Compilation of longitudinal microbiota data and hospitalome from hematopoietic cell transplantation patients

Abstract

The impact of the gut microbiota in human health is affected by several factors including its composition, drug administrations, therapeutic interventions and underlying diseases. Unfortunately, many human microbiota datasets available publicly were collected to study the impact of single variables, and typically consist of outpatients in cross-sectional studies, have small sample numbers and/or lack metadata to account for confounders. These limitations can complicate reusing the data for questions outside their original focus. Here, we provide comprehensive longitudinal patient dataset that overcomes those limitations: a collection of fecal microbiota compositions (>10,000 microbiota samples from >1,000 patients) and a rich description of the "hospitalome" experienced by the hosts, i.e., their drug exposures and other metadata from patients with cancer, hospitalized to receive allogeneic hematopoietic cell transplantation (allo-HCT) at a large cancer center in the United States. We present five examples of how to apply these data to address clinical and scientific questions on host-associated microbial communities.

Conflict of interest statement

M.R.M.v.d.B. and J.U.P. received financial support from Seres Therapeutics. M.-A.P/ has received honoraria from AbbVie, Bellicum, Bristol-Myers Squibb, Incyte, Merck, Novartis, Nektar Therapeutics, and Takeda; has received research support for clinical trials from Incyte, Kite (Gilead) and Miltenyi Biotec; and serves on data and safety monitoring boards for Servier and Medigene and scientific advisory boards for MolMed and NexImmune.

Figures

Fig. 1

Timeline of clinical events and microbiota composition for a representative patient (PatientID = 1511) receiving hematopoietic cell transplantation at MSK. Day 0 is the day the patient received the infusion of hematopoietic cells (day of HCT). Negative days represent pre-transplant days and positive days represent post-transplant days. The data shown here are for the period from day −5 to day +21.

Fig. 2

Projection of all microbiota samples from MSK patients receiving HCT onto a two-dimensional space using t-SNE (t-distributed stochastic neighbor embedding). White lines with arrows: microbiota compositional trajectory of the PatientID = 1511 also shown in Fig. 1.

Fig. 3

Relative frequency of antibiotics administered to HCT patients at MSK by oral (a) and intravenous (b) routes. Antibiotics are grouped by their categories and displayed in the same color within each group.

Fig. 4

Effects of orally (a) and intravenously (b) administered antibiotics on microbes. Each ASV was labeled by its lowest taxonomy level that is not unclassified.

Fig. 5

A compilation of cases of positive blood culture infections for the bacteria analyzed in previous publications,. Here the period ranging from day −15 to day + 35 around the day of HCT (day 0) is shown, and we highlight the cases of Enterococcus (in green) and Escherichia (in red) analyzed below in Fig. 6.

Fig. 6

(a) The average abundance of Enterococcus is higher in patients who got Enterococcal bloodstream infections (n = 79) than in patients who did not (n = 940), especially in the critical period of two weeks after the transplant (day 0, where ‘Day’ is relative to the nearest allo-HCT transplant). (b) The average abundance of bacteria of the genus Escherichia is higher in patients who got a bloodstream infection by that genus (n = 52) than in patients who did not (n = 967), especially in the critical period of two weeks after the transplant (day 0). (c) The hazard ratio calculated for the risk of bloodstream infection after the patient was detected with an intestinal domination. These analyses were previously done by defining intestinal domination at an abundance threshold of 30% domination,. The results shown here reveal that domination redefined at an abundance threshold as small as 1% still increases the risk of bloodstream infection by Enterococcus. (d) The presence in the stool is even a stronger predictor of bloodstream infection for the case Escherichia, for which even levels of 0.1% have a significant association.