ABSTRACT NUMBER: ESOC2026LB131 THREE-MONTH HEMORRHAGIC PROGRESSION ON 7-TESLA MRI IN CEREBRAL AMYLOID ANGIOPATHY: A BATMAN SUBSTUDY

Abstract Background and aims Dutch-type (D-CAA) and sporadic cerebral amyloid angiopathy (sCAA) are characterized by cerebral microbleeds (CMBs) and cortical superficial siderosis (cSS). The BATMAN trial was a randomized, placebo-controlled trial of minocycline that investigated inflammatory cerebrospinal fluid biomarkers. Here, we aimed to assess short-term hemorrhagic radiological progression and the effect of minocycline on CMB progression. Methods The BATMAN trial included 58 participants who were randomized to minocycline or placebo. Participants underwent 7-Tesla MRI at baseline and after three months (n=46). Descriptive statistics were used to evaluate the frequency and extent of progression of CMBs and cSS. We used quasi-Poisson regression to estimate adjusted relative risks (aRRs) for an increase in the number of CMBs, with treatment group as independent variable and baseline microbleed count as a covariate. Results All patients had one or more CMBs at baseline. During follow-up, progression of microbleeds was observed in 42 (91%) of 46 patients (range 1-73, median 6). There was no difference in the aRR between the minocycline and placebo groups (aRR = 0.96, 95% CI 0.66-1.39; p=0.82). Among 34 participants with cSS at baseline, progression of cSS during follow-up was observed in 10 (29%) participants, while no incident cases occurred among those without baseline cSS. Conclusions Radiological progression of hemorrhagic markers occurred in the majority of patients within three months, underscoring the progressive nature of CAA. Minocycline had no effect on the progression of cerebral microbleeds. This study shows that CMBs, used as a marker of CAA progression, are sensitive to change even in short-duration therapeutic trials. Conflict of interest Figure 1 - belongs to Results Figure 2 - belongs to Conclusions