A Study of JNJ-63733657 in Participants With Early Alzheimer's Disease: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of JNJ-63733657 in Participants With Early Alzheimer's Disease (NCT04619420), a randomized phase 2 study. The primary outcome is Double-blind (DB) Treatment Period: Change From Baseline in Integrated Alzheimer's Disease Rating Scale for Mild Cognitive Impairment (iADRS-MCI) Total Score at Week 104.

2. Clinical Question

3. Trial Design

The registry describes a Phase 2 randomized, quadruple-masked, parallel-group study with placebo control and a long-term extension treatment period. The stated purpose was to assess the efficacy and safety of JNJ-63733657 in participants with early Alzheimer's disease. The listed enrollment was 523. The registry lists a start date of 2021-01-06, a primary completion date of 2025-10-17, and a status of terminated. The reason for termination is not reported in the registry record.

The double-blind treatment arms reported for the primary endpoint were JNJ-63733657 1000 mg, JNJ-63733657 3000 mg, and placebo. According to the arm descriptions, treatment or matching placebo was given by intravenous infusion every 4 weeks during the double-blind period. Participants could continue into the long-term extension. The full extension protocol, including the details omitted from the shortened arm descriptions, is not reported in the registry record.

The primary endpoint was planned for the double-blind treatment period and measured change from baseline to Week 104. The primary outcome analysis population was described as the full analysis set of randomized participants who received at least one dose of double-blind study intervention, with the analyzed count depending on evaluability for this outcome. The registry reports 142 analyzed participants in the 1000 mg arm, 132 in the 3000 mg arm, and 137 in the placebo arm.

The registry reports sponsor analyses using mixed models for repeated measures (MMRM), with differences of least-squares means as the comparison measure. More detail on model covariates, the repeated-measures covariance structure, handling of intercurrent events, and the original statistical analysis plan is not reported in the registry record. The trial’s original sample-size assumptions and justification are also 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.

Only the first 6 of 7 groups are shown.

Double-Blind Treatment (Week 0-Week 245)

Reasons for not completing this period:

LTE Period (Day 1 of LTE up to 57 Weeks)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: FAS included all randomized participants who received at least 1 dose of DB study intervention. Here, 'N' (overall number of participants analyzed) signifies the number of participants evaluable for this outcome measure. This outcome measure was planned to be analyzed for DB treatment period only.

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 FAS included all randomized participants who received at least 1 dose of DB study intervention.

6. Primary Endpoint

The registry lists the primary outcome as Double-blind (DB) Treatment Period: Change From Baseline in Integrated Alzheimer's Disease Rating Scale for Mild Cognitive Impairment (iADRS-MCI) Total Score at Week 104, measured over Baseline (Day 1), Week 104.

The primary endpoint was change from baseline in the Integrated Alzheimer's Disease Rating Scale for Mild Cognitive Impairment (iADRS-MCI) total score at Week 104. The measure combines cognition and function into one score. Its cognitive component is the ADAS-Cog13 total score, which ranges from 0 to 85, with higher scores indicating worse cognitive performance. Its daily-function component is the ADCS-ADL-MCI total score, which ranges from 0 to 53, with lower scores indicating worse daily function. The composite ranges from 0 to 138, and lower iADRS-MCI scores indicate worse performance.

Combining cognitive and functional information can provide a single outcome that reflects more than one aspect of disease-related change. It also means that the composite’s direction and interpretation depend on how its component scores are combined: a decline in the composite corresponds to worse performance. In this study, the endpoint was chosen to assess clinical decline over the double-blind treatment period, with baseline recorded at Day 1 and the main assessment at Week 104.

The posted group means are changes from baseline, not the groups’ absolute scores at Week 104. The mean changes were -11.3 for 1000 mg, -11.6 for 3000 mg, and -12.1 for placebo. These summaries describe average change among participants analyzed in each arm; they do not by themselves show how individual participants changed or establish a treatment effect. Comparisons between arms, their uncertainty, and the sponsor’s repeated-measures analyses are needed to assess how compatible the results are with a difference between treatment and placebo.

7. Results Posted in the Registry

• Double-Blind (DB) Treatment Period: JNJ-63733657 1000 mg: mean -11.30 versus -12.10 on control; difference 0.80 (95% CI -2.15 to 3.75), standardized effect d = 0.06; z-test p = 0.595.
• Double-Blind (DB) Treatment Period: JNJ-63733657 3000 mg: mean -11.60 versus -12.10 on control; difference 0.50 (95% CI -2.79 to 3.79), standardized effect d = 0.04; z-test p = 0.766.

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.

Double-blind (DB) Treatment Period: Change From Baseline in Alzheimer's Disease Assessment Scale - Cognitive, 13-item Version (ADAS-Cog13) Total Score at Week 104

• 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:
  • Double-Blind (DB) Treatment Period: JNJ-63733657 1000 mg: mean 6.00 versus 6.10 on control; difference -0.10 (95% CI -1.76 to 1.56), standardized effect d = -0.01; z-test p = 0.906.
  • Double-Blind (DB) Treatment Period: JNJ-63733657 3000 mg: mean 6.50 versus 6.10 on control; difference 0.40 (95% CI -1.54 to 2.34), standardized effect d = 0.05; z-test p = 0.686.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Double-Blind (DB) Treatment Period: JNJ-63733657 1000 mg and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Double-Blind (DB) Treatment Period: JNJ-63733657 3000 mg and the control. The interval gives the range of effects that remain compatible with the data.

Double-blind (DB) Treatment Period: Change From Baseline in Alzheimer's Disease Cooperative Study Activities of Daily Living for Mild Cognitive Impairment (ADCS-ADL-MCI) Total Score at Week 104

• 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:
  • Double-Blind (DB) Treatment Period: JNJ-63733657 1000 mg: mean -5.20 versus -6.40 on control; difference 1.20 (95% CI -0.73 to 3.13), standardized effect d = 0.14; z-test p = 0.224.
  • Double-Blind (DB) Treatment Period: JNJ-63733657 3000 mg: mean -5.10 versus -6.40 on control; difference 1.30 (95% CI -0.58 to 3.18), standardized effect d = 0.16; z-test p = 0.176.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Double-Blind (DB) Treatment Period: JNJ-63733657 1000 mg and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Double-Blind (DB) Treatment Period: JNJ-63733657 3000 mg and the control. The interval gives the range of effects that remain compatible with the data.

Double-blind (DB) Treatment Period: Change From Baseline in Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) Total Scale Index Score at Week 88

• 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:
  • Double-Blind (DB) Treatment Period: JNJ-63733657 1000 mg: mean -3.80 versus -4.70 on control; difference 0.90 (95% CI -1.02 to 2.82), standardized effect d = 0.11; z-test p = 0.359.
  • Double-Blind (DB) Treatment Period: JNJ-63733657 3000 mg: mean -4.70 versus -4.70 on control; difference 0.00 (95% CI -1.81 to 1.81), standardized effect d = 0.00; z-test p = 1.000.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Double-Blind (DB) Treatment Period: JNJ-63733657 1000 mg and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Double-Blind (DB) Treatment Period: JNJ-63733657 3000 mg and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists a broad set of secondary outcomes covering cognition, daily function, neuropsychological status, dementia rating, and neuropsychiatric symptoms. Twenty-five secondary outcomes are posted. The available detailed results include Week 104 changes in ADAS-Cog13 and ADCS-ADL-MCI, and Week 88 change in RBANS Total Scale Index. The posted comparisons for these continuous outcomes generally have intervals that include zero. The secondary analyses therefore do not supply a consistent, precise pattern of separation between either dose and placebo that would materially change the reading of the primary result.

These measures assess domains related to, but distinct from, the composite primary endpoint. ADAS-Cog13 focuses on cognition, ADCS-ADL-MCI on daily activities, and RBANS on neuropsychological performance. Looking across related measures can help describe whether a primary finding is accompanied by similar patterns in component domains. In this record, the secondary results should be treated as supporting context rather than independent confirmation of a treatment effect.

Many outcomes and comparisons create multiplicity: when numerous hypotheses are tested, some apparently notable results can arise by chance even if there are no true differences. The registry record does not report a multiplicity adjustment for the secondary results. Their p-values and intervals should therefore not be interpreted as if each were the sole prespecified test. The number of outcome measures and analyses makes caution especially important, and a secondary result would carry less weight without a clearly reported testing hierarchy, adjustment strategy, and confirmatory pattern.

Because secondary outcomes have different scales, assessment times, and meanings, they should not be reduced to one pooled conclusion. Nor should they be used to override the primary endpoint simply because one measure happens to look more favorable. The registry’s secondary outcome list provides useful breadth, but the missing details about multiplicity and the absence of precise, consistent contrasts limit the strength of conclusions that can be drawn from the secondary findings.

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

• Double-blind (DB) Treatment Period: Change From Baseline in Alzheimer's Disease Assessment Scale - Cognitive, 13-item Version (ADAS-Cog13) Total Score at Week 104
• Double-blind (DB) Treatment Period: Change From Baseline in Alzheimer's Disease Cooperative Study Activities of Daily Living for Mild Cognitive Impairment (ADCS-ADL-MCI) Total Score at Week 104
• Double-blind (DB) Treatment Period: Change From Baseline in Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) Total Scale Index Score at Week 88
• Double-blind (DB) Treatment Period: Change From Baseline in Each of the 5 Individuals RBANS Indices at Week 88
• Double-blind (DB) Treatment Period: Change From Baseline in Each of the 12 Subtests Comprising the RBANS at Week 88
• Double-blind (DB) Treatment Period: Change From Baseline in Clinical Dementia Rating-Sum of Boxes (CDR-SB) Score at Week 104
• Double-blind (DB) Treatment Period: Change From Baseline in Neuropsychiatric Inventory (NPI) Total Score at Week 104
• Double-blind (DB) Treatment Period: Percentage of Participants With Progression From Clinical Dementia Rating-Global Score (CDR-GS) 0 to 0.5 or Higher, 0.5 to 1 or Higher, and 1 to 2 or Higher

9. Statistical Interpretation

10. Statistical Methods Explained

What does a mean change describe?

A mean change summarizes the average difference between baseline and the specified follow-up assessment for participants analyzed in an arm. For the primary endpoint, the posted means are changes from baseline to Week 104. A negative mean indicates that the average score was lower at follow-up than at baseline. Since lower iADRS-MCI scores indicate worse performance, the direction of those changes corresponds to worsening on the composite measure. The mean does not describe every participant: some may have changed more, less, or in a different direction. The standard deviation conveys how spread out participant-level changes were around the arm mean, rather than the precision of the mean itself.

What is a between-arm mean difference?

A between-arm mean difference compares the average outcome in one treatment arm with the average in the placebo arm. In the recalculations, the reported contrasts use the posted arm means. A positive contrast means the treatment-arm change was numerically higher than the placebo-arm change. For this composite, a higher score indicates better performance relative to a lower score, but the contrast must still be interpreted with its uncertainty and with the analysis method in mind. A mean difference is not a percentage change, and it does not indicate how many participants improved or worsened.

The recalculated contrasts are unadjusted comparisons based on the summary statistics. They do not account for repeated measurements over time, baseline covariates, or other model features unless those are already reflected in the posted summaries. The displayed recalculation should therefore be understood as a transparent comparison of the arm-level summaries, not as a reproduction of the sponsor’s complete primary analysis. The registry record does not report the full specifications needed to independently reproduce the MMRM.

Why use a repeated-measures model?

In general, a mixed model for repeated measures uses outcome observations collected at multiple visits to estimate treatment-group differences over time. It can use the pattern of repeated outcome data rather than relying only on one final-visit comparison. In general, such models can accommodate participants with incomplete visit records under assumptions about the missing-data process. The exact assumptions, covariates, visit terms, covariance structure, and missing-data procedures for this study are not reported in the registry record, so the method label alone does not establish which details were used.

The sponsor reports differences of least-squares means from MMRM. In general, least-squares means are model-based group estimates, often adjusted for terms included in the model. They are not necessarily identical to the raw arithmetic means from the arm summaries. This explains why the sponsor’s reported contrasts can differ from unadjusted recalculations without either number being a simple transcription of the other. The registry record identifies MMRM and the reported contrast, but does not provide sufficient specification to assess every model assumption or reproduce the fitted estimates.

How should a confidence interval be read?

In general, a confidence interval expresses the range of estimates compatible with the data and the statistical procedure at a stated confidence level. The sponsor reports 95% confidence intervals. Their width reflects uncertainty: wider intervals allow a broader range of possible differences, while narrower intervals provide greater precision. An interval that includes zero means that a zero difference is among the values compatible with the data under that procedure. It does not prove that the groups are identical. An interval that includes values in both directions also signals uncertainty about which group had the higher underlying outcome.

The unadjusted recalculated intervals are based on posted summary statistics. They are useful as a check on the scale and rough precision of the arm contrasts, but they should not replace the sponsor’s model-based primary analysis. The registry record does not specify a confidence level alongside each recalculated interval, so those should be read as the intervals associated with the recalculation rather than described as the sponsor’s reported 95% intervals.

What does a p-value add?

In general, a p-value measures how unusual data at least as incompatible with a specified null hypothesis would be if that null hypothesis and the analysis assumptions were true. It is not the probability that the null hypothesis is true, nor is it a measure of the size or clinical importance of a difference. The recalculated unadjusted p-values and sponsor-reported MMRM p-values arise from different analyses, so they need not be identical. For the primary endpoint, the reported intervals include zero and the p-values are not small under conventional significance testing. The registry record does not report whether or how multiplicity adjustment was applied to the primary comparisons.

Why do dispersion and sample size matter?

In general, greater variability among individual outcomes makes a group mean less precise when other features are held constant. The posted primary standard deviations are 12.23, 14.58, and 12.87 for the 1000 mg, 3000 mg, and placebo arms, respectively. The analyzed arm counts are 142, 132, and 137. These sample sizes and standard deviations contribute to the uncertainty in unadjusted comparisons. They do not alone determine the precision of the MMRM estimates, which also depends on the repeated measurements and model specification.

General statistical principles also distinguish statistical uncertainty from clinical importance. A narrow interval around a small difference might exclude many larger effects, while a wide interval can remain compatible with a range of effects. Whether a difference is clinically meaningful requires a prespecified interpretation threshold or other clinical context. Such a threshold is not reported in the registry record. Accordingly, the posted estimates describe statistical contrasts, not a definitive judgment about clinical importance.

11. Confidence Intervals and Precision

The sponsor’s MMRM estimate for 1000 mg versus placebo is 1, with a 95% confidence interval from -2.01 to 4.08. For 3000 mg versus placebo, the estimate is 0, with a 95% confidence interval from -3.09 to 3.04. Each interval includes zero and extends to values on both sides. This means the model-based results do not pin down the direction of the difference; they remain compatible with a range of possible contrasts.

The unadjusted recalculations also give intervals spanning zero: -2.148 to 3.748 for 1000 mg versus placebo, and -2.791 to 3.791 for 3000 mg versus placebo. These intervals are based on the posted group summaries, rather than on the sponsor’s repeated-measures model. Their widths indicate that the recalculated contrasts are not estimated with high precision relative to the small point estimates. They should not be treated as interchangeable with the sponsor’s reported confidence intervals.

A confidence interval is not a range that contains the true effect with certainty. In general, it describes uncertainty under a statistical procedure and its assumptions. The sponsor’s intervals are explicitly reported at the 95% level. The registry record does not report the clinical threshold that would distinguish a meaningful from a trivial difference on this endpoint, so the intervals cannot by themselves settle clinical importance.

Precision is also specific to the endpoint, analysis population, time point, and model. The primary outcome analyzed fewer participants than the overall enrollment, and the counts differ by arm. The reasons for those analysis counts and the detailed missing-data methods are not reported in the registry record. The intervals therefore communicate uncertainty in the reported contrasts, while the registry leaves some aspects of the analysis and data availability insufficiently described for a fuller assessment.

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

The 80% power column gives the illustrative number of participants per arm under which a study would have an 80% chance of detecting the specified effect if that effect and the planning assumptions were correct. The 90% power column gives the corresponding number per arm for a 90% chance. These are design calculations under stated assumptions, not estimates of the chance that this completed trial found a true effect. They are not observed power.

The rows for the full observed effect, 75% of the observed effect, and 50% of the observed effect show how the planning requirement changes when the target difference is smaller relative to the observed variability. Smaller effects are harder to distinguish from random variation and therefore require more participants. The sample-size curve should be read in the same planning context: it illustrates how the required number changes with the target effect, not what happened to participants in this trial. Because the observed effects used as planning inputs are small in some comparisons, the resulting illustrative counts can be very large. They should not be interpreted as evidence about the trial’s original intent or as a recommendation for a future study without additional design information.

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 = 0.800; pooled standard deviation σ = 12.548.
2. n = 2 × (1.96 + 0.842)² × 12.548² / 0.800² = 3,862.1, rounded up to 3,863 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 <- 0.799999999999999                      # observed difference in means
sigma <- 12.5483033933095                     # 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(4619420)
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 = 3,863 (80% power), 5,171 (90% power)
 75% of observed effect: n per arm = 6,867 (80% power), 9,192 (90% power)
 50% of observed effect: n per arm = 15,449 (80% power), 20,682 (90% power)
power.t.test: 3863.1 
Simulated power at n = 3,863 per arm: 0.801 (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 registry reports sponsor analyses using MMRM and differences of least-squares means. For the primary endpoint, the reported contrasts are 1 for 1000 mg versus placebo and 0 for 3000 mg versus placebo, each with a 95% confidence interval spanning zero and p=0.986. These are the sponsor’s model-based results for the primary comparisons.

The unadjusted recalculations use the posted arm means and standard deviations. They give mean differences of 0.8 and 0.5, respectively, with intervals that also include zero. The small numerical differences between these estimates and the sponsor’s estimates are expected when methods and estimands differ: an unadjusted comparison of summary means is not the same as an MMRM comparison of least-squares means. The registry record does not give sufficient model specifications to reproduce the MMRM from the posted summaries.

The recalculations are a descriptive cross-check, not a replacement for the sponsor’s analysis. The sponsor’s model-based results are the relevant reported analyses for the planned repeated-measures comparison. Both approaches indicate substantial uncertainty around the between-arm contrasts, but their estimates and p-values should not be combined or treated as if they came from a single analysis.

15. Multiplicity, Missing Data and Analysis Populations

The registry reports two dose-versus-placebo primary comparisons and a larger set of secondary outcomes. Testing several contrasts increases the chance of at least one apparently unusual result under a collection of null hypotheses. In general, multiplicity procedures control or otherwise address that increased error risk by specifying which tests are confirmatory and how evidence is assessed across them. The registry record does not report whether a multiplicity adjustment, gatekeeping procedure, or testing hierarchy was used for the primary and secondary outcomes. The sponsor-reported p-values should therefore be read with that reporting gap in mind.

The registry identifies MMRM as the method for the sponsor’s primary reported analyses, but does not report the full statistical model specification. Details such as covariates, repeated-measures covariance structure, handling of intercurrent events, and sensitivity analyses are not reported in the registry record. Without these details, readers cannot fully evaluate model assumptions or reproduce the primary estimates. The same limitation applies to interpreting how model-based estimates relate to participants with incomplete follow-up.

The outcome description defines the primary analysis population as randomized participants who received at least one dose of double-blind intervention and were evaluable for the outcome. The outcome-specific analyzed counts are given, but reasons that participants were not evaluable and the amount or pattern of missing data are not reported in the registry record. The approach used to handle missing outcome measurements is not reported in the registry record. In general, missingness can affect estimates if the availability of measurements is related to outcomes or treatment, so transparent reporting of missing-data assumptions and sensitivity analyses is important.

The registry also does not report the original sample-size justification, planned effect size, or the trial’s original power assumptions. The separate planning illustrations based on observed effects do not fill that gap. Finally, the record lists the study as terminated but does not report the reason. Without the reason and the full analysis plan, it is not possible to determine from this record whether termination affected the planned conduct or interpretation of the analyses.

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 serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: fall (2.9%), covid-19 (2.9%), pneumonia (2.5%), osteoarthritis (2.3%), urinary tract infection (1.5%).

During the double-blind treatment period, the registry reports serious adverse events in 34 of 171 participants at risk in the 1000 mg group, 42 of 175 in the 3000 mg group, and 43 of 175 in the placebo group. Deaths reported for those groups were 1, 2, and 2, respectively. These are counts of participants, with the corresponding at-risk group sizes; they are not measures of event severity beyond the serious adverse event classification and do not establish why an event occurred.

These safety counts are descriptive. They do not by themselves show whether differences between groups are caused by treatment, and they do not capture the full clinical context, timing, duration, or relatedness assessments for individual events. The registry also reports long-term extension groups, but extension treatment histories differ from the original double-blind assignments and should not be directly interpreted as randomized comparisons without further details. The registry record does not report enough information to assess comparative safety statistically or to attribute deaths or serious events to an intervention.

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

The posted primary results show average declines in iADRS-MCI scores in the 1000 mg, 3000 mg, and placebo arms through Week 104. Since lower scores indicate worse performance, the negative mean changes describe worsening on the composite measure across all three groups. The between-group estimates are small relative to their uncertainty, and both sponsor-reported 95% confidence intervals include zero. The unadjusted recalculations lead to the same broad statistical reading: the data do not precisely establish a difference between either dose and placebo on this endpoint.

This does not prove that the treatment groups are equivalent. Failure to establish a difference is not evidence that the true difference is exactly zero. The intervals remain compatible with contrasts in either direction, and the registry does not report a clinically meaningful threshold for interpreting their size. The results should not be converted into a claim that a treatment works or does not work. They describe what the posted endpoint data and analyses indicate, with uncertainty.

The outcome is a composite of cognition and daily function, so a group-level average cannot reveal the full range of individual experiences or determine which component drove a change. The results also apply to the outcome population and double-blind period reported in the registry, not automatically to every enrolled participant or to the long-term extension. The analyzed counts are lower than overall enrollment, and the reasons and missing-data procedures are not reported in the registry record.

The study’s terminated status is an important part of its registry context, but the reason is not reported in the registry record. It cannot be used by itself to infer why the trial stopped or to reinterpret the numerical estimates. A complete assessment would require additional information about the protocol, analysis plan, missing data, multiplicity, and termination. The registry results are informative as posted, but those limits constrain how strongly the findings can be generalized or explained.

18. Why This Trial Matters Statistically

This trial illustrates why a clinical endpoint’s group means are only the beginning of statistical interpretation. All arms had negative mean change on the primary composite, but the question is whether the average change differed between treatment and placebo. Comparing each arm’s change with baseline alone would not answer that randomized comparison. The between-arm estimates and their intervals are essential because they show both the estimated contrast and the uncertainty around it.

It also demonstrates the distinction between a raw summary comparison and a model-based repeated-measures analysis. The unadjusted recalculations are transparent and useful for understanding the posted arm summaries, while the sponsor’s MMRM estimates reflect a different analysis method. Differences between these estimates do not automatically imply an error; they underscore why analysis methods and their specifications need to be reported clearly.

The interval widths matter as much as the point estimates. The primary contrasts are close to zero, but the intervals also extend to values on either side. Statistical non-significance does not establish equivalence, and a p-value does not communicate the full range of plausible differences. For researchers, this is a practical example of interpreting estimates and uncertainty together rather than using a significance threshold as the whole conclusion.

Finally, the record highlights the importance of transparency around trial status, missing data, multiplicity, and sample-size planning. The posted design illustrations based on observed effects are not the original design justification, and the registry does not report several details needed to evaluate the analysis fully. Careful separation of reported results, recalculations, and unreported design information is central to responsible reading of a clinical trial.

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 leaves important design and analysis details unreported. It does not provide the original sample-size justification or the assumptions used to plan the trial. The reason the study is listed as terminated is also not reported, limiting interpretation of the trial’s conduct and context.

The sponsor identifies MMRM as the primary analysis method, but the full model specification is not reported. Covariates, covariance structure, detailed treatment of repeated visits, sensitivity analyses, and methods for handling missing outcome values are not reported in the registry record. The outcome-specific analyzed counts are available, but reasons for differences between those counts and overall enrollment are not reported.

The registry does not report a clinical threshold for a meaningful iADRS-MCI difference or a multiplicity strategy for the multiple comparisons and secondary outcomes. This makes it difficult to judge clinical importance or the confirmatory status of secondary findings. The posted summary statistics support simple unadjusted recalculations, but these cannot reproduce the sponsor’s repeated-measures model.

Safety results are reported as counts and at-risk group sizes, without enough detail in this record to establish causal attribution or fully compare groups. Extension groups also reflect differing treatment histories, so their counts should not be read as direct randomized comparisons without additional analysis. These limitations do not invalidate the posted estimates, but they narrow the conclusions that can be drawn from the registry alone.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study of JNJ-63733657 in Participants With Early Alzheimer's Disease — NCT04619420.
• Registry record last updated 2026-11-05; status last verified 2026-10.
• Results first posted 2026-11-05.
• 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

The registry describes a Phase 2 randomized, quadruple-masked, parallel-group study of JNJ-63733657 versus placebo in participants with early Alzheimer's disease. The primary endpoint was change from baseline in iADRS-MCI total score at Week 104. The posted means were -11.3 for 1000 mg, -11.6 for 3000 mg, and -12.1 for placebo. Sponsor-reported MMRM comparisons had 95% confidence intervals that included zero for both doses versus placebo. The study enrolled 523 participants and is listed as terminated; the reason is not reported in the registry record.