An antifungal reshaped the gut microbiome and tracked with milder IBD in a small study

Most talk of the gut "microbiome" means bacteria, but the gut also hosts fungi, and fungal imbalance has been implicated in inflammatory bowel disease (IBD) [s1]. A study published in Nature Medicine on 1 September tests an idea that follows from that: if fungi such as Candida contribute to IBD, does treating them help? The early answer is intriguing and heavily qualified [s1].

A clever entry point

The researchers used a real-world observation as their lever. In IBD patients with mild oral thrush — a mouth infection caused by Candida overgrowth — Candida albicans strains are shared between the mouth and the gut, suggesting the two populations are linked [s1]. That let them compare two antifungal strategies that reach different places [s1].

The study was a prospective observational comparison, not a randomised trial. One group received a mouth-only antifungal — "swish-and-spit" nystatin, which the authors label oral nystatin fungal targeting, or ORNT (18 patients) [s1]. The other received fluconazole, an antifungal absorbed through the gut, labelled gastrointestinal and oral fluconazole fungal targeting, or GIFT (35 patients) [s1]. In all, 53 patients with mild-to-moderate ulcerative colitis or Crohn's disease took part [s1].

What happened

The contrast between the two arms is the useful part. Fluconazole, but not nystatin, effectively reduced the intestinal Candida burden and reshaped the gut fungal community [s1]. Because nystatin acts in the mouth and fluconazole reaches the gut, that difference points to the gut fungal population, rather than the mouth, as the relevant target [s1].

The knock-on effects extended beyond fungi. Fluconazole treatment was accompanied by increased bacterial diversity, expansion of short-chain-fatty-acid-producing taxa, restoration of anti-inflammatory microbial metabolites, and durable shifts in cross-kingdom microbial networks — that is, changes in how bacteria and fungi interact [s1]. And these microbiome and metabolomic changes coincided with improved disease activity indices and a decreased risk of disease progression over the eight-week follow-up period [s1].

The word "coincided" is load-bearing

This is where caution has to come in, and the authors' own language supplies it. The clinical improvements "coincided with" the microbiome changes [s1]. That is an association across time, in an unblinded, non-randomised study of 53 people followed for eight weeks. It does not establish that clearing gut Candida caused the improvement, and several ordinary explanations — regression to the mean, co-treatment, the natural waxing and waning of IBD, or expectancy effects in an unblinded design — are not ruled out by this design.

The sample is also small and the follow-up short. Fifty-three patients split across two groups, watched for two months, can show a signal and a mechanism but cannot show durable clinical benefit or safety at the scale that would change treatment.

Why the two-arm design is the clever part

The comparison of nystatin against fluconazole is what gives the study more traction than a single-group before-and-after would. Nystatin, swished and spat, acts essentially in the mouth; fluconazole is absorbed and reaches the gut [s1]. Because the two drugs differ mainly in where they act, the finding that only the gut-reaching one reduced intestinal Candida and reshaped the gut fungal community points the mechanism at the gut population rather than the mouth [s1]. That is a stronger inference than "we gave an antifungal and things improved," even though neither arm was a true placebo control.

It also illustrates a subtlety in microbiome research: showing that an intervention changes the microbiome is comparatively easy, and this study does it convincingly for the fungal and bacterial communities alike [s1]. Showing that the change is what improved the patient is the hard part, and it is the part an eight-week observational study leaves open.

What the authors claim, and what it is worth

The paper's stated contributions are appropriately framed: it demonstrates the feasibility of mycobiome-based patient stratification — using a patient's fungal profile to decide who might be treated — and provides evidence that targeted antifungal therapy can reshape the intestinal microbiome [s1]. Those are claims about feasibility and mechanism, not a recommendation to give people with IBD an antifungal.

Its value is as a proof of concept and a hypothesis engine: it identifies a plausible, testable intervention (gut-active antifungal therapy in a fungal-defined subgroup) and the biology that might explain it. The test that matters next is a randomised, blinded, adequately powered trial measuring clinical outcomes, not microbiome shifts.

This article summarises a single small observational study and does not offer medical advice. Antifungal drugs have their own risks and interactions and are not a self-directed treatment for IBD.

Sources

• Antifungal therapy improves microbiome dynamics in inflammatory bowel disease, Nature Medicine, 1 September 2026