Dynamics of the Gut Mycobiome in Patients with Ulcerative Colitis

Abstract

Background and Aims:

Intestinal fungi have been implicated in the pathogenesis of ulcerative colitis (UC), however it remains unclear if fungal composition is altered during active vs quiescent disease.

Methods:

We analyzed clinical and metagenomic data from the Study of a Prospective Adult Research Cohort with Inflammatory Bowel Disease (SPARC IBD), available via the IBD Plexus Program of the Crohn’s & Colitis Foundation. We evaluated the fungal composition of fecal samples from 421 patients with UC during clinical activity and remission. Within a longitudinal subcohort (n=52), we assessed for dynamic taxonomic changes across alterations in clinical activity over time. We examined if fungal amplicon sequence variants and fungal-bacterial relationships were altered during activity vs remission. Finally, we classified activity in UC using a supervised machine learning random forest model trained on fungal abundance data.

Results:

During clinical activity, the relative abundance of genus Candida was increased 3.5-fold (p-adj < 1 × 10⁻⁴) compared to during remission. Patients with longitudinal reductions in clinical activity demonstrated parallel reductions in Candida relative abundance (p<0.05). Candida relative abundance correlated with Parabacteroides diastonis, Faecalibacterium prausnitzii and Bacteroides dorei relative abundance (p<0.05) during remission, however these correlations were disrupted during activity. Fungal abundance data successfully classified patients with active or quiescent UC (AUC ~0.80), with Candida relative abundance critical to the success of the model.

Conclusions:

Clinical activity in UC is associated with an increased relative abundance of Candida, cross-sectionally and dynamically over time. The role of fecal Candida as a target for therapeutics in UC should be evaluated.

Graphical Abstract

INTRODUCTION

Ulcerative colitis (UC) is a form of inflammatory bowel disease (IBD) that primarily affects the colon, leading to frequent relapses, hospitalizations, surgeries, and increased lifetime morbidity¹. The pathogenesis of UC is likely driven by disruptions in host-microbiome homeostasis, with alterations in specific bacterial taxa and microbial products linked to colonic inflammation². However, therapies directed at modulating the gut microbiome, including probiotics, antibiotics, or fecal microbial transplantation (FMT) have had modest and inconsistent effects in treating UC, presumably due to inter-individual differences in an incompletely characterized microbiome^{3, 4}. Specifically, the role of non-bacterial microbial kingdoms in UC pathogenesis – notably fungi – and the effect of manipulating their populations on disease course is poorly understood.

Gut fungi – and the fungal genus Candida – have been previously implicated in the pathogenesis of ulcerative colitis (UC). Dysbiosis of the fungal microbiome (mycobiome) has been observed in inflammatory bowel disease (IBD) patients compared to healthy individuals, with UC patients showing relative increases in Candida⁵. Furthermore, genetic polymorphisms in fungal antigen-sensing genes, including Dectin-1, have been linked to severe forms of UC⁶. Oral gavage of Candida in mice exacerbates Th-17 mediated inflammation⁷, and filamentous forms of Candida can activate the inflammasome and induce colonic Th17 responses⁸. Interestingly, patients with UC undergoing fecal microbial transplantation (FMT) who have higher fecal Candida prior to transplant, demonstrate favorable responses to the microbial therapy⁹. We also previously showed that Candida is enriched during endoscopic activity vs remission in fecal samples among a small cohort of patients with UC (n=53)¹⁰.

Despite these findings, it remains unclear whether the gut mycobiome changes during active UC compared to quiescent disease. In this study, we investigated potential associations between the mycobiome and UC by performing an expanded, longitudinal secondary analysis of clinical and metagenomic metadata obtained from a prospective cohort (Study of a Prospective Adult Research Cohort with Inflammatory Bowel Disease, or SPARC)¹¹. We hypothesized that Candida would be enriched during clinical activity. By identifying potential culprit fungal taxa associated with inflammation, we hoped to characterize fungal-UC phenotypes that could support personalized approaches to therapies, including FMT, probiotics, or antifungal treatments.

METHODS

Cohort Description

The cohort of patients included in this study was derived from SPARC IBD, a geographically diverse longitudinal research cohort of IBD patients utilizing standardized data and biosample collection methods and processing techniques¹¹. Demographics and collection methods of this cohort have been previously described¹¹.

Inclusion/Exclusion Criteria for Study Cohort

Cross-sectional:

For this study, we included 421 patients from the SPARC cohort who had a history of UC, underwent colonoscopy, and had available ITS2 fungal mycobiome sequencing data [Supp Fig 1]. Patients were either clinically active or in clinical remission.

Longitudinal:

Among the 421 patients in the cross-sectional cohort, we examined the subset of patients (n=52) who contributed 2 serial fecal samples over time. Patients either had clinical activity or clinical remission at baseline, then had repeat clinical assessment at follow-up.

Definitions

A two-item patient reported outcome (PRO-2) is a validated practical index of disease activity in UC that includes a rectal bleeding score and stool frequency score¹². Clinical activity was defined as PRO-2 ≥ 2. Clinical remission was defined by PRO-2 ≤ 1. See supplementary methods for additional definitions.

Microbial Preparation and Sequencing Analysis

For details regarding microbial isolation, library prep, sequencing analysis, see supplementary methods.

Analyses

Relative abundance, fractional prevalence, diversity, and differential abundance

To evaluate the relative abundance, we excluded rare organisms, defined as taxa that constituted less than 1% of the total abundance. Proportions were calculated by dividing the number of reads assigned to each taxon by the total number of reads. Fractional prevalence was calculated by dividing the number of patients containing a specific taxon by the total number of patient samples. Alpha diversity of the mycobiome was assessed using observed ASVs and Shannon diversity indices. For beta-diversity, weighted UniFrac was used to generate a distance matrix, which was then ordinated using Non-metric Multidimensional Scaling (NMDS). A negative binomial model (DESeq2 [version 3.15])¹³ was used to calculate the differential abundance of specific taxa across comparison groups in patients with UC, adjusted for potential confounding effects of age, gender, immunosuppressive therapy, steroid use, antibiotic use, and probiotic use. Dynamic changes in microbial abundance were assessed over longitudinal samples obtained at 2 timepoints across clinical activity categories. For heterogeneity and machine learning analysis, see supplementary methods.

Statistical Analysis

We conducted all analyses and figure preparation using R studio (version 2023.03.0)¹⁴ or QIIME2¹⁵. We performed a Wilcoxon rank-sum test (p<0.05) to evaluate whether alpha diversity indices differed significantly between comparators. To evaluate significance for beta-diversity, permutational ANOVA analysis was used to calculate significance between groups (p<0.05). For differential abundance analysis, p-values were corrected for multiple testing using the Benjamini and Hochberg method¹⁶. See supplemental methods for additional statistical analyses.

ETHICAL CONSIDERATIONS

Ethical approval for the study was obtained from the Tufts Medical Center Institutional Review Board, Boston, MA.

RESULTS

Clinical cohort

Of 467 patients with ITS2 sequencing of the mycobiome, 421 patients had 2-item patient-reported outcome scores (PRO-2) obtained concurrently with stool specimen collection (median interval = 0 months). 104 patients (25%) had clinically active disease, with either mild-moderate clinically active disease (n=73, PRO-2 of 2–3) or severe clinically active disease (n=31, PRO-2 of 4–6), while 317 patients (75%) were in clinical remission (PRO-2 ≤ 1). Study cohort characteristics are given in Table 1. Longitudinal fecal sampling was available among 52 patients, who each contributed 2 fecal samples (total of 104 samples), over a median of 4 months (IQR 2–8) [Supp Table 1]. Within this subcohort (n=52), at baseline, 37 patients were in clinical remission (71%) and 15 patients had clinical activity (29%).

Table 1 –

Study Cohort

			Clinical Status
		Total Cohort	Clinical Activity	Clinical Remission	p
n		421	104	317
Age, mean (SD)		46.4 (15.2)	46.4 (15.7)	46.4 (15.0)	0.97
Gender, female, frequency		227 (53.9%)	57 (54.8%)	170 (53.5%)	0.83
Disease Extent, n, (%)	E1	30 (7.1%)	11 (10.6%)	19 (6%)	0.11
	E2	89 (21.1%)	18 (17.3%)	71 (22.4%)	0.27
	E3	226 (53.7%)	53 (51%)	173 (54.6%)	0.52
	unknown	76 (18.1%)	22 (21.2%)	54 (17%)	0.34
Disease duration, years, median (IQR)		11 (5–20)	10 (4–20)	12 (6–20)	0.81
PRO-2 score, median (IQR)		0 (0–1)	3 (2–4)	0 (0–0)	<0.005
Rectal bleeding score, median (IQR)		0 (0–0)	1 (0–3)	0 (0–0)	<0.005
Stool Frequency score, median (IQR)		0 (0–1)	2 (1–3)	0 (0–0)	<0.005
Physician’s Global Assessment Score, median (IQR)		0 (0–1) [n=288]	1 (1–2) [n=72]	0 (0–0) [n=216]	<0.005
Mayo endoscopic score, median (IQR)		1 (0–2) [n=117]	2 (1–3) [n=30]	0 (0–1) [n=87]	<0.005
Interval, Fecal sample collection and PrO-2 scoring, months, median (IQR)		0 (0–0)	0 (0–0)	0 (0–0)	0.71
Interval, fecal sample collection and PGA Scoring, months, median (IQR)		0 (0–0)	0 (0–0)	0 (0–0)	0.87
Interval, fecal sample collection and MES scoring, months, median (IQR)		0 (0–0)	0 (0–0)	0 (0–1)	<0.05
Longitudinal sampling available		104	25	79	-
Immunosuppressive exposure		300 (71.3%)	72 (69.2%)	228 (71.9%)	0.59
Vedolizumab		102 (24.2%)	33 (31.7%)	69 (21.7%)	<0.05
Anti-TNF		156 (37.1%)	35 (33.7%)	121 (38.1%)	0.41
Thiopurine / methotrexate		136 (32.3%)	29 (27.9%)	107 (33.6%)	0.27
Ustekinumab		19 (4.5%)	12 (11.5%)	7 (2.2%)	<0.005
Tofacitinib		28 (6.7%)	13 (12.5%)	15 (4.7%)	<0.05
Tacrolimus / cyclosporine		20 (4.8%)	3 (2.9%)	17 (5.3%)	0.42
Concurrent steroid use (n= 321)		29 (9.0%)	17 (22.4%)	12 (4.8%)	<0.005
Concurrent antibiotic use (n=255)		42 (16.5%)	14 (36.8%)	28 (12.9%)	<0.005
Concurrent probiotic use (n=281)		64 (22.8%)	16 (27.1%)	48 (21.6%)	0.37

Relative abundance and prevalence of fungal phylum and genera in ulcerative colitis

We assessed the relative abundance of fungal phylum across all samples [Fig 1A]. We found that most fungal amplicon sequence variants (ASVs) belonged to the phylum Ascomycota (86% of ASVs), with a minority classified into the phylum Basidiomycota (3%) [Fig 1B]. Utilizing BLASTN, unidentifiable ASVs (11%) represented low prevalence fungi or dietary contaminants [Supp Table 2]. We assessed the relative abundance of fungal genera across all samples [Fig 1C]. 40% of ASVs belonged to genus Saccharomyces, 30% to Candida, 7% to Penicillium, 3% to Rhodotorula, 3% to Agaricus, with the remaining (17%) belonging to unidentified genera [Fig 1D]. Utilizing BLASTN, 2 fungal ASVs were resolvable to Geotrichum candidum and Candida glabrata [Supp Table 2].

Fig 1 –

Prevalence and abundance of fungal taxa among patients with UC. (A) Fractional prevalence and total abundance of fungal phyla. X-axis shows the proportion of total samples with detected phylum, y-axis shows the phylum microbial counts. (B) Relative abundance of fungal phyla. X-axis shows relative abundance, y-axis shows fungal phylum name. (C) Fractional prevalence and total abundance of fungal genera. X-axis shows the proportion of samples with detected genus, y-axis shows the total genus microbial counts. (D) Relative abundance of fungal genus. X-axis shows relative abundance, y-axis shows fungal genus name.

Diversity of the mycobiome in clinically active and remission ulcerative colitis

Among 421 patients with UC, the alpha and beta diversity of the mycobiome was similar during clinical activity compared to remission (mean Shannon diversity index = 0.94 vs 0.93, p=0.42, beta diversity: p=0.92, PERMANOVA) [Supp Fig2AB].

Candida is increased in relative abundance during clinical activity in ulcerative colitis

Among 421 patients with UC, during clinical activity, genus Candida showed 2.8-fold higher relative abundance compared to during clinical remission (p-adj<0.001) [Fig 2A]. Genera Agaricus and Rhodotorula showed lower relative abundances during clinical activity compared to during remission (p-adj < 0.001) [Fig 2AB]. After adjusting for the effects of age, gender, and immunosuppressive exposure, UC patients with clinical activity showed a 3.5-fold higher relative abundance of Candida compared to remission (p-adj < 1 × 10⁻⁴), with a 2.6-fold lower relative abundance of Agaricus during clinical activity (p-adj < 1 × 10⁻⁴) [Fig 2C]. After adjusting for concurrent steroid use, UC patients with clinical activity showed a 2.7-fold higher relative abundance of Candida compared to remission (p-adj < 0.001). Similarly, adjusting for concurrent antibiotic use, UC patients with clinical activity showed a 2.3-fold higher relative abundance of Candida compared to remission (p-adj < 0.001). While adjusting for probiotic use still resulted in UC patients with clinical activity showing a 1.8-fold increased relative abundance of Candida compared to remission, this result was less significant (p-adj<0.07). In a stratified analysis, there was no difference in Candida relative abundance among patients in remission or activity who did or did not take probiotics.

Fig 2 –

Differential abundance of fungal genera in clinical activity vs remission among patients with UC. (A) Differential log 2-fold changes in fungal genera among patients in clinical activity vs clinical remission. (B) Relative abundance of highly abundant fungal genera in patients with remission [R] vs activity [A]. (C) Differentially abundant fungal genera, after adjusting for age, sex, and immunosuppressive exposure.

Candida differential relative abundance increases with worsening disease severity index in ulcerative colitis

The relative abundance of Candida linearly increased with PRO-2 score (ranging from 0–6, p-adj < 0.05) [Supp Fig 3A]. Additionally, we found a trend towards increased Candida relative abundance with endoscopic severity (p-adj=0.15) [Supp Fig 3B]. Candida differential relative abundance was not significantly affected by immunosuppressive exposure (p-adj=0.23) [Supp Fig 3C]. Furthermore, Candida relative abundance was not significantly affected by concurrent steroid (p-adj=0.78), antibiotic (p-adj=0.33) or probiotic use (p-adj=0.56) [Supp Fig 4].

Candida longitudinal dynamics are perturbed by changes in clinical activity in ulcerative colitis

Candida relative abundance was significantly altered across 2 serial timepoints (median of 4 months, IQR 2–8) among all patients with longitudinal fecal sampling (n=52 patients) (p<0.05) [Fig 3A]. Patients who had no change in clinical activity (n=39, median time interval of 4 months, IQR=2–8) had stable Candida relative abundance across time (p = 0.32) [Fig 3B]. Patients who experienced a change in clinical activity (n=13, either from clinical remission to activity or vice versa, median time interval of 5 months, IQR=3–9) also experienced a change in Candida relative abundance over time (p<0.05) [Fig 3C].

Fig 3 –

Longitudinal relative abundance of Candida across disease activity. Relative abundance of Candida at timepoint 1 and 2 in patients with UC (A) in entire longitudinal cohort (n=52) (B) with no change in clinical status (n=39) (C) with shift in clinical status (n=13) (D) with stable clinical remission (n=33) (E) with clinical activity evolving to clinical remission (n=9) (F) with persistent clinical activity (n=6).

Among patients persistently in clinical remission (n=33), there was no difference in the overall low Candida mean relative abundance between timepoints (0.02 vs 0.06, p=0.72) [Fig 3D]. Among patients with persistent clinical activity (n=6), there was no difference in the overall high Candida mean relative abundance between timepoints (0.28 vs 0.40, p =0.29) [Fig 3F]. Among patients with clinical activity who then developed clinical remission (n=9), Candida mean relative abundance concomitantly decreased from high to low (0.43 vs 0.02, p<0.05) [Fig 3E]. Among patients in clinical remission who then developed clinical activity (n=4), Candida was not detectable in any samples.

Among clinically active patients who maintained activity (n=6) or progressed to remission (n=9), there was no significant difference in baseline steroid use (2/5 vs 0/6 patients, p=0.18), antibiotic use (2/2 vs 3/7 patients, p=1.0), or probiotic use (1/1 vs 3/6 patients, p=1.0). Among patients in remission who maintained remission (n=33) or progressed to activity (n=4), there was no significant baseline difference in steroid use (0/33 vs 1/4, p=0.11), antibiotic use (5/22 vs 0/2, p=1.0), or probiotic use (3/22 vs 1/2 patients, p=0.31).

Heterogeneity among Candida amplicon sequence variants

Within the entire patient cohort, we identified 5 unique Candida ASVs [Fig 4A]. Three of these ASVs represent different strains of Candida albicans (ASV1–3), while 1 ASV represents Candida tropicalis (ASV4) or Candida glabrata (ASV5). Candida albicans was detected in 108 patients (28.1%, ASVs 1–3), while Candida tropicalis was detected in 2 patients (0.5%) and Candida glabrata was detected in 9 patients (2%) [Fig 4B]. Upon differential abundance analysis, Candida albicans (combined aggregate of ASVs 1–3) demonstrated a 3-fold increase during clinical activity vs remission (p-adj < 1.1 × 10⁻³) while Candida glabrata and Candida tropicalis were of low prevalence and did not exhibit significant differences [Fig 4C].

Fig 4 –

Heterogeneity in Candida amplicon sequence variants (ASVs) across patients with UC. (A) Fractional prevalence (x-axis) and total counts (y-axis) of ASVs attributed to Candida genus. (B) Relative abundance of each Candida ASV C) Relative abundance of each Candida ASV during clinical remission vs activity.

Classification of ulcerative colitis utilizing a supervised machine learning model

A supervised random forest machine learning model was trained and tested on the fungal sequencing data from the patient cohort achieved a testing area under the curve (AUC) of ~0.80, with overall accuracy of 81% in classifying clinically quiescent vs active UC [Fig 5AB]. Saccharomyces and Candida demonstrated the highest feature importance to the model (0.22 and 0.20, respectively) [Fig 5C] [Supp Table 3], with individual performance characteristics for each predictor provided in [Supp Fig 5].

Fig 5 -.

Classification of clinical activity in ulcerative colitis utilizing a supervised machine learning model trained on fungal taxa abundance data. Y-axis represents the true positive rate while x-axis represents false positive rate, with (A) overall micro-averaging (dark blue squares) and macro-averaging (light blue squares) (B) per class, with prediction of quiescence (pink line) and prediction of activity (black line). (C) Heat map demonstrates the feature importance of each predictor taxa contributing to the model, with light squares demonstrating high sequence counts and dark squares demonstrating low sequence counts.

Candida-bacterial interkingdom relationships

Within the entire cohort (n=421), Candida relative abundance was positively correlated with the relative abundances of bacterial genera Parabacteroides diastonis (+0.15, p<0.005), and negatively correlated with Eubacterium hallii (−0.10, p<0.05) and Bifidobacterium adolescentis (−0.11, p<0.05) [Supp Fig 6A]. During clinically active disease (n=93), Candida relative abundance did not significantly correlate with any bacterial taxa [Supp Fig 6B]. However, during remission (n=284), Candida relative abundance was positively correlated with Parabacteroides diastonis (+0.16, p<0.05), Faecalibacterium prausnitzii (+0.14, p<0.05) and Bacteroides dorei (+0.13, p<0.05) [Supp Fig 6C].

DISCUSSION

Previous deep-sequencing profiles of several IBD cohorts have consistently shown a high relative abundance of Candida in IBD patients compared to healthy controls. However, how Candida and other fungi are altered during inflammation in UC patients has not been thoroughly examined. In this large, secondary analysis of a prospective adult cohort of 421 UC patients, we found that the relative abundance of Candida increased 3-fold in patients with clinical activity. This association is strengthened by the large sample size, adjustment for medication use, the relative high abundance and prevalence of Candida across patient samples, the use of a validated definition of clinical activity, and dynamic assessment over longitudinally sampled patients. Furthermore, fecal samples and clinical indices were obtained concurrently in time. Overall, these findings substantiate a link between Candida and inflammatory activity in UC, lending support to the hypothesis that the mycobiome may prove useful as a target for microbial manipulation to improve disease outcomes in UC.

Mouse models have previously shown that oral gavage of Candida albicans worsens inflammation in dextran sodium sulfate (DSS) colitis⁷, mediated through Dectin-1 receptors¹⁷ and downstream fine tuning of Th1/Th17 balance in the colon¹⁸. Complementary to these mechanistic studies, our study provides epidemiologic data linking Candida to colonic inflammation in UC.

The current study is further bolstered by assessment of the mycobiome dynamically over time. Longitudinal changes in the mycobiome in humans have been few, with none performed to date in IBD or UC. Candida has been previously reported to be more temporally stable on repeated measurements within the same person¹⁹, suggesting that it is more likely to be a persistent resident commensal in the human gut. Our finding that the relative abundance of Candida decreases in patients who evolve from clinical activity to clinical remission supports that Candida populations may dynamically parallel the degree of colonic inflammation.

The current study also examined the amplicon sequence variant composition of genus Candida. ASVs representing Candida albicans predominated and were significantly linked to inflammatory status. A prior study found that gut C. albicans isolates from human colonic samples demonstrated high genetic variability coinciding with altered pathogenic transcriptional programs²⁰. Further assessment of ASV heterogeneity and strain-level analyses in future studies may provide additional insight into Candida abundance, genetic variation, and UC pathogenesis. In our analysis of fungal-bacterial correlations, during conditions of quiescence, Candida abundance positively correlated with Parabacteroides diastonis, Faecalibacterium prausnitzii, and Bacteroides dorei, genera known to induce regulatory responses in the gut^21–23. Candida can influence gut bacterial assembly through metabolic competition and collaboration^24–26. Given these findings, it remains possible that during activity, the expansion of Candida is associated with bacterial dysbiosis and a loss of anti-inflammatory bacteria such as Faecalibacterium.

Our study also utilized fungal abundance data to train a supervised machine learning random forest model, to classify patients into active or quiescent UC. The model achieved an AUC of ~0.80 using cross-sectional microbiome data from highly abundant fungal taxa. Future studies could combine fungal-bacterial features to improve the operating characteristics of an externally validated model.

Given our observed association of Candida relative abundance with clinical activity in UC, the implication of our study is that treating patients with anti-fungal therapy may contribute to inducing remission. In a proof-of-concept, Candida albicans reduced FMT efficacy in a mouse model of Clostridium difficile infection, with antifungal therapy restoring FMT response²⁵. A randomized controlled trial of fluconazole in Candida colonized patients with UC led to clinical and biochemical improvements over placebo²⁷. Selecting patients for antifungal therapy alongside standard medications is a promising consideration in future trials, especially in refractory or severe UC.

Our study is limited by lack of dietary intake history, as nutritional components can influence fungal colonization²⁸. For example, though we found enriched Agaricus (mushrooms) during remission, this may reflect ingestion of edible mushrooms, rather than increases in a true gut colonizer²⁹. Another shortcoming is that the study primarily rested on ITS2-based sequencing. Clinical data was also limited by the smaller subcohort size among longitudinal samples, the absence of histologic data, and lack of reporting on antifungal use. We were additionally limited by the availability of concurrent endoscopic scoring, however our use of the validated PRO-2 allowed for longitudinal comparisons, and reflects a practical and clinically meaningful association between PRO-2 and Candida that can be tested in subsequent investigations. Future studies that combine large-scale gut compositional and functional assessments will better translate host-microbe associations towards generating mechanistic insights that may power clinically meaningful microbial interventions.

Overall, we report that among a large, prospective, well-characterized cohort of 421 patients, the abundance of genus Candida is significantly associated with clinical activity. The strength of this association is underlined even after adjustment for immunosuppressive exposure, antibiotic use, steroid use, and probiotic use. ASVs belonging to Candida albicans strongly linked to inflammatory activity. Interkingdom relationships between Candida and anti-inflammatory bacterial taxa evident during clinical quiescence were disrupted during clinical activity. Supervised machine learning, resting primarily on Candida and Saccharomyces abundance features, achieved a promising ability to classify patients with active vs quiescent UC. In the light of prior observations supporting a mechanistic role for Candida in colonic inflammation, this report substantiates a relationship between Candida and clinical activity in UC, opening the door for trials exploring antifungal, probiotic, or microbial transplantation therapeutics alongside conventional treatments to optimize outcomes in ulcerative colitis.

DATA TRANSPARENCY STATEMENT:

ITS2 (Internal Transcribed Spacer 2) sequencing, clinical data, metagenomic data, and tissue pathology was obtained from the Crohn’s & Colitis Foundation. For access to ITS2 and clinical metadata, researchers may contact the Crohn’s and Colitis Foundation and obtain data through IBD plexus (https://www.crohnscolitisfoundation.org/research/current-research-initiatives/ibd-plexus

What You Need to Know.

Background

The fungal gut microbiome has been previously implicated in the pathogenesis of inflammatory bowel disease. Candida, in particular, can drive gut inflammatory responses and has been shown to be increased in IBD patients compared to healthy people.

Findings

This study found that stool Candida was increased during inflammation in ulcerative colitis compared to remission. Furthermore, Candida decreased over time as patients moved from active disease towards quiescence.

Implications for Patient care

Elevated Candida is linked to active ulcerative colitis, and future studies might evaluate if treating Candida may contribute to improved outcomes.

ABBREVIATIONS:

NIH

National Institutes of Health

ITS2

Internal Transcribed Spacer 2

IBD

Inflammatory Bowel Disease

UC

ulcerative colitis

AUC

area under the curve

FMT

fecal microbial transplantation

SPARC-IBD

Study of a Prospective Adult Research Cohort with Inflammatory Bowel Disease

PRO-2

2-item patient reported outcome

RBS

rectal bleeding score

SFS

stool frequency score

MES

Mayo endoscopic subscore

TNF

tumor necrosis factor

JAK

Janus kinase

QIIME2

Quantitative Insights Into Microbial Ecology 2

ASV

amplicon sequence variant

NMDS

Non-metric Multidimensional Scaling

BLASTN

Nucleotide Basic Local Alignment Search Tool

DSS

dextran sodium sulfate