Specific gut microbiome members are associated with distinct immune markers in pediatric allogeneic hematopoietic stem cell transplantation

Abstract

Background: Increasing evidence reveals the importance of the microbiome in health and disease and inseparable host-microbial dependencies. Host-microbe interactions are highly relevant in patients receiving allogeneic hematopoietic stem cell transplantation (HSCT), i.e., a replacement of the cellular components of the patients' immune system with that of a foreign donor. HSCT is employed as curative immunotherapy for a number of non-malignant and malignant hematologic conditions, including cancers such as acute lymphoblastic leukemia. The procedure can be accompanied by severe side effects such as infections, acute graft-versus-host disease (aGvHD), and death. Here, we performed a longitudinal analysis of immunological markers, immune reconstitution and gut microbiota composition in relation to clinical outcomes in children undergoing HSCT. Such an analysis could reveal biomarkers, e.g., at the time point prior to HSCT, that in the future could be used to predict which patients are of high risk in relation to side effects and clinical outcomes and guide treatment strategies accordingly.

Results: In two multivariate analyses (sparse partial least squares regression and canonical correspondence analysis), we identified three consistent clusters: (1) high concentrations of the antimicrobial peptide human beta-defensin 2 (hBD2) prior to the transplantation in patients with high abundances of Lactobacillaceae, who later developed moderate or severe aGvHD and exhibited high mortality. (2) Rapid reconstitution of NK and B cells in patients with high abundances of obligate anaerobes such as Ruminococcaceae, who developed no or mild aGvHD and exhibited low mortality. (3) High inflammation, indicated by high levels of C-reactive protein, in patients with high abundances of facultative anaerobic bacteria such as Enterobacteriaceae. Furthermore, we observed that antibiotic treatment influenced the bacterial community state.

Conclusions: We identify multivariate associations between specific microbial taxa, host immune markers, immune cell reconstitution, and clinical outcomes in relation to HSCT. Our findings encourage further investigations into establishing longitudinal surveillance of the intestinal microbiome and relevant immune markers, such as hBD2, in HSCT patients. Profiling of the microbiome may prove useful as a prognostic tool that could help identify patients at risk of poor immune reconstitution and adverse outcomes, such as aGvHD and death, upon HSCT, providing actionable information in guiding precision medicine.

Keywords: 16S rRNA gene profiling; B cells and NK cells; Data integration; Gut microbiota; HSCT; Human beta-defensin 2; Immune reconstitution; Pediatric cancer; Ruminococcaceae; acute GvHD.

Fig. 1

Monitoring of the host immune system and intestinal microbiome in hematopoietic stem cell transplantation (HSCT). a Study outline. A cohort of 37 pediatric recipients of allogeneic HSCT was monitored prior to HSCT, at the time point of HSCT, and post-HSCT (median follow-up time 5.2 years). A range of patient characteristics as well as disease outcomes, immune markers, immune cell counts, and intestinal patterns of microbial community composition were recorded at the noted time points (up to 12 months post-HSCT). See Table S1 for details regarding the patient characteristics. b Plasma hBD2 concentrations over time and in comparison to healthy controls (young adults). The y-axis was log10-transformed for better visualization. Zeros were replaced with 1 to avoid undefined values on the log-transformed axis. Asterisks indicate whether the concentrations at each respective time point were significantly different from any of the other time points (showing the maximum significance level). c Bacterial alpha-diversity, measured by inverse Simpson index, of the intestinal microbiome shown with log10-transformed y-axis. d Relative abundances over time of the 8 most abundant taxonomic families in the dataset (comprising 98 fecal samples). *P < 0.05; **P < 0.01; and ***P < 0.001

Fig. 2

Correlations between immune markers, immune cell counts, and outcomes in patients undergoing HSCT. a Pairwise Spearman’s correlation between immune markers and immune cell counts in HSCT patients (n = 37) that were determined to be significant in a permutational multivariate analysis of variance using (microbial) distance matrices (adonis) (See Table S2). Positive and negative correlations are represented by red or blue circles, respectively, and the size of circles and intensity of color refer to the strength of the correlation. Correlations that are significant (P ≤ 0.05) are indicated by a black outline of the circle. b Natural killer (NK) and total B cell (mature and immature) reconstitution in month + 2 with respect to the maximum acute GvHD (aGvHD) grade (0–I vs. II–IV) in HSCT patients (n = 37). hBD2_pre, hBD2_w0, hBD2_w1, hBD2_w2 plasma human beta-defensin 2 concentration pre-HSCT, on the day of HSCT, and in weeks + 1 and + 2, respectively; mono_pre, mono_w3 monocyte counts pre-HSCT and in week + 3, respectively; neutro_m3 neutrophil count in month + 3; CD8⁺_m1 CD8⁺ T cell counts in month + 1; Age recipient age at time point of transplantation; NK_m1, NK_m2 natural killer cell counts in months + 1 and + 2, respectively; B_m2, mat_B_m2, immat_B_m2 all, mature, and immature B cell counts in month + 2; CD4⁺_m2 CD4⁺ T cell counts in month + 2; Citr_pre, Citr_w1 plasma citrulline levels pre-HSCT and in week + 1, respectively; CRP, CRP_w1, CRP_w5, CRP_w6, CRP_m3, CRP_m6 C-reactive protein levels at time points simultaneous to microbiome characterization, in weeks + 1, + 5, and + 6, and in months + 3 and + 6, respectively. *P < 0.05; **P < 0.01, and ***P < 0.001

Fig. 3

Sparse partial least squares (sPLS) regression of immune parameters and intestinal bacterial taxa during HSCT. a Correlation circle plot for the first two sPLS dimensions with correlations displayed for > 0.2/< − 0.2. The two grey circles indicate correlation coefficient radii at 0.5 and 1.0. Bacterial operational taxonomic units (OTUs) are displayed as circles, and are colored according to the cluster they are affiliated with (cluster 1: blue; cluster 2: orange; cluster 3: grey). Variables projected closely to each other are positively correlated. Variables projected diametrically opposite from each other are negatively correlated. Variables situated perpendicularly to each other are not correlated. The variance explained by the OTUs is 4.12% on component 1, and 4.79% on component 2. The variance explained by the clinical variables is 16.1% on component 1, and 17.19% on component 2. b Clustered image map (CIM) of the first two sPLS dimensions, displaying pairwise correlations between OTUs (bottom) and clinical variables (left). Red and blue indicate positive and negative correlations, respectively. Hierarchical clustering (clustering method: complete linkage, distance method: Pearson’s correlation) was performed within the mixOmics cim() function based on the sPLS regression model. An overview of the OTU abundances in the individual samples is provided in Figure S3, and a list of the individual OTUs and their cluster-affiliation is provided in Table S3. c Loading plots of OTUs with maximum contributions on the first (left) and second (right) component, respectively. The bars are colored according to the cluster they are affiliated with. The family-affiliation for each respective OTU is indicated by color (for legend see b). Abbreviations of variables are the same as in Fig. 2

Fig. 4

Canonical correspondence analysis (CCpnA) of immune parameters and intestinal bacterial taxa in patients undergoing HSCT. Triplot showing dimension 1 and 2 of the CCpnA that included continuous clinical variables (arrows), categorical variables (+), and OTUs (circles). Samples are depicted as triangles. OTUs with a correlation of > 0.2/< − 0.2 in the sPLS analysis were included in the CCpnA model. Only the variables and OTUs with a score > 0.2/< − 0.2 in at least one CCpnA dimension are shown. The OTUs in the CCpnA plot are colored according to the cluster they were affiliated with in the sPLS-based hierarchical clustering analysis, and the ellipses present an 80% confidence interval, assuming normal distribution. For visualization purposes, this plot is a section focussing on the categorical and continuous variables contributing to the ordination. The full size version of the CCpnA triplot, including all samples and OTUs, is presented in Figure S2A. Abbreviations of variables are the same as in Fig. 2

Fig. 5

Bacterial community state types over time in patients undergoing allogeneic HSCT. Patients are grouped into four outcome groups: (1) patients who developed no or mild aGvHD (grade 0–I) and survived vs. (2) patients who did not survive; (3) patients who developed moderate to severe aGvHD (grade II–IV) and survived vs. (4) patients who did not survive. The day of commencement of aGVHD grade II–IV and the day of death post-HSCT are displayed to the right. Patients with moderate to severe aGvHD (groups 3 and 4) most often harbored the Lactobacillaceae-dominated community state type 1 (CST 1), especially at late time points. CST2, dominated by Ruminococcaceae and Lachnospiraceae, did not persist after HSCT in any of the patients in groups 3 and 4. A detailed overview of the CSTs is provided in Additional file 7: Figure S4, and information about individual OTUs and their cluster-affiliation is provided in Additional file 5: Table S3

Fig. 6

Longitudinal profiles of microbial community composition and immune markers in non-survivors with aGvHD. Abundances of Lactobacillaceae increased predominantly after aGvHD onset in patients who died during the follow-up period. Patient P24 developed chronic GvHD on day + 187, relapsed on day + 548, and died on day + 784 due to graft rejection after re-transplantation. Patient P26 died on day + 602 after a relapse on day + 442 followed by re-transplantation on day + 518. Patient P30 died on day + 192 after relapse on day + 77. Patient P26 and P30 had no reported bacterial infections during the depicted period. InvSimpson inverse Simpson diversity index, hBD2 human beta-defensin 2, aGvHD acute graft-versus-host-disease, inf bacterial infection