Work overview

Section 02 of 09

Case Presentation

Complement‐Dominant Relapse of Mixed Autoimmune Hemolytic Anemia After Rituximab Response

Naru Tomigaki, Seiji Kakiuchi, Akimasa Sakamoto, Shutaro Fujioka, Isamu Harima, Hiroaki Akiyama, Ryotaro Niwa, Ikumi Takagi, Yoko Kozuki, Yoshiharu Miyata, Sou Tanaka, Hiroyuki Tsuji, and Nobuko Iwata · 2026

Contents

Section 02 of 09

  1. 01Introduction
  2. 02Case Presentation
  3. 03Discussion
  4. 04Conclusion
  5. 05Section 5
  6. 06Funding
  7. 07Ethics Statement
  8. 08Consent
  9. 09Conflicts of Interest
Text size
Work overview

Section 2 of 9

Case Presentation

Naru Tomigaki, Seiji Kakiuchi, Akimasa Sakamoto, Shutaro Fujioka, Isamu Harima, Hiroaki Akiyama, Ryotaro Niwa, Ikumi Takagi, Yoko Kozuki, Yoshiharu Miyata, Sou Tanaka, Hiroyuki Tsuji, and Nobuko Iwata · about 11 minutes

Day 0 denotes the initial presentation on January 8, 2025. The initial diagnostic course through Day 120 was reported previously, and the present report describes the subsequent longitudinal follow‐up in the same patient [5]. A woman in her 80s presented with progressive exertional dyspnea and hemolytic anemia. At presentation, hemoglobin was 8.6 g/dL and lactate dehydrogenase (LDH) was 875 U/L, with reduced haptoglobin and reticulocytosis. The direct antiglobulin test (DAT) was positive for both IgG and C3d, and no erythrocyte agglutination was observed on the peripheral blood smear. Cold agglutinin testing was not performed. Serum immunofixation and serum free light‐chain testing were not performed at presentation; both were first performed on Day 50. Computed tomography showed pulmonary emboli without lymphadenopathy, and ultrasonography identified bilateral soleal vein thrombosis. The patient was initially diagnosed with warm AIHA and received prednisolone 50 mg/day (Table 1; Figure 1A).

A: Clinical and serologic findings at selected time points
Parameter | Day 0 Initial presentation | Day 43 Cold‐triggered relapse | Day 245 Stable follow‐up | Days 385–406 Pre‐sutimlimab evaluation | Day 441 During sutimlimab
Hemoglobin, g/dL | 8.6 | 9.7 | 12.1 | 9.6 | 12.3
LDH, U/L | 875 | 675 | 211 | 373 | 203
Total bilirubin, mg/dL | 1.80 | 1.28 | 0.54 | 1.10 | 0.46
Haptoglobin, mg/dL | 3 | ND | 27 | 2 | 23
Reticulocytes, % | 10.3 | ND | 2.1 | 2.9 | 1.4
Peripheral blood smear | No erythrocyte agglutination | Prominent erythrocyte agglutination with occasional spherocytes | ND | ND | ND
Polyspecific DAT | 1+ | Weakly positive | ND | 1+ | Weakly positive
DAT, anti‐IgG | Weakly positive | Negative | Negative | Negative | Weakly positive
DAT, anti‐C3d | Weakly positive | Weakly positive | Weakly positive | Weakly positive | Weakly positive
Cold agglutinin titer | Not performed | 1:65,536 | ND | 1:65,536 | 1:131,072
Direct agglutination | Not performed | Positive at both RT and 37°C | ND | 37°C, 2+; RT, positive | 37°C, 2+; RT, 3+
Serum IgM, mg/dL | 153 | ND | ND | 100 on Day 385 | 125
Treatment | Prednisolone,50 mg/day initiated | Prednisolone,25 mg/day | Prednisolone,5 mg/day | Prednisolone, 5 mg/day;first sutimlimab dose (6.5 g) | Fourth sutimlimab dose (6.5 g); prednisolone reduced to 2.5 mg/day
B: Bone marrow and monoclonal protein evaluation
Parameter | Initial evaluation (Day 50) | Relapse evaluation (Days 376–385) | Treatment‐phase serum evaluation (Days 427–441)
Bone marrow histology | No overt B‐cell lymphoma, plasma cell neoplasm, lymphoplasmacytic infiltration, or lymphoid aggregates | Normocellular marrow with relative erythroid hyperplasia; CD20‐positive B cells accounted for approximately 1% of nucleated cells, and CD138‐positive plasma cells accounted for 2%–5% of nucleated cells; no significant κ/λ imbalance by dual‐color in situ hybridization; no lymphoplasmacytic infiltration or lymphoid aggregates | Not repeated
Bone marrow flow cytometry | CD45‐gated lymphoid fraction, 14.4%; CD19‐positive cells, 45.2%; CD20‐positive cells, 44.6%; κ‐positive cells, 33.6%; λ‐positive cells, 6.4%; κ:λ ratio, 5.3:1 | Previously identified CD20‐positive κ‐restricted population no longer detectable by flow cytometry; κ‐positive events, 0.8%; λ‐positive events, 0.4% | Not repeated
Serum protein electrophoresis | ND | ND | Suspected M‐peak measuring 0.896 g/dL on Day 427 and 0.891 g/dL on Day 441
Serum immunofixation | Negative | Not performed | IgG‐κ and IgM‐κ monoclonal proteins identified on Day 441
Serum free light chains | κ, 9.4 mg/L; λ, 8.7 mg/L; κ/λ ratio, 1.08 | κ, 11.8 mg/L; λ, 10.2 mg/L; κ/λ ratio, 1.16 on day 385 | κ, 9.9 mg/L; λ, 10.4 mg/L; κ/λ ratio, 0.95 on Day 441
Molecular and clonality testing | MYD88 L265P testing, next‐generation sequencing, and IGH clonality analysis were not performed at any time point.
Imaging and systemic findings | CT showed no lymphadenopathy; subsequent CT on Day 72 showed no lymphadenopathy or splenomegaly | No palpable lymphadenopathy or hepatosplenomegaly; no fever, night sweats, or weight loss; contemporaneous CT or PET was not performed | No new systemic findings were documented

FIGURE 1: Longitudinal clinical course from initial presentation through complement‐dominant relapse and sutimlimab treatment. (A) Summary of the previously reported initial course in the same patient and the subsequent stable follow‐up. The patient initially presented with hemolytic anemia on Day 0 and was diagnosed with warm autoimmune hemolytic anemia based on direct antiglobulin test positivity for both IgG and C3d. Prednisolone was initiated. Following cold‐triggered relapse on Day 43, the markedly elevated cold agglutinin titer, cold‐reactive erythrocyte agglutination, and complement‐dominant direct antiglobulin test findings led to reassessment as mixed autoimmune hemolytic anemia. Bone marrow flow cytometry on Day 50 identified a κ‐restricted CD20‐positive B‐cell population, whereas serum immunofixation was negative. Rituximab 375 mg/m2 was administered weekly for four doses on Days 58, 65, 72, and 79. Hemolysis improved by Days 92–120, and the patient remained asymptomatic without transfusion or treatment modification through Day 245. (B) Changes in hemoglobin and lactate dehydrogenase levels at selected time points from Days 245 to 469. Laboratory evidence of recurrent hemolysis emerged on Day 315, followed by clinically evident relapse after cold exposure during the subsequent winter on Day 343. Relapse‐phase bone marrow examination and flow cytometry were performed on Day 376; the previously identified κ‐restricted CD20‐positive population was no longer detectable by flow cytometry. Sutimlimab 6.5 g was administered on Days 406, 413, 427, 441, 455, and 469, as indicated by the blue arrowheads. Within 7 days after the first dose, hemoglobin increased from 9.6 to 11.6 g/dL and lactate dehydrogenase decreased from 373 to 230 U/L without red blood cell transfusion. (C) Serial direct antiglobulin test results, cold agglutinin titers, complement measurements, serum IgM concentrations, and direct agglutination findings during relapse and sutimlimab treatment. Negative results are shown as “−,” whereas tests that were not determined are shown as “ND.” AIHA, autoimmune hemolytic anemia; C3, complement component 3; C4, complement component 4; CH50, total hemolytic complement activity; DAT, direct antiglobulin test; FCM, flow cytometry; Hb, hemoglobin; IFE, immunofixation electrophoresis; LDH, lactate dehydrogenase; ND, not determined; PSL, prednisolone; RT, room temperature; w+, weakly positive.

FIGURE 1: Longitudinal clinical course from initial presentation through complement‐dominant relapse and sutimlimab treatment. (A) Summary of the previously reported initial course in the same patient and the subsequent stable follow‐up. The patient initially presented with hemolytic anemia on Day 0 and was diagnosed with warm autoimmune hemolytic anemia based on direct antiglobulin test positivity for both IgG and C3d. Prednisolone was initiated. Following cold‐triggered relapse on Day 43, the markedly elevated cold agglutinin titer, cold‐reactive erythrocyte agglutination, and complement‐dominant direct antiglobulin test findings led to reassessment as mixed autoimmune hemolytic anemia. Bone marrow flow cytometry on Day 50 identified a κ‐restricted CD20‐positive B‐cell population, whereas serum immunofixation was negative. Rituximab 375 mg/m2 was administered weekly for four doses on Days 58, 65, 72, and 79. Hemolysis improved by Days 92–120, and the patient remained asymptomatic without transfusion or treatment modification through Day 245. (B) Changes in hemoglobin and lactate dehydrogenase levels at selected time points from Days 245 to 469. Laboratory evidence of recurrent hemolysis emerged on Day 315, followed by clinically evident relapse after cold exposure during the subsequent winter on Day 343. Relapse‐phase bone marrow examination and flow cytometry were performed on Day 376; the previously identified κ‐restricted CD20‐positive population was no longer detectable by flow cytometry. Sutimlimab 6.5 g was administered on Days 406, 413, 427, 441, 455, and 469, as indicated by the blue arrowheads. Within 7 days after the first dose, hemoglobin increased from 9.6 to 11.6 g/dL and lactate dehydrogenase decreased from 373 to 230 U/L without red blood cell transfusion. (C) Serial direct antiglobulin test results, cold agglutinin titers, complement measurements, serum IgM concentrations, and direct agglutination findings during relapse and sutimlimab treatment. Negative results are shown as “−,” whereas tests that were not determined are shown as “ND.” AIHA, autoimmune hemolytic anemia; C3, complement component 3; C4, complement component 4; CH50, total hemolytic complement activity; DAT, direct antiglobulin test; FCM, flow cytometry; Hb, hemoglobin; IFE, immunofixation electrophoresis; LDH, lactate dehydrogenase; ND, not determined; PSL, prednisolone; RT, room temperature; w+, weakly positive.

Cold‐triggered relapse occurred on Day 43 while the patient was receiving prednisolone 25 mg/day. The cold agglutinin titer was 1:65,536, anti‐IgG DAT was negative, anti‐C3d DAT remained weakly positive, and direct agglutination testing was positive at both room temperature and 37°C, prompting reassessment as mixed AIHA [5].

Bone marrow examination on Day 50 showed no overt lymphoid or plasma cell neoplasm. Flow cytometry identified a κ‐restricted CD20‐positive B‐cell population, whereas serum immunofixation was negative and the serum free light‐chain ratio was normal (Table 1; Figure 2A,B). Rituximab 375 mg/m2 was administered on Days 58, 65, 72, and 79. Hemoglobin increased to 11.2 g/dL and LDH decreased to 267 U/L by Day 92. On Day 120, hemoglobin was 11.1 g/dL and LDH was 268 U/L. Prednisolone was tapered to 5 mg/day by Day 73.

FIGURE 2: Bone marrow flow cytometry before and after rituximab. Flow cytometric analysis of bone marrow mononuclear cells gated on the lymphoid population is shown. (A, B) Before rituximab administration, a distinct CD20‐positive B‐cell population with κ light‐chain predominance was identified. Within the CD45‐gated lymphoid fraction, CD20‐positive cells accounted for 44.6%, and κ‐positive and λ‐positive cells accounted for 33.6% and 6.4%, respectively (κ:λ ratio, 5.3:1), consistent with a κ‐restricted B‐cell population. Pretreatment flow‐cytometric findings from the same patient were reported previously [5]. The panels shown here were generated from the original flow‐cytometry output. (C, D) At relapse after rituximab treatment, the previously identified CD20‐positive population was no longer detectable by flow cytometry. Only minimal κ‐positive and λ‐positive events remained (0.8% and 0.4%, respectively). These paired analyses showed that the previously detected κ‐restricted population was no longer identifiable by marrow flow cytometry despite subsequent winter relapse. Percentages shown within the plots are quadrant values from the original flow cytometry output, whereas the percentages reported in the text for the pretreatment specimen refer to the CD45‐gated lymphoid fraction.

FIGURE 2: Bone marrow flow cytometry before and after rituximab. Flow cytometric analysis of bone marrow mononuclear cells gated on the lymphoid population is shown. (A, B) Before rituximab administration, a distinct CD20‐positive B‐cell population with κ light‐chain predominance was identified. Within the CD45‐gated lymphoid fraction, CD20‐positive cells accounted for 44.6%, and κ‐positive and λ‐positive cells accounted for 33.6% and 6.4%, respectively (κ:λ ratio, 5.3:1), consistent with a κ‐restricted B‐cell population. Pretreatment flow‐cytometric findings from the same patient were reported previously [5]. The panels shown here were generated from the original flow‐cytometry output. (C, D) At relapse after rituximab treatment, the previously identified CD20‐positive population was no longer detectable by flow cytometry. Only minimal κ‐positive and λ‐positive events remained (0.8% and 0.4%, respectively). These paired analyses showed that the previously detected κ‐restricted population was no longer identifiable by marrow flow cytometry despite subsequent winter relapse. Percentages shown within the plots are quadrant values from the original flow cytometry output, whereas the percentages reported in the text for the pretreatment specimen refer to the CD45‐gated lymphoid fraction.

From Day 120 through Day 245, the patient remained asymptomatic without transfusion, treatment modification, infection, or clinically relevant cold exposure. Hemoglobin remained 11.8–12.7 g/dL and LDH 211–239 U/L. On Day 245, total bilirubin was 0.54 mg/dL, haptoglobin was 27 mg/dL, reticulocytes were 2.1%, anti‐IgG DAT was negative, and anti‐C3d DAT remained weakly positive (Figure 1A,B).

Laboratory evidence of recurrent hemolysis emerged on Day 315, followed by cold‐associated exertional dyspnea and fatigue during the subsequent winter. Between Days 315 and 385, hemoglobin decreased from 11.1 to 10.5 g/dL and LDH increased from 260 to 350 U/L. DAT findings were predominantly C3d‐positive, and erythrocyte agglutination persisted (Figure 1B,C).

Relapse‐phase marrow examination on Day 376 showed erythroid hyperplasia, approximately 1% CD20‐positive B cells, and 2%–5% CD138‐positive plasma cells without significant κ/λ imbalance, lymphoplasmacytic infiltration, or lymphoid aggregates. The previously identified κ‐restricted CD20‐positive population was no longer detectable by flow cytometry (Table 1; Figure 2C,D). The patient had no palpable lymphadenopathy, hepatosplenomegaly, fever, night sweats, or weight loss. Contemporaneous computed tomography or positron emission tomography was not performed.

By Day 385, hemoglobin was 10.5 g/dL, LDH was 350 U/L, total bilirubin was 0.93 mg/dL, haptoglobin was 3 mg/dL, and reticulocytes were 4.1%. Anti‐IgG DAT was negative, anti‐C3d DAT was weakly positive, and serum IgM was 100 mg/dL. Serum free light chains were κ: 11.8 mg/L and λ: 10.2 mg/L, with a κ/λ ratio of 1.16. A quadrivalent meningococcal conjugate vaccine (MenACWY‐TT; MenQuadfi) and a 23‐valent pneumococcal polysaccharide vaccine (PPSV23; Pneumovax 23) were administered on Day 387 before complement inhibition.

Immediately before the first sutimlimab dose on Day 406, hemoglobin had decreased to 9.6 g/dL, and LDH had increased to 373 U/L. Total bilirubin was 1.10 mg/dL, haptoglobin was 2 mg/dL, and reticulocytes were 2.9%. The polyspecific DAT was 1+, anti‐IgG was negative, and anti‐C3d remained weakly positive. The cold agglutinin titer was 1:65,536, and direct agglutination was positive at 37°C (2+) and room temperature. Complement measurements showed C3 of 120 mg/dL, C4 of 12 mg/dL, and total hemolytic complement activity (CH50) of 31 U/mL. Prednisolone remained at 5 mg/day without dose escalation.

Sutimlimab 6.5 g was administered on Days 406, 413, 427, 441, 455, and 469. Within 7 days after the first dose, hemoglobin increased from 9.6 to 11.6 g/dL and LDH decreased from 373 to 230 U/L without red blood cell transfusion. Total bilirubin decreased to 0.48 mg/dL, haptoglobin increased to 16 mg/dL, and exertional dyspnea and fatigue improved. The polyspecific DAT, anti‐IgG DAT, and anti‐C3d DAT were each 1+ on Day 413, indicating persistence of serologic abnormalities despite the early hematologic response (Figure 1B,C).

During treatment, the cold agglutinin titer remained 1:131,072, while C4 increased and CH50 became suppressed. Serum protein electrophoresis showed suspected peaks of 0.896 g/dL on Day 427 and 0.891 g/dL on Day 441; immunofixation on Day 441 identified IgG‐κ and IgM‐κ monoclonal proteins, with a normal serum free light‐chain ratio. Direct agglutination and DAT reactivity persisted during the early hematologic response.

By Day 455, hemoglobin was 12.7 g/dL and DAT had become negative. Prednisolone was discontinued, and hemoglobin remained 12.8 g/dL with LDH 201 U/L on Day 469.