Time saved in activities of daily living and whole‐brain volume: Post hoc analysis of a randomized feasibility trial of gamma oscillation treatment in participants with mild or moderate Alzheimer's disease

Abstract

INTRODUCTION

Gamma oscillations in the brain are necessary for normal cognitive function, sensory processing, and memory consolidation, and are reduced in Alzheimer's disease (AD). In a 6 month, randomized, feasibility trial in participants with mild‐to‐moderate AD (OVERTURE [NCT03556280], n = 76), a non‐invasive method for sensory‐evoked brain gamma oscillations outperformed sham on the secondary outcomes of slowing decline on the Alzheimer's Disease Cooperative Study–Activities of Daily Living (ADCS‐ADL) functional scale, magnetic resonance imaging measures of whole brain volume and the Mini‐Mental State Examination (MMSE) cognitive outcome, despite not showing statistical significance on the primary outcome (Mild and Moderate Alzheimer's Disease Composite [MADCOMS]), a composite cognitive‐functional score. In this post hoc analysis of OVERTURE, we evaluated the effects of investigational sensory‐evoked gamma oscillation treatment in terms of time saved, as an estimate of slowing in disease progression, on ADCS‐ADL, MMSE, and whole‐brain volume.

METHODS

Disease trajectories based on the ADCS‐ADL, MMSE, and whole‐brain volume changes from baseline within each treatment group were constructed using mixed‐effects models. Horizontal projection from active to sham arm yielded time saved from baseline at each visit. Data from the open label extension (OLE) phase of the OVERTURE study have also been used to analyze the time‐saving effect of active treatment in an extended period.

RESULTS

Compared to sham, time savings of 4.83, 4.59, and 4.09 months over 6 months of active treatment on ADCS‐ADL, MMSE, and whole‐brain atrophy were observed in the randomized controlled trial phase. When including the OLE phase, time savings of 8.66, 10.00, and 7.48 months over 14.64, 15.98, and 13.46 months of active treatment on ADCS‐ADL, MMSE, and whole‐brain atrophy were observed relative to the sham group.

DISCUSSION

These findings suggest that further exploration of the effect of evoked gamma oscillations in participants with mild‐to‐moderate AD, as well as the evaluation of treatment effects using time saved, is merited.

Highlights

• Evoked gamma oscillation slows functional loss and brain atrophy in Alzheimer's disease.

• Slowing of functional and cognitive decline and brain atrophy worsening can be expressed as time saved.

• Evoked gamma oscillation saves 4.83 months of progression in activities of daily living, 4.59 months of progression in Mini‐Mental State Examination, and 4.09 months of decline in whole‐brain volume over 6 months.

1. INTRODUCTION

Alzheimer's disease (AD) is a progressive neurodegenerative disease associated with relentless decline of cognitive and functional capacity and brain atrophy. Gamma oscillations, neuronal activity in the range of 30 to 100 Hz that are necessary for sensory processing, cognition, and memory consolidation, are perturbed in patients with AD. ¹ Sensory stimulation is an innovative investigational AD treatment that non‐invasively evokes gamma oscillations detectable by electroencephalography using combined visual and auditory sensory onramps. The OVERTURE feasibility study (NCT03556280) evaluated once‐daily 1 hour investigational device use in participants with mild to moderate AD based on the National Institute on Aging–Alzheimer's Association diagnostic criteria. Treatment was found to be safe and well tolerated, with maintained blinding and > 85% average compliance of completers. Although there was no significant difference versus sham on the primary clinical outcome (Mild and Moderate Alzheimer's Disease Composite [MADCOMS]), a significant preservation of function (Alzheimer's Disease Cooperative Study–Activities of Daily Living [ADCS‐ADL]), cognition (Mini‐Mental State Examination [MMSE]), and whole‐brain volume was observed. ²

A previous feasibility trial (NCT04055376) in participants with mild probable AD reported that sensory‐evoked gamma oscillation treatment for 3 months may reduce hippocampal atrophy; preserve or strengthen functional connectivity in the default mode network and medial visual network, respectively; and improve verbal memory (face–name association, delayed recall). ³ Similarly, another early feasibility trial (NCT03543878) of sensory‐evoked gamma oscillation in participants with prodromal AD also suggested that gamma oscillation treatment for 8 weeks could strengthen functional connectivity and alter immune factors in the brain. ⁴ A post hoc analysis of this trial revealed changes in myelin and synaptic biology, as well as lipid transportation through cerebrospinal fluid proteomics. ⁵ Post hoc analyses of OVERTURE demonstrated reductions of occipital cortical thinning, total and regional white matter atrophy, and myelin content loss, ⁶ together with conservation of corpus callosum structure ⁷ in participants treated with evoked gamma oscillation, versus those receiving sham treatment. These analyses provide preliminary data supporting the treatment effects of sensory‐evoked gamma oscillation over various treatment durations and suggest diverse biological mechanisms resulting in the preliminary efficacy observed in active treated participants in measures of cognition, function, and brain structure. Prospective, longer studies in participants with AD at different stages would be required to further elucidate the mechanisms of evoked gamma oscillation treatment.

Because AD is a progressive disease, changes in cognition, function, behavior, brain volume, or biomarker levels can be expressed through their temporal dimension. In participants with mild AD, cognitive decline has been thought to precede functional decline in the context of AD intervention, ⁸ , ⁹ , ¹⁰ but once functional symptoms are common, particularly in the moderate stage, the two may decline in parallel. ¹¹

The timing and magnitude of changes in ADCS‐ADL and brain volume outcomes are noteworthy observations in OVERTURE. A reliable transformation of these outcomes into their respective time component tests (TCTs), either as a descriptive measure or as an inferential test, without a significant introduction of bias or reduction of power, has been previously described. ¹² , ¹³ , ¹⁴ This transformation allows for evaluation of treatment benefits in terms of time saved, defined as a difference in the time needed to reach the same point of decline for both the treatment and control groups. This metric reduces the complexity of data interpretation and describes therapeutic benefits in terms relevant to patients, caregivers, and clinicians. Furthermore, TCTs can be combined to create composite time‐focused outcomes that can be more sensitive to the underlying pathology than the original psychometric scales. ¹³

The goal of this post hoc analysis was to estimate time savings of evoked gamma oscillation treatment versus sham in participants with mild to moderate AD in terms of daily functions and whole‐brain volume over the duration of the randomized phase, but also extended through the open label extension (OLE) phase, until the active decline matches, or “catches up to,” the sham decline at the end of the randomized phase.

2. METHODS

2.1. Data source

We used patient‐level data from OVERTURE (NCT03556280), a randomized (2:1), sham‐controlled, 6 month, feasibility trial (N = 76) designed to assess initial safety, tolerability, and efficacy of evoked gamma oscillations in participants with mild to moderate AD. ² Despite the failure to demonstrate a significant effect on its primary outcome (MADCOMS), the OVERTURE study yielded significant effects on secondary outcomes, including ADCS‐ADL, MMSE, and whole‐brain volume. Patient‐level data of ADCS‐ADL, MMSE, and magnetic resonance imaging (MRI)–based brain volumetry were used in this post hoc analysis.

RESEARCH IN CONTEXT

1. Systematic review: The authors searched PubMed and relevant conference proceedings (e.g., Alzheimer's Association International Conference) for literature on evoked gamma oscillations as treatment for Alzheimer's disease (AD), as well as on the time saved concept and the time component test. Relevant publications are cited.

2. Interpretation: Despite its small size and failure to demonstrate a significant effect on its primary efficacy outcome (MAD), the OVERTURE feasibility study (N = 76) yielded significant effects on the activities of daily living, whole‐brain volume, and Mini‐Mental State Examination consistent with a broader treatment effect. We converted the point and volumetry differences between the two groups into time saved with gamma oscillation evoking device versus sham, thereby presenting the findings using an intrinsically intuitive parameter (time).

3. Future directions: We propose that both evoked gamma oscillation and time saved analyses deserve the attention of the AD research community.

2.2. Missing data handling

A mixed model was used for all endpoints. Mixed models handle missing data by using the correlation structure from the random effects to relate later visits to earlier visits within subjects and to relate subjects within sites. In this model, completed participants who have similar baseline characteristics represent participants who drop out. No explicit imputation was performed.

2.3. Disease trajectory modeling

Time savings on ADCS‐ADL, MMSE, whole‐brain volume, hippocampal volume, and ventricular volume were assessed based on least squares mean changes from baseline. Given the differences in scheduled visits for different measurements, various models were used to estimate progression over time. All models were consistent with those in the original publication reporting results from the OVERTURE trial. ²

ADCS‐ADL progression was estimated with a mixed‐effects model described below. The model included treatment group, study day, baseline ADCS‐ADL, baseline MMSE, and age at baseline as main effects as well as interaction terms for treatment‐by‐study‐day and baseline‐ADCS‐ADL‐by‐study‐day. Clinical site and participant were included in the model as random intercept effects.

MMSE progression was estimated using a mixed‐effects model that included fixed effects for treatment, age, and baseline MMSE, and a random intercept for clinical site. Because MMSE was only measured once after baseline, no random effect for subject or “time” variable was included.

Progression in whole‐brain, hippocampal, and ventricular volumes was estimated using a mixed model including treatment group, study day, baseline of the relevant volumetric endpoint, baseline MMSE, and age at baseline as main effects as well as interaction terms for treatment‐by‐study‐day and baseline‐volumetric‐endpoint‐by‐study‐day. Clinical site and participant were included in the model as random intercept effects. ¹⁵ , ¹⁶

2.4. Time savings estimation

Time saved in the active treatment group versus the sham group was estimated as described previously. ¹² , ¹³ , ¹⁴ Briefly, trajectory models for each outcome were developed using mixed models. The estimated final value of the active arm for each outcome was projected horizontally onto the corresponding outcome's sham progression (purple line in Figure 1A). The time point at which the sham arm achieves that same value is then subtracted from the final time point to create the “time saved” values (Figure 1A). This same process was performed for active arm values at other time points, and the time saved for each time point was then plotted against time in the study. This difference between disease time curves between the active and sham treatment groups corresponds to the time saved. Standard errors (SEs) for the projected disease time values for each arm, and for their difference (i.e., time saved), were calculated as described previously. ¹² , ¹³ , ¹⁴

FIGURE 1.

Comparison of time saved estimation in randomized phase to time saved including OLE phase. A, An illustration of time saved effect whereby treatment delays disease progression. At time point Y, the change on the clinical outcome over time is less with a treatment than with sham (or with natural disease progression). At this time point, the decline on the clinical outcome with treatment (A) was reached at time point X with sham. The difference between X and Y is the time saved with treatment (adapted from Dickson et al. ¹² ). B, An illustration of time saved effect for both randomized phase and OLE phase. OLE, open label extension; RCT, randomized controlled trial

2.5. Time saving estimation with OLE

The OLE phase of the study provides an opportunity to estimate time savings with active treatment beyond the end of the original randomized phase of the study. To use these data, we calculated the time needed for the active group to progress to the same level as the sham group at the end of the randomized phase (6 months). A horizontal line is drawn from the end of the sham trajectory at the end of the randomized phase, forward to the point at which it meets the active group trajectory during the OLE. The length of this line represents the time savings with treatment beyond 6 months (Figure 1B). Further details of the statistical methodology for this analysis are included in the supporting information.

The main difference between time‐saving estimates with only the randomized controlled trial (RCT) phase and with both the RCT phase and the OLE phase is the projection direction between the sham group and the active group. For time saving, we compare the differences in time for the sham group and the active group to reach the same level of progression. With a 6 month RCT phase only, because the active group progressed more slowly than the sham group, we projected the 6 month active progression back to the sham group trajectory (upper black arrow in Figure 1B). When the OLE data are included, it extends the available data for the active group to a line up to 18 months (6 month RCT + 12 month OLE). With the OLE extension, the progression on all the outcomes in the active group exceeded the progression of the 6 month sham group, so the 6 month sham progression was projected forward onto the OLE active trajectory (lower black arrow in Figure 1B).

2.6. AD stage transitions

AD stages were defined in terms of MMSE with stage boundaries being defined as 19 (highest moderate score) and 13 (highest severe score). For example, an individual who started with a score of ≥ 20 (mild) would be considered having progressed to a new stage when their score reaches ≤ 19 (moderate). In the OVERTURE trial MMSE values were recorded at screening/baseline visits and at month 6 visits. Linear interpolation, assuming that the screening value was observed on study day 0, was used for each participant to estimate the time when they transitioned between stages within the treatment period. The interpolated transition times were then summarized by treatment arm using a Kaplan–Meier plot.

3. RESULTS

3.1. Baseline characteristics

Participant characteristics in the OVERTURE trial ² were well balanced in ADCS‐ADL and MRI whole‐brain volume between study arms (Table S1 in supporting information). Imbalances between active and sham groups at baseline were observed in age and MMSE score, as shown in the original OVERTURE publication. ² Age and baseline MMSE were included as covariates in the model, which allows the model to account for these imbalances at baseline.

3.2. Time saved over 6 months

After 6 months of treatment, in the sham group, estimated decline on ADCS‐ADL was 6.61 points greater than the estimated decline in the active treatment group (95% confidence interval [CI]: 2.98–10.23; p = 0.0004), and the corresponding estimated loss in whole‐brain volume was greater by 11.93 cm³ (95% CI: 3.87–19.99 cm³; p = 0.0049; Table S2 in supporting information). ADCS‐ADL differences between the sham and active arms were significant at 3 months, without adjustment for multiplicity (Table S2).

Converted to a time scale, these results indicate that, over the 6 month follow‐up, the active treatment saved 4.83 months of ADCS‐ADL decline, compared to sham treatment (95% CI: 2.08–7.58 months; p = 0.0006), reflecting minimal progression in the active group compared to rapid progression in the sham group (Table S3 in supporting information). ADCS‐ADL change over time (Figure 2A), disease‐time progression versus calendar time (Figure 2B), and time savings over 6 months of treatment (Figure 2C) are shown.

FIGURE 2.

Treatment effects on ADCS‐ADL and whole‐brain volume. ADCS‐ADL estimated score change from baseline (A), disease time versus calendar time (B), and time saved over 6 months (C) with active treatment versus sham; whole‐brain volume estimated score change from baseline (D), disease time versus calendar time (E), and time saved over 6 months (F) with active treatment versus sham; (B) and (E) show the change from baseline for active and sham, standardized to the sham decline, which has 1 month of disease time (progression) for every month of the study. Data spread was shown as either SE bands (A), (B), (D), and (E) or as 95% confidence bands (C), (F). Bars in panels (A) and (D) represent raw means and SE. ADCS‐ADL month 3 change from baseline Active n = 35, Sham n = 20; and month 6 change from baseline Active n = 29, Sham n = 14. Whole‐brain volume month 3 change from baseline Active n = 31, Sham n = 15; and month 6 change from baseline Active n = 30, Sham n = 19. ADCS‐ADL, Alzheimer's Disease Cooperative Study–Activities of Daily Living; CI, confidence interval; LS, least squares; SE, standard error

For whole‐brain volume loss, the active treatment saved 4.09 months of decline over the 6 month follow‐up, compared to the sham treatment (95% CI 1.49–6.69 months; p = 0.0021). Between‐group differences in brain volumetry were not significant at 3 months, in either volume loss (Table S2) or time saved (Table S3). Figures of whole‐brain volume change over time (Figure 2D), disease‐time progression versus calendar time (Figure 2E), and time savings over 6 months of treatment (Figure 2F) show a protective effect of gamma frequency neuromodulation on whole‐brain volume.

Similar figures are shown for MMSE, MRI hippocampal volume, and ventricular volume (Figure S1 in supporting information). A forest plot of time saved treatment effects on these previously mentioned outcomes is shown in Figure 3A, to illustrate the overall broad treatment effect during the randomized phase of the OVERTURE study.

FIGURE 3.

Treatment effects on time saved with RCT phase and OLE phase on clinical outcomes and volumetrics: (A), time saved over only RCT phase, (B) time saved including both RCT phase and OLE phase. ADCS‐ADL, Alzheimer's Disease Cooperative Study–Activities of Daily Living; CI, confidence interval; MMSE, Mini‐Mental State Examination; OLE, open label extension; RCT, randomized controlled trial

3.3. Time saved including OLE phase

Inclusion of the OLE data in calculations of time savings allows estimation of the time savings over a longer period compared to the original 6 month randomized phase of the study. The period over which time savings can be estimated for each outcome is the time at which the active arm progresses to the same outcome value as the sham group had at the end of 6 months; see illustration in Figure 1B. For the ADCS‐ADL and whole‐brain volume, this period is > 12 months but < 15 months, and the estimated time savings are 8.7 months for ADCS‐ADL and 7.5 months for whole‐brain volume (Figure 4). For MMSE this period is almost 16 months and the estimated time savings is 10.0 months (Figures 3B and S2 in supporting information). For hippocampal volume and ventricular volume, this period is between 7 months and 8 months, and the estimated time saving is 1.2 months for hippocampal volume and 1.4 months for ventricular volume (Figures 3B and S2).

FIGURE 4.

Changes in (A) ADCS‐ADL scores and (B) whole‐brain volume (cm³) over time in the active and sham arms. Dashed lines illustrate where the active least‐square mean and SE values are in time relative to the sham decline. Data are shown as least‐square mean change from baseline ± SE. ADCS‐ADL, Alzheimer's Disease Cooperative Study–Activities of Daily Living; OLE, open label extension; RCT, randomized controlled trial; SE, standard error

The estimated time savings for the functional performance on the ADCS‐ADL, cognitive performance on the MMSE, and MRI volumetric changes on whole‐brain volume are all between 7 and 10 months over ≈ 12 to 15 months of treatment. A forest plot (Figure 3B) showed the overall more robust time‐saving effect with active treatment when including prolonged treatment (OLE phase) compared to only including the 6 month randomized phase. The cognitive outcome MMSE became significantly in favor of treatment when the OLE phase data are included.

3.4. AD stage transition

To explore AD transition rates, we estimated the time until stage conversion in the active group versus the sham group. MMSE scores of 20 to 26 were defined as mild stage, from 14 to 19 as moderate stage, and < 14 as severe stage. The Kaplan–Meier plot shows an earlier transition to a more severe stage of disease across the 6 months of the trial, with ≈ 36% of patients progressing to a more severe stage in the sham arm compared to ≈ 25% of patients progressing to a more severe stage in the active arm (Figure 5).

FIGURE 5.

Kaplan–Meier plot of AD stage transition rates by MMSE category. Lines indicate the probability of worsening at least one AD stage. There was no adjustment for baseline differences. Shaded regions represent the 95% confidence intervals about the probability of worsening at least one AD stage. Vertical lines represent censoring. AD, Alzheimer's disease; MMSE, Mini‐Mental State Examination

3.5. Clinical interpretation of time saved

To better understand how to interpret time savings on cognition and function, one MMSE item, “Orientation to Place,” and two ADCS‐ADL items, “Usual Common Pastimes Performance” and “Usual Conversation Participation,” were selected to illustrate clinical meaningfulness, because they were the clinical items showing the most time savings on the MMSE and ADCS‐ADL.

The orientation to place item asks participants “where are we: state, country, town, hospital, and floor?”, with one point awarded for each correct item. In the active group, 36.9% of patients improved in “Orientation to Place,” compared to 13.6% in the sham arm, which indicated more awareness of state, country, town, hospital, and floor locations. Conversely, for worsening, these percentages were 21.6% for active patients, compared to 50.0% for placebo patients. The score change breakdown can be seen in Figure 6A and Table S4 in supporting information.

FIGURE 6.

Illustration for changes in orientation to place, common pastimes, and conversation participation. ADCS‐ADL, Alzheimer's Disease Cooperative Study–Activities of Daily Living; MMSE, Mini‐Mental State Examination

Among ADCS‐ADL items, better performance on the items “Usual Common Pastimes Performance” and “Usual Conversation Participation” suggests prolonged maintenance of function in performing hobbies and participating in conversations. The rating scores for these items are as shown below:

Usual common pastimes performance

1. No or don't know (0)

2. With help (1)

3. With supervision (2)

4. Without supervision or help (3)

Usual conversation participation

1. No or don't know (0)

2. Rarely or never spoke (1)

3. Said things unrelated to topic (2)

4. Said things related to topic (3)

For “Usual Common Pastimes Performance,” in the active group, 6.4% of patients worsened compared to 50.0% of patients in the sham group, with worsening most often being a change from “without supervision or help” to “with supervision” Examining the worsening difference between active and sham reveals that 43.6% of participants in the active group remained stable or improved in their performance of pastimes, who would have otherwise lost a point or more if untreated. Figure 6B shows the active arm distribution of change is shifted to the right of the sham arm distribution. “Usual Common Pastimes Performance” had improved percentages of 29.0% for active, which was most often changing from “with supervision” to “without supervision or help,” compared to 0% for sham.

“Usual Conversation Participation Performance” had improved percentages of 22.6% for active, which was most often changing from “rarely or never spoke” to “usually said things that were related to the topic” compared to 7.1% for sham. In the active group, 9.7% of patients worsened on this item compared to 42.8% who worsened in the sham group, meaning that 33.1% of the active participants remained stable or improved in their performance in conversation participation who would have otherwise lost a point or more if untreated (Figure 6C and Table S4). Participants with worsening most often correspond to a change from “usually said things that were related to the topic” to “rarely or never spoke”.

4. DISCUSSION

In OVERTURE, positive treatment effects on meaningful outcomes (ADCS‐ADL and whole‐brain volume) were observed in participants with mild‐to‐moderate AD, and these findings correspond to meaningful time savings, or longer maintenance of a milder stage of disease. These findings suggest a broad effect of evoked gamma oscillation treatment, which should be confirmed in future studies. Although cognitive decline often appears before functional decline during the mild AD stage in the context of AD intervention, ⁸ , ⁹ , ¹⁰ in moderate disease, cognition often declines in parallel with function, ¹¹ consistent with the similar time savings seen on ADCS‐ADL and MMSE in this study. Regarding MRI brain volumetry, it has been reported that hippocampal atrophy can be evident at the earliest mild cognitive impairment stage, whereas whole‐brain atrophy accelerates at later stages of the disease. ¹⁷ This may explain why there was a significant time savings benefit on whole‐brain volume but not on hippocampal volume in OVERTURE. Brain volume outcomes and clinical outcomes are known to be correlated cross‐sectionally in the natural history of AD, and recently, a high consistency between clinical outcomes and brain volume outcomes in AD trials has been reported when treatment effects are large and likely to be clinically meaningful, ¹⁸ which suggests that preservation of brain function takes place simultaneously with the observable maintenance of brain structure (i.e., MRI brain volume).

In this post hoc analysis, we opted to investigate OVERTURE findings further by means of TCTs, that is, by converting changes over time in ADCS‐ADL, MMSE, whole‐brain volume, hippocampal volume, and ventricular volume into time saved with treatment versus control, as described previously. ¹² , ¹³ , ¹⁴ In addition, a similar approach that includes OLE data also demonstrated longer time savings with longer treatment, particularly for the ADCS‐ADL, MMSE, and whole‐brain volume. The TCT approach reduces the complexity of data interpretation without adding bias or sacrificing statistical power, while providing estimates that are of great value to patients with AD, their caregivers, and their physicians.

The TCT approach is not the only time‐based analysis of clinical trial data. For example, Raket ¹⁵ used progression models for repeated measures (PMRM) to estimate the disease progression curve. A simple PMRM analysis of the OVERTURE dataset also resulted in time savings estimates convincingly favoring treatment (data not shown). For the estimation method of “time saved,” there are different approaches as well; for time saved in the RCT phase, we projected time saved onto the sham disease progression trajectory; however, the sham group could be projected to treatment disease progression trajectory, as was done with the OLE phase data. ¹⁶ Using these two approaches and longer follow‐up time with the OLE phase increased the robustness of the time saving estimates on the sensory evoked gamma oscillation treatment. In addition, a TCT approach based on individual participant data has been presented recently ¹⁹ which allows for the analysis of response rates in terms of reaching a clinically meaningful threshold.

In addition to presenting estimates of time saved, an item‐level description of time savings using Orientation to Place from the MMSE and Usual Common Pastimes Performance and Usual Conversation Participation Performance from the ADCS‐ADL was explored. These illustrations show that time savings can correspond to improving or maintaining individuals at a stable level on an item level in contrast to individuals in the sham group who show more loss of cognitive or functional performance. While the illustrations were based on specific items to clarify interpretation, time savings always correspond to an improvement or maintenance at stable level across the entire scale and will often correspond to specific item preservation as shown above.

A key limitation of our analysis is its post hoc nature: it is exploratory, not hypothesis‐testing, and all p values should be considered nominal, as no adjustments for multiple comparisons were made. Although baseline age and MMSE were included as covariates in the mixed models for each of the endpoints, it should be acknowledged that this modeling does not completely exclude the possibility that there is still unaccounted bias due to baseline imbalances in unmeasured characteristics. In addition, OVERTURE was a relatively small study, and its findings need to be confirmed in another prospective and larger clinical trial.

In conclusion, we have shown that the effects of evoked gamma oscillation treatment in participants with mild‐to‐moderate AD can be expressed in a meaningful way in terms of time saved, demonstrating a methodology that should be of value in further research.

CONFLICT OF INTEREST STATEMENT

R.K., L.L., M.H., J.R., C.S., and C.H. are employees of Cognito Therapeutics. M.S. has consulted for Eisai, Lilly, NeuroTherapia, Signant Health, Novo Nordisk, Alzheon, Athira, Anavex, Cognito Therapeutics, GSK, AbbVie, and serves on the board of Cervomed. Pentara Corporation consults for more than 30 companies in the Alzheimer's disease space, including Cognito Therapeutics. B.H., J.N.J., S.D., M.M., J.C., and C.M. are employees of Pentara Corporation; S.H. is Pentara's owner and CEO. Author disclosures are available in the supporting information.

CONSENT TO PARTICIPATE

Trial participants provided written informed consent. The trial was approved by an independent ethics committee, and the committee's approval was obtained before the trial was initiated.

Supporting information

Supporting Information

Supporting Information

Kern R, Haaland B, Nicodemus‐Johnson J, et al. Time saved in activities of daily living and whole‐brain volume: Post hoc analysis of a randomized feasibility trial of gamma oscillation treatment in participants with mild or moderate Alzheimer's disease. Alzheimer's Dement. 2025;11:e70118. 10.1002/trc2.70118

DATA AVAILABILITY STATEMENT

Restrictions apply to the data that support the findings of this study. The trial is registered at ClinicalTrials.gov as NCT03556280.