Anselamimab Misses Primary Endpoint Overall but Cuts Mortality 62% in Kappa AL Amyloidosis

Why This Study Matters

Light chain (AL) amyloidosis is a rare, life-threatening disease in which misfolded free immunoglobulin light chains aggregate into fibrils that deposit in organs — most consequentially the heart — driving high early mortality among patients diagnosed at advanced stages. Roughly 79% of patients carry the lambda light chain isotype and 21% the kappa isotype. Existing anti-plasma cell dyscrasia (anti-PCD) regimens suppress the underlying clonal plasma cells but do not directly clear amyloid fibrils already deposited in organs, leaving targeted fibril-clearing therapy a critical unmet need in this ultrarare disease.

Anselamimab (formerly CAEL-101) is a chimeric IgG1 kappa monoclonal antibody designed to bind selectively to amyloid fibrils and accelerate their clearance, independent of — and complementary to — anti-PCD chemotherapy. The CARES program was designed to test whether adding anselamimab to standard-of-care anti-PCD therapy improves survival and reduces cardiovascular morbidity in patients with advanced (Mayo stage IIIa/IIIb) AL amyloidosis.

Study Design

CARES comprised two parallel, multicenter, international, double-blind, placebo-controlled phase III trials — CAEL101-301 (Mayo stage IIIb; NCT04504825) and CAEL101-302 (Mayo stage IIIa; NCT04512235) — conducted across 19 countries. Data from both trials were pooled for the primary analysis, performed 18 months after the last patient was enrolled in each trial. Patients were randomly assigned 2:1 via centralized interactive response technology, stratified by geographic region, to receive intravenous anselamimab (1,000 mg/m²) or placebo, in addition to standard-of-care anti-PCD therapy (cyclophosphamide, bortezomib, and dexamethasone [CyBorD], with daratumumab permitted where available). Study drug was administered once every 7 (±1) days for four infusions, then once every 14 (±2) days.

The primary endpoint — originally a simple time-to-mortality analysis — was amended during the trial to a hierarchical composite of time to all-cause mortality (ACM) and frequency of cardiovascular hospitalization (CVH), analyzed via the stratified Finkelstein-Schoenfeld test with win ratio estimation, because improving standard-of-care outcomes meant the originally prespecified number of mortality events was unlikely to be reached in a reasonable time. A prespecified analysis by light chain isotype (kappa vs. lambda) was built into the protocol given the potential for differential antibody binding to kappa versus lambda fibrils.

Patient Population

The trials enrolled 406 newly diagnosed, treatment-naive adults with European modification of Mayo 2004 stage IIIa or IIIb AL amyloidosis. Of these, 72 patients (17.7%) had the kappa light chain isotype and 328 (80.8%) had the lambda isotype (4 patients, 1.5%, had missing isotype data). Baseline demographics and disease characteristics were generally balanced between treatment arms in the overall population, though the kappa isotype subgroup showed some baseline imbalances — for example, more anselamimab-treated kappa patients were classified as Mayo stage IIIb (47.9% vs. 29.2% on placebo) with higher median NT-proBNP (8,215.1 ng/L vs. 5,456.0 ng/L).

Primary Endpoint Results

In the hierarchical primary analysis (ACM assessed first, then CVH frequency), anselamimab was not associated with a statistically significant benefit over placebo in the overall population — win ratio 1.11 (95% CI, 0.83-1.50; P=.332). None of the secondary endpoints reached significance in the overall population either. However, a prespecified isotype analysis revealed a marked divergence between the kappa and lambda populations.

ACM = all-cause mortality; CVH = cardiovascular hospitalization. "Nominal" P-values reflect that isotype subgroup analyses were exploratory and not adjusted for multiplicity.

In the kappa isotype population, anselamimab-treated patients also showed a median improvement of 10.37 points (95% CI, -9.38 to 26.56) in the Kansas City Cardiomyopathy Questionnaire-Overall Score (KCCQ-OS) at week 50 versus placebo, and a favorable trend in 6-minute walk test distance (+18.00 m; 95% CI, -71.60 to 95.92 m). No improvement was observed in the lambda isotype population on either measure, and hematologic response (based on reduction of the difference between involved and uninvolved free light chains) was comparable between arms overall.

Subgroup Analyses

Within the kappa isotype population, benefit varied further by disease stage:

Trends favoring anselamimab for the primary composite end point were observed across all evaluable subgroups (≥15 patients) within the kappa isotype population (forest plot analysis). The authors caution that with only 72 kappa isotype patients total, the subgroup is underpowered for the stratified Finkelstein-Schoenfeld test, and a potential survivor-bias effect complicates interpretation of the CVH component (only patients who survived and tied on the mortality comparison contributed to the CVH comparison).

Safety Profile

Safety was comparable between the anselamimab and placebo arms in the overall safety population (271 anselamimab-treated, 134 placebo-treated patients), and anselamimab-treated kappa isotype patients showed a safety profile generally consistent with the overall population.

The most frequently reported treatment-emergent adverse events (occurring in ≥25% of patients in either arm) were diarrhea, peripheral edema, constipation, nausea, COVID-19, hypotension, cough, and fatigue. Arm-by-arm rates for these individual events were reported in the online Data Supplement, which was not available in the full-text source used for this article and is therefore not reproduced here. The authors note that most commonly reported TEAEs are anticipated in the context of underlying AL amyloidosis, common comorbidities, and concomitant anti-PCD therapy.

Interpretation and Broader Context

Anselamimab did not meet its hierarchical primary endpoint in the overall CARES population, formally a negative phase III result. However, in the prespecified kappa isotype subgroup — representing 17.7% of enrolled patients — anselamimab produced nominally significant reductions in all-cause mortality (62%) and cardiovascular hospitalization (71%), alongside a quality-of-life improvement on the KCCQ-OS, with no corresponding benefit observed in the lambda isotype population.

Although this study was negative for its primary endpoint, the efficacy signal in kappa isotype AL amyloidosis was strong and has biological plausibility. Future trials restricting enrollment to this subset of patients should be performed to validate this finding.
— Jonathan W. Friedberg, MD, Editor-in-Chief, Journal of Clinical Oncology

The authors propose a biological rationale: anselamimab was raised against a kappa isotype immunogen, and preclinical work has suggested differential binding affinity to kappa versus lambda amyloid fibrils, which may explain the isotype-specific clinical effect. Because light-chain isotype is already routinely determined as part of standard AL amyloidosis diagnostic workup, the authors argue this could support a readily identifiable target population for a future, dedicated confirmatory trial. The kappa isotype cohort in CARES (n=72) is described by the authors as the largest kappa AL amyloidosis population studied to date, and it uniquely included Mayo stage IIIb patients, who have been excluded from other AL amyloidosis trials.