Food cobalamin malabsorption (FCM) represents a prevalent, often underdiagnosed, etiology of vitamin B12 deficiency, particularly within an aging population. Unlike pernicious anemia, an autoimmune disorder targeting intrinsic factor, FCM stems from the impaired release of cobalamin from food proteins, primarily due to age-related gastric changes, medication-induced gastric hypochlorhydria, metformin, or non-immune atrophic gastritis. The clinical presentation of FCM mirrors that of general cobalamin deficiency, encompassing a spectrum of neurological (peripheral neuropathy, cognitive decline), hematological (megaloblastic anemia), and gastrointestinal (glossitis, anorexia) manifestations. Given the potential for irreversible neurological sequelae, early detection and intervention are paramount. High-dose oral cobalamin (125–250 mcg daily) has demonstrated efficacy, offering a convenient and cost-effective alternative to parenteral administration.

Drug-induced and idiosyncratic cytopenias, including anemia, neutropenia, and thrombocytopenia, present significant challenges in fields like immunohematology and internal medicine. These conditions are often unpredictable, multifactorial, and can arise from a complex interplay of drug reactions, immune abnormalities, and other poorly understood mechanisms. In many cases, the precise triggers and underlying factors remain unclear, making diagnosis and management difficult. However, advancements in artificial intelligence (AI) are offering new opportunities to address these challenges. With its ability to process vast amounts of clinical, genomic, and pharmacovigilance data, AI can identify patterns and risk factors that may be missed by traditional methods. Machine learning algorithms can refine predictive models, enabling earlier detection and more accurate risk assessments. Additionally, AI’s role in enhancing patient engagement—through tailored monitoring and personalized treatment strategies—ensures more effective follow-up and improved clinical outcomes for patients at risk of these potentially life-threatening conditions. Through these innovations, AI is paving the way for a more proactive and personalized approach to managing drug-induced cytopenias.

Drug-induced agranulocytosis is a rare but life-threatening adverse reaction associated with numerous non-chemotherapy drugs. Management relies on immediate drug withdrawal, infection control, and, in selected patients, administration of granulocyte-colony stimulating factor (G-CSF). This review summarizes current knowledge on the determinants of epidemiology, clinical presentation, hematologic and biologic features, comorbidities, and outcomes influencing the decision to introduce G-CSF in drug-induced agranulocytosis. Evidence from observational studies and meta-analyses suggests that G-CSF shortens neutropenia duration and hospitalization, although its impact on mortality remains uncertain. The decision to use G-CSF should consider initial neutrophil count, presence of severe infection or sepsis, age, and comorbidities. Despite the accumulated experience, randomized controlled trials are still lacking, and treatment algorithms remain empirical.

Febrile neutropenia is a medical emergency requiring rapid and rigorous management considering the risk of severe infection. Febrile neutropenia is a frequent complication in patients receiving chemotherapy. Initial assessment is vital in order to decide the follow-up (outpatient treatment, conventional care or intensive care unit) and to initiate adapted antibiotic therapy as soon as possible. Depending on the infectious syndrome (clinically documented, microbiologically documented or undocumented), antibiotic therapy should be adapted or discontinued, considering the recovery from aplasia and absence of fever. Antifungal agents are not systematically used and its use should be discussed according to the context. G-CSF should be used prophylactically, and is not an adjunct to antibiotic therapy for febrile neutropenia. Granulocyte transfusions are exceptionnally indicated and its use should be discussed on a case-by-case basis. Standard precautions are adequate for the majority of patients, with the exception of specific situations.

An 83-year-old woman was admitted to the emergency department for a 7-day history of fatigue and progressive cyanosis in the feet and hands after cold exposure despite physical protective measures. Upon arrival, the patient presented with necrotic cutaneous lesions in both hands and distal lower extremities. Upon admission, hemoglobin was 7.6 g/dL and laboratory tests were consistent with cold agglutinin disease (CAD), the presence of monoclonal IgM, and flow cytometry consistent with lymphoplasmacytic lymphoma, but MYD88 L265P mutation was negative. The patient required blood transfusion, resulting in stabilized hemoglobin and a decrease in markers of hemolysis. Treatment with aspirin 250 mg daily and intravenous iloprost 0.5 mL/h was initiated with a poor clinical response at day 4. Amputation was required. Plasma exchange was performed and chemotherapy with rituximab and bendamustine was initiated. The clinical course was marked by further necrosis, prompting discussions regarding an additional amputation that was not performed considering the high surgical risk and refusal by the patient. Supportive treatment was initiated, and the patient expired one month after hospital admission.

Idiosyncratic drug-induced severe neutropenia and agranulocytosis are rare but potentially life-threatening adverse drug reactions. They cause profound neutrophil depletion after exposure to non-cytotoxic medications. Advances in supportive care have substantially reduced mortality. Long-term management nevertheless remains challenging. Clinicians must decide whether a structurally or pharmacologically related drug can be prescribed safely (cross-reactivity). They must also decide whether the culprit drug can ever be reintroduced (rechallenge). Current evidence supports an immune-mediated pathogenesis. It involves reactive drug metabolites, hapten formation, human leukocyte antigen (HLA)-restricted antigen presentation, and T-cell activation. The strength of evidence for cross-reactivity varies considerably across drug classes. Clinically meaningful cross-reactivity is well established for antithyroid thionamides. Evidence for beta-lactam antibiotics and thienopyridine antiplatelet agents remains limited, resting largely on case reports and pharmacovigilance data. For most other medications, evidence is insufficient to support generalized class-wide avoidance. Evidence supporting drug rechallenge is similarly heterogeneous. The most robust data concern clozapine. Carefully selected patients with previous neutropenia—but generally not agranulocytosis—may successfully undergo rechallenge under intensive hematologic monitoring, often with granulocyte colony-stimulating factor (G-CSF) support. Outside this setting, deliberate rechallenge is rarely justified. It should be reserved for exceptional clinical circumstances when no effective therapeutic alternative exists. This review summarizes current mechanistic insights. It evaluates the available evidence for cross-reactivity and rechallenge across major drug classes, discusses the emerging role of pharmacogenomics, and proposes a practical framework to guide individualized clinical decision-making. Permanent discontinuation of the culprit drug remains the standard approach. Advances in immunology, pharmacogenomics, and international pharmacovigilance registries are progressively enabling more personalized risk assessment.