A Study to Investigate the Efficacy, Safety, and Tolerability of JNJ-42165279 in Adolescent and Adult Participants With Autism Spectrum Disorder: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Investigate the Efficacy, Safety, and Tolerability of JNJ-42165279 in Adolescent and Adult Participants With Autism Spectrum Disorder (NCT03664232), a randomized phase 2 study. The primary outcome is Change From Baseline to Day 85 in the Autism Behavior Inventory (ABI) Core Domain Score (Social Communication and Restrictive Behaviour).

2. Clinical Question

The primary outcome combines social communication and repetitive/restrictive behavior scores. A lower score indicates less severe symptoms on the ABI scoring description, so a negative change from baseline represents improvement within a group. The statistical comparison of interest is the difference in change between groups, rather than whether either group changed from baseline by itself.

The study was registered as a randomized, double-blind, parallel-group, multicenter, placebo-controlled phase 2 trial. Its stated purpose was to evaluate efficacy, safety, and tolerability during 12 weeks of treatment. The registry reports enrollment of 78, while the primary outcome analysis population is defined separately and the posted arm summaries report analyzed participant counts for the BID groups.

The core question is therefore about the estimated contrast between JNJ-42165279 and placebo on a continuous symptom score at the specified assessment time. The posted results include a sponsor-reported repeated-measures model and a simple recalculation from arm-level means and standard deviations. Those analyses use different approaches, so their estimates should be read in their own contexts rather than treated as interchangeable. Neither the comparison nor its uncertainty establishes an individual participant's response.

3. Trial Design

This was a phase 2 randomized, double-blind, parallel-group, multicenter study of JNJ-42165279 versus placebo in adolescent and adult participants with autism spectrum disorder. The registry lists the study as completed. The study start date is 2018-11-07 and the primary completion date is 2022-10-17. The planned treatment period described in the registry was 12 weeks.

Participants in the JNJ-42165279 BID arm were to self-administer 25 mg tablets orally twice daily. Participants in the placebo BID arm were to self-administer matching placebo tablets orally twice daily. The registry also lists QD groups in the adverse-event results, but the primary efficacy comparison described in the posted results is between Placebo BID and JNJ-42165279 25 mg BID. The results reported for those efficacy groups should not be conflated with the additional QD safety groups.

Enrollment in the registry is 78. The primary efficacy population is not defined simply as everyone enrolled: the registry says the full analysis set at the BID dose included randomized participants who received at least one dose at that dose and had a baseline and at least one postbaseline efficacy assessment. For the primary summary, the posted analyzed counts are 25 in the Placebo BID arm and 26 in the JNJ-42165279 25 mg BID arm. Those counts describe the analysis of this endpoint, not the overall registered enrollment.

The primary endpoint was change from baseline to Day 85 in the ABI core domain score, described as the combined social communication and repetitive/restrictive behavior domains. The sponsor's reported analysis used a mixed effects model for repeated measures and gave a least-squares mean difference. A separate unadjusted recalculation compares the posted group means using the summary statistics. The registry record does not report the trial's original sample-size assumptions or a detailed statistical analysis plan.

The listed secondary outcomes cover several ABI domains, other behavior scales, clinician and self-rated assessments, and global severity or improvement measures. The record lists 20 outcome measures and 3 analyses, with 16 secondary outcomes posted. Such breadth gives context to the primary endpoint, but results across multiple measures require care because the record does not report a multiplicity adjustment strategy.

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: The full analysis set twice daily (FAS-BID) included all randomized participants who had received at least 1 dose of study agent at the BID dose and had both a baseline and at least 1 postbaseline efficacy assessment. Here, N (number of participants analyzed) signifies participants who were analyzed for this outcome measure. As per study statistical analysis plan (SAP), data for this outcome measure was planned to be collected and analyzed only on the participants who received BID dosing.

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: The full analysis set included all randomized participants who received at least 1 dose of study agent at either dose and had both a baseline and at least 1 postbaseline efficacy assessment.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline to Day 85 in the Autism Behavior Inventory (ABI) Core Domain Score (Social Communication and Restrictive Behaviour), measured over Baseline (Day 1) to Day 85.

The primary endpoint is change from baseline to Day 85 in the Autism Behavior Inventory core domain score, which combines social communication and repetitive/restrictive behavior. The ABI is described as a 62-item questionnaire used to track outcomes in autism spectrum disorder. Items assess either quality of a behavior or its frequency, and responses are scored on scales from 0 to 3. For frequency, the anchors run from never to very often; for quality, from without help to not at all. Higher scores indicate more severe symptoms.

The two core domains comprise 23 social communication items and 15 repetitive/restrictive behavior items. Domain and sub-domain scores are calculated as the average of non-missing items. The core score is calculated by combining the social communication and repetitive/restrictive behavior domain scores with weighting based on the total number of items in those domains. The resulting score ranges from 0 to 3.

Change from baseline is useful because it expresses how the measured score moved over the treatment period relative to each participant's starting point. Negative changes indicate improvement according to the registry's scoring description. The between-group comparison asks whether average change differed between the treatment and placebo groups. That contrast is more directly relevant to a randomized comparison than interpreting within-group change alone, since changes in both groups may reflect factors other than treatment assignment.

The ABI core score focuses on the two domains named in the primary endpoint, while the secondary outcomes assess additional domains and other instruments. It is a continuous measure, so the reported mean difference describes an average contrast; it does not indicate how many participants experienced a particular degree of change.

7. Results Posted in the Registry

• JNJ-42165279 25 mg BID: mean -0.26 versus -0.20 on control; difference -0.06 (95% CI -0.25 to 0.13), standardized effect d = -0.18; z-test p = 0.534.

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.

Self Global Impression of Improvement (Self GI-I) Scale Score at Day 85

• 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:
  • JNJ-42165279 25 mg BID: mean 3.41 versus 3.03 on control; difference 0.38 (95% CI -0.15 to 0.91), standardized effect d = 0.37; z-test p = 0.158.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between JNJ-42165279 25 mg BID and the control. The interval gives the range of effects that remain compatible with the data.

The registered secondary outcomes span ABI mood and anxiety, challenging behavior, and self-regulation domains; ABC subscales; ABI short-form and clinician-interview domains; clinical global severity; and repetitive behavior scores. The registry lists 16 secondary outcomes as posted. This breadth means the secondary measures address several related but distinct aspects of behavior and assessment, rather than serving as interchangeable versions of the primary endpoint.

The available secondary summary-level comparison is for Self Global Impression of Improvement at Day 85. It has a positive mean difference, but the registry record does not state the scale anchors needed to determine the meaning of a higher score. Its unadjusted interval includes zero. The record does not provide sponsor-reported analyses for this secondary measure in the information available for interpretation, so it should not be treated as a confirmed secondary finding.

In relation to the primary result, the secondary information does not provide a clear basis for changing the interpretation of the ABI core endpoint. The primary estimate itself is uncertain, and the secondary measure also has uncertainty. Nor is it possible to make a reliable claim about a consistent pattern across all secondary outcomes from the available summary information. The list of outcomes alone does not indicate whether effects aligned in direction or differed in magnitude.

As a general statistical concern, examining many outcomes creates more opportunities for apparently notable findings to arise by chance. Unadjusted secondary p-values do not account for the full set of comparisons. A multiplicity procedure can control error across a defined family of tests, but the registry record does not report whether a procedure was planned or used. Secondary results can be informative and hypothesis-generating, but they carry less confirmatory weight than a prespecified primary analysis when multiplicity handling is not reported.

Interpretation should therefore consider the secondary endpoints as supporting context, not as a collection of independent confirmations. Differences in instrument, domain, assessment method, and scale also matter. Without detailed analysis plans and complete outcome-specific results, neither selective emphasis on one measure nor a claim of broad consistency is justified.

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

• Change From Baseline to Day 85 in the ABI Mood and Anxiety Domain Score
• Change From Baseline to Day 85 in the ABI Challenging Behavior Domain Score
• Change From Baseline to Day 85 in the ABI Self-Regulation Domain Score
• Change From Baseline to Day 85 in the Aberrant Behavior Checklist (ABC) Subscales Scores
• Change From Baseline to Day 71 in the Autism Behavior Inventory-Short Form (ABI-S) Domains Scores
• Change From Baseline to Day 85 in the Autism Behavior Inventory-Clinician Interview (ABI-C) Domains Scores
• Change From Baseline to Day 85 in the Clinical Global Impression-Severity (CGI-S) Scale Score
• Change From Baseline to Day 85 in the Repetitive Behavior Scale - Revised (RBS-R) Subscale Score

9. Statistical Interpretation

10. Statistical Methods Explained

What does the primary estimate compare?

The primary endpoint is a continuous outcome: change in ABI core domain score from baseline to Day 85. The basic group-summary approach compares the average change in the JNJ-42165279 BID arm with the average change in the placebo BID arm. The reported unadjusted mean difference is -0.06, obtained by contrasting the posted group means. Since more negative ABI change indicates improvement, the negative sign points toward a larger average reduction in the treatment arm in this sample. The sign does not by itself indicate whether the estimated contrast is precise or distinguishable from zero.

As a general statistical concept, a mean difference is expressed in the units of the outcome. It summarizes an average contrast, not the range of individual responses. Participants can have changes that differ substantially from the group mean, and a group-average effect does not tell the reader how outcomes are distributed among individuals.

Why is there more than one estimate?

The registry reports the sponsor's primary analysis as a mixed effects model for repeated measures, with a least-squares mean difference of -0.05. Such a model is generally used for repeated measurements over time and can estimate group contrasts while accounting for the structure of those repeated observations. The registry record does not give the model specification, covariance structure, covariates, or detailed handling of missing outcome observations. Those details cannot be inferred from the method name.

The recalculated estimate uses the posted arm-level summary values and is an unadjusted comparison. It does not reproduce the sponsor's model. It therefore serves as a straightforward check on the direction and approximate size of the contrast in the available summaries, while the sponsor's model remains the trial-reported analysis. Because the discrepancy flag is false, there is no indicated discrepancy requiring a special qualification; the difference between estimates should still be understood as a consequence of different analysis approaches and summaries.

As a general statistical concept, a model-based least-squares mean may differ from a raw group mean contrast because it reflects the model's treatment of repeated observations and any included terms. The registry's reported analysis does not give enough detail to identify which specific features account for the difference in this trial. It would be inappropriate to assert that a particular adjustment was used when it is not reported.

How should the p-values be understood?

The unadjusted recalculation gives p = 0.534. The sponsor's reported analysis gives p =0.284. A p-value summarizes how compatible the observed data are with a specified null model under the assumptions of the analysis. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of an effect. A larger p-value does not prove there is no effect; it signals that the analysis does not provide strong evidence against its null model.

These two p-values are not direct substitutes for each other. They arise from different analyses, and the sponsor's p-value belongs to the repeated-measures model. The registry identifies the reported comparison as a superiority analysis, but the record does not report whether any other inferential procedures or adjustments were part of a wider plan. The results should therefore be interpreted alongside the estimate and confidence interval, not by a threshold alone.

Why does variability matter?

The ABI endpoint is summarized using means and standard deviations. Standard deviation describes the spread of participant scores around the group mean. As a general statistical concept, greater within-group variability makes it harder to estimate a between-group mean difference precisely when group sizes are otherwise similar. The primary arm summaries report standard deviations of 0.4 for placebo and 0.274 for JNJ-42165279. The recalculation uses the posted summaries rather than participant-level data.

The posted standard errors are 0.08 for placebo and 0.054 for JNJ-42165279. A standard error describes uncertainty in an estimated mean, rather than variation among individual scores. Standard errors contribute to the uncertainty of a contrast and hence to its confidence interval. The registry record does not post participant-level values, so additional distributional checks or alternative analyses cannot be performed from the summaries alone.

What is the role of the analysis population?

The registry defines the BID full analysis set as randomized participants who received at least one dose at the BID dose and had both a baseline and at least one postbaseline efficacy assessment. This definition excludes people who do not meet those conditions from that endpoint analysis, even though they may be included in overall enrollment or other analyses. The arm-level analyzed counts are 25 and 26 for the primary endpoint. These are analysis counts, not a replacement for the registered enrollment figure.

As a general concept, an analysis population definition affects what comparison the results represent. Randomization establishes the assigned groups, but restricting an analysis to participants with postbaseline data can make the analyzed population differ from all randomized participants. Whether that affects the estimate depends on reasons for missingness and how missing data are handled. The registry record does not report enough details to evaluate those questions fully.

What can and cannot be inferred?

The reported mean difference and interval address an average group contrast on the ABI score. They do not establish a treatment mechanism, prove a clinical benefit, or describe the outcome of every participant. The registry record does not report a clinically important difference threshold for the ABI core score, so the statistical contrast cannot be classified against such a threshold using the posted information.

As a general statistical concept, a confidence interval conveys a range of estimates reasonably compatible with the data under the analysis assumptions. Its width reflects precision, and its location reflects the direction and size of estimates. The interval should be considered with the selected analysis, endpoint definition, measurement timing, and population. It cannot eliminate uncertainty about unreported model details, missing observations, or the broader set of outcomes examined.

11. Confidence Intervals and Precision

The recalculated primary mean difference is -0.06, with a 95% confidence interval from -0.249 to 0.129. This interval includes zero and spans values on either side of it. Its width indicates that the summary data do not pin down the group contrast precisely. The point estimate is only one location within that range; it should not be read as if it were known without uncertainty.

As a general statistical explanation, a confidence interval is a range produced by a method that, over repeated studies under its assumptions, would contain the true parameter at the stated confidence level at the corresponding rate. It does not mean there is a 95% probability that this particular fixed interval contains the true effect. Interpretation also depends on whether the analysis assumptions are reasonable and whether the interval corresponds to the question of interest.

The sponsor reports an 80% confidence interval from -0.17 to 0.07 for the least-squares mean difference. This is narrower than a confidence interval at a higher confidence level would generally be, because lower confidence coverage permits a narrower interval under a given method. The sponsor interval also includes zero. It is tied to the sponsor's mixed effects model, so its bounds should not be substituted for those of the unadjusted recalculation.

Precision depends on the variability of the outcome, the amount of information available for analysis, and the chosen analysis. For the primary comparison, the posted arm summaries show standard deviations of 0.4 and 0.274, with analyzed counts of 25 and 26. These summaries give context for why uncertainty remains, but they do not permit a full evaluation of the repeated-measures model or data patterns over time.

The secondary Self Global Impression of Improvement comparison also has an interval that includes zero, extending from -0.147 to 0.907. Its endpoint uses a different scale, and the registry does not state the scale anchors. Precision should not be compared across endpoints without considering their units and meaning. In both cases, uncertainty is relevant: the point estimates do not alone show how well the data distinguish among plausible group differences.

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 sample-size figures are planning illustrations based on the observed effect, not the original design assumptions for this trial. For these continuous-outcome comparisons, the figures are participants per arm, not total events. The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is the stated fraction of the observed effect. The registry record does not report the trial's own original sample-size assumptions or justification.

For planning for 100% of the observed primary effect, the illustration gives 509 participants per arm for 80% power and 682 participants per arm for 90% power. For planning for 75% of the observed effect, it gives 905 participants per arm for 80% power and 1,211 participants per arm for 90% power. For planning for 50% of the observed effect, it gives 2,035 participants per arm for 80% power and 2,725 participants per arm for 90% power.

The 80% power column means the illustrated design would have an 80% chance of detecting the specified effect under the stated assumptions if that effect were the true effect. The 90% column describes a higher chance of detection under those same assumptions. These are design calculations, not probabilities that the trial result is correct. Smaller effects require more participants to detect reliably, which is why the planning counts rise as the assumed fraction of the observed effect falls.

A sample-size curve conveys the same relationship visually: the required number of participants depends on the effect size being targeted and the selected power. These calculations use the observed effect as their starting point, so they are sensitive to the uncertainty in that estimate. This is not observed power. The figures are illustrations for future planning under explicit assumptions, not evidence that the completed trial was adequately or inadequately sized according to its own protocol.

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 = -0.060; pooled standard deviation σ = 0.342.
2. n = 2 × (1.96 + 0.842)² × 0.342² / -0.060² = 508.7, rounded up to 509 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 <- -0.06                      # observed difference in means
sigma <- 0.341571996175665                     # 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(3664232)
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 = 509 (80% power), 682 (90% power)
 75% of observed effect: n per arm = 905 (80% power), 1,211 (90% power)
 50% of observed effect: n per arm = 2,035 (80% power), 2,725 (90% power)
power.t.test: 509.7 
Simulated power at n = 509 per arm: 0.798 (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 reported primary analysis is a mixed effects model for repeated measures. It reports a least-squares mean difference of -0.05, an 80% confidence interval from -0.17 to 0.07, and p =0.284. The interval includes zero. This is the analysis reported by the trial and should take precedence as the sponsor's own result.

The recalculated contrast is -0.06, with an unadjusted p-value of 0.534 and a 95% confidence interval from -0.249 to 0.129. It is calculated from posted summary statistics and is not a reconstruction of the repeated-measures model. The sponsor's estimate and recalculated estimate are both negative, but they have different inferential summaries and confidence levels. They should not be merged into a single analysis.

The registry record does not provide the model specification or enough information to reproduce the sponsor's least-squares estimate. The recalculation is therefore best read as a simple, unadjusted comparison that offers context for the posted group means. It cannot replace the sponsor's analysis, establish why the model estimate differs, or resolve uncertainty in the results. The discrepancy flag is false; no discrepancy between the sponsor analysis and recalculation is identified in the record.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists numerous outcome measures, including the primary ABI core domain score and secondary scales or domains. Testing multiple outcomes raises a multiplicity question: as more comparisons are made, the chance of at least one apparently positive result can increase even when no underlying effect exists. This is a general statistical issue, not evidence that any particular result in this trial is false.

The registry record does not report a multiplicity adjustment strategy or identify a family of hypotheses for which error control was planned. It also does not provide enough detail to determine whether the secondary outcomes were hierarchical, exploratory, or otherwise governed by a prespecified testing sequence. Consequently, unadjusted secondary p-values should be treated cautiously and should not be read as independent confirmatory evidence alongside the primary result.

The primary endpoint analysis is reported as a mixed effects model for repeated measures, but the registry record does not give the model specification, covariates, covariance structure, or detailed missing-data procedure. Those omissions limit an independent assessment of the analysis assumptions. The posted description of the full analysis set requires randomized participants to have received at least one BID dose and to have baseline and postbaseline efficacy assessments. It does not describe how participants missing later measurements were handled within the model.

As a general concept, repeated-measures methods can use available measurements over time, but their validity depends on model assumptions about the data and missingness. The method name alone does not establish which assumptions were used or whether they are appropriate. The registry record also does not report sensitivity analyses that would show how robust results are to alternative missing-data assumptions.

The registry record does not report the trial's original sample-size justification, interim analyses, or a detailed statistical analysis plan. Their absence from the record does not establish that such procedures were not conducted; it means they cannot be assessed from the registry information. This uncertainty is separate from the uncertainty quantified by the outcome confidence intervals.

16. 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 5% reporting threshold.

Most frequent non-serious adverse event terms

The adverse-event results list serious adverse events and other adverse events separately for placebo and JNJ-42165279 groups at both QD and BID dosing. The registry reports zero participants affected by serious adverse events in each listed group. It also reports zero deaths in each listed group. These are counts as recorded, not evidence that rare harms are impossible or that the groups have identical safety profiles.

For other adverse events, the posted affected counts are 5 for Placebo QD, 7 for Placebo BID, 7 for JNJ-42165279 25 mg QD, and 3 for JNJ-42165279 25 mg BID. The at-risk counts listed for those groups are 7, 30, 8, and 32, respectively. These QD and BID safety groups should remain distinct from the primary efficacy comparison, which concerns the BID groups.

The registry gives no top serious adverse-event terms. Counts alone do not provide event type, timing, severity beyond the serious classification, or a comparative estimate adjusted for exposure. The posted figures therefore describe what was recorded in these groups, but do not support broader safety conclusions beyond the limited registry summary.

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

The primary result estimates an average difference between JNJ-42165279 and placebo in change on the ABI core domain score from baseline to Day 85. The sponsor's repeated-measures estimate is negative, and the unadjusted recalculation is also negative. Because the ABI description defines negative change as improvement, both estimates point toward a greater mean reduction in the treatment group. Their confidence intervals include zero, so the posted results do not clearly distinguish the average group contrast from no difference.

This does not establish that the treatment works, nor does it establish that it does not work. An interval spanning zero reflects uncertainty about the direction and size of the contrast. It does not demonstrate equivalence, because the registry record does not report an equivalence margin or a non-inferiority framework. It also does not show that every participant experienced the group-average change.

The endpoint measures a defined set of ABI behaviors and domains. It is not a direct measure of every aspect of autism spectrum disorder, and the registry does not report a clinically important difference threshold for this core score. Statistical uncertainty therefore cannot be translated into a conclusion about the practical importance of the observed point estimate using the record alone.

The secondary measures add context but do not settle the primary question. The available secondary comparison has uncertainty, the scale anchors are not stated, and the record does not report multiplicity handling. The collection of outcomes should not be treated as a set of independent replications of the primary comparison.

The safety summary records no serious adverse events or deaths in the listed groups, alongside counts for other adverse events. These figures are descriptive and limited to the registry's reported groups and counts. They do not prove the absence of risk or support advice for individual treatment decisions. Overall, the results are best understood as an imprecise estimate from a completed phase 2 trial, with limits in the registry detail available for evaluating analysis choices, missingness, and the wider set of outcomes.

18. Why This Trial Matters Statistically

This trial is statistically informative because it illustrates the distinction between a point estimate and the uncertainty around it. The primary mean difference is close to zero relative to the width of its confidence interval, and the sponsor-reported interval also includes zero. A result should not be reduced to whether a p-value crosses a conventional threshold: the size, direction, precision, analysis method, and endpoint meaning all contribute to interpretation.

It also demonstrates why sponsor-reported analyses and simple recalculations must be distinguished. The sponsor used a mixed effects model for repeated measures, while the recalculation uses posted arm summaries. Their estimates are similar in direction, but the methods and confidence levels differ. A reproducible calculation from summaries can be a useful check without replacing the trial's own analysis or claiming access to the full underlying data.

The primary endpoint is a continuous composite score built from defined ABI domains. Understanding its construction and sign convention is essential: negative change denotes improvement, but an estimated between-group contrast still needs to be interpreted with its uncertainty. This prevents a within-group change from being mistaken for evidence of a treatment-specific difference.

The registry's list of secondary outcomes also highlights the statistical importance of multiplicity and complete reporting. A broad set of outcomes can provide a richer picture, but unadjusted findings across multiple measures require cautious interpretation. Finally, planning calculations based on an observed effect are not the same as the original trial design rationale or observed power. Keeping those concepts separate helps avoid overstating what the trial can establish.

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 registry record gives limited detail about the sponsor's repeated-measures model. It names the method and reports a least-squares mean difference, but does not report the model specification, covariance structure, covariates, or detailed missing-data assumptions. Those omissions prevent a full independent assessment or reproduction of the primary analysis.

The primary analysis population requires at least one postbaseline efficacy assessment as well as baseline data and exposure to BID dosing. The record does not describe how many participants lacked assessments at particular time points, why data were missing, or how missing observations were handled. This makes it difficult to assess whether the analyzed population and model assumptions could affect the estimated contrast.

The registry record does not report the original sample-size rationale, interim analyses, a detailed analysis plan, or a multiplicity strategy. The listed secondary outcomes cover several measures, but the available information does not establish whether results were prespecified in a hierarchy or adjusted across comparisons. Therefore, secondary findings have limited confirmatory interpretation.

The primary endpoint's clinical interpretation is constrained by the absence of a reported clinically important difference threshold. The record describes score construction and direction, but does not establish what magnitude of change would be meaningful to participants or clinicians. The scale anchors for the posted Self Global Impression of Improvement comparison are also not stated.

The safety summary is limited to counts by listed QD and BID groups and does not provide detailed event narratives or exposure-adjusted comparisons. Finally, the primary result concerns a particular BID comparison, endpoint, and assessment time in a phase 2 study. The registry summary alone does not establish how well the finding generalizes beyond that study context.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study to Investigate the Efficacy, Safety, and Tolerability of JNJ-42165279 in Adolescent and Adult Participants With Autism Spectrum Disorder — NCT03664232.
• Registry record last updated 2025-04-23; status last verified 2025-04.
• Results first posted 2025-04-23.
• 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

NCT03664232 is a completed randomized, double-blind, parallel-group, multicenter phase 2 study of JNJ-42165279 versus placebo in adolescent and adult participants with autism spectrum disorder. The registry reports enrollment of 78 and a 12-week treatment period. The primary endpoint was change from baseline to Day 85 in the ABI core domain score.

The sponsor's mixed effects model for repeated measures reported a least-squares mean difference of -0.05, with an 80% confidence interval from -0.17 to 0.07 and p =0.284. An unadjusted recalculation from posted summaries estimated a mean difference of -0.06, with a 95% confidence interval from -0.249 to 0.129 and p-value 0.534. Both intervals include zero.

The registry lists multiple secondary outcomes and safety results for QD and BID groups. The record does not report the original sample-size assumptions, detailed model specification, missing-data procedure, or multiplicity strategy. Interpretation is consequently limited by uncertainty in the estimated effects and by the statistical details not reported in the registry record.