Tumor Microbiome Diversity and Composition Influence Pancreatic Cancer Outcomes.

Most patients diagnosed with resected pancreatic adenocarcinoma (PDAC) survive less than 5 years, but a minor subset survives longer. Here, we dissect the role of the tumor microbiota and the immune system in influencing long-term survival. Using 16S rRNA gene sequencing, we analyzed the tumor microbiome composition in PDAC patients with short-term survival (STS) and long-term survival (LTS). We found higher alpha-diversity in the tumor microbiome of LTS patients and identified an intra-tumoral microbiome signature (Pseudoxanthomonas-Streptomyces-Saccharopolyspora-Bacillus clausii) highly predictive of long-term survivorship in both discovery and validation cohorts. Through human-into-mice fecal microbiota transplantation (FMT) experiments from STS, LTS, or control donors, we were able to differentially modulate the tumor microbiome and affect tumor growth as well as tumor immune infiltration. Our study demonstrates that PDAC microbiome composition, which cross-talks to the gut microbiome, influences the host immune response and natural history of the disease.

Influence of oral microbiome on longitudinal patterns of oral mucositis severity in patients with squamous cell carcinoma of the head and neck.

BACKGROUND: This study investigated the influence of oral microbial features on the trajectory of oral mucositis (OM) in patients with squamous cell carcinoma of the head and neck. METHODS: OM severity was assessed and buccal swabs were collected at baseline, at the initiation of cancer treatment, weekly during cancer treatment, at the termination of cancer treatment, and after cancer treatment termination. The oral microbiome was characterized via the 16S ribosomal RNA V4 region with the Illumina platform. Latent class mixed-model analysis was used to group individuals with similar trajectories of OM severity. Locally estimated scatterplot smoothing was used to fit an average trend within each group and to assess the association between the longitudinal OM scores and longitudinal microbial abundances. RESULTS: Four latent groups (LGs) with differing patterns of OM severity were identified for 142 subjects. LG1 has an early onset of high OM scores. LGs 2 and 3 begin with relatively low OM scores until the eighth and 11th week, respectively. LG4 has generally flat OM scores. These LGs did not vary by treatment or clinical or demographic variables. Correlation analysis showed that the abundances of Bacteroidota, Proteobacteria, Bacteroidia, Gammaproteobacteria, Enterobacterales, Bacteroidales, Aerococcaceae, Prevotellaceae, Abiotrophia, and Prevotella_7 were positively correlated with OM severity across the four LGs. Negative correlation was observed with OM severity for a few microbial features: Abiotrophia and Aerococcaceae for LGs 2 and 3; Gammaproteobacteria and Proteobacteria for LGs 2, 3, and 4; and Enterobacterales for LGs 2 and 4. CONCLUSIONS: These findings suggest the potential to personalize treatment for OM. PLAIN LANGUAGE SUMMARY: Oral mucositis (OM) is a common and debilitating after effect for patients treated for squamous cell carcinoma of the head and neck. Trends in the abundance of specific microbial features may be associated with patterns of OM severity over time. Our findings suggest the potential to personalize treatment plans for OM via tailored microbiome interventions.

Easy NanoString nCounter data analysis with the NanoTube.

SUMMARY: The NanoTube is an open-source pipeline that simplifies the processing, quality control, normalization and analysis of NanoString nCounter gene expression data. It is implemented in an extensible R library, which performs a variety of gene expression analysis techniques and contains additional functions for integration with other R libraries performing advanced NanoString analysis techniques. Additionally, the NanoTube web application is available as a simple tool for researchers without programming expertise. AVAILABILITY AND IMPLEMENTATION: The NanoTube R package is available on Bioconductor under the GPL-3 license (https://www.bioconductor.org/packages/NanoTube/). The R-Shiny application can be downloaded at https://github.com/calebclass/Shiny-NanoTube, or a simplified version of this application can be run on all major browsers, at https://research.butler.edu/nanotube/. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

A unified mediation analysis framework for integrative cancer proteogenomics with clinical outcomes.

MOTIVATION: Multilevel molecular profiling of tumors and the integrative analysis with clinical outcomes have enabled a deeper characterization of cancer treatment. Mediation analysis has emerged as a promising statistical tool to identify and quantify the intermediate mechanisms by which a gene affects an outcome. However, existing methods lack a unified approach to handle various types of outcome variables, making them unsuitable for high-throughput molecular profiling data with highly interconnected variables. RESULTS: We develop a general mediation analysis framework for proteogenomic data that include multiple exposures, multivariate mediators on various scales of effects as appropriate for continuous, binary and survival outcomes. Our estimation method avoids imposing constraints on model parameters such as the rare disease assumption, while accommodating multiple exposures and high-dimensional mediators. We compare our approach to other methods in extensive simulation studies at a range of sample sizes, disease prevalence and number of false mediators. Using kidney renal clear cell carcinoma proteogenomic data, we identify genes that are mediated by proteins and the underlying mechanisms on various survival outcomes that capture short- and long-term disease-specific clinical characteristics. AVAILABILITY AND IMPLEMENTATION: Software is made available in an R package (https://github.com/longjp/mediateR). SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Healthcare provider perceptions and reported practices regarding opioid prescription for patients with chronic cancer pain.

PURPOSE: Data indicates that clinicians might be under-prescribing opioids for patients with chronic cancer pain, and this could impact adequate pain management. Few studies have sought to understand healthcare provider (HCP) perceptions and practices regarding the prescription of opioids for chronic cancer pain. We assessed HCP perceptions and practices regarding opioid prescription for patients with chronic cancer pain since the onset of the COVID-19 pandemic. METHODS: An anonymous cross-sectional survey was conducted among 186 HCPs who attended an opioid educational event in April 2021 and 2022. RESULTS: Sixty-one out of 143 (44%) opioid prescribers reported reluctance to prescribe opioids for chronic cancer pain. In a multivariate logistic model, younger participants (log OR - 0.04, 95% CI - 0.085, - 0.004; p = 0.033) and pain medicine clinicians (log OR - 1.89, CI - 3.931, - 0.286; p = 0.034) were less reluctant, whereas providers who worry about non-medical opioid use were more reluctant to prescribe opioids (log OR 1.58 95% CI 0.77-2.43; p < 0.001). Fifty-three out of 143 (37%) prescribers had experienced increased challenges regarding opioid dispensing at pharmacies, and 84/179 (47%) of all respondents reported similar experience by their patients. Fifty-four out of 178(30%) were aware of opioid-related harmful incidents to patients or their families, including incidents attributed to opioid misuse by a household or family member. CONCLUSION: A considerable number of opioid prescribers were reluctant to prescribe opioids for patients with chronic cancer pain. Many reported challenges regarding dispensing of opioids at the pharmacies. These may be unintended consequences of policies to address the opioid crisis. Future measures should focus on addressing regulatory barriers without undermining the gains already made to combat the opioid crisis.

A machine learning-based method for automatically identifying novel cells in annotating single-cell RNA-seq data.

MOTIVATION: Single-cell RNA sequencing (scRNA-seq) has been widely used to decompose complex tissues into functionally distinct cell types. The first and usually the most important step of scRNA-seq data analysis is to accurately annotate the cell labels. In recent years, many supervised annotation methods have been developed and shown to be more convenient and accurate than unsupervised cell clustering. One challenge faced by all the supervised annotation methods is the identification of the novel cell type, which is defined as the cell type that is not present in the training data, only exists in the testing data. Existing methods usually label the cells simply based on the correlation coefficients or confidence scores, which sometimes results in an excessive number of unlabeled cells. RESULTS: We developed a straightforward yet effective method combining autoencoder with iterative feature selection to automatically identify novel cells from scRNA-seq data. Our method trains an autoencoder with the labeled training data and applies the autoencoder to the testing data to obtain reconstruction errors. By iteratively selecting features that demonstrate a bi-modal pattern and reclustering the cells using the selected feature, our method can accurately identify novel cells that are not present in the training data. We further combined this approach with a support vector machine to provide a complete solution for annotating the full range of cell types. Extensive numerical experiments using five real scRNA-seq datasets demonstrated favorable performance of the proposed method over existing methods serving similar purposes. AVAILABILITY AND IMPLEMENTATION: Our R software package CAMLU is publicly available through the Zenodo repository (https://doi.org/10.5281/zenodo.7054422) or GitHub repository (https://github.com/ziyili20/CAMLU). SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

CNGPLD: case-control copy-number analysis using Gaussian process latent difference.

MOTIVATION: Cross-sectional analyses of primary cancer genomes have identified regions of recurrent somatic copy-number alteration, many of which result from positive selection during cancer formation and contain driver genes. However, no effective approach exists for identifying genomic loci under significantly different degrees of selection in cancers of different subtypes, anatomic sites or disease stages. RESULTS: CNGPLD is a new tool for performing case-control somatic copy-number analysis that facilitates the discovery of differentially amplified or deleted copy-number aberrations in a case group of cancer compared with a control group of cancer. This tool uses a Gaussian process statistical framework in order to account for the covariance structure of copy-number data along genomic coordinates and to control the false discovery rate at the region level. AVAILABILITY AND IMPLEMENTATION: CNGPLD is freely available at https://bitbucket.org/djhshih/cngpld as an R package. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Bayesian hierarchical quantile regression with application to characterizing the immune architecture of lung cancer.

The successful development and implementation of precision immuno-oncology therapies requires a deeper understanding of the immune architecture at a patient level. T-cell receptor (TCR) repertoire sequencing is a relatively new technology that enables monitoring of T-cells, a subset of immune cells that play a central role in modulating immune response. These immunologic relationships are complex and are governed by various distributional aspects of an individual patient's tumor profile. We propose Bayesian QUANTIle regression for hierarchical COvariates (QUANTICO) that allows simultaneous modeling of hierarchical relationships between multilevel covariates, conducts explicit variable selection, estimates quantile and patient-specific coefficient effects, to induce individualized inference. We show QUANTICO outperforms existing approaches in multiple simulation scenarios. We demonstrate the utility of QUANTICO to investigate the effect of TCR variables on immune response in a cohort of lung cancer patients. At population level, our analyses reveal the mechanistic role of T-cell proportion on the immune cell abundance, with tumor mutation burden as an important factor modulating this relationship. At a patient level, we find several outlier patients based on their quantile-specific coefficient functions, who have higher mutational rates and different smoking history.

Attempts to Understand Oral Mucositis in Head and Neck Cancer Patients through Omics Studies: A Narrative Review.

Oral mucositis (OM) is a common and clinically impactful side effect of cytotoxic cancer treatment, particularly in patients with head and neck squamous cell carcinoma (HNSCC) who undergo radiotherapy with or without concomitant chemotherapy. The etiology and pathogenic mechanisms of OM are complex, multifaceted and elicit both direct and indirect damage to the mucosa. In this narrative review, we describe studies that use various omics methodologies (genomics, transcriptomics, microbiomics and metabolomics) in attempts to elucidate the biological pathways associated with the development or severity of OM. Integrating different omics into multi-omics approaches carries the potential to discover links among host factors (genomics), host responses (transcriptomics, metabolomics), and the local environment (microbiomics).

Lead-Time Trajectory of CA19-9 as an Anchor Marker for Pancreatic Cancer Early Detection.

BACKGROUND & AIMS: There is substantial interest in liquid biopsy approaches for cancer early detection among subjects at risk, using multi-marker panels. CA19-9 is an established circulating biomarker for pancreatic cancer; however, its relevance for pancreatic cancer early detection or for monitoring subjects at risk has not been established. METHODS: CA19-9 levels were assessed in blinded sera from 175 subjects collected up to 5 years before diagnosis of pancreatic cancer and from 875 matched controls from the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial. For comparison of performance, CA19-9 was assayed in blinded independent sets of samples collected at diagnosis from 129 subjects with resectable pancreatic cancer and 275 controls (100 healthy subjects; 50 with chronic pancreatitis; and 125 with noncancerous pancreatic cysts). The complementary value of 2 additional protein markers, TIMP1 and LRG1, was determined. RESULTS: In the PLCO cohort, levels of CA19-9 increased exponentially starting at 2 years before diagnosis with sensitivities reaching 60% at 99% specificity within 0 to 6 months before diagnosis for all cases and 50% at 99% specificity for cases diagnosed with early-stage disease. Performance was comparable for distinguishing newly diagnosed cases with resectable pancreatic cancer from healthy controls (64% sensitivity at 99% specificity). Comparison of resectable pancreatic cancer cases to subjects with chronic pancreatitis yielded 46% sensitivity at 99% specificity and for subjects with noncancerous cysts, 30% sensitivity at 99% specificity. For prediagnostic cases below cutoff value for CA19-9, the combination with LRG1 and TIMP1 yielded an increment of 13.2% in sensitivity at 99% specificity (P = .031) in identifying cases diagnosed within 1 year of blood collection. CONCLUSION: CA19-9 can serve as an anchor marker for pancreatic cancer early detection applications.

Myelodysplastic/myeloproliferative neoplasms-unclassifiable with isolated isochromosome 17q represents a distinct clinico-biologic subset: a multi-institutional collaborative study from the Bone Marrow Pathology Group.

Classification of myeloid neoplasms with isolated isochromosome i(17q) [17p deletion with inherent monoallelic TP53 loss plus 17q duplication] is controversial. Most cases fall within the WHO unclassifiable myelodysplastic/myeloproliferative neoplasms (MDS/MPN-U) category. The uniformly dismal outcomes warrant better understanding of this entity. We undertook a multi-institutional retrospective study of 92 adult MDS/MPN-U cases from eight institutions. Twenty-nine (32%) patients had isolated i(17q) [MDS/MPN-i(17q)]. Compared to MDS/MPN without i(17q), MDS/MPN-i(17q) patients were significantly younger, had lower platelet and absolute neutrophil counts, and higher frequency of splenomegaly and circulating blasts. MDS/MPN-i(17q) cases showed frequent bilobed neutrophils (75% vs. 23%; P = 0.03), hypolobated megakaryocytes (62% vs. 20%; P = 0.06), and a higher frequency of SETBP1 (69% vs. 5%; P = 0.002) and SRSF2 (63% vs. 5%; P = 0.006) mutations that were frequently co-existent (44% vs. 0%; P = 0.01). TP53 mutations were rare. The mutation profile of MDS/MPN-U-i(17q) was similar to other myeloid neoplasms with i(17q) including atypical chronic myeloid leukemia, chronic myelomonocytic leukemia, myelodysplastic/myeloproliferative neoplasm with ring sideroblasts and thrombocytosis, myelodysplastic syndrome and acute myeloid leukemia, with frequent concomitant SETBP1/SRSF2 mutations observed across all the diagnostic entities. Over a median follow-up of 52 months, patients with MDS/MPN-i(17q) showed a shorter median overall survival (11 vs. 28 months; P < 0.001). The presence of i(17q) retained independent poor prognostic value in multivariable Cox-regression analysis [HR 3.686 (1.17-11.6); P = 0.026] along with splenomegaly. We suggest that MDS/MPN-i(17q) warrants recognition as a distinct subtype within the MDS/MPN-U category based on its unique clinico-biologic features and uniformly poor prognosis.

A semi-mechanistic dose-finding design in oncology using pharmacokinetic/pharmacodynamic modeling.

While a number of phase I dose-finding designs in oncology exist, the commonly used ones are either algorithmic or empirical model-based. We propose a new framework for modeling the dose-response relationship, by systematically incorporating the pharmacokinetic (PK) data collected in the trial and the hypothesized mechanisms of the drug effects, via dynamic PK/PD modeling, as well as modeling of the relationship between a latent cumulative pharmacologic effect and a binary toxicity outcome. This modeling framework naturally incorporates the information on the impact of dose, schedule and method of administration (e.g., drug formulation and route of administration) on toxicity. The resulting design is an extension of existing designs that make use of pre-specified summary PK information (such as the area under the concentration-time curve [AUC] or maximum serum concentration [Cmax ]). Our simulation studies show, with moderate departure from the hypothesized mechanisms of the drug action, that the performance of the proposed design on average improves upon those of the common designs, including the continual reassessment method (CRM), Bayesian optimal interval (BOIN) design, modified toxicity probability interval (mTPI) method, and a design called PKLOGIT that models the effect of the AUC on toxicity. In case of considerable departure from the underlying drug effect mechanism, the performance of the design is shown to be comparable with that of the other designs. We illustrate the proposed design by applying it to the setting of a phase I trial of a γ-secretase inhibitor in metastatic or locally advanced solid tumors. We also provide R code to implement the proposed design.

A Comprehensive Search of Non-Canonical Proteins in Non-Small Cell Lung Cancer and Their Impact on the Immune Response.

There is substantial interest in mining neoantigens for cancer applications. Non-canonical proteins resulting from frameshift mutations have been identified as neoantigens in cancer. We investigated the landscape of non-canonical proteins in non-small cell lung cancer (NSCLC) and their induced immune response in the form of autoantibodies. A database of cryptoproteins was computationally constructed and comprised all alternate open reading frames (altORFs) and ORFs identified in pseudogenes, noncoding RNAs, and untranslated regions of mRNAs that did not align with known canonical proteins. Proteomic profiles of seventeen lung adenocarcinoma (LUAD) cell lines were searched to evaluate the occurrence of cryptoproteins. To assess the immunogenicity, immunoglobulin (Ig)-bound cryptoproteins in plasmas were profiled by mass spectrometry. The specimen set consisted of plasmas from 30 newly diagnosed NSCLC cases, pre-diagnostic plasmas from 51 NSCLC cases, and 102 control plasmas. An analysis of LUAD cell lines identified 420 cryptoproteins. Plasma Ig-bound analyses revealed 90 cryptoproteins uniquely found in cases and 14 cryptoproteins that had a fold-change >2 compared to controls. In pre-diagnostic samples, 17 Ig-bound cryptoproteins yielded an odds ratio ≥2. Eight Ig-bound cryptoproteins were elevated in both pre-diagnostic and newly diagnosed cases compared to controls. Cryptoproteins represent a class of neoantigens that induce an autoantibody response in NSCLC.

ProgPerm: Progressive permutation for a dynamic representation of the robustness of microbiome discoveries.

BACKGROUND: Identification of features is a critical task in microbiome studies that is complicated by the fact that microbial data are high dimensional and heterogeneous. Masked by the complexity of the data, the problem of separating signals (differential features between groups) from noise (features that are not differential between groups) becomes challenging and troublesome. For instance, when performing differential abundance tests, multiple testing adjustments tend to be overconservative, as the probability of a type I error (false positive) increases dramatically with the large numbers of hypotheses. Moreover, the grouping effect of interest can be obscured by heterogeneity. These factors can incorrectly lead to the conclusion that there are no differences in the microbiome compositions. RESULTS: We translate and represent the problem of identifying differential features, which are differential in two-group comparisons (e.g., treatment versus control), as a dynamic layout of separating the signal from its random background. More specifically, we progressively permute the grouping factor labels of the microbiome samples and perform multiple differential abundance tests in each scenario. We then compare the signal strength of the most differential features from the original data with their performance in permutations, and will observe a visually apparent decreasing trend if these features are true positives identified from the data. Simulations and applications on real data show that the proposed method creates a U-curve when plotting the number of significant features versus the proportion of mixing. The shape of the U-Curve can convey the strength of the overall association between the microbiome and the grouping factor. We also define a fragility index to measure the robustness of the discoveries. Finally, we recommend the identified features by comparing p-values in the observed data with p-values in the fully mixed data. CONCLUSIONS: We have developed this into a user-friendly and efficient R-shiny tool with visualizations. By default, we use the Wilcoxon rank sum test to compute the p-values, since it is a robust nonparametric test. Our proposed method can also utilize p-values obtained from other testing methods, such as DESeq. This demonstrates the potential of the progressive permutation method to be extended to new settings.

Cytoplasmic Cyclin E Expression Predicts for Response to Neoadjuvant Chemotherapy in Breast Cancer.

BACKGROUND: Pathologic complete response (pCR) has been shown to be associated with favorable outcomes in breast cancer. Predictors of pCR could be useful in guiding treatment decisions regarding neoadjuvant therapy. The objective of this study was to evaluate cyclin E as a predictor of response to neoadjuvant chemotherapy in breast cancer. METHODS: Patients (n = 285) with stage II-III breast cancer were enrolled in a prospective study and received neoadjuvant chemotherapy with anthracyclines, taxanes, or combination of the two. Pretreatment biopsies from 190 patients and surgical specimens following chemotherapy from 192 patients were available for immunohistochemical analysis. Clinical and pathologic responses were recorded and associated with presence of tumor infiltrating lymphocytes, cyclin E, adipophilin, programmed cell death-ligand 1, and elastase staining and other patient, tumor and treatment characteristics. RESULTS: The pCR rate was significantly lower in patients with cytoplasmic cyclin E staining compared with those who had no cyclin E expression (16.1% vs 38.9%, P = 0.0005). In multivariable logistic regression analysis, the odds of pCR for patients who had cytoplasmic negative tumors was 9.35 times (P value < 0.0001) that compared with patients with cytoplasmic positive tumors after adjusting for ER, PR, and HER2 status. Cytoplasmic cyclin E expression also predicts long-term outcome and is associated with reduced disease free, recurrence free, and overall survival rates, independent of increased pretreatment tumor infiltrating lymphocytes. CONCLUSIONS: Cyclin E independently predicted response to neoadjuvant chemotherapy. Hence, its routine immunohistochemical analysis could be used clinically to identify those breast cancer patients expected to have a poor response to anthracycline/taxane-based chemotherapy.

NExUS: Bayesian simultaneous network estimation across unequal sample sizes.

MOTIVATION: Network-based analyses of high-throughput genomics data provide a holistic, systems-level understanding of various biological mechanisms for a common population. However, when estimating multiple networks across heterogeneous sub-populations, varying sample sizes pose a challenge in the estimation and inference, as network differences may be driven by differences in power. We are particularly interested in addressing this challenge in the context of proteomic networks for related cancers, as the number of subjects available for rare cancer (sub-)types is often limited. RESULTS: We develop NExUS (Network Estimation across Unequal Sample sizes), a Bayesian method that enables joint learning of multiple networks while avoiding artefactual relationship between sample size and network sparsity. We demonstrate through simulations that NExUS outperforms existing network estimation methods in this context, and apply it to learn network similarity and shared pathway activity for groups of cancers with related origins represented in The Cancer Genome Atlas (TCGA) proteomic data. AVAILABILITY AND IMPLEMENTATION: The NExUS source code is freely available for download at https://github.com/priyamdas2/NExUS. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

aPCoA: covariate adjusted principal coordinates analysis.

SUMMARY: In fields, such as ecology, microbiology and genomics, non-Euclidean distances are widely applied to describe pairwise dissimilarity between samples. Given these pairwise distances, principal coordinates analysis is commonly used to construct a visualization of the data. However, confounding covariates can make patterns related to the scientific question of interest difficult to observe. We provide adjusted principal coordinates analysis as an easy-to-use tool, available as both an R package and a Shiny app, to improve data visualization in this context, enabling enhanced presentation of the effects of interest. AVAILABILITY AND IMPLEMENTATION: The R package 'aPCoA' and Shiny app can be accessed at https://cran.r-project.org/web/packages/aPCoA/index.html and https://biostatistics.mdanderson.org/shinyapps/aPCoA/.

Bayesian compositional regression with structured priors for microbiome feature selection.

The microbiome plays a critical role in human health and disease, and there is a strong scientific interest in linking specific features of the microbiome to clinical outcomes. There are key aspects of microbiome data, however, that limit the applicability of standard variable selection methods. In particular, the observed data are compositional, as the counts within each sample have a fixed-sum constraint. In addition, microbiome features, typically quantified as operational taxonomic units, often reflect microorganisms that are similar in function, and may therefore have a similar influence on the response variable. To address the challenges posed by these aspects of the data structure, we propose a variable selection technique with the following novel features: a generalized transformation and z-prior to handle the compositional constraint, and an Ising prior that encourages the joint selection of microbiome features that are closely related in terms of their genetic sequence similarity. We demonstrate that our proposed method outperforms existing penalized approaches for microbiome variable selection in both simulation and the analysis of real data exploring the relationship of the gut microbiome to body mass index.

Compositional zero-inflated network estimation for microbiome data.

BACKGROUND: The estimation of microbial networks can provide important insight into the ecological relationships among the organisms that comprise the microbiome. However, there are a number of critical statistical challenges in the inference of such networks from high-throughput data. Since the abundances in each sample are constrained to have a fixed sum and there is incomplete overlap in microbial populations across subjects, the data are both compositional and zero-inflated. RESULTS: We propose the COmpositional Zero-Inflated Network Estimation (COZINE) method for inference of microbial networks which addresses these critical aspects of the data while maintaining computational scalability. COZINE relies on the multivariate Hurdle model to infer a sparse set of conditional dependencies which reflect not only relationships among the continuous values, but also among binary indicators of presence or absence and between the binary and continuous representations of the data. Our simulation results show that the proposed method is better able to capture various types of microbial relationships than existing approaches. We demonstrate the utility of the method with an application to understanding the oral microbiome network in a cohort of leukemic patients. CONCLUSIONS: Our proposed method addresses important challenges in microbiome network estimation, and can be effectively applied to discover various types of dependence relationships in microbial communities. The procedure we have developed, which we refer to as COZINE, is available online at https://github.com/MinJinHa/COZINE .

Gut Microbiome Signatures Are Predictive of Infectious Risk Following Induction Therapy for Acute Myeloid Leukemia.

BACKGROUND: The majority of studies that provide insights into the influence of the microbiome on the health of hematologic malignancy patients have concentrated on the transplant setting. Here, we sought to assess the predictive capacity of the gastrointestinal microbiome and its relationship to infectious outcomes in patients with acute myeloid leukemia (AML). METHODS: 16s rRNA-based analysis was performed on oral swabs and stool samples obtained biweekly from baseline until neutrophil recovery following induction chemotherapy (IC) in 97 AML patients. Microbiome characteristics were correlated with clinical outcomes both during and after IC completion. RESULTS: At the start of IC, higher stool Shannon diversity (hazard ratio [HR], 0.36; 95% confidence interval [CI], .18-.74) and higher relative abundance of Porphyromonadaceae (HR, 0.36; 95% CI, .18-.73) were associated with increased probability of remaining infection-free during neutropenia. A baseline stool Shannon diversity cutoff of <2 had optimal operating characteristics for predicting infectious complications during neutropenia. Although 56 patients received therapy >72 hours with a carbapenem, none of the patients had an infection with an extended spectrum ß-lactamase-producing organism. Patients who received carbapenems for >72 hours had significantly lower α-diversity at neutrophil recovery (P = .001) and were approximately 4 times more likely to have infection in the 90 days following neutrophil recovery (HR, 4.55; 95% CI, 1.73-11.93). CONCLUSIONS: Our results suggest that gut microbiome evaluation could assist with infectious risk stratification and that improved targeting of antibiotic administration during IC could decrease subsequent infectious complications in AML patients.Baseline microbiome diversity is a strong independent predictor of infection during acute myeloid leukemia induction chemotherapy (IC) among clinical and microbiome covariates. Higher baseline levels of Porphyromonadaceae appear protective against infection, while carbapenem use is associated with consequences to the microbiome and infection susceptibility post-IC.