Blood immune proteins linked to treatment efficacy, side effects

Findings raise prospect of biomarkers for personalized therapy

Professors Kang Dong-woo (left) and Byun Ki-hwan of the Department of Psychiatry at Seoul St. Mary's Hospital, the Catholic University of Korea. [Seoul St. Mary's Hospital]
Professors Kang Dong-woo (left) and Byun Ki-hwan of the Department of Psychiatry at Seoul St. Mary's Hospital, the Catholic University of Korea. [Seoul St. Mary's Hospital]

South Korean researchers have found that levels of immune proteins in the blood before antibody treatment for Alzheimer's disease may be linked to both treatment efficacy and the occurrence of side effects.

A research team led by Professor Kang Dong-woo of the Memory Disorder Clinic in the Department of Psychiatry at Seoul St. Mary's Hospital, the Catholic University of Korea, said Tuesday it had confirmed associations between pre-treatment blood levels of complement proteins C1q and C3 and patients' treatment responses and side effects. Professor Byun Ki-hwan participated in the study as first author.

The findings showed that higher pre-treatment C1q levels were associated with smaller reductions in amyloid buildup in the brain after treatment. Conversely, patients with lower C3 levels tended to develop amyloid-related imaging abnormalities, or ARIA, a known side effect, more frequently.

Amyloid-targeting antibody treatment works by helping to clear amyloid beta that has accumulated in the brain. It is applied at the mild cognitive impairment or mild dementia stage of Alzheimer's disease, with the goal of slowing the progression of cognitive decline. Because the treatment can cause ARIA — brain swelling or microbleeds detectable on MRI scans — regular MRI monitoring is required during the course of therapy.

The research team conducted a retrospective analysis of real-world clinical data from 62 early-stage Alzheimer's patients who began lecanemab treatment at Seoul St. Mary's Hospital. Of those, 34 patients who underwent amyloid positron emission tomography, or PET, scans both before treatment and 26 weeks after were also assessed for changes in amyloid levels.

Among the PET-tracked group, the mean centiloid value — a measure of amyloid burden in the brain — fell from 67.8 before treatment to 38.1 at 26 weeks. The average reduction was 29.7 centiloids, representing a 43.8 percent decrease from pre-treatment levels.

C1q and C3 are components of the "complement" immune protein system. C1q initiates the complement cascade, while C3 plays a central role in the subsequent reaction process. The team analyzed the associations between pre-treatment levels of the two proteins and post-treatment amyloid reduction and ARIA occurrence.

Among all 62 patients, ARIA occurred in seven, or 11.3 percent — five with mild cases and two with moderate cases. Because the study was a single-institution exploratory analysis, further validation through large-scale, multi-center research is needed. Should the results be replicated, C1q and C3 could serve as candidate biomarkers for evaluating treatment response and safety.

"The significance of this study lies in connecting blood test data and brain imaging obtained from real clinical settings to confirm that complement proteins C1q and C3 may each be associated with amyloid reduction and ARIA in anti-amyloid antibody treatment," Kang said. "We will verify the clinical utility of these findings through follow-up research involving multiple institutions."

The findings were published in the international journal Alzheimer's & Dementia: Translational Research & Clinical Interventions (TRCI).

Meanwhile, as of May 2026, Seoul St. Mary's Hospital had administered antibody treatment to slow Alzheimer's disease progression in 200 cases — a milestone achieved through the neurology and psychiatry departments each operating independently as prescribing entities.


woo@heraldcorp.com