SGLT2 Inhibitors Tied to Hemoglobin Gains in MDS, Myelofibrosis Anemia

Sodium-glucose cotransporter 2 (SGLT2) inhibitors were associated with hemoglobin increases of about 1.0 to 1.6 g/dL in small observational studies of patients with myelodysplastic syndromes (MDS) or myeloproliferative neoplasms (MPN), but the evidence is too thin to support their use for anemia, according to a narrative review published in Anemia.¹

The study authors found no randomized controlled trials in these disorders and also explained that context matters. A rise in hemoglobin may be a therapeutic response in patients with anemia, but in MPN subtypes with normal or high baseline counts, it may instead signal erythrocytosis.

Anemia Burden and Treatment Gaps in MDS and Myelofibrosis

Anemia is the most frequent cytopenia in MDS and affects about one-third to one-half of patients with myelofibrosis at diagnosis, the authors noted. Erythropoiesis-stimulating agents and luspatercept (Reblozyl; Bristol Myers Squibb) help some patients, but responses are often modest or short-lived, and Janus kinase inhibitors used in myelofibrosis can worsen anemia early in treatment, the authors noted.

Empagliflozin (Jardiance; Boehringer Ingelheim/Eli Lilly), dapagliflozin (Farxiga; AstraZeneca), and canagliflozin (Invokana; Janssen Pharmaceuticals) are widely used in diabetes, heart failure, and kidney disease, where trials have recorded higher hemoglobin and hematocrit.¹ A JAMA meta-analysis of 85 randomized trials with 70,543 participants, published in October 2026, found hemoglobin was 0.60 g/dL higher with SGLT2 inhibitors than with control (95% CI, 0.55-0.66), although the investigators called the clinical significance unclear.² A 2025 meta-analysis showing similar gains in chronic kidney disease.³

Mechanistic studies attribute these gains to increased erythropoietin, better kidney oxygenation, and lower hepcidin rather than to fluid loss alone, the authors explained. Searching the literature through June 2026, the review authors found reports covering about 58 exposed patients with MDS or MPN.¹

Hemoglobin Response to SGLT2 Inhibition in MDS and Myelofibrosis

The most relevant MDS data came from a retrospective study of 11 patients with anemia who had MDS or overlap MDS/MPN. Among the 9 who did not need regular transfusions, median hemoglobin climbed 1.2 g/dL from baseline to peak, and 6 of 9 (67%) gained at least 1 g/dL. Neither of the 2 transfusion-dependent patients responded, and 4 patients were on stable MDS- or anemia-directed therapy, which complicates interpretation.

In myelofibrosis, hemoglobin rose in 14 of 16 patients (88%) in a cohort with a median age of 74 years and a median baseline hemoglobin of 11 g/dL. The median gain was 1.0 g/dL (range, 0-3.0), and 5 of the 6 patients (83%) who started below 10 g/dL improved by at least 1 g/dL.

Erythrocytosis Risk in Polycythemia Vera and Essential Thrombocythemia

In MPN subtypes characterized by high blood counts, the picture differed. Hemoglobin rose by about 1.5 g/dL among 11 patients with essential thrombocythemia, most without anemia, and in one case canagliflozin preceded a jump in hemoglobin from about 14.5 to 16.9 g/dL that led to a polycythemia vera diagnosis.

“This distinction is important when evaluating SGLT2 inhibitors across MPN subtypes, because an erythropoietic effect may be desirable in myelofibrosis-associated anemia but potentially problematic in polycythemia vera or essential thrombocythemia,” the authors wrote.

Class risks include genital and urinary tract infections, volume depletion, and rare euglycemic diabetic ketoacidosis. The authors proposed a hypothesis-generating algorithm that favors patients with anemia and residual marrow capacity and calls for checking hemoglobin and hematocrit every 2 to 4 weeks after initiation. They wrote that SGLT2 inhibitors “should not currently be considered standard therapy for anemia in MDS or MPN but rather as a potential adjunct in carefully selected patients.”

The evidence comes mostly from small, uncontrolled, heterogeneous reports, and publication bias favoring positive responses could not be excluded, the authors acknowledged.

“Although hemoglobin increases have been reported in some anemic patients, these findings are hypothesis-generating and cannot currently prove therapeutic efficacy or causality,” they wrote.

They called for phase 2 trials that identify predictors of response and track outcomes such as transfusion independence.

References

1. Mohamed SF, Abdulgayoom M, Al-Mashdali AF, Rabadi M, Abdelmahmuod EA, Yassin M. SGLT2 inhibitors and hemoglobin improvement in myelodysplastic syndromes and myeloproliferative neoplasms: mechanistic rationale and emerging clinical evidence. Anemia. 2026;2026(1):4918154. doi:10.1155/anem/4918154

2. Yang Y, Yang C, Wang Y, et al. Sodium-glucose cotransporter 2 inhibitor use and levels of hemoglobin, markers of iron metabolism, and erythropoiesis. JAMA. Published online October 24, 2026. doi:10.1001/jama.2026.19921

3. Grossi G. Meta-analysis finds SGLT2 inhibitors boost hemoglobin, hematocrit in CKD. AJMC^®. April 8, 2025. Accessed November 11, 2026. https://www.ajmc.com/view/meta-analysis-finds-sglt2-inhibitors-boost-hemoglobin-hematocrit-in-ckd