain in Alcohol-dependent Subjects After Presentation of Alcohol-related Cues., measured over 14 days (Baseline to Day 14).

The primary endpoint measures the change from baseline to Day 14 in BOLD signal activation values detected in brain reward circuitry when participants were presented with alcohol-related cues. BOLD is a functional imaging signal that reflects changes associated with brain activity; it is not itself a direct measure of craving, drinking behavior, or treatment benefit. The registry gives the unit as percentage (%) of change and identifies the summary as a mean with standard deviation.

The study description connects the endpoint to a specific research question: whether VIVITROL attenuates or blocks increases in BOLD signal in response to alcohol-related cues. In general, a cue-related imaging endpoint can help researchers investigate a biological response to a stimulus under controlled conditions. It does not, by itself, establish what participants experience outside the scanning setting or what happens to later drinking outcomes.

The registry notes that, as was standard among fMRI studies at the study site at the time, a change in BOLD signal in the range of 5% to 6% in anterior cingulate measured using a 3T magnet was considered highly significant in block-designed experiments. The registry does not explain how that site-level reference relates numerically to the posted group summaries. Interpretation should therefore keep the endpoint’s stated units and analysis context in view rather than treat the benchmark as a direct threshold for the reported between-group estimate.

7. Results Posted in the Registry

• VIVITROL 380 mg: mean 0.04 versus -0.04 on control; difference 0.08 (95% CI 0.05 to 0.12), standardized effect d = 2.00; z-test p = <0.001.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Change From Baseline in Obsessive-Compulsive Drinking Scale (OCDS) Score in Alcohol-dependent Subjects

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • VIVITROL 380 mg: mean -8.50 versus -7.30 on control; difference -1.20 (95% CI -5.54 to 3.14), standardized effect d = -0.20; z-test p = 0.588.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between VIVITROL 380 mg and the control. The interval gives the range of effects that remain compatible with the data.

Change From Baseline in Daily Craving Score in Alcohol-dependent Subjects (Actiwatch Data)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • VIVITROL 380 mg: mean -0.85 versus -0.79 on control; difference -0.06 (95% CI -0.74 to 0.62), standardized effect d = -0.06; z-test p = 0.869.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between VIVITROL 380 mg and the control. The interval gives the range of effects that remain compatible with the data.

Across the two secondary outcomes, the posted pattern is less decisive than the primary imaging comparison: the recalculated mean-difference intervals include zero, and the unadjusted p-values are 0.588 and 0.869. The outcomes concern change in Obsessive-Compulsive Drinking Scale score and daily craving score from baseline to Day 28. They measure different aspects of drinking-related experience than the primary Day 14 fMRI response, so their results should be regarded as related evidence rather than direct replications of the primary endpoint.

The secondary findings do not establish that there is no difference between groups. In general, a non-significant test can reflect a small or uncertain effect, limited information, or both. The registry’s planning illustrations for these outcomes are based on the observed effects and are not evidence about the actual study’s design adequacy. The original sample-size rationale for each endpoint is not reported in the registry record.

Unadjusted secondary results carry less weight because multiple outcomes create more opportunities for apparently noteworthy findings by chance. In general, a multiplicity strategy can control the interpretation of a family of tests, but the registry does not report whether such a strategy was used. These results should therefore be read as exploratory context rather than a definitive set of independent confirmations. The record reports two secondary outcomes, while the overall record also lists five outcome measures and three analyses; it does not explain the complete relationship among those entries. The posted secondary pattern offers no clear corroboration of the primary imaging result, but it also cannot settle the broader clinical question.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Change From Baseline in Obsessive-Compulsive Drinking Scale (OCDS) Score in Alcohol-dependent Subjects
• Change From Baseline in Daily Craving Score in Alcohol-dependent Subjects (Actiwatch Data)

9. Statistical Interpretation

10. Statistical Methods Explained

In general, a mean summarizes the center of a continuous outcome by averaging observed values. The registry identifies the primary and secondary outcomes as means and reports standard deviations for the arms. A standard deviation describes the spread of individual observations around a group mean; it is not the uncertainty in the mean difference itself. The spread reported for the primary outcome is 0.036 in the VIVITROL group and 0.049 in the placebo group. These summaries give context for the variability among analyzed participants, but do not reveal the shape of the outcome distributions or whether there were influential observations.

What does the mean difference represent?

In general, a mean difference subtracts one group mean from another, with the direction depending on the stated comparison. For the primary outcome, the recalculated comparison is VIVITROL 380 mg relative to placebo. The posted means are 0.041 and -0.044, respectively, and the recalculated mean difference is 0.085. The positive direction means that the average change value was higher in the VIVITROL group than in the placebo group. That direction is a description of the outcome scale, not an automatic interpretation as benefit or harm. The clinical meaning of a higher value is not fully defined by the registry.

What does a confidence interval add?

In general, a confidence interval gives a range of values compatible with the observed data under the assumptions of the method. It conveys estimation uncertainty and is more informative about magnitude than a p-value alone. The recalculated primary interval excludes zero, so it is consistent with a positive between-group mean difference under the calculation used. The sponsor’s reported analysis does not include an interval in the registry. The recalculated intervals are unadjusted and should not be treated as a replacement for a complete prespecified analysis.

What does a p-value tell the reader?

In general, a p-value describes how unusual data at least as incompatible with a null hypothesis would be if that null hypothesis and the test assumptions were correct. It is not the probability that the null hypothesis is true, nor does it quantify the probability that a result will replicate. The sponsor reports a one-sided t-test for the primary comparison. The recalculated unadjusted p-value is a separate result. The registry does not report the sponsor test’s full specification, including its null formulation, or explain why a one-sided test was used. That missing context matters when comparing the sponsor’s result with a recalculation using summary statistics.

In general, a t-test evaluates a difference in means relative to its estimated uncertainty. Its validity depends on assumptions concerning the observations and the way the comparison was formed. A summary-statistic recalculation cannot verify those assumptions from individual participant data. It also cannot reproduce adjustments, covariates, or data handling that may have been used in a fuller analysis unless those details are reported. The registry does not report a covariate-adjusted model for the posted primary result.

Why distinguish the randomized result from the open-label period?

In general, randomization supports a comparison between assigned groups by helping balance measured and unmeasured characteristics on average at the time of assignment. Masking can reduce the influence of expectations on participant experience, outcome assessment, or study conduct. This trial is listed as randomized and quadruple-masked, but the registry does not define which four roles were masked or explain the procedures used to maintain masking. The optional open-label period differs: participants could opt to continue and receive VIVITROL, so it does not preserve the same randomized placebo comparison. Its observations should not be combined with the double-blind contrast as if they arose from the same design.

In general, missing outcome data can affect both the number of observations and the interpretation of a comparison. The registry says that the primary analysis included randomized, dosed participants with Day 14 fMRI and that no data were imputed. This tells readers that the reported analysis did not fill in missing values through an imputation procedure, but it does not say why imaging data were unavailable or whether missingness differed by group. The analysis population is therefore important context for the primary result. The registry’s descriptions of the secondary outcome analysis populations are reflected in their outcome-specific group counts, but reasons for unavailable measurements are not reported.

What can an unadjusted recalculation establish?

In general, a recalculation from group means, standard deviations, and sample sizes can offer a simple, transparent comparison based on posted summary statistics. It is not necessarily identical to the sponsor’s analysis. The primary recalculation is identified as unadjusted, and its p-value is unadjusted. It does not account for unreported analytic choices, multiplicity, or missing-data mechanisms. The record indicates no discrepancy, so the sponsor’s own analysis takes precedence; the recalculation remains a complementary way to understand the direction, magnitude, and approximate precision represented by the posted summaries.

General statistical interpretation also distinguishes statistical evidence from clinical importance. A result can be statistically separated from zero without showing that the difference is meaningful to participants, durable, or related to a clinical outcome. Here the primary endpoint is fMRI activation following alcohol-related cues, while the secondary measures concern drinking-related self-report and daily craving scores. Each answers a different question. The registry does not report a threshold for a clinically important between-group difference on the primary endpoint, and the imaging-site benchmark is not shown to be a treatment-effect threshold.

11. Confidence Intervals and Precision

In general, a confidence interval communicates how precisely an effect is estimated under the calculation’s assumptions. A narrower interval indicates less statistical uncertainty about the range of compatible values, while a wider interval indicates more. It does not account automatically for bias, unreported design choices, or uncertainty introduced by selective reporting. For the primary comparison, the recalculated interval lies above zero, indicating that the posted summary statistics support a positive mean difference under the unadjusted calculation. The registry does not give a sponsor-reported confidence interval for this outcome.

The primary analysis uses 15 participants in the VIVITROL group and 13 in the placebo group. The observed group standard deviations are 0.036 and 0.049. The estimate and its interval should be read in light of those group counts and the fact that the registry states no data were imputed. The interval quantifies sampling uncertainty from the recalculation; it does not resolve uncertainty about why other enrolled participants lack a Day 14 fMRI result or whether the analyzed participants differ systematically from those not analyzed.

For the secondary comparisons, the recalculated confidence intervals include zero. In general, an interval that includes zero means the posted data do not clearly distinguish the observed mean difference from no difference at the confidence level used in the calculation. It does not establish equivalence or rule out effects that may matter. The registry does not report a prespecified equivalence margin. The secondary estimates therefore remain uncertain, and their interpretation is further limited by the absence of a reported multiplicity strategy.

Precision is not the same as importance. Even a relatively precise estimate would need a clinically interpretable scale and a relevant outcome to support a conclusion about practical meaning. The primary imaging measure has a registry-described site benchmark, but the record does not explain its direct comparability to the reported group summaries. Precision also cannot correct limitations of the outcome’s scope: an fMRI response to cues is not itself a measure of subsequent drinking or sustained change.

12. 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 figures are illustrations based on the observed effect, not the trial’s original design rationale. For these continuous outcomes, the figures are participants per arm, not total events. The listed assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; each planning scenario uses the stated fraction of the observed effect. The original trial’s own sample-size assumptions are not reported in the registry record.

Read each scenario as a planning question: if a future study aimed to detect the full observed effect, a smaller number of participants per arm is indicated than if it aimed to detect only a fraction of that effect. For the primary comparison, the planning illustration lists 4 participants per arm for 80% power and 6 per arm for 90% power when planning for 100% of the observed effect. For 75% of the observed effect, the corresponding figures are 7 and 10 per arm; for 50%, they are 16 and 22 per arm.

In general, 80% power means an 80% chance of detecting an effect of the specified size under the planning assumptions if that effect is real; 90% power describes a higher chance under those assumptions. It does not mean that a study has an 80% or 90% probability of being correct. Smaller effects are harder to detect and require more participants per arm under otherwise unchanged assumptions. The illustrations are not observed power and should not be read as evidence that the completed study was adequately sized for its original objective.

13. 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 = 0.085; pooled standard deviation σ = 0.042.
2. n = 2 × (1.96 + 0.842)² × 0.042² / 0.085² = 3.9, rounded up to 4 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 <- 0.085                      # observed difference in means
sigma <- 0.0424970587217516                     # 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(511836)
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 = 4 (80% power), 6 (90% power)
 75% of observed effect: n per arm = 7 (80% power), 10 (90% power)
 50% of observed effect: n per arm = 16 (80% power), 22 (90% power)
power.t.test: 5.1 
Simulated power at n = 4 per arm: 0.658 (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.

14. Analyses Reported by the Sponsor

The sponsor’s primary analysis is reported as a one-sided t-test, with p < 0.001. The registry does not give the sponsor’s corresponding effect estimate, confidence interval, test statistic, or detailed model specification. The recalculation uses the posted group means, standard deviations, and group counts to produce an unadjusted mean difference and an unadjusted p-value. It is a simple comparison from summaries, not a reconstruction of every possible feature of the sponsor’s analysis.

The registry record marks no discrepancy between the sponsor analysis and the recalculation. Under the stated interpretation for these results, the sponsor’s own analysis takes precedence. The recalculated result is useful as a transparent companion summary of magnitude and precision, but it should not be presented as correcting or replacing the sponsor’s test. In particular, the sponsor’s use of a one-sided test and the rationale for that direction are not explained in the record. The original statistical plan is not reported in the registry record, so it is not possible to establish from the posted information whether the one-sided specification was prespecified.

The sponsor’s p-value and the recalculated interval answer related but distinct questions. The p-value summarizes evidence under the sponsor’s reported test; the recalculated interval describes uncertainty around a mean difference under the recalculation’s assumptions. The registry lacks enough detail to reconcile the methods beyond noting that both analyses indicate a separated primary comparison.

15. Multiplicity, Missing Data and Analysis Populations

The primary analysis population is defined as all randomized, dosed participants who had fMRI on Day 14, with no data imputed. This definition is important because it identifies who contributed to the primary result, but it does not show how many participants were randomized to each group before outcome availability, why some participants lacked imaging, or whether reasons differed by group. Those details are not reported in the registry record. Without them, readers cannot assess whether the analyzed population remains comparable by randomized assignment or whether missing outcome data could have affected the estimate.

In general, imputation is one way to address missing values, but not imputing data does not remove the possibility of bias from missingness. A complete-case analysis can be informative when the analyzed participants represent the randomized groups appropriately; whether that condition holds cannot be determined from the posted information. The registry’s secondary outcomes have their own outcome-specific counts, and it does not provide a detailed account of missingness or analysis populations for each measure beyond those counts.

Multiplicity arises when several outcomes or tests are examined. In general, testing multiple hypotheses without an adjustment can increase the chance that at least one result appears statistically persuasive by chance. This trial lists multiple outcome measures and analyses, including the primary imaging outcome and secondary craving-related measures. The registry does not report a multiplicity adjustment, a hierarchy for testing outcomes, or whether the primary endpoint was the sole confirmatory test. Consequently, the unadjusted secondary p-values should be interpreted cautiously, and the relationship between the number of analyses and the sponsor’s reported primary test is not fully documented.

The registry also does not report interim analyses, stopping rules, a complete prespecified statistical analysis plan, or details about handling protocol deviations. These omissions do not show that such procedures were absent; they mean their role cannot be assessed from the registry record. The same restraint applies to subgroup analyses and covariate adjustment: unless described in the record, their use should not be assumed.

16. 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 1% reporting threshold.

Most frequent non-serious adverse event terms

For the double-blind period, the registry reports 0 serious adverse events among 16 participants at risk in the VIVITROL 380 mg group and 0 among 15 at risk in the placebo group. In the optional open-label VIVITROL period, it reports 0 serious adverse events among 22 participants at risk. These are counts for the registry’s stated periods and at-risk groups. They do not provide a complete picture of event timing, exposure, or the nature of non-serious events.

The registry reports other adverse events affecting 6 of 16 participants in the double-blind VIVITROL group, 10 of 15 in the double-blind placebo group, and 7 of 22 in the optional open-label period. Specific serious adverse-event terms are not reported in the registry record. Death counts are also not reported in the registry record.

In general, zero observed serious events in a small group does not establish that the risk is absent. The open-label group is optional and is not a randomized comparison, so its count should not be compared with the blinded groups as though allocation and follow-up were equivalent. These registry figures describe what was recorded for the listed study periods; they do not support broader safety conclusions beyond those observations.

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

The primary result describes a difference in average change in an fMRI signal after alcohol-related cues among the analyzed participants. The VIVITROL group mean was 0.041 and the placebo group mean was -0.044; the recalculated mean difference was 0.085. The recalculated interval excludes zero, and the sponsor reported a one-sided t-test with p < 0.001. These findings indicate a statistically separated primary imaging comparison in the posted analysis.

That result does not, by itself, show that VIVITROL reduces alcohol craving, changes drinking behavior, or improves a clinical outcome. The endpoint is a brain activation measure collected at Day 14, not a direct assessment of later alcohol use. The study also posted secondary measures of obsessive-compulsive drinking symptoms and daily craving at Day 28, but their unadjusted comparisons are not statistically decisive. The registry does not report whether changes in the imaging measure are validated as predictors of clinical outcomes.

Nor does statistical separation establish that the magnitude is clinically important. The record notes an imaging-site reference range of 5% to 6% in anterior cingulate for block-designed experiments, but it does not explain whether that benchmark can be directly compared with the reported between-group summaries. A threshold for a clinically meaningful treatment difference on this endpoint is not reported in the registry record.

The result also does not settle how well the finding would generalize beyond the analyzed participants or study setting. The primary analysis included participants with Day 14 fMRI and did not impute missing data. The reasons for unavailable imaging and the full original analysis plan are not reported. The appropriate conclusion is therefore limited: the posted summaries show a primary imaging difference under the reported analyses, while its clinical meaning, robustness, and relationship to subsequent drinking outcomes remain uncertain.

18. Why This Trial Matters Statistically

This trial illustrates why statistical interpretation should separate an endpoint’s measured quantity from the broader clinical question that motivates it. The primary outcome is an fMRI response to alcohol-related cues, while the clinical context is treatment-seeking adults with alcohol dependence. A statistically distinct imaging result is evidence about the measured signal in the analyzed group; it does not automatically translate into a change in craving or drinking.

It also demonstrates the value of presenting both an estimate and its uncertainty. A p-value can indicate evidence against a null hypothesis under a specified test, but an effect estimate and confidence interval help readers assess direction and precision. Here the sponsor’s reported analysis gives a one-sided p-value without an estimate or interval, while the recalculation from posted summaries supplies a mean difference and interval. The two forms of information are complementary, and the sponsor’s analysis remains the primary reported test.

The trial further emphasizes the importance of analysis populations. Enrollment is 31, whereas the primary outcome counts are 15 and 13, and the registry defines the primary population by Day 14 imaging availability. No imputation was used. Understanding how such outcome availability relates to randomization is essential when evaluating whether a comparison retains the protections of a randomized design.

Finally, secondary outcomes and safety counts need careful framing. The secondary results are not strong confirmations of the primary signal, and no multiplicity approach is reported. Zero serious adverse events in the listed groups is a factual count, not proof of no risk. Together, these points show how a registry result can be statistically informative yet remain limited in clinical interpretation and certainty.

19. 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 result is based on a specific imaging outcome at Day 14 among randomized, dosed participants with available fMRI. The registry states that no data were imputed but does not report reasons for missing imaging, how many participants were initially assigned to each group, or whether missingness differed between groups. This limits assessment of selection into the analyzed population.

The outcome is a BOLD signal response to alcohol-related cues, not a direct behavioral or clinical endpoint. Although the registry describes a site-level reference for anterior cingulate signal changes, it does not explain whether that reference is directly comparable to the posted summary values or define a clinically meaningful treatment difference. The relationship between imaging change and later craving or drinking is not reported.

The sponsor’s analysis is identified as a one-sided t-test, but the registry does not report its full specification, effect estimate, confidence interval, or rationale for the direction of testing. The recalculation is unadjusted and based on posted summary statistics, so it cannot verify individual-level assumptions or unreported analytic choices. The record indicates no discrepancy, but does not provide a complete statistical analysis plan.

The optional open-label period differs from the randomized comparison and is subject to participant choice. Its safety count should not be treated as a randomized efficacy estimate. The registry also does not report a multiplicity strategy for the multiple outcomes and analyses, nor interim-analysis details. These are limitations of what can be assessed from the record, not evidence that procedures were or were not undertaken.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: ALK21-018: Effects of Medisorb® Naltrexone (VIVITROL®) on Alcohol Craving in Treatment-seeking, Alcohol-dependent Adults — NCT00511836.
• Registry record last updated 2017-07-11; status last verified 2017-06.
• Results first posted 2010-12-07.
• 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.

23. Record Summary

NCT00511836 is a completed randomized, quadruple-masked, parallel-group phase 4 study sponsored by Alkermes, Inc., enrolling 31 participants with alcohol dependence. It compared a single double-blind administration of VIVITROL 380 mg with placebo and assessed change in cue-related BOLD activation from baseline to Day 14. The primary analysis included participants with Day 14 fMRI and used no imputation.

The posted primary means were 0.041 for VIVITROL and -0.044 for placebo, with a recalculated mean difference of 0.085. The sponsor reported a one-sided t-test with p < 0.001; the registry reports no discrepancy with the recalculation. Two secondary outcomes measured at Day 28 have unadjusted comparisons whose confidence intervals include zero. Participants could opt into an additional open-label VIVITROL period. The results describe an imaging comparison, and the registry does not establish its clinical meaning or report a complete statistical plan.