The Gut Microbiota in Patients with Polycythemia Vera: Novel Vs. Conventional Treatments

Background: The human gut microbiota (the population of microorganisms present) is important for digestion of food but also for development of the host immune system and protection against pathogens. Changes in the gut microbiota are linked to several inflammatory diseases such as diabetes, atopic diseases and Alzheimer's disease.

Polycythemia vera (PV) is one of the Philadelphia chromosome negative classical myeloproliferative neoplasms (MPNs), which also include essential thrombocythemia (ET) and primary myelofibrosis (PMF). MPNs are increasingly recognized as inflammatory driven diseases.

The role of the gut microbiota in patients with MPNs is largely unknown. In a small study (n=25) the microbiota of MPN patients had higher levels of Prevotellaceae compared to healthy controls, and differed significantly in composition between patients treated with hydroxyurea and ruxolitinib. Since MPNs are likely to be driven by chronic inflammation and the gut microbiota influences the immune system, investigations of the PV-microbiota are highly relevant.

We compared the microbiota in a cross-sectional study of patients with PV stratified into five different treatment groups.

Method and Material: Patients above 18 years diagnosed with PV, according to the 2016 World Health Organization (WHO) classification, were invited to participate in the study. The exclusion criteria were: pregnancy, use of antibiotics within the last 2 months, change in treatment within the last 3 months or inability to understand the oral and written information.

Clinical and biochemical data for each patient were collected retrospectively and included co-morbidities, smoking status, anti-inflammatory treatment, hypertension, haematological parameters, haematological treatment, body mass index (BMI), among others.

Stool samples, no more than 6 hours old were stored at -80°C. DNA was extracted by using the EMAG® Nucleic acid extraction system, [bioMérieux] according to the manufacturer's instructions. The bacterial microbiota was characterized by amplicon-based next generation sequencing of the V3-V4 region of the 16S ribosomal unit using a MiSeq instrument, Illumina. BION was used to assign taxonomic classification.

The patients were divided into 5 groups according to treatment: no treatment (n=18), hydroxyurea (n=33), PEG-interferon-α2 (IFN) (n=23), IFN combined with ruxolitinib (COMBI) (n= 21) and patients treated with other combinations e.g. ruxolitinib, anagrelide, hydroxyurea combined with IFN hydroxyurea combined anagrelide, or hydroxyurea combined with ruxolitinib, (n=11).

The alpha diversity was measured using the Shannon diversity index, and compared with a pairwise Wilcoxon test adjusted for multiple testing. Beta diversity (difference between the samples) was visualised by a PCoA plot, and compared using an ANOSIM test. Differential abundance analysis was performed by Linear discriminant analysis Effect Size (LEfSe).

Results: In total, 116 patients with PV were included. Of these, 106 fulfilled the inclusion criteria (49 women and 57 men) and had a median age of 68 years (range: 31 - 85). The five treatment groups did not differ in alpha diversity. The bacterial composition differed slightly between IFN group and no treatment group (p=0.032, R=0.075), and between IFN and COMBI group (p= 0.031, R=0.054).

Patients from the no treatment group had a higher relative abundance of the Bacteroides genus (39%) compared to the IFN group (14.7%) (p=0.003) and hydroxyurea group (23.1%) (p=0.047), but not the COMBI group (30.1%). A lower abundance of the Bacteroides genus was found in the IFN group compared to the COMBI group (p=0.004) and compared to the hydroxyurea group (p=0.04). Due to the small number of patients treated with other combinations, these were not included in the differential abundance analysis.

Conclusions: Among the five treatment groups in patients with PV, the alpha diversity of the microbiota were similar, but the relative abundance of the Bacteroides genus in patients not receiving any treatment compared to patients treated with IFN and hydroxyurea was higher. A lower abundance of Bacteroides genus was seen in the IFN group compared to the COMBI group and compared to the hydroxyurea group.

Whether these differences are explained as a treatment response to IFN or clinical parameters, such as comorbidities, age JAK2 allele burden or BMI need further investigations.