Washed microbiota transplantation improves haemoglobin levels in anaemia of chronic disease

Anaemia of chronic disease (ACD) is the second most common type of anaemia and lacks an effective treatment. Patients with anaemia are reported to have altered gut microbial profiles, which may affect erythropoiesis. Here, we investigated the gut microbial features of patients with ACD and determined whether regulating gut microbiota using washed microbiota transplantation (WMT) was effective in treating ACD.


| BACKGROUND
Anaemia is an important global health issue, with an agestandardised prevalence of 23%, which accounts for 8.8% of world years lived with disability. 1,23][4] Unfortunately, as eradication of the underlying causes of ACD is difficult and ACD does not respond to treatment with erythropoietin, effective therapeutic strategies are lacking for patients with ACD. 4 Haematopoietic stem and progenitor cells have been detected in the human intestine indicating that the gut is involved in erythropoiesis. 5Emerging evidence has also emphasised that gut microbiota plays a role in haematopoiesis.7][8] Further, interventions, such as antibiotics and probiotics, which modulate the gut microbiota, influence iron absorption and the number of haematopoietic stem cells in the bone marrow. 9,10The gut microbiota also impacts the progression and treatment of cancer, chronic kidney disease and autoimmune diseases, [11][12][13] which are responsible for ACD.Thus, targeting gut microbiota may hold promise in the treatment of ACD.
Faecal microbiota transplantation (FMT), involving the administration of a faecal suspension from a healthy donor into a patient's gastrointestinal tract, is an effective treatment to restore the balance of gut microbiota. 14FMT has been used for the treatment of diseases related to gut microbiota dysbiosis in China, including Clostridioides difficile infection, irritable bowel syndrome, inflammatory bowel disease and adverse effects (AEs) of cancer treatment (e.g.radiation enteritis and immune checkpoint inhibitor-associated colitis) 15,16 ; however, its use in clinical practice is limited due to significant AEs, such as infections and deaths. 17Washed microbiota transplantation (WMT) is an improved methodology of FMT based on the automated purification and washing process, in which faecal microbiota are extracted using an automatic purification system, followed by washing the microbiota pellet three times to reduce FMT-related AEs. 18,19In this study, we aimed to evaluate the efficacy of WMT in patients with ACD.

| Study design and patients
This was a retrospective, real-world study conducted in accordance with the guidelines of the Declaration of Helsinki.
The study was approved by the Ethics Committees of the First Affiliated Hospital of Guangdong Pharmaceutical University (No. 2022-76).Written informed consent was obtained from all patients or their legal representative if they were unable to provide consent.
Patients who underwent WMT at our hospital from 1 December 2016 and 31 July 2022, were eligible for this study.The key inclusion criteria for patients with ACD group were (1) adults with ACD and (2) receipt of WMT more than once.The key inclusion criteria for the noanaemia group were (1) adults without anaemia, bleeding or any of the underlying causes of ACD and (2) receipt of WMT more than once.The key exclusion criteria for the ACD group were (1) patients with anaemia caused by iron deficiency, folate, or vitamin B12 deficiency, acute bleeding or unexplained cause; (2) patients receiving treatment for anaemia (such as iron supplements, folic acid, erythropoietin or blood transfusion) and (3) patients with an escalation of therapy targeting the chronic disease causing the anaemia during WMT.

| Sample collection
Sterile containers were provided to collect the stool samples.The inner part of the stools from healthy controls, patients before each WMT course and stool donors for WMT was collected within about 10 min after defecation.Samples were stored in faecal DNA preservation tubes (Invitek, Germany) at −80°C until further processing.

| Colonic transendoscopic enteral tubing procedure
The colonic transendoscopic enteral tubing (TET) technique was performed as previously described. 20Briefly, after the colonoscope reached the caecum, a TET tube was inserted into the caecum via the paraffin-lubricated colonoscope channel.Then, the colonoscope was carefully withdrawn, while keeping the TET tube in the caecum.Thereafter, the colonoscope was reinserted into the colon and the tube was fixed on the wall using one to four endoscopic clips.

| WMT procedure
Donors were screened and washed microbiota suspensions were prepared as previously described. 21Briefly, medical history was collected and blood tests and faecal examinations for pathogens were performed as part of donor screening.Prior to providing stool samples, donors were asked to consume a fibre-rich diet and avoid a high-fat diet for 1 week.Regular exercise was also encouraged for the donors.Fresh faecal samples from healthy donors were homogenised in normal saline and microfiltered using an automatic microbiota purification system (GenFMTer, FMT Medical).The microbiota pellet was then washed three times and resuspended in normal saline.Patients received 120 mL of freshly washed microbiota suspension for 3 consecutive days through a nasojejunal tube (middle gastrointestinal tract) or a TET tube (lower gastrointestinal tract).The mode of receiving the suspension was decided based on patients' disease conditions and choices.Washed microbiota suspensions from various healthy donors were randomly allocated to patients.Patients with leucopenia (white blood cells <4 × 10 9 /L), caused mainly by radiotherapy and chemotherapy, were closely monitored for symptoms, including fever, diarrhoea and abdominal pain, after WMT for the early identification of any WMT-related infection.

| Data collection
Data were collected from electronic medical records of enrolled patients before and after WMT including demographic data, body mass index, smoking status, alcohol consumption, medical history, medication use, indication for WMT, route of WMT delivery (middle or lower gastrointestinal tract), AEs of WMT, treatment of WMT-related AEs and haematological parameters, including red blood cell (RBC), haemoglobin (HGB), reticulocyte count, mean corpuscular volume, mean corpuscular HGB concentration and mean corpuscular HGB.

| Definitions
Alcoholism was defined as weekly alcohol consumption exceeding 140 g for females and 210 g for males.Anaemia was diagnosed when the HGB level was <120 g/L in women and < 130 g/L in men.Anaemia was diagnosed as ACD based on underlying alterations in iron homeostasis along with clinical evidence of inflammation (including cancer, infections, immune-mediated diseases, inflammatory diseases, chronic kidney diseases, congestive heart failure, chronic pulmonary, obesity, aging and critical illness). 4The primary endpoint was the HGB level restored to normal in patients with ACD.AEs of WMT (such as fever, diarrhoea, abdominal pain and fatigue) were evaluated based on the clinical judgement of physicians.The effect of WMT on haematological parameters was assessed as follows: △haematological parameter = haematological parameter after WMT minus haematological parameter at baseline.

| 16S rRNA gene microbiome sequencing analysis
A total of 74 stool samples, including 34 from healthy controls, 10 from patients with ACD before WMT, 9 from matched patients with ACD after WMT and 21 from WMT donors were used for microbiome sequencing.Microbiome 16S rRNA gene sequencing was conducted by Majorbio Bio-Pharm Technology Co. Ltd. (Shanghai, China), as previously described. 21In brief, bacterial genomic DNA was extracted from faecal samples using the E.Z.N.A.® soil DNA Kits (Omega Bio-Tek) according to the manufacturer's instructions.The concentration and quality of extracted DNA were determined using a NanoDrop 2000 spectrophotometer (ThermoFisher Scientific).The 16S V3-V4 region was amplified using polymerase chain reaction with primers 338F (5′-ACTCC TAC GGG AGG CAGCAG-3′) and 806R (5′-GGACT ACH VGG GTW TCTAAT-3′).Amplicons were visualised using 2% agarose gel electrophoresis.Amplicons were then sequenced on the Illumina MiSeq sequencing platform (PE300).Sequence data were submitted to the NCBI (National Center for Biotechnology Information) Sequence Read Archive under project accession, PRJNA 909475.
Raw reads were de-multiplexed and quality filtered using fastp (version 0.19.6) and merged using FLASH (version 1.2.7). 22,23De-replicated reads were clustered into operational taxonomic units (OTUs) at 97% identity using UPARSE 7.1. 24A representative sequence for each OTU was selected based on its most abundant read.Sample read data were rarefied to 20,000 prior to further analysis, which still yielded an average Good's coverage of 99.09%.The taxonomies of representative OTU sequences were analysed using the Ribosomal Database Project Classifier (version 2.2) with a 70% confidence threshold.Microbial functions were predicted using the FAPROTAX database.The Majorbio Cloud platform (https://cloud.majorbio.com) was used for bioinformatic analysis.

| Statistical analysis
Power and sample size calculations were performed using virtual power and sample size calculators (http:// power andsa mples ize.com/).Statistical analyses were performed using Prism 9 (GraphPad).Categorical variables are expressed as numbers and percentages, normally distributed continuous variables as mean and standard deviations and non-normally distributed continuous variables as median and interquartile range.One sample t-tests or Wilcoxon signed-rank tests were applied for null hypotheses.Student's t-test or Mann-Whitney U test was performed for comparing independent groups, as appropriate.Categorical variables were compared using chi-square or Fisher's exact tests, where appropriate.A p value of < .05 was considered statistically significant.

| Gut microbiota profiles of healthy controls and patients with ACD
To investigate possible differences in gut flora between patients with ACD and healthy controls, we conducted 16S rRNA gene sequencing.Ten patients with ACD (age, 69.81 ± 9.15 years, 3 males) and 34 healthy controls (age, 50.65 ± 13.11 years, 19 males) were included in the profile analysis.A comparison of the genus-level taxonomic profiles between the two groups is presented in Figure 1A.Alpha diversity analysis showed significantly lower Sobs, Chao and Ace index values in patients with ACD than in healthy controls, while there was no marked difference in the Shannon index (Figure 1B).Principal coordinate analysis (PCoA) based on Jaccard abundance distances revealed obvious differences in the gut microbiota composition between patients with ACD and healthy controls (Figure 1C).Relative to healthy controls (participants without anaemia), patients with ACD had obvious changes in genus-level relative abundances, including lower relative abundances of Faecalibacterium, Lachnoclostridium, Lachnospiraceae NK4A136 group and Butyricicoccus genera (Figure 1D).Notably, a correlation heatmap revealed that the relative abundances of two butyrate-producing genera (Lachnospiraceae NK4A136 group and Butyricicoccus) were positively correlated with the HGB level (Figure 1E), indicating the protective effects of these two genera in ACD.Further functional prediction analysis using FAPROTAX showed that the fumarate respiration, nitrite ammonification and nitrate reduction pathways were upregulated in patients with ACD (Figure 1F).These results suggest an alteration of gut microbiota composition and function in patients with ACD.Hence, we aimed to target gut microbiota as a treatment option for ACD.

| Effects of WMT on haematological parameters in patients with ACD
Based on the inclusion and exclusion criteria, 37 patients with ACD were eligible for the analysis (Figure 2); demographic and clinical data are summarised in Table 1.Functional bowel disorder was the most common indication for WMT, accounting for 64.86% (n = 24) of cases, followed by inflammatory bowel disease (n = 2), and AEs during cancer treatment (n = 2).Among patients with ACD, 37, 18 and 11 completed two, three and four WMT courses, respectively.Median times for the intervals between the first and second, second and third, and third and fourth rounds of WMT were 35.00 (32.50-40.50)days, 38.00 (36.00-63.00)days, and 97.50 (88.25-135.75)days, respectively (Figure 3A).
After the first, second and third WMT rounds, HGB levels were restored to normal in 27.02% (10 out of 37, p < .001),27.78% (5 out of 18, p < .001)and 36.37%(4 out of 11, p = .004) of patients with ACD, respectively.Further quantitative analysis showed that the HGB levels gradually increased after the first (△HGB: 2.92 ± 7.14, p = .018),second (△HGB: 3.28 ± 6.32, p = .042)and third (△HGB: 3.73 ± 7.21, p = .117)round of WMT in patients with ACD (Figure 3B), although RBC and reticulocyte levels were not significantly changed (Figure 3C,D).These findings indicate that WMT not only stopped the reduction of HGB levels but also increased it in patients with ACD.

| Clinical factors associated with the effects of WMT on patients with ACD
Next, we aimed to determine the clinical factors that influenced the efficacy of WMT in ACD.Among patients with ACD, 13 received WMT through the middle gastrointestinal tract route and 24 via the lower gastrointestinal tract route.The effects of WMT on haematological parameters did not differ significantly between patients who underwent WMT through different delivery routes (Figure 4A).
Among the 37 patients with ACD, 13 had anaemia due to older age (called 'anemia of the elderly') and 24 had anaemia caused by various types of chronic disease, including coronary heart disease, chronic kidney disease, cancer and immune-mediated diseases (Table 2).Relative to patients with anaemia of the elderly, those with anaemia caused by the chronic disease had greater improvements in HGB (△HGB: 4.75 ± 6.90 vs. −.46 ± 6.53, p = .032)and RBC (△HGB: .15[−.06 to .27] vs. .02[−.24 to .14],p = .047;Figure 4B) after WMT treatment.

| WMT-related AE prevalence in patients with ACD
As safety endpoints, we analysed WMT-related AEs in patients with ACD.A total of 105 WMT procedures were performed, and WMT-related AEs were observed during two WMT procedures, including one case of fever and one of fatigue, with an overall prevalence of 1.90%.Fatigue in one patient was self-limited, and resolved without additional therapy, and the fever in another patient was resolved after treatment with antipyretics.No serious AEs were observed.

| Gut microbiota profiles in patients with ACD before and after WMT
The gut microbiota profiles in patients with ACD before and after WMT were compared using 16S rRNA gene sequencing analysis to investigate the potential mechanism of the WMT effect in ACD.The relative genus-level abundances of gut microbiota in healthy donors and patients with ACD before and after WMT are shown in Figure 5A.Alpha diversity analyses based on Sobs, Chao, Ace and Shannon indices showed enhanced gut microbiota richness and diversity in patients with ACD after WMT, although the differences were not statistically significant (Figure 5B).The analysis of beta diversity using PCoA showed altered gut microbiota composition in patients with ACD after WMT (Figure 5C).Additionally, several harmful genera, including Escherichia-Shigella and Klebsiella, were significantly decreased after WMT (Figure 5D).Further, butyrate-producing bacteria, Lachnospiraceae NK4A136 group and Butyricicoccus, which were significantly decreased in patients with ACD and positively correlated with HGB levels (Figure 1D,E), were markedly recovered after WMT (Figure 5D).Further functional analysis suggested that the fumarate T A B L E 1 Demographic and clinical characteristics of patients at baseline.and nitrite ammonification pathways, which were upregulated in patients with ACD (Figure 1F), were significantly downregulated after WMT (Figure 5E).

| Effects of WMT on haematological parameters in patients without anaemia
We also investigated whether WMT had harmful effects on patients without anaemia.According to the inclusion criteria, 53 patients without anaemia, bleeding or any underlying causes of ACD were enrolled in the noanaemia group (Figure 2).No obvious impact of WMT on haematological parameters was detected in these patients (Figure 6).

| DISCUSSION
The gut microbial composition and function in patients with ACD were found to be altered based on our analysis.
Targeting gut microbiota through WMT not only halted but also reversed the decline of HGB levels in patients with ACD.This effect was particularly seen in those patients whose ACD was not caused by anaemia of the elderly.Mechanistically, WMT modified the gut microbiota composition; restored Lachnospiraceae NK4A136 group and Butyricicoccus abundances, which were decreased in ACD and positively correlated with HGB level and downregulated gut microbial functions that were upregulated in ACD.Furthermore, WMT was safe and did not have an adverse impact on the HGB levels in patients without anaemia.To the best of our knowledge, this is the first clinical study to assess the therapeutic role of WMT in ACD.
The association between anaemia and gut microbiota has been widely reported.A study involving 10 pregnant women with iron deficiency anaemia and 10 healthy pregnant women showed that the microbial composition in patients with anaemia differed significantly from that in healthy controls. 7A similar finding was also observed in children with anaemia 8 ; however, the relationship between ACD and gut microbiota is poorly In the present study, decreased microbial richness and markedly altered gut microbial composition and function were noted in patients with ACD.The abundance of the butyrate-producing bacteria, Butyricicoccus, was reported to be lower in women with iron deficiency anaemia than in healthy women. 25A similar observation was noted in our study wherein patients with ACD had reduced the abundance of the microbiota from the Lachnospiraceae NK4A136 group, and Butyricicoccus genera.These results indicate that the gut microbiota may contribute to ACD pathogenesis and could, therefore, serve as a therapeutic target.Abbreviation: ACD, anaemia of chronic disease.

Number Percentage (%)
The role of gut microbiota in treating anaemia has been widely reported. 9,26Supplementation with the probiotic strain Lactobacillus plantarum 299v led to a clear increase in iron absorption in women. 27,28Besides, treatment using probiotics Bifidobacterium and Lactobacillus significantly increased serum iron concentrations in patients with early-stage chronic kidney disease (estimated glomerular filtration rate range, 60-45 mL/min/1.73m 2 ) 29 ; however, there is no convincing evidence to support the clinical benefits of probiotics in increasing HGB levels in patients with anaemia.An anaemia mouse model study revealed that transplantation of faecal microbiota from healthy controls partially improved the bone marrow suppression induced by antibiotic administration. 10In this study, we showed that WMT is effective in elevating HGB levels in patients with ACD, suggesting the clinical utility of WMT as a treatment for ACD.
We now describe the potential mechanisms by which WMT affects ACD.(1) Systematic inflammation driven by chronic diseases, aging and microbial compounds is a distinct feature of ACD.1][32][33] FMT can significantly reduce the serum levels of multiple cytokines, including IL-6 and interferon-γ, 34,35 which may contribute to improvements in bone marrow haematopoietic function.Furthermore, an animal study by Zeng et al.  showed that healthy gut microbiota from young mice transplanted into aged mice significantly increased the proportion and the absolute number of haematopoietic stem cells by decreasing the concentration of inflammatory cytokines secreted by bone marrow myeloid cells. 369][40] Thus, WMT may improve haematopoietic function via treating the primary diseases that contribute to ACD. (3) Hepcidin, a hepatic hormone that inhibits ferroportin iron-export activity and plays a central role in ACD pathophysiology, 32,33 is regulated by commensal gut bacteria. 41Therefore, WMT may also influence hepcidin expression and subsequently restore iron homeostasis.
Upon subgroup analysis, we found that patients with anaemia of the elderly achieved substantially less benefit from WMT than those with anaemia caused by chronic disease, possibly due to the decline in the activity of haematopoietic cells in the bone marrow of older adults. 42,43lthough several studies have suggested that patients receiving WMT through the lower gastrointestinal route experienced better therapeutic effects, 21,44 no significant differences in the effects of WMT were observed on haematological parameters between patients with ACD receiving WMT through different delivery routes; nevertheless, these results may be due to the relatively small sample size in each subgroup.
Safety is another important concern in WMT.The prevalence of WMT-related AEs (1.90%) was much lower in our study compared to the FMT-related AEs (19%) reported previously. 17Additionally, the AEs in our study were mild and relieved after symptomatic or no treatment.Furthermore, no adverse impacts of WMT on haematological parameters were detected in patients without anaemia.These findings indicate that WMT is a safe treatment modality for ACD.
WMT has already been widely used to modify the gut microbiota in patients with various diseases. 21,45In this study, we found clear alterations in gut microbial composition and a trend toward increasing richness and diversity of the gut microbiota in patients with ACD after WMT.Further, the relative abundances of two butyrateproducing bacteria, Lachnospiraceae NK4A136 group and Butyricicoccus, which were markedly decreased in patients with ACD and positively correlated with HGB concentration were recovered after WMT.7][48] The study by Zeng et al. also proved the importance of Lachnospiraceae and Butyricicoccus in mouse models. 36These results indicate that Lachnospiraceae NK4A136 group and Butyricicoccus are important in WMT for ACD treatment, and could serve as novel therapeutic targets.Besides, the gut microbiota pathways of fumarate respiration, nitrite ammonification and nitrate reduction, were upregulated in patients with ACD, whereas the fumarate respiration and nitrite ammonification pathways were downregulated after WMT, further supporting the idea that these microbial pathways contribute to ACD.However, the mechanism of how the two pathways affect ACD requires further investigation.
The limitations of this research should be highlighted.First, the demographic characteristics of patients with ACD and healthy controls differed, which may explain the differences in gut microbiota profiles.Second, some potential confounders, especially the diet, which affected the nutritional status and erythropoiesis of patients, were not taken into consideration.Third, considering the small sample size and the short follow-up duration, some subgroup analyses were difficult to perform, and the maintenance duration of the effect of WMT on ACD was not assessed sufficiently.Fourth, given that the faecal samples were collected and preserved using faecal DNA preservation tubes (which contained the protective solution to prevent the degradation of nucleic acid), live gut bacteria, particularly Lachnospiraceae NK4A136 group and Butyricicoccus could not be acquired for further studies.Therefore, clinical studies are needed that involve large sample sizes, a longer follow-up duration and functional experiments of the two butyrate-producing bacteria isolated from faecal samples.

| CONCLUSIONS
Patients with ACD exhibit differences in gut microbial composition and function relative to healthy controls.WMT is an effective and safe treatment for ACD that acts by modifying gut microbial composition, restoring butyrateproducing bacteria and regulating gut microbial functions.

F I G U R E 1
Gut microbiota profiles of patients with ACD and healthy donors.(A) Bar plot showing relative abundances of gut microbiota at the genus level.(B) Box plots of gut microbiota alpha diversity indices (Sobs, Chao, Ace and Shannon).(C) PCoA of bacterial communities at the genus level.(D) Bar plot of the relative abundances of the top 15 differential genera.(E) Heatmap showing correlations of genus-level abundances and haematological parameters.(F) Functional prediction analysis plot generated using FAPROTAX.Data in box plots are presented as median and interquartile range.*p < .05;**p < .01;***p < .001.ACD, anaemia of chronic disease; PCoA, principal coordinate analysis.

F I G U R E 3
Effects of WMT on haematological parameters in patients with ACD.(A) Study design.(B-D) Effects of WMT on HGB, RBC and Ret in patients with ACD.△haematological parameter = haematological parameter after WMT minus haematological parameter at baseline.Dot plots show the mean ± standard deviation.*p < .05.ACD, anaemia of chronic disease; HGB, haemoglobin; RBC, red blood cell count; Ret, reticulocyte count; WMT, washed microbiota transplantation.

F I G U R E 4
Clinical factors associated with effects of WMT on patients with ACD.(A) Effects of WMT on haematological parameters in patients with ACD who underwent WMT through the middle or lower gastrointestinal tract route.(B) Effects of WMT on haematological parameters in patients with ACD caused by old age or chronic disease.△haematological parameter = haematological parameter after WMT minus haematological parameter at baseline.Dot plots show the mean ± standard deviation.*p < .05.ACD, anaemia of chronic disease; WMT, washed microbiota transplantation.T A B L E 2 Main comorbidity related to anaemia in patients with ACD.

F I G U R E 5
Gut microbiota profiles in patients with ACD before and after WMT.(A) Bar plot showing the relative abundance of gut microbiota at the genus level.(B) Box plots of gut microbiota alpha diversity indices (Sobs, Chao, Ace and Shannon).(C) PCoA of bacterial communities at the genus level.(D) Bar plot of relative abundances of the top 15 differential genera.(E) Functional prediction analysis plot generated using FAPROTAX.Data in box plots are presented as median and interquartile range.*p < .05;**p < .01.ACD, anaemia of chronic disease; PCoA, principal coordinate analysis; WMT, washed microbiota transplantation.

F I G U R E 6
Effects of WMT on haematological parameters in patients without anaemia.(A-C) Effects of WMT on HGB, RBC and Ret in patients without anaemia of chronic disease.△haematological parameter = haematological parameter after WMT minus haematological parameter at baseline.HGB, haemoglobin; RBC, red blood cell count; Ret, reticulocyte count; WMT, washed microbiota transplantation.Dot plots show the mean ± standard deviation.
of the study population.ACD, anaemia of chronic disease; WMT, washed microbiota transplantation.