GAIN Trial: Phase 2/3 Study of COR388 in Subjects With Alzheimer's Disease: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for GAIN Trial: Phase 2/3 Study of COR388 in Subjects With Alzheimer's Disease (NCT03823404), a randomized phase 2, phase 3 study. The primary outcome is Alzheimer's Disease Assessment Scale-Cognitive Subscale 11 (ADAS-Cog 11).

2. Clinical Question

3. Trial Design

The GAIN trial was a randomized, double-blind, placebo-controlled study of COR388 in people with Alzheimer’s disease. The registry identifies it as a Phase 2/3 trial, with randomized allocation, quadruple masking and a parallel-group model. These design features describe how the study was organized; they do not, on their own, establish the size or certainty of any treatment effect.

The record lists three regimens: COR388 80 mg twice daily, COR388 40 mg twice daily and placebo twice daily. Both active regimens were oral capsules. The study description says that the trial assessed efficacy, safety and tolerability in people with mild to moderate Alzheimer’s disease dementia. Enrollment was 643. The sponsor is Cortexyme Inc., and the registry status is completed.

The registry lists the study start as 2019-03-28 and primary completion as 2022-01-01. These dates give the recorded study timeline, but do not indicate when individual participants began or finished treatment. The registry record does not report treatment duration for each participant, adherence, or the detailed schedule of study visits beyond the primary outcome’s baseline-to-Week-48 timeframe.

Randomization is intended to reduce systematic differences between treatment groups at the time of assignment. Masking is intended to limit the influence of knowledge about assigned treatment on participants’ experiences, clinical assessments or study conduct. The record calls the study quadruple-masked but does not specify in the information available which roles were masked. A parallel design assigns participants to a regimen that is compared with the other groups, rather than describing a crossover between regimens.

The posted primary efficacy population is the Intent-to-Treat Population. The registry record does not provide the operational definition used for that population, details of the analysis model, or the handling of participants who discontinued or had missing measurements. Those omissions limit how fully the reported design can be connected to an analysis of the primary outcome.

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: Randomized subjects were stratified by baseline Mini-Mental State Examination (MMSE) score (MMSE ≥12 and ≤18, and MMSE ≥19 and ≤24) and Apolipoprotein E genotype (ApoE4 positive either homozygous or heterozygous vs. all others) to assure balanced distribution of mild and moderate Alzheimer's disease and a balanced distribution of ApoE4 subjects, across treatment arms.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Intent-to-Treat Population

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: Safety Population

6. Primary Endpoint

The registry lists the primary outcome as Alzheimer's Disease Assessment Scale-Cognitive Subscale 11 (ADAS-Cog 11), measured over Baseline to Week 48.

The primary endpoint was the ADAS-Cog 11 total score, a continuous measure of cognitive function. The registry describes 11 items covering orientation, verbal memory, language and praxis, which are areas of function commonly impaired in Alzheimer’s disease. The stated scale ranges from 0 to 70, with higher scores indicating greater disease severity.

The primary timeframe was baseline to Week 48. A measure taken at baseline provides a reference point for assessing change over follow-up. For this endpoint, the direction of a score change matters: in general, a lower score indicates less measured severity on the scale, while a higher score indicates greater severity. The meaning of a change also depends on its size and uncertainty, neither of which can be determined from the endpoint description alone.

A cognitive scale is used to quantify performance through a consistent set of assessments rather than relying only on a general impression. Its value as an outcome depends on how it is administered, the completeness of follow-up, and the analysis used to compare groups. The registry record identifies the ADAS-Cog 11 as the primary efficacy measure but does not provide a single reported estimate of the treatment-group comparison for baseline to Week 48. It therefore does not allow the magnitude or precision of a primary cognitive result to be described.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does randomization contribute?

Randomization assigns participants to study groups by a chance-based process. In general, its purpose is to make the groups comparable at the start with respect to both measured and unmeasured characteristics, although chance imbalances can remain. It supports a fairer comparison than a design in which treatment choice is tied to prognosis or other participant characteristics. Random assignment does not ensure that follow-up, adherence or measurement is identical across groups, and it does not by itself determine how the outcome should be analyzed.

The registry describes a randomized parallel-group design with three regimens: COR388 80 mg twice daily, COR388 40 mg twice daily and placebo twice daily. A parallel design compares outcomes among groups assigned to different regimens during the study. The record does not provide the randomization sequence-generation method, allocation concealment details or any stratification factors, so those aspects of the allocation process are not reported in the registry record.

What does masking mean?

The record describes the trial as double-blind in its study description and identifies masking as quadruple. Masking is intended to reduce the chance that expectations influence participant reports, clinical assessments or study conduct. The information available does not specify which participant, investigator, assessor or other roles were masked, or whether masking was maintained throughout follow-up. It also does not describe any assessment of whether masking was successful. Those details are not reported in the registry record.

How should a continuous outcome be summarized?

The primary outcome is a continuous score, and the record identifies the mean as its summary measure and the standard deviation as its measure of dispersion. The mean describes the arithmetic average of scores in a group. The standard deviation describes how widely individual scores tend to vary around that average; it is not the uncertainty of the group mean or the uncertainty of a between-group comparison.

To interpret a comparison, readers generally need the group summaries and an estimate of the difference between groups, along with an uncertainty measure such as a confidence interval. Baseline values and follow-up values may be analyzed in different ways, such as comparing change scores or adjusting follow-up scores for baseline. Those approaches can differ in assumptions and precision. The registry record does not report the primary-outcome model or specify whether the analysis compared change scores, adjusted follow-up scores or used another approach.

Why does the analysis population matter?

The registry identifies the primary outcome population as the Intent-to-Treat Population. In general, an intent-to-treat approach aims to analyze participants according to their randomized assignment, which preserves the comparison created by randomization. The exact definition and implementation can vary, especially when outcome measurements are missing. The registry record does not define the analysis population in further detail or explain how participants with no Week 48 score were handled.

Missing outcomes can affect interpretation if the probability of missingness is related to treatment assignment, health status or the unobserved outcome. Statistical methods can use assumptions to address missing data, but those assumptions need to be described and assessed. The registry record does not report a missing-data method, sensitivity analysis or reasons for missing endpoint measurements.

What is the role of uncertainty and over-dispersion?

Uncertainty reflects the fact that a study observes a sample rather than every person to whom the result might be relevant. A confidence interval is one way to express that uncertainty around an estimate. Its width is influenced by the amount of information and the variability in the outcome, as well as the analysis method. A narrow interval indicates greater precision under the stated method; a wide interval indicates less precision. No primary-outcome confidence interval is reported in the registry information.

For a continuous score, variation among participants is commonly described with a standard deviation. In other kinds of outcomes, observed variation can exceed what a simple statistical model assumes; this is called over-dispersion. In general, failing to account for over-dispersion can make uncertainty estimates too small and conclusions too confident. The registry record identifies the primary outcome as continuous and does not report an over-dispersion method or a model where that issue is directly relevant.

What do the sample-size assumptions mean?

The record describes a planning illustration based on the observed effect, using a two-sided significance level of 0.05, equal allocation between two arms and a normal approximation. It says the effect to detect is a stated fraction of the observed effect. For time-to-event outcomes, the described planning quantity is the total number of events; for other outcomes, it is participants per arm. These are planning assumptions, not evidence about the number of participants actually needed to establish a result and not observed power.

Planning illustrations depend on the assumed effect and analysis conditions. A smaller effect to detect generally requires more information than a larger effect, all else being equal. Such an illustration does not establish that the trial was designed around that effect, because the assumptions are explicitly described as distinct from the trial’s own design. The registry record does not report the trial’s sample-size justification, its own target assumptions or a detailed statistical analysis plan. No numerical planning estimates are included in the registry information available for this analysis.

How should significance testing be understood?

In general, a statistical test evaluates data against a specified null hypothesis under assumptions about the design and model. A p-value does not measure the probability that a treatment is effective, nor does it describe the size or clinical importance of an effect. Interpretation requires the estimate, its uncertainty and the analytic context. The registry information available does not report a primary-outcome p-value or a complete account of the tests used for the primary comparison.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 8 outcome measures and 2 analyses. The primary endpoint is the ADAS-Cog 11, and secondary outcomes include the Clinical Dementia Rating-Sum of Boxes (CDR-SB), Mini-Mental State Examination (MMSE) and Neuropsychiatric Inventory (NPI). Considering multiple outcomes, comparisons or time points can increase the chance of obtaining at least one apparently notable result by chance. Statistical plans may address this with prespecified testing sequences, adjusted significance thresholds or other approaches. The registry record does not report whether a multiplicity adjustment was used, which outcomes or time points were included in any testing hierarchy, or whether the two analyses refer to separate populations or models.

The record indicates that the primary outcome has several categories or time points and is not reduced to one comparison. That makes the absence of a single primary estimate especially important: a reader cannot tell from this information which contrast was intended as the main basis for inference or how results across categories or time points were handled. The detailed analysis plan is not reported in the registry record.

The primary population is named as the Intent-to-Treat Population, but the record does not report how missing ADAS-Cog 11 measurements were handled, how many participants contributed endpoint data, or whether sensitivity analyses examined alternative missing-data assumptions. Missingness can reduce precision and can also bias a comparison if participants with missing measurements differ in relevant ways from those with observed measurements. The reasons for missing measurements and the method used to address them are not reported in the registry record.

The record also does not report interim analyses, stopping rules or the full statistical analysis plan. Those omissions do not establish that such procedures were or were not used; they mean the registry information does not permit their evaluation.

11. 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: diarrhoea (0.9%), covid-19 (0.9%), pneumonia (0.9%), dehydration (0.9%), cholecystitis (0.9%).

The registry reports serious adverse event counts and other adverse event counts for the three study groups, with an at-risk denominator listed for each type of count. For COR388 80 mg twice daily, the record reports 25 affected by serious adverse events, 139 affected by other adverse events and 5 deaths, with 214 at risk for each category. For COR388 40 mg twice daily, it reports 20 affected by serious adverse events, 150 affected by other adverse events and 1 death, with 212 at risk for each category. For placebo twice daily, it reports 19 affected by serious adverse events, 88 affected by other adverse events and 0 deaths, with 217 at risk for each category.

The record lists diarrhoea, COVID-19, pneumonia, dehydration, cholecystitis and liver function test abnormal among the top serious adverse event terms, each at 0.9%. These terms and percentages are reported without a group-specific attribution in the information available for this analysis. The registry record does not provide event timing, exposure-adjusted rates or a statistical comparison for these events.

These are descriptive counts, not proof that a study regimen caused an event. The categories “serious” and “other” should not be combined or treated as measures of the same clinical severity. Differences in the counts may reflect chance, exposure, follow-up or other factors not detailed in the registry record. The listed denominators do not by themselves explain duration of exposure or completeness of safety surveillance.

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

The registry record establishes that the GAIN trial was a randomized, masked, placebo-controlled Phase 2/3 study with two COR388 dose regimens and a placebo group. It also identifies the primary cognitive measure, the population used for that outcome and the baseline-to-Week-48 timeframe. These details clarify the intended comparison and the type of outcome being assessed.

The available information does not provide a single primary-outcome comparison. Therefore, it cannot show the direction or magnitude of the ADAS-Cog 11 difference between either COR388 group and placebo, or how precisely such a difference was estimated. No efficacy conclusion should be inferred from the fact that the trial is completed or from its design alone. A study can be well randomized and still have a result that is uncertain, small, large or not reported in the registry record.

The safety counts document reported serious adverse events, other adverse events and deaths in each group. They are useful as a description of what the registry reports, but do not establish causation. Assessing comparative safety would require further information, including definitions, event timing, treatment exposure, follow-up and an analysis that accounts for the study design.

In general, statistical evidence is not the same as clinical importance. A numerical difference may be estimated precisely yet be small in practical terms, or it may be imprecise enough to remain compatible with a range of effects. Interpretation also depends on the outcome’s measurement properties, missing data, multiplicity and the analysis plan. Those elements are not sufficiently reported for a full evaluation of the primary efficacy finding in this registry record.

The record can support a careful description of the trial’s question and design, but it does not support a claim that COR388 did or did not change cognition relative to placebo. The distinction is important: absence of a reported comparison is not evidence that the groups had identical outcomes, and it is not evidence of a particular effect.

13. Why This Trial Matters Statistically

The GAIN trial illustrates why knowing a study’s design is only one part of statistical interpretation. Randomization, masking, a placebo comparator and parallel groups describe safeguards and the structure of a comparison. To assess the primary efficacy finding, readers also need the actual outcome estimates, their uncertainty and a clear account of how the analysis was conducted. Those primary comparison details are not reported in the registry information available for this analysis.

The selected endpoint is a continuous cognitive score assessed from baseline to Week 48. Continuous measures can convey more information than a simple event/no-event outcome, but interpretation depends on the scale’s direction, its variability and the treatment-group contrast. The registry gives the scale range and explains that higher scores indicate greater disease severity, yet does not provide a single result that connects those measurement details to a comparative estimate.

The study also has multiple outcome measures and the primary outcome is described as having several categories or time points. Such a setting makes prespecification and multiplicity handling relevant. Without the analysis plan and a reported primary comparison, readers cannot determine which contrast was treated as decisive or how uncertainty across outcomes and time points was managed.

Safety reporting raises related questions. Counts of serious adverse events, other adverse events and deaths provide a useful descriptive record, but meaningful comparisons require context about exposure, follow-up and event definitions. A count alone is not a causal estimate.

Statistically, this trial record is a reminder to distinguish a trial’s intended question, its reported endpoints and the evidence available to answer that question. Transparent estimates and methods are necessary to judge both precision and limitations; design labels alone cannot substitute for them.

14. 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 principal limitation for interpretation is that the registry record does not provide one primary-outcome comparison. It identifies ADAS-Cog 11 as the primary measure and gives the baseline-to-Week-48 timeframe, but notes that the outcome has several categories or time points and is not reduced to one comparison. No primary estimate, confidence interval, standard deviation or p-value is included in the information available for this analysis. The magnitude and precision of the primary efficacy result therefore cannot be evaluated.

The record does not report the statistical model, the exact contrasts tested, any multiplicity adjustment or the analysis plan. Although it names the Intent-to-Treat Population, it does not define the population operationally or describe how missing measurements, discontinuations or deviations from assigned treatment were handled. These omissions limit assessment of bias and reproducibility.

The registry reports adverse event counts, but the information available does not describe event timing, exposure duration, definitions beyond the serious classification, or comparative statistical analyses. The top serious event terms are listed with percentages but without group-specific attribution in the registry record used for this description. As a result, the counts cannot by themselves support causal or exposure-adjusted comparisons.

The record’s planning description is an illustration based on the observed effect, not the trial’s own design justification. The trial’s sample-size rationale and full statistical plan are not reported in the registry record. These limitations concern what can be concluded from the available registry information; they do not establish what additional information may exist outside that record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: GAIN Trial: Phase 2/3 Study of COR388 in Subjects With Alzheimer's Disease — NCT03823404.
• Registry record last updated 2023-02-23; status last verified 2023-01.
• Results first posted 2023-02-23.
• 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.

18. Record Summary

The GAIN trial was a completed randomized, double-blind, placebo-controlled Phase 2/3 study of COR388 in people with mild to moderate Alzheimer’s disease dementia. It used a parallel-group design with quadruple masking and compared COR388 80 mg twice daily, COR388 40 mg twice daily and placebo twice daily. The registry reports enrollment of 643 and identifies Cortexyme Inc. as sponsor.

The primary efficacy outcome was ADAS-Cog 11 total score in the Intent-to-Treat Population, assessed from baseline to Week 48. Higher scores indicate greater disease severity. The registry record does not report one primary comparison result because the outcome has several categories or time points. It lists secondary outcomes of CDR-SB, MMSE and NPI, as well as safety counts by group. Details of the primary analysis, missing-data handling and multiplicity adjustment are not reported in the registry record.