Cold agglutinin disease: rare anemia gets first approved drug as treatment map redrawn

Cold agglutinin disease has long been one of hematology’s quiet corners: a rare anemia in which antibodies turn against the patient’s own red blood cells whenever body temperature dips. A comprehensive review now published in Blood Research pulls together what decades of study have established about the condition and how a wave of new drugs is rewriting its management. The central message is that CAD is not one disease but two processes fused together, and that treating it well means attacking both.

The first process is a low-grade clonal expansion of B cells in the bone marrow. These cells churn out a monoclonal immunoglobulin M autoantibody of the kappa light-chain type, characteristically built from the IGHV4-34 gene rearrangement and lacking the MYD88L265P mutation. The second process is complement activation. When blood flows through the cooler peripheral circulation, the IgM binds the I antigen on red blood cells and switches on the classical complement pathway in the sequence C1, C4, C2 and C3. The resulting C3b coating marks the cells for destruction by Kupffer cells in the liver, producing mostly extravascular hemolysis. Only when the cascade runs to completion, forming the membrane attack complex C5b-9, does intravascular hemolysis occur, and that happens chiefly during acute exacerbations.

This dual architecture explains a long-standing clinical puzzle: why the standard tools of warm autoimmune hemolytic anemia, corticosteroids and splenectomy, largely fail in CAD. Warm AIHA is driven by IgG antibodies and Fc-receptor-mediated clearance in the spleen; CAD hemolysis is complement-driven and predominantly hepatic. The review states plainly that steroids and splenectomy are generally ineffective and not recommended, a point with immediate practical consequences for clinicians who reach for familiar remedies.

The distinction between primary CAD and cold agglutinin syndrome is equally consequential. CAS arises secondary to infections such as Mycoplasma pneumoniae or Epstein-Barr virus, or to B-cell lymphoma and other disorders, and is managed by treating the underlying cause. Infection-associated cold agglutinins are typically polyclonal and transient, resolving with the inciting illness, whereas lymphoma-associated cases often carry monoclonal antibodies from an underlying clonal disorder. The 2022 World Health Organization classification formally recognized CAD-associated lymphoproliferative disease as a distinct entity, separate from lymphoplasmacytic lymphoma and Waldenström macroglobulinemia, on the strength of its nodular CD20-positive marrow infiltration, minimal plasmacytic differentiation, absent perinodular mast cells, IGHV4-34 restriction and MYD88L265P negativity.

Epidemiologically, CAD is rare but far from benign. Estimated prevalence runs 5 to 20 per million population, with an annual incidence of roughly 0.5 to 1.9 per million, and it accounts for about 15 to 25 percent of all AIHA cases. The disease strikes older adults, with a median age at diagnosis near 67 to 72 years and a female predominance. Survival is relatively preserved: a large multinational study put median survival at approximately 16 years from diagnosis, close to the roughly 18.4-year life expectancy of the age-matched general population, meaning morbidity rather than mortality dominates the clinical picture. That morbidity is substantial. Around 90 percent of patients report cold-induced circulatory symptoms such as acrocyanosis of fingers, toes, ears and nose, Raynaud-like phenomena and livedo reticularis, which can substantially impair daily activities. Importantly, these symptoms are not complement-mediated, so complement inhibitors do not address them. More than half of patients need red blood cell transfusions over the course of disease, with seasonal worsening in colder months and febrile exacerbations common.

Thrombosis adds a further burden. Patients show an increased incidence of venous thromboembolism, with a reported hazard ratio of approximately 3.3 relative to the general population, and arterial thrombotic events are also more frequent. Complement-driven endothelial activation is thought to contribute to this prothrombotic state, and the review recommends thromboprophylaxis in high-risk situations such as acute hemolytic exacerbations and the perioperative period.

Diagnosis follows a logical sequence. First, hemolysis is confirmed through elevated lactate dehydrogenase and indirect bilirubin, reticulocytosis, reduced haptoglobin and a peripheral smear that may show erythrocyte agglutination. Second, a monospecific direct antiglobulin test differentiates the pattern: CAD characteristically shows strong C3d positivity with negative or only weakly positive IgG, and occasional mixed patterns with weak IgG do not by themselves exclude the diagnosis. Third, subtyping distinguishes warm AIHA, predominantly IgG-positive, from cold AIHA, predominantly C3d-positive with a cold agglutinin titer of at least 1:64. Finally, primary CAD is separated from secondary CAS by evaluating for infection, autoimmunity or lymphoma, with bone marrow examination confirming the clonal process. A working diagnosis requires chronic hemolysis, the characteristic DAT pattern, a titer of at least 1:64 at 4 degrees Celsius and the absence of overt malignant lymphoma; bone marrow findings, though valuable, are not mandatory for every patient.

Practical diagnostic details can make or break accuracy. Blood should be collected and transported at 37 degrees Celsius, because cooling causes in vitro agglutination that can spuriously elevate mean corpuscular volume and mean corpuscular hemoglobin concentration while lowering the measured red cell count; these artifacts resolve on rewarming. A polyspecific DAT alone is insufficient. Thermal amplitude, the temperature range over which the antibody acts, predicts clinical severity better than titer: roughly 28 to 30 degrees Celsius or higher marks a pathogenic threshold, while a high titer with low thermal amplitude may be clinically silent. Secondary causes must be actively excluded by screening for Mycoplasma, EBV, cytomegalovirus and lymphomas including diffuse large B-cell lymphoma, marginal zone lymphoma and LPL/WM.

Treatment has bifurcated along the disease’s two axes. B-cell-directed therapy targets the clone and offers durable, treatment-free remission. Rituximab monotherapy at 375 mg/m2 weekly for four weeks yields overall response rates of about 50 to 60 percent with low complete response rates and a median response duration of about 11 months, suiting frail or elderly patients and allowing repetition at relapse. Bendamustine plus rituximab is a preferred first-line option for many patients without other illnesses, achieving overall response rates of 71 to 78 percent, complete responses of 40 to 53 percent and, notably, five-year sustained remission in approximately 77 percent in the Nordic prospective experience, at the cost of grade 3-4 neutropenia in roughly one-third of patients. Real-world data reinforce this: a recent single-center series of strictly defined primary CAD, confirmed MYD88L265P-negative by digital droplet PCR, achieved responses in all treated patients with a median response duration of 90 months. Complement inhibition works on the other axis. Sutimlimab, a humanized anti-C1s antibody, was the first approved drug for CAD. In the phase 3 CARDINAL trial of 24 transfusion-dependent patients, it met its composite primary endpoint in 54 percent, raised mean hemoglobin from 8.6 to 11.2 g/dL, rendered 71 percent transfusion-free and improved fatigue, with responses evident within the first week. The phase 3 CADENZA trial in 42 non-transfusion-dependent patients confirmed this, with the primary endpoint achieved in 72.7 percent versus 15 percent on placebo and hemoglobin of at least 12 g/dL in 50 percent versus 7 percent. Because sutimlimab does not touch the clone, disease activity recurs on discontinuation, making therapy continuous; vaccination against encapsulated bacteria, including Neisseria meningitidis, Streptococcus pneumoniae and Haemophilus influenzae type b, should be completed at least two weeks before the first dose, though no meningococcal infections occurred in the pivotal trials.

The pipeline is crowded. Riliprubart, a next-generation anti-C1s antibody, produced rapid and sustained hemoglobin improvement after a single dose in a phase 1b study, with responses maintained for around 15 weeks and the prospect of dosing as infrequently as every 12 weeks. Iptacopan, an oral factor B inhibitor, achieved its primary endpoint in 50 percent of a small phase 2 cohort with a mean hemoglobin increase of 2.2 g/dL at week 12 and LDH normalization in all patients. Pegcetacoplan, a subcutaneous C3 inhibitor, raised hemoglobin by a median of 2.4 g/dL at week 48 in its CAD cohort but has not advanced to pivotal development for the disease. For relapsed or refractory cases, daratumumab, ibrutinib and bortezomib offer options, with bortezomib achieving an overall response rate of about 32 percent and a median response duration of roughly 16 months in a GIMEMA phase 2 study limited by neuropathy. Terminal C5 inhibition with eculizumab is poorly suited because hemolysis lies upstream of C5. The review also sketches a sequential strategy, using a complement inhibitor first as a bridge to rapid hemolysis control, then B-cell-directed therapy for clonal induction, followed by careful tapering, a framework consistent with proposed algorithms but still awaiting prospective validation. Access remains uneven: in some health systems, including Korea, sutimlimab is not yet approved and bendamustine-based therapy is not formally reimbursed under a CAD indication, while cost-effectiveness analyses indicate continuous sutimlimab markedly exceeds conventional willingness-to-pay thresholds whereas fixed-duration chemoimmunotherapy delivers durable responses at substantially more favorable cost. Ongoing trials of zanubrutinib, povetacicept, OM336, GL-0719 and CD19-directed CAR T-cell therapy may deepen clonal control, but the review’s bottom line stands: durable outcomes require clone eradication, not merely hemolysis control, and broader access remains an unmet need.

Subject of Research: Pathophysiology, diagnosis, and evolving therapies of cold agglutinin disease, a complement-mediated autoimmune hemolytic anemia caused by a clonal B-cell disorder

Article Title: Cold agglutinin disease: pathophysiology, diagnosis, and the evolving therapeutic landscape

Article References: Choi, D.-H. (2026). Cold agglutinin disease: pathophysiology, diagnosis, and the evolving therapeutic landscape. Blood Research, 61(1), Article 49. https://doi.org/10.1007/s44313-026-00169-8

Image Credits: AI Generated

DOI: 10.1007/s44313-026-00169-8

Keywords: cold agglutinin disease, autoimmune hemolytic anemia, complement inhibition, sutimlimab, rituximab, bendamustine, direct antiglobulin test, classical complement pathway, cold agglutinin syndrome, riliprubart, iptacopan, B-cell lymphoproliferative disorder