Immune-mediated hemolytic anemia (IMHA), which is characterized by antibody-mediated red blood cell (RBC) destruction, can manifest as either a primary or secondary condition and is generally triggered by an underlying stimulus to the immune system (3,5,11). Precursor-targeted immune-mediated anemia (PIMA) is a rare disorder characterized by inefficient erythropoiesis associated with the immune-mediated destruction of erythroid precursors within the bone marrow. This condition has several different presentations, including nonregenerative immune-mediated anemia (NRIMA), nonregenerative immune-mediated hemolytic anemia (NRIMHA), and pure red cell aplasia (PRCA) (1-3,6,7).

Published studies on the clinical findings, survival, and outcomes of cats with PIMA remain limited. In one study involving 15 cats with PRCA and NRIMA, the disease generally developed in cats under 3 years of age, with remission achieved 12-42 days after commencing immunosuppressive therapy. The mortality rate was 27% (3). Further, one prior case report described a presumptive diagnosis of PIMA in a cat with NRIMA and concurrent gastrointestinal lymphoma (8). The hallmark of PIMA is ineffective erythropoiesis due to suspected immune-mediated targeting of erythroid precursors. At the same time, the characteristic symptoms include erythroid hypercellularity, persistent nonregenerative anemia, and occasional cytological evidence of selective phagocytosis of erythroid precursors, known as rubriphagocytosis. PIMA frequently responds to immunosuppressive therapy and is associated with IMHA, further supporting its immune-mediated pathogenesis (2). The reported mortality rate of PRCA disease is 23.5-33%, and although remission can be achieved with immunosuppressive therapy, the relapse rate is high, at 27-75% (3).

Herein, we describe the case of a 1-year-old Scottish Fold cat diagnosed with PIMA.

A 1-year-old, 4.1 kg, spayed female Scottish Fold cat presented with pale mucous membranes. Initial evaluation at a primary veterinary hospital revealed normocytic, normochromic, moderate anemia with hematocrit (Hct) of 15.5% (reference interval, 27-47%), mean corpuscular volume (MCV) of 52 fL (reference interval, 40-55 fL), mean corpuscular hemoglobin concentration (MCHC) of 31.2 g/dL (reference interval, 31-36 g/dL). The cat was tentatively diagnosed with immune-mediated hemolytic anemia (IMHA) and prescribed prednisolone (PDS) at 2 mg/kg PO BID, mycophenolate mofetil (MMF), clopidogrel, and doxycycline. After two weeks, the Hct increased to 35.4%. However, tapering the PDS dose to 0.7 mg/kg PO BID led to a drop in Hct to 20.3%. MMF was replaced with cyclosporine (25 mg/day, PO, SID), and the PDS dose was increased to 1.5 mg/kg PO BID, resulting in a rise in Hct to 39.3%. Despite tapering efforts, anemia recurred whenever the PDS dose was below 1 mg/kg PO BID, with Hct falling below 26%. After six months of treatment adjustments, the patient was referred for further evaluation while receiving PDS (1 mg/kg, PO, BID).

Upon further evaluation, physical examination revealed pale pink dry mucous membranes, tachycardia, and abdominal distension. The manual packed cell volume (PCV) measured by microhematocrit tube centrifugation was 22%, with a total solid volume of 9 g/dL. A complete blood count (CBC) was performed using an automated hematology analyzer (IDEXX ProCyte DX, IDEXX Laboratories Inc., USA), along with a blood film examination stained with Diff-Quick, revealed macrocytic normochromic regenerative mild anemia, based on the following findings: Hct 22.1% (reference interval, 30.3-52.3%), MCV 67 fL (reference interval, 35.9-53.1 fL), MCHC 32.1 g/dL (reference interval, 28.1-35.8 g/dL), reticulocyte count 265.7 ×10³/μL (reference interval, 3-50 ×10³/μL). Further, a stress leukogram (neutrophil count 15.9 ×10³/μL [reference interval, 2.30-10.29 ×10³/μL], monocyte count 1.68 ×10³/μL [reference interval, 0.05-0.67 ×10³/μL]), with a left shift (1,368 band neutrophils/μL) was identified (Table 1). Examination of ×1,000 oil immersion fields revealed 2 spherocytes and 3 polychromatophils (Fig. 1). At the same time, no agglutination was observed in a saline agglutination test (SAT) involving the mixing of 4 drops of saline with 1 drop of blood. Counting of 300 white blood cells revealed 73 nucleated red blood cells, yielding a corrected white blood cell count of 23.90 ×10³/μL (reference interval, 2.87-17.02 ×10³/μL).

Figure 1. Peripheral blood smears. Numerous small spherical RBCs were identified as spherocytes (arrows). Although feline RBCs generally lack central pallor, the smaller RBCs, in this case, were interpreted as spherocytes. Polychromatophils (arrowheads) can also be observed. ×1,000, Diff-Quik staining. Scale bar = 10 μm.

Table 1 . Selected laboratory results during the patient’s initial visit.

Parameters	Reference interval	Case
Manual packed cell volume	30-52 (%)	22
Red blood cells	6.54-12.20 (×1012/L)	3.3
Hematocrit	30.3-52.3 (%)	22.1
Hemoglobin concentration	9.8-16.2 (g/dL)	7.1
Mean corpuscular volume	35.9-53.1 (fL)	67
Mean corpuscular hemoglobin concentration	28.1-35.8 (g/dL)	32.1
Red blood cell distribution width	15.0-27.0 (%)	31.3
Reticulocytes	3.0-50.0 (×103/μL)	265.7
White blood cells	2.87-17.02 (×109/L)	29.64
Segmented neutrophils	2.30-10.29 (×109/L)	15.9
Band neutrophils	0-1,000 (cells/μL)	1,368
Lymphocytes	0.92-6.88 (×109/L)	4.001
Monocytes	0.05-0.67 (×109/L)	1.684
Eosinophils	0.17-1.57 (×109/L)	0.842
Basophils	0.01-0.26 (×109/L)	0.105
Corrected WBC	5.5-19.5 (×109/L)	23.902
Platelets	151-600 (×109/L)	447
Alanine aminotransferase	12-130 (U/L)	142
Glucose	74-159 (mg/dL)	238
Triglyceride	10-100 (mg/dL)	>375
Fructosamine	191-349 (μmol/L)	465

To rule out infectious hemolytic anemia, a multiplex polymerase chain reaction (PCR) analysis (Feline anemia panel, IDEXX Laboratories, Westbrook, Maine, USA) was performed, which yielded negative results for Anaplasma spp., Bartonella spp., Cytauxzoon felis, Ehrlichia spp., and feline hemotropic Mycoplasma. Serological tests were also negative for feline leukemia virus (FeLV) antigen and feline immunodeficiency virus (FIV) antibody (SNAP, Feline Triple Test, IDEXX Laboratories). Urinalysis and fecal examination revealed no abnormalities except glucosuria. Abdominal ultrasonography showed no causes of anemia, such as tumors, inflammation, or intra-abdominal bleeding. The biochemistry profile indicated hyperglycemia (238 mg/dL; reference interval, 74-159 mg/dL), elevated fructosamine (465 μmol/L; reference interval, 191-349 μmol/L), and hypertriglyceridemia (above the range; reference interval, 10-100 mg/dL), which might due to steroid-induced diabetes mellitus. Hyperlactatemia (3.4 mmol/L; reference interval, 0-2.5 mmol/L) was also observed (Table 1).

Based on the above results, nonregenerative anemia was suspected; however, a definitive diagnosis was challenging due to the cat's 6-month history of PDS use: as the PDS dose was tapered, the anemia worsened (Hct, 14.6%), with a substantially reduced reticulocyte count of 61.8 ×10³/μL observed on day 7. Simultaneously, spherocytes increased to 4 in the high-power field (×1,000) (Table 2). However, regeneration improved with increasing PDS dosage (Fig. 2).

Figure 2. Changes in hematologic parameters during the therapeutic course in a cat with anemia.

Table 2 . Blood smear results obtained over a period of several days, indicating the presence of immune-mediated hemolytic anemia and the extent of regeneration.

Parameters	Initial visit	7th day	After steroid tapering for 5 days	2 months after medication for PDS & chlorambucil
Hct (%)	22.1	14.6	14.7	35
PCV (%)	22	17	12	28
RBC (1012/L)	3.3	2.85	2.45	6.48
Polychromatophil (cells/hpf)	3	1.5	3	0
Spherocyte (cells/hpf)	2	4	4.5	1
Nucleated RBC (cells/100 WBC)	24.3	6	10	0

Bone marrow (BM) aspiration was performed at the proximal humerus on day 7 to ascertain the cause of persistent anemia and inadequate regeneration. Marked cellularity was observed, with no evidence of dysplasia (Fig. 3A). The myeloid to erythroid (M:E) ratio was low (0.35:1), indicating erythroid hypercellularity. It contains primarily erythroid precursors, with a moderate left shift characterized by increased rubriblasts. The erythroid series maturation was complete, with moderate polychromatophils in the background (Fig. 3B). The myeloid series maturation was complete and morphologically unremarkable. Megakaryocytes appeared to be adequately increased in both mature and immature forms. Real-time PCR for FeLV provirus in the BM samples was negative.

Figure 3. Cytological evaluation of the bone marrow. (A) At low magnification, the cellularity was determined to be 90%, indicating hypercellular bone marrow. A heterogeneous pattern was observed with various cell lines present. ×200, Diff-Quik stain. Scale bar = 50 μm. (B) Erythroid lineage cells with intensely basophilic cytoplasm and a high N:C ratio are predominantly observed. The M:E ratio was confirmed to be 0.35:1. ×1,000, Diff-Quik stain. Scale bar = 10 μm.

Eventually, the cat was diagnosed with PIMA for the following reasons: history of normocytic, normochromic anemia at the time of initial presentation to the primary hospital, persistent anemia due to an inadequate regenerative response, absence of agglutination in SAT, no evidence of underlying disease, and bone marrow findings supportive of ineffective erythropoiesis, marked by erythroid hyperplasia with increased rubriblasts and no dysplastic features in other cell lines. The cat was subsequently maintained on immunosuppressive therapy comprising PDS (1 mg/kg, PO, SID) and chlorambucil (2 mg/cat, PO, EOD).

One month after the initiation of PDS and chlorambucil therapy, further examination revealed a hematocrit of 18.3%, while the reticulocyte count measured by an automated hematology analyzer was 60.8 ×10³/μL. Subsequently, while maintaining a hematocrit of over 30%, the PDS gradually decreased (Fig. 2). However, two months after discontinuation of PDS, relapse occurred (Fig. 2). Following anemia relapse, MMF and cyclosporine were added to the treatment regimen; however, no improvement in anemia was observed, and regeneration was not confirmed. Follow-up was discontinued at the owner's request.

In this case, the diagnosis of suspected PIMA was primarily based on the history of normocytic, normochromic anemia identified in the initial blood test, the bone marrow examination results, and the patient's response to immunosuppressive therapy. PIMA is characterized by the immune-mediated destruction of erythroid precursors within the bone marrow, resulting in ineffective erythropoiesis and persistent anemia despite apparent regenerative efforts.

Bone marrow samples are classified as having erythroid hyperplasia when BM cellularity is normal or increased and the M:E ratio is below 1.2:1. Erythroid hypoplasia is defined as normal or decreased BM cellularity with an M:E ratio of > 2.16:1. Pure red cell aplasia is defined as the presence of severe nonregenerative anemia, with a marked increase in the M:E ratio above 75:1 (8). In a previous study of bone marrow evaluations conducted in 30 cats with severe nonregenerative anemia lasting more than 5 days without identifiable causes, erythroid hyperplasia was predominant, noted in 80% of cases (24/30 cats) (6). Rubriphagocytosis and myelofibrosis are also common (10,11). However, in another investigation of feline PIMA cases, rubriphagocytosis was observed in 4/30 cats, while myelofibrosis was confirmed in 4/22 cats (6). In the present case, severe anemia persisted for more than six months. After excluding other diseases, bone marrow evaluation revealed an M:E ratio of 0.35:1, indicating erythroid hyperplasia. The examination revealed increased rubriblasts and a left shift in erythroid precursors, suggesting potential maturation arrest at the progenitor cell level. The complete maturation of the erythroid series and moderate numbers of polychromatophils is likely attributed to prolonged immunosuppressive therapy, including PDS. Crucially, no dysplasia was observed in cell lines, a key differentiation factor between PIMA and other marrow disorders such as myelodysplastic syndromes. Furthermore, real-time PCR for the FeLV provirus in the bone marrow samples was negative, ruling out FeLV-associated anemia.

Spherocytosis was observed in 2% of PIMA cases (2 out of 66 dogs) (2). In one study, dogs diagnosed with PIMA and marked spherocytosis (>5/hpf) suggested hemolysis due to the presence of different antibodies or overlapping epitopes in erythroid progenitors and mature RBCs (4). Spherocytosis has not yet been reported in feline PIMA cases. In this case, the observed spherocytes may have resulted from antibody-coated RBCs binding to the macrophage Fc portion in peripheral blood, independent of bone marrow progenitor cells. However, the presence of spherocytes doesn't necessarily confirm IMHA. As previously mentioned, evaluating spherocytes in cats can be challenging, so these cells may not be true spherocytes.

Reticulocytes were not evaluated during the initial blood test at a local hospital six months prior. However, considering the confirmed normocytic normochromic moderate anemia at that time, the possibility of absent regeneration was considered. During the initial visit to our clinic, the cat took immunosuppressive medication, making accurate assessment challenging. Nonetheless, when the immunosuppressive medication was discontinued, a trend towards a decreased reticulocyte count was observed, in addition to worsening anemia without a sufficient regenerative response. Additionally, bone marrow examination in this case revealed a distinct decrease in the M:E ratio to 0.35:1. Based on the observed prolonged severe anemia, inadequate regenerative response, and insufficient erythropoiesis, feline PIMA was suspected.

Glucocorticoids are generally considered the primary treatment option for feline PIMA (3,5,6,7,9), whereas second-line immunosuppressive options include cyclosporine, chlorambucil, and MMF (3,6,7,9). In one study, glucocorticoid monotherapy was administered to seven feline PIMA patients, achieving remission in six cats, with relapse in three cases. Another cat in this study achieved remission without subsequent relapse following treatment with a combination of glucocorticoids and cyclosporine. Among the six cats treated with chlorambucil as an adjuvant to glucocorticoids, three showed remissions, while three exhibited no medication response (3). In a separate feline PIMA study, two cats treated with MMF showed no increase in HCT (6). The relapse rate for feline PIMA is reported to be high, ranging from 27% to 77% (3,6). While the reasons for introducing a second immunosuppressant and the choice of the agent were not explicitly clear, and relapses were frequent, most cats responded to treatment modifications. In one study, all three relapsing cats responded to treatment modifications (3), with the same observed in 10/13 cats in another study (6). In the present case, the patient experienced severe anemia despite prior treatment with PDS, MMF, and cyclosporine. Given the insufficient response to this medication, the decision was made to change the treatment to chlorambucil and subsequently initiate therapy.

Despite the challenging interpretation of apparent nonregenerative anemia due to prior medication history, the diagnosis and treatment of this patient suspected of having feline PIMA were facilitated based on the observed pattern of anemia and bone marrow examination results. However, despite the use of PDS, cyclosporine, and MMF at a local hospital, anemia persisted. An immunosuppressive treatment was attempted by switching to chlorambucil, which resulted in successful management lasting more than a year. This case highlights the importance of bone marrow examination in patients with persistent anemia. Furthermore, it emphasizes that treatment modification can be beneficial when the response to previous medication is suboptimal.

This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (2020R1C1C1008675).

The authors have no conflicting interests.