Restoring Immune Balance: Low-Dose IL-2 Shows Promise in Alzheimer’s Research

The neuroinflammatory cascade in Alzheimer’s disease (AD) has long been considered more consequence than cause. But growing evidence now positions immune dysfunction—particularly regulatory T cell (Treg) impairment—as a modifiable contributor to disease progression.

A new phase 2a trial published in Alzheimer’s Research & Therapy offers compelling proof-of-concept that restoring Treg activity with low-dose IL-2 can alter inflammatory mediators and CSF biomarkers in mild to moderate AD, with early signals of clinical benefit.

Led by Faridar et al. at Houston Methodist, this double-blind, placebo-controlled study randomized 38 patients to receive subcutaneous IL-2 (1 million IU/day for five days) every four weeks (q4wks) or every two weeks (q2wks), or placebo over a 21-week treatment phase. The primary aim was safety; secondary and exploratory endpoints included Treg expansion, inflammatory cytokine modulation, CSF biomarker shifts, and cognitive scales.

Biologically Active, Clinically Tolerable

No serious adverse events or treatment-related deaths occurred. Injection site erythema and eosinophilia were more common with IL-2, but both were manageable. Importantly, 100 percent of participants completed treatment, supporting tolerability.

Both IL-2 schedules significantly expanded Tregs, but the q4wks regimen produced more durable effects:

• Treg percentage and FOXP3 expression remained elevated throughout the 21-week period in q4wks, while q2wks waned after week nine.
• Treg suppressive function improved in both arms but without significant differences between dosing frequencies.

These findings suggest that more frequent dosing may paradoxically exhaust Tregs—a phenomenon reported in preclinical models.

IL-2 q4wks treatment reduced key pro-inflammatory mediators:

• CCL2 and CCL11, chemokines involved in microglial activation and immune cell trafficking, were significantly suppressed only in the q4wks arm.
• IL-15, which promotes effector T cell activity, dropped in both arms but more consistently with q4wks dosing.
• IL-4, an anti-inflammatory cytokine associated with T2 polarization, increased significantly only with IL-2 q4wks.

CSF biomarkers paralleled these changes:

• Aβ42 levels rose in the q4wks group versus placebo (p=0.045), a potential sign of reduced amyloid sequestration in plaques.
• Neurofilament light chain (NfL) levels remained stable with IL-2 q4wks but increased 217 pg/mL with placebo (p=0.060), suggesting neuroprotection.
• No significant effects were seen on CSF p-tau181 or GFAP levels.

While underpowered for efficacy, clinical outcomes trended favorably:

• ADAS-Cog scores improved modestly with IL-2 q4wks (−0.45 points) versus worsening with q2wks (+5.16) and placebo (+4.48) (p=0.061).
• CDR-SB and CGIC also favored the q4wks group, though not significantly.

Limitations and Considerations

This was a small, single-center, exploratory trial not designed or powered to establish clinical efficacy. Baseline age differences between groups—especially the older IL-2 q2wks cohort—may have influenced immune responsiveness. Follow-up was limited to nine weeks post-treatment, offering minimal insight into durability. The trial’s exclusion of late-stage AD and the lack of diversity in race/ethnicity may also limit generalizability.

Larger, longer studies are needed to validate clinical impact and determine optimal dosing cadence.

A Step Forward for Immune-Based Therapies in AD

The findings underscore that low-dose IL-2, particularly at four-week intervals, may engage a disease-modifying mechanism in AD by restoring Treg function and altering central and peripheral inflammation. Importantly, the effects were seen in a well-characterized, biomarker-confirmed cohort of early AD, setting the stage for larger, longer studies to test durability and clinical impact.

Reference
Faridar A, Gamez N, Li D, et al. Low-dose interleukin-2 in patients with mild to moderate Alzheimer’s disease: a randomized clinical trial. Alzheimers Res Ther. 2025;17:146. doi:10.1186/s13195-025-01791-x