The goal of this clinical trial is to learn if transcranial photobiomodulation (t-PBM), a light-based brain therapy, can help improve memory and cognitive skills in older adults with amnestic mild cognitive impairment (aMCI). This condition involves memory problems that are greater than normal aging and may increase the chance of developing Alzheimer's disease.

The main questions this study aims to answer are:

1. Does t-PBM, when used together with cognitive training, improve memory and cognitive skills?
2. Does t-PBM change how certain brain areas communicate with each other, especially in the default mode network (DMN)?

Researchers will compare:

t-PBM plus cognitive training to sham (inactive) t-PBM plus the same cognitive training to see if the active light treatment leads to better cognitive improvement and healthier brain activity.

Participants will:

• Provide a blood sample so the research team can create a genetic profile;
• Complete cognitive tests before and after the 4-week program;
• Meet with a dietitian before and after the program so the research team can make sure diet stays consistent and does not influence brain results;
• Have a brain fMRI scan before and after treatment to measure brain connectivity changes;
• Take part in eight sessions of cognitive training;
• Receive either active t-PBM or sham t-PBM during these sessions.

Amnestic mild cognitive impairment (aMCI) is associated with early disruptions in large-scale brain networks, including the default mode network (DMN), which often shows inefficient or abnormally high connectivity in individuals at increased risk for Alzheimer's disease. Human studies examining the impact of transcranial photobiomodulation (t-PBM) on DMN connectivity and cognitive performance in aMCI remain limited, and few controlled trials have assessed Group × Time interaction effects. This study was designed to address this gap by evaluating whether t-PBM produces measurable changes in resting-state functional connectivity and cognitive outcomes when combined with cognitive rehabilitation.

The trial uses a randomized, sham-controlled, pretest-posttest design. Participants with aMCI were randomly assigned to receive either active t-PBM paired with a cognitive rehabilitation program or sham t-PBM with the same cognitive rehabilitation. The intervention was delivered across eight sessions. Cognitive assessments and resting-state fMRI scans were conducted before the first session and after the final session to evaluate changes in cognitive performance and DMN connectivity. The experimental model allows for the direct evaluation of treatment-related neural changes while accounting for natural variation and practice effects.

To minimize confounding factors, dietary stability was monitored. All participants completed a dietitian visit at baseline and follow-up to document eating habits and reduce the influence of diet-related changes on cognitive or imaging outcomes. In addition, peripheral venous blood samples were collected at baseline for genetic analysis. Genotyping targeted APOE alleles (rs429358, rs7412) and the COMT Val158Met (rs4680) polymorphism. DNA extraction followed silica membrane-based protocols, and allelic discrimination was performed using TaqMan-based real-time PCR. Quality assurance included blinded laboratory processing, random duplicates, call rate thresholds (>95%), and Hardy-Weinberg equilibrium criteria. These genetic data were incorporated into exploratory models to examine whether genotype moderated baseline brain connectivity or response to the intervention.

Neuroimaging was conducted using a GE SIGNA Hero 3.0 Tesla MRI system. High-resolution T1-weighted structural images were acquired using an axial MPRAGE sequence (TR = 2745 ms, TE = 3.01 ms, TI = 1020 ms, flip angle = 8°, 1 mm slice thickness, voxel size 0.44 × 0.44 × 1 mm, 225 slices). Resting-state functional MRI data were collected using an axial multiband echo-planar imaging sequence (TR = 2000 ms, TE = 22 ms, flip angle = 20°, voxel size 3.125 × 3.125 × 3 mm, 45 slices, matrix 64 × 64, 200 volumes; approximately 6 minutes 40 seconds). Participants were instructed to remain still, keep their eyes closed, and avoid focused mental activity during the scan. These data enable reliable estimation of DMN functional connectivity before and after the intervention.

The study integrates cognitive rehabilitation outcomes, functional neuroimaging, dietary monitoring, and genetic profiling to explore neurobiological and behavioral effects of t-PBM in aMCI. By evaluating network-level changes and potential individual differences in treatment response, the trial aims to improve understanding of whether t-PBM can serve as a network-targeted adjunctive intervention for individuals with early cognitive impairment.

Inclusion Criteria:

• Age 55 years or older
• Diagnosis of amnestic Mild Cognitive Impairment (aMCI)

Exclusion Criteria:

• Presence of a neurological disorder
• Presence of a severe psychiatric disorder
• Having a pacemaker or other implanted medical device that is not MRI-compatible

De-identified individual participant data (IPD), including genetic, neuroimaging, and cognitive assessment data, will not be shared outside the primary research team due to ethical, legal, and institutional restrictions related to participant privacy and data protection.

Participants and their caregivers were masked to whether they received active t-PBM or sham stimulation. The clinician who diagnosed and referred participants into the study was also masked to group assignment. The sham procedure produced the same visual cues and procedures as active t-PBM but delivered no therapeutic light. The device operator was aware of assignment due to technical requirements. The researcher responsible for data collection and analysis was not masked. Participants and caregivers were informed of their group assignment only after the study was completed.