Gut microbiota differences in MPN

Changes in gut microbiota composition are associated with a wide range of inflammatory conditions. It is therefore likely that gut changes are involved in the development of myeloproliferative neoplasms (MPN).

Here, we summarize a study by Eickhardt-Dalbøge et al.¹ in European Journal of Hematology investigating the overall association of gut microbiota across MPN subtypes and somatic mutations.

Study design¹

• Overall, 227 patients with MPN aged ≥18 years were enrolled between November 2018 and August 2021; 42 healthy controls (HC) were also included
• Patients had a confirmed diagnosis of essential thrombocythemia (ET), polycythemia vera (PV), preprimary myelofibrosis (MF), MF or post PV/ET-MF, and were grouped according to sub-diagnoses and Janus kinase (JAK)2^V617F/calreticulin mutation status.
• Gut microbiota was analyzed using next-generation sequencing

Key findings¹

The median observed bacterial richness of gut microbiota was higher in patients with MPN vs HC (Figure 1).

Figure 1. Median observed significant differences in the richness of gut microbiota* 

CALR, calreticulin; ET, essential thrombocythemia; HC, healthy control; JAK, Janus kinase; MPN, myeloproliferative neoplasia.
*Adapted from Eickhardt-Dalbøge, et al.¹

• The alpha diversity (as measured by the median Inverse Simpson index) did not differ between patients with MPN and HC
  • Patients with MPN and JAK2^V617F positive had a significantly lower median Inverse Simpson index vs calreticulin positive patients and HC (21.8 vs 26.3, and 29, respectively, p < 0.05)
• The bacterial composition in patients with ET resembled that of HC the most, compared with those from patients with PV, preprimary MF, MF, or post PV/ET-MF
• Patients with MPN had a significantly lower abundance of Firmicutes vs HC (52% vs 59%, p = 0.01) but higher Actinobacteriota (4% vs 2%, p < 0.001)
• Patients with a high JAK2^V617F allele burden had a higher relative abundance of Akkermansia vs patients with a lower allele burden (5% vs 1%, p < 0.01)