Haematopoietic stem cell transplantation induces severe dysbiosis in intestinal microbiota of paediatric ALL patients

Gastrointestinal complications are common after allogeneic haematopoietic stem cell transplantation (allo-HSCT). Intensive pre-transplant conditioning and GvHD damage the intestinal epithelium. Also intensive therapy not including HSCT predisposes patients to intestinal injury.

The intestinal microbiota consists of ~1000 bacterial species.1 The critical role of intestinal microbiota in the success of allo-HSCT has been addressed in a few studies (reviewed in refs 2, 3) mainly focusing on adult patients. We report here an in-depth profiling of the intestinal microbiota in five paediatric ALL patients treated with allo-HSCT. In our longitudinal study, we also followed the microbiota of six ALL patients receiving conventional induction therapy without HSCT, and compared microbiota recovery in the two patient groups.

Paediatric ALL patients at the Helsinki and Tampere University Hospitals were enrolled into the study. All patients entering HSCT received bone marrow grafts, in four cases from HLA-identical sibling donors (Table 1). The sibling donors were also enrolled as controls for the microbiota of healthy children and adolescents. The study was approved by the Ethical Committee of the Helsinki University Hospital and the Review Board of Tampere University Hospital. A written informed consent was obtained from the participants’ parents and from the participants 12 years of age or older.

Table 1 Characteristics of the enrolled patients and donors and fecal sample collection times.

We collected fecal samples from the HSCT patients once during or before conditioning and six times post HSCT (Table 1). From the non-HSCT patients, the day-0 fecal sample was obtained at mid-consolidation (a 9-week phase with peroral mercaptopurine and three high-dose methotrexate infusions 3 weeks apart), and thereafter 3–6 samples were collected (Table 1). One fecal sample was collected from each sibling donor within 2–6 days before or after the bone marrow harvest. All samples were frozen at −80 °C within 24 h after defecation.

Total bacterial DNA was extracted from the samples as described.4 The bar-coded pyrosequencing method was applied to microbiota profiling. The V4–V6 region of the 16S rRNA gene was PCR-amplified using a universal bacterial primer pair (F515 5′-IndexTermTGYCAGCMGCCGCGGTA-3′; 1061 R 5′-IndexTermTCACGRCACGAGCTGACG-3′). The PCR products were purified, pooled in equal amounts, and sequenced using a Genome sequencer FLX+ (Roche) in the Eurofins MWG (Germany). The trimming of sequences was performed in Mothur v.1.31.2. The sequencing provided 1.86 million reads, of which the 540 611 (29%) high-quality sequences were aligned and classified into bacterial taxa as described.4

Microbiota diversity in the donors, determined as inverse Simpson diversity index (SDI), varied from 4.1 to 12.1 (mean SDI 8.2; Figure 1a). This is comparable to previous data on healthy children’s microbiota.1, 5 In patient samples the diversity was markedly lower, with the mean SDI 1.7–5.8 in the HSCT group and 3.5–4.5 in the non-HSCT group, calculated from samples within comparable timeframes (Figure 1a). No statistically significant differences in SDI were noted between the patient groups within any timeframe. The mean SDI in the patient samples increased gradually, but remained lower than in the healthy controls up to the end of the study period that lasted ~6 months. The most striking difference to the healthy controls was seen in the HSCT group immediately (1–8 days) after transplantation (P<0.05, Kruskall–Wallis test with Dunn’s multiple comparisons).

Figure 1

Diversity and composition of intestinal microbiota in ALL patients treated with allo-HSCT or with conventional chemotherapy. Identification of bacteria in fecal samples collected at indicated time intervals was done by pyrosequencing of 16 S rRNA gene. Healthy control microbiota was determined from samples collected from the sibling donors of HSCT grafts (D1–D4). (a) Diversity of microbiota indicated as boxplots with mean inverse Simpson diversity index (SDI) calculated from samples of HSCT-treated patients (middle panel) and patients treated with conventional ALL chemotherapy (right panel). For comparison of patient groups, the diversity indices were calculated from pre-HSCT sample (Allo-HSCT group) or 0-sample at consolidation (No HSCT group), and from samples within the indicated timeframes post-HSCT or after 0-sample. In cases where two samples from one patient were within the timeframe, the mean SDI of these samples was used in the analysis. The mean SDI of microbiota of HSCT graft donors is also indicated (left panel). (b,c) Relative proportions of bacterial phyla (b) and genera (c) in individual donor samples (left panels) and patient samples are indicated. Upper right panels: samples from the HSCT-treated patients, with numbers indicating the sampling time before and after HSCT in days. Lower right panels: samples from the non-HSCT-treated patients, with numbers indicating the start of sample collection at consolidation phase (0) and later sampling times in days after this. DNA isolation from the day 12 post-HSCT sample of patient #2 and from the day 0-sample of patient #8 was not successful, and two samples found norovirus positive (patients #7 and #8, one in each sample series) were discarded.

The dominant microbial phyla in the donors were Firmicutes (44–84% of microbiota), Actinobacteria (4–47%) and Bacteroidetes (3–18%), while the proportion of Proteobacteria was low (0.1–0.3%) (Figure 1b). In patient samples, the relative proportion of Proteobacteria was markedly higher, up to 6% in some samples from three patients in the non-HSCT group, and even 60% or over 90% in two samples from patient #3 in the HSCT group (Figure 1b).

The most dominant genera in the donors (representing over 10% of the microbiota in at least one donor) were Bifidobacterium, Bacteroides, Blautia and Faecalibacterium (Figure 1c). Altogether, 14–17 bacterial genera were observed within each healthy subject, with the most abundant genus comprising up to 40% of the microbiota. Enterococcus, Staphylococcus, Enterobacter, Bacteroides and unclassified genera of Lachnospriraceae were rare in donor samples, while they constituted over 90% of microbiota in the HSCT patient samples taken less than three weeks post-HSCT (Figure 1c). Conversely, the genus Feacalibacterium, which harbors the beneficial species F. prausnitzii, was undetectable in many patient samples. Most of the later post-HSCT samples began to resemble those of healthy children, with an increased relative abundance of members of normal intestinal microbiota, such as Blautia, Bifidobacterium and Roseburia (Figure 1c).

Our results support previous studies that correlate immunosuppression, severe GvHD, and other HSCT complications with a decreased diversity of the intestinal microbiota and increased colonization with the genus Enterococcus and the phylum Proteobacteria.6, 7, 8, 9, 10 Abundance of these microbial groups increases the risk for severe bloodstream infections,6 emphasizing the importance of their detection. The early post-HSCT samples of patient #3, who had a severe, steroid-refractory grade IV GvHD, were dominated by Proteobacteria (Enterobacter and other, unclassified Enterobacteriacae) (Figures 1b and c). The genus Enterococcus, in turn, dominated in at least one sample of all other HSCT patients (Figure 1c). In the non-HSCT group, Enterococcus constituted nearly 70% of the microbiota in the day-1 sample of patient #6, and sporadical dominance of given microbial genera was noted also in some other samples, but less often than in the HSCT group (Figure 1c).

In the patients #4 and #5, who had no signs of GvHD, the microbiota stabilized more rapidly than in the other HSCT patients (Figure 1c). It was also of note that the pre-HSCT samples of patients #4 and #5 were rich in Bacteroidetes (Figure 1b). A marked relative abundance of Bacteroidetes in the pre-HSCT samples of non-GvHD patients was also reported by Biagi et al.8 It will be interesting to see whether pre-HSCT microbial profiling could be used as a biological marker in the assessment of HSCT complication risks.

Due to the limited number of patients in our cohort, the age or gender distribution could not be matched, and the mean age was higher in the HSCT group (Table 1). Thus, our comparison of the microbial diversities between the groups has to be considered descriptive. However, as the diversity of the microbiota in general increases with the growth of the child,5 the noted trend towards a lower diversity in the immediate post-HSCT period is likely to reflect the toxic impact of the initial phase of HSCT on the intestine.

The main focus in our study was on the long-term characterization of individual variation in the microbiota composition and recovery after HSCT, with the follow-up time reaching 6 months (173–242 days). The only previous follow-up study completed with paediatric patients so far showed a temporal disruption of gut microbiota after HSCT in 10 patients and an emerging recovery towards the pre-HSCT microbial diversity during a follow-up period of 51–103 days.8 Our results are in line with this, and the extended follow-up of our study provides evidence that in some patients, the recovery of diversity can take 5 months or even longer.

During HSCT, the intestinal ecosystem is predisposed to dysbiosis by various factors, including the post-HSCT phase of parenteral nutrition, and the often extensive antimicrobial medication.11, 12, 13 Also chemotherapy as such renders the intestinal epithelium vulnerable to damage, and chemotherapy-induced neutropenia increases susceptibility to infections. All patients in our cohort were on trimethoprim/sulphamethoxazole prophylaxis and received broad-spectrum antibiotics for the treatment of bacterial infections. The HSCT patients received antifungal prophylaxis and treatment for suspected fungal infections during cytopenia. Four non-HSCT patients also received antifungal medication. The antibiotic exposures of each patient are detailed in Supplementary Figure S1. As expected, the reduction in SDI often coincided with the administration of antimicrobials, but this was not the sole reason for the fluctuations in microbial diversity. A recent study described a reduced microbial diversity already at the diagnosis of paediatric ALL, with even more reduction among antibiotic-treated patients.14 Thus, a complex set of factors ranging temporally from the pre-diagnosis period to maintenance seems to affect the microbiota of ALL patients.

In conclusion, our long-term follow-up study shows that the intestinal microbiota balance is markedly disturbed by allo-HSCT in the paediatric setting. Considerable reduction in microbial diversity and temporal variation in the phyla and genera were typical in fecal samples from HSCT patients. Reduced diversity was noted, to some extent, also among the non-HSCT patients, indicating that intensive ALL therapy in general disturbs the microbiota. However, HSCT seems to induce the most significant and enduring dysbiosis, warranting monitoring of intestinal microbiota during patient follow-up.


  1. 1

    Lozupone CA, Stombaugh JI, Gordon JI, Jansson JK, Knight R . Diversity, stability and resilience of the human gut microbiota. Nature 2012; 489: 220–230.

  2. 2

    Chen Y, Zhao Y, Cheng Q, Wu D, Liu H . The role of intestinal microbiota in acute graft-versus-host disease. J Immunol Res 2015; 2015: 145859.

  3. 3

    Zama D, Biagi E, Masetti R, Gasperini P, Prete A, Candela M et al. Gut microbiota and hematopoietic stem cell transplantation: where do we stand? Bone Marrow Transplant 2017; 52: 7–14.

  4. 4

    Wacklin P, Tuimala J, Nikkila J, Sebastian T, Makivuokko H, Alakulppi N et al. Faecal microbiota composition in adults is associated with the FUT2 gene determining the secretor status. PLoS ONE 2014; 9: e94863.

  5. 5

    Nylund L, Satokari R, Salminen S, de Vos WM . Intestinal microbiota during early life-impact on health and disease. Proc Nutr Soc 2014; 73: 457–469.

  6. 6

    Taur Y, Xavier JB, Lipuma L, Ubeda C, Goldberg J, Gobourne A et al. Intestinal domination and the risk of bacteremia in patients undergoing allogeneic hematopoietic stem cell transplantation. Clin Infect Dis 2012; 55: 905–914.

  7. 7

    Jenq RR, Ubeda C, Taur Y, Menezes CC, Khanin R, Dudakov JA et al. Regulation of intestinal inflammation by microbiota following allogeneic bone marrow transplantation. J Exp Med 2012; 209: 903–911.

  8. 8

    Biagi E, Zama D, Nastasi C, Consolandi C, Fiori J, Rampelli S et al. Gut microbiota trajectory in pediatric patients undergoing hematopoietic SCT. Bone Marrow Transplant 2015; 50: 992–998.

  9. 9

    Holler E, Butzhammer P, Schmid K, Hundsrucker C, Koestler J, Peter K et al. Metagenomic analysis of the stool microbiome in patients receiving allogeneic stem cell transplantation: loss of diversity is associated with use of systemic antibiotics and more pronounced in gastrointestinal graft-versus-host disease. Biol Blood Marrow Transplant 2014; 20: 640–645.

  10. 10

    Taur Y, Pamer EG . The intestinal microbiota and susceptibility to infection in immunocompromised patients. Curr Opin Infect Dis 2013; 26: 332–337.

  11. 11

    Ubeda C, Pamer EG . Antibiotics, microbiota, and immune defense. Trends Immunol 2012; 33: 459–466.

  12. 12

    Barker CC, Anderson RA, Sauve RS, Butzner JD . GI complications in pediatric patients post-BMT. Bone Marrow Transplant 2005; 36: 51–58.

  13. 13

    Styczynski J, Gil L . EBMT Paediatric Diseases Working Party. Prevention of infectious complications in pediatric HSCT. Bone Marrow Transplant 2008; 42 (Suppl 2): S77–S81.

  14. 14

    Rajagopala SV, Yooseph S, Harkins DM, Moncera KJ, Zabokrtsky KB, Torralba MG et al. Gastrointestinal microbial populations can distinguish pediatric and adolescent Acute Lymphoblastic Leukemia (ALL) at the time of disease diagnosis. BMC Genom 2016; 17: 635.

Download references


Patients and sibling donors are thanked for participationin the study. Pia Valle, Satu Ranta, Annamari Aitolahti, Sisko Lehmonen and Paula Salmelainen are thanked for assistance. The study was supported by the SalWe Research Program for IMO (Tekes grant 648/10), and by the EVO Medical Research Fund of the Finnish Red Cross Blood Service.

Author information

Correspondence to J Mättö.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Additional information

Supplementary Information accompanies this paper on Bone Marrow Transplantation website

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Lähteenmäki, K., Wacklin, P., Taskinen, M. et al. Haematopoietic stem cell transplantation induces severe dysbiosis in intestinal microbiota of paediatric ALL patients. Bone Marrow Transplant 52, 1479–1482 (2017). https://doi.org/10.1038/bmt.2017.168

Download citation

Further reading