Case Report: Pernicious anemia-associated pancytopenia responds to immunotherapy

Introduction

Pancytopenia arises from diverse causes, including marrow failure, nutritional deficiency, autoimmune disease, and hematologic malignancy. Accurate diagnosis requires careful synthesis of clinical, morphologic, and laboratory data to prevent misclassification and delayed recognition of reversible conditions. Among potential nutritional etiologies, vitamin B₁₂ deficiency is a frequent cause of pancytopenia, yet remains underrecognized. In addition, measurement of serum vitamin B₁₂ by competitive luminescence binding assays is vulnerable to interference by high titer anti–intrinsic factor antibodies present in pernicious anemia, potentially producing falsely normal results and confounding the diagnosis (1, 2).

There are numerous reports of vitamin B₁₂ deficiency and pernicious anemia as causes of megaloblastic anemia with an MDS-like picture. These cases, however, typically feature low serum B₁₂ levels, with or without posterior column neurological findings, and rapid correction with replacement therapy. Here, we report a diagnostically misleading presentation of pernicious anemia related to immune-mediated assay interference in B₁₂ measurement, along with sustained hematologic responses to immunosuppressive therapy alone, emphasizing the importance of metabolic confirmation of B₁₂ sufficiency when evaluating unexplained cytopenias.

Case description

A 67-year-old Latina with a medical history notable for type II diabetes mellitus, thyroid nodules, and hypertension presented to the hematology clinic at the request of her primary care physician for acute pancytopenia noted on a recent presentation to the Emergency Department. She described gradual development of severe fatigue, dyspnea on exertion, easy bruising, and chest wall discomfort, without neurologic or constitutional (“B”) symptoms. There was a family history of gastric cancer (father, developed in his 8^th decade). Physical examination was unremarkable, apart from ecchymoses related to blood punctures.

Laboratory results: White blood cell count (WBC) 2.6 x 10⁹/L, absolute neutrophil count (ANC) 0.9 K/uL, hemoglobin (Hb) 76 g/L, hematocrit (hct) 20.3%, mean cell volume (MCV) 100.9 fL), and platelets (Plt) 90 x 10⁹/L. Evidence of marked hemolysis: total bilirubin 2.6 mg/dL (reference range (ref.) 0.0-1.2 mg/dL) with unconjugated (indirect) bilirubin 2.0 mg/dL, lactate dehydrogenase (LDH) 5575 U/L (ref. 90–180 U/L), haptoglobin <10 mg/dL (ref. 30–200 mg/dL), and reticulocyte index 0.57. The direct antiglobulin test (DAT) and an enhanced DAT (“Super Coombs”, which detects low-level or atypical immunoglobulin and/or complement binding to red blood cells; performed at a reference laboratory, Versiti) were negative. Serum folate was 14.8 ng/mL (ref. >5.8 ng/mL) and total vitamin B12 level 410 pg/mL (ref. 181–914 pg/mL).

Peripheral blood smear revealed macrocytic, normochromic red blood cells with polychromasia, occasional target cells, megaloblastic forms, and red blood cell agglutination. Bone marrow evaluation revealed dyserythropoiesis, myeloid and megakaryocyte dysplasia, without an increase in blasts or ring sideroblasts (M:E 0.5). Cytogenetics, myeloid FISH panel, and NGS studies (Neogenomics Myeloid Disorders Profile, 63 gene panel) were all negative for known MDS-related genetic alterations. A bone marrow failure NGS assay (Invitae bone marrow failure syndromes panel, 115 genes) was negative. Both B- and T-cell clonality assays of peripheral blood lymphocytes were also negative.

Serum erythropoietin was 194 mU/mL, less than expected for a hematocrit (Hct) of 20%. Serum protein electrophoresis, free light-chain assay, hemoglobin electrophoresis, and red cell enzymopathy studies were all normal.

The patient was prescribed prednisone, 1 mg/kg daily for two weeks, for concern for MDS with an associated immune-mediated hemolytic process, in addition to exogenous erythropoietin, with complete resolution of her cytopenias within four months (Figure 1). However, three months after a prednisone taper, she developed recurrence of pancytopenia (WBC 3.0 x 10⁹/L, ANC 1.5 K/uL, Hb 95 g/L, MCV 92.1 fL, and Plt 140 x 10⁹/L) with hemolysis (unconjugated bilirubin 1.0 mg/dL, LDH 765 U/L, and haptoglobin < 10 mg/dL). Prednisone was resumed at 1 mg/kg, and the patient received weekly rituximab 375 mg/m² for four doses for suspected refractory immune-mediated hemolysis, with marked improvement in her cytopenias.

Four months later, her cytopenias again recurred, and she developed unintentional weight loss of 3 kg. Computed tomography (CT) of the chest, abdomen, and pelvis showed no organomegaly or lymphadenopathy. Flow cytometry studies for evidence of paroxysmal nocturnal hemoglobinuria were negative. On repeat bone marrow examination, trilineage dysplasia was present with greater than 20% dysplastic elements in each lineage, without increased blasts. There was an increase in erythroid elements with megaloblastic maturation, including numerous binucleated forms. The megakaryocytic lineage featured small hyposegmented forms and occasional elements with separated nuclei. Granulocytes featured aberrant nuclear segmentation (Figures 2A–C). Iron staining did not show increased iron stores, but ring sideroblasts now accounted for 11% of the erythroid elements (Figure 2D). The trephine core biopsy was consistent with a hypercellular bone marrow for age (Figure 2E). Cytogenetics and NGS studies were again negative. The peripheral blood smear demonstrated numerous poikilocytes, including macro-ovalocytes, occasional schistocytes, and spherocytes (Figure 2F).

Immunotherapy with rituximab was resumed for five weekly doses of 375 mg/m², followed by eight monthly doses. Her counts again recovered (Figure 1). However, one month after monthly dosing was initiated, the patient presented with her fourth episode of worsening cytopenias and hemolysis (WBC 1.7 x 10⁹/L, ANC 0.7 K/uL, Hb 85 g/L (MCV 90.6 fL), Plt 74 x 10⁹/L, unconjugated bilirubin 3.8 mg/dL, LDH 9413 U/L, and haptoglobin < 10 mg/dL). One mg/kg of prednisone was added, and the patient once again presented significant improvement in cytopenias and hemolysis markers. Given the non-sustained response to immunosuppression and the marked erythroid responses, nutritional studies were reevaluated. Vitamin B₁₂ levels were 101 pg/mL with an elevated serum methylmalonic acid (MMA) of 29 umol/L (ref. 0.00-0.40 umol/L). Anti-intrinsic factor antibody (anti-IF Ab) testing was positive, confirming a diagnosis of pernicious anemia. Immunosuppression was stopped, and the patient started parenteral vitamin B₁₂ with complete and sustained resolution of cytopenias. Notably, the patient reported significant anxiety related to the initial diagnosis of MDS and expressed profound relief following clarification of a reversible cause.

Follow-up and outcomes

Following initiation of parenteral vitamin B₁₂ replacement, the patient experienced rapid, complete, and sustained normalization of blood counts. At last follow-up (28 months; December 2025), she remains clinically well without recurrence of cytopenias. Vitamin B₁₂ therapy was well tolerated, and no adverse events were observed. The patient declined follow-up bone marrow examination. Gastric biopsies performed during surveillance endoscopy revealed extensive intestinal metaplasia, which is being closely monitored.

Discussion

The differential diagnosis of pancytopenia in adults is broad and includes hematologic malignancy, bone marrow failure disorders, vitamin deficiencies, and autoimmune conditions. This case illustrates a multistep diagnostic trap in which the immune biology of pernicious anemia resulted in misinterpretation of laboratory testing, bone marrow morphology, and therapeutic response, ultimately producing a convincing but incorrect diagnosis of progressive MDS.

Vitamin B₁₂ deficiency and pernicious anemia are well-recognized MDS mimics. Classically, they present as macrocytic anemia, megaloblastosis, and low serum B₁₂ levels, prompting timely replacement therapy. In severe cases, profound ineffective erythropoiesis can result in elevated LDH, indirect hyperbilirubinemia, and low haptoglobin, which can simulate hemolytic anemia or marrow failure syndromes. Vitamin B₁₂ deficiency has also been reported to produce marked dysplastic morphologic changes—including pseudo–Pelger-Huët neutrophils, nuclear-cytoplasmic asynchrony, and even cytogenetic abnormalities—that can be indistinguishable from MDS at initial evaluation (2). Several case reports and series have described patients initially suspected of having MDS due to pancytopenia and marrow dysplasia, who were ultimately found to have reversible B₁₂ deficiency, with hematologic normalization following repletion therapy (3).

However, this case differs in critical aspects from the previously reported literature. First, serum vitamin B₁₂ levels were repeatedly normal at presentation, delaying suspicion for deficiency. Second, the patient demonstrated reproducible hematologic responses to erythropoietin initially, then corticosteroids and anti-CD20 therapy—strongly reinforcing an immune-mediated or clonal hematologic diagnosis, atypical for pernicious anemia. These laboratory and clinical features masked a reversible nutritional deficiency for more than a year. To our knowledge, sustained clinical responses to immunotherapy such as rituximab have not been previously emphasized in pernicious anemia.

This case raises two important clinical questions: whether the cause of her presentation could have been determined sooner, with less invasive testing, and why there were both clinical and laboratory responses to glucocorticoids and rituximab.

The first question centers on the limitations of commercially available vitamin B₁₂ testing. There is no gold standard test for the measurement of serum B₁₂ levels. Studies by Devalia et al. suggest that use of a cutoff value of serum B₁₂ less than 200 ng/L would diagnose 97% of true vitamin deficiency (4). Additionally, significant inter-laboratory variability in serum concentrations occurs (5). Notably, the majority of laboratories measure B₁₂ using rapid competitive luminescence binding assays, where a known quantity of labelled vitamin B₁₂ tracer is added to the patient serum sample and competes with the patient’s endogenous B₁₂ for binding to intrinsic factor provided as an assay reagent. The measured signal is inversely proportional to the patient’s B₁₂ concentration. In pernicious anemia, anti–intrinsic factor antibodies present in the patient’s serum may bind to the assay intrinsic factor reagent, thereby preventing binding of the labelled B₁₂ tracer and producing spuriously normal or elevated results. Analytical mechanisms and potential interferences of this and related competitive assays have been described and illustrated in detail previously (5–7).

While most vitamin B₁₂ assays now include a pretreatment denaturation step to inactivate these antibodies, it is suspected that this inactivation step may fail in some patients with high antibody titers (5, 6). Had MMA levels been measured at initial presentation, the correct diagnosis would have been made sooner.

The second question—why immunosuppression was effective—likely reflects partial treatment of the autoimmune pathophysiology underlying pernicious anemia, presumably partially treating the metabolic deficiency. Pernicious anemia is an immune-mediated disease; however, unlike other autoimmune conditions that are treated with immunosuppression, treatment is with B₁₂ administration. Several reports describe clinical, laboratory, and histopathologic responses of pernicious anemia to oral glucocorticoids (7–11). We suspect that immunotherapy for “immune-mediated” hemolysis partially treated the immunopathogenesis of the underlying pernicious anemia, confounding assessment of treatment response and delaying reassessment of the diagnosis.

Conclusion

This case underscores the potential limitations of relying solely on serum vitamin B12 measurements in patients with unexplained cytopenias. Current common laboratory techniques used to measure serum B₁₂ concentrations may be significantly altered by the presence of high-titer anti-intrinsic factor antibodies, as previously described in patients with pernicious anemia. Moreover, transient clinical improvement following immunosuppressive therapy can obscure the correct diagnosis by partially treating the autoimmune basis of pernicious anemia without correcting the metabolic consequence. If there is clinical suspicion for pernicious anemia, measurement of MMA should be considered when B₁₂ levels are normal or low-normal.

The evaluation of pancytopenia requires a search for primary and secondary causes. Marrow dysplasia can be due to primary MDS or secondary causes, including B₁₂ deficiency. Notably, primary MDS without somatic gene mutations, as now routinely determined by NGS gene panels or typical cytogenetic changes, is uncommon (<10%) (12), and should prompt consideration of other causes.

As occurred here, a misdiagnosis of MDS may confer substantial emotional trauma and potentially lead to unnecessary treatments with hematopoietic growth factors and/or hypomethylating agents.

Strengths and limitations

This report is limited by its single-patient design and may not capture the full spectrum of presentations of pernicious anemia with immune features. Additionally, assay interference may vary across laboratory platforms. However, the strength of this case lies in its longitudinal follow-up, repeated molecular evaluation, and reproducible immunotherapeutic responses, which together illuminate a clinically relevant diagnostic failure mode.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: John Strouboulis, King’s College London, United Kingdom

Reviewed by: Tayler Van Denakker, Icahn School of Medicine at Mount Sinai, United States

Ilker Çordan, Konya City Hospital, Türkiye