Antifungal pill held up against IV therapy on survival in phase 3

On 21 October 2026 the drug makers F2G and Shionogi reported results of OASIS, a randomised, adjudicator- and sponsor-blinded phase 3 trial of olorofim, an oral antifungal that is not yet approved anywhere, and the data were presented as a late-breaking report at the IDWeek 2026 conference the next day. The trial enrolled 225 adults with invasive aspergillosis, a mould infection of the lungs or other organs, whose infection was refractory to, or unsuitable for, the usual first-line drugs, the mould-active azoles. Olorofim was compared with intravenous liposomal amphotericin B, followed by the doctor's choice of standard therapy. The main question was how many people were alive six weeks later.

About one in four died within six weeks, in both groups

The primary endpoint was death from any cause by day 42. It occurred in 23.8% of patients on olorofim (36 of 151) and in 24.3% on the comparator (18 of 74), a difference of −0.5 percentage points with a 95% confidence interval from −13.1 to +10.8. The trial was designed to show non-inferiority, that olorofim is not worse than the comparator by more than a margin of 20 percentage points agreed in advance with the US and European regulators. By that rule it succeeded.

By day 84, 32.5% had died on olorofim and 29.7% on the comparator, a difference of +2.7 points (95% CI −10.6 to +15.0). The overall response, which requires improvement in symptoms, scans and fungal tests at once, was 26.5% against 31.1% at day 42 and 30.5% against 28.4% at day 84.

Not worse is a different claim from better

The width of that interval matters more than the headline. At day 42 the data are compatible both with 13.1 fewer deaths per 100 patients on olorofim and with 10.8 more. At day 84 the point estimate leans slightly towards the comparator. The trial does not show that olorofim helps people live longer than current therapy. It shows that the trial met its pre-specified non-inferiority criterion, with a wide margin.

The clearer difference is in side effects. At least one adverse event judged related to treatment occurred in 35.8% of patients on olorofim and in 63.9% on the comparator, serious treatment-related events in 4.0% against 18.1%. Most of the gap was kidney harm: 1.3% against 22.2%. Liver injury judged at least possibly related to treatment was similar, 4.6% against 4.2%, and the companies name it an important safety signal for olorofim. People on olorofim left hospital earlier, after an average of 17 days against 28.

Who these patients are

The OASIS participants were largely immunocompromised: about 79% had a blood cancer and/or a stem cell transplant, about 11% had received an organ transplant, and about 40% had very low white cell counts around the time they entered the trial. They were enrolled because a mould-active azole was inappropriate or had failed: known or suspected resistance of the mould, intolerance, interactions with their other medicines, infection that broke through azole prophylaxis, or failure of earlier azole therapy. For this group the existing alternatives are given into a vein, such as amphotericin B; in this trial drug-related kidney adverse events were reported in 22.2% of the comparator group.

Olorofim belongs to a new class called orotomides. In laboratory experiments on fungal isolates it inhibited fungal dihydroorotate dehydrogenase, an enzyme of pyrimidine biosynthesis, a different target from existing antifungal classes. An earlier single-arm phase 2b study in 202 patients with few or no treatment options, published in The Lancet Infectious Diseases in 2025, found a successful overall response in 28.7% at day 42 and liver enzyme rises at least possibly due to the drug in 10%. That study had no comparison group; OASIS provides the randomised comparison it lacked.

What still stands between the result and patients

These results come from the companies' release and a conference presentation. A peer-reviewed paper with the full analysis has not yet appeared, and the release gives no subgroup analysis or detailed baseline results. The trial was funded by the manufacturer. Blinding covered the committee that judged outcomes and the sponsor, so it was not a double-blind trial. The comparison was with amphotericin B followed by standard care, not with the azoles that most patients receive first, so it says nothing about people for whom azoles work.

F2G says it expects to resubmit its application to the US Food and Drug Administration by the end of 2026, and Shionogi plans filings in Europe and Asia. Olorofim already has orphan and breakthrough therapy designations in the US and orphan designation in the EU, which can speed review but say nothing about the result. For a patient or family facing this infection after cancer treatment or a transplant, treatment options are a question to discuss with the treating haematologist or infectious disease specialist.

This material is for information only and does not replace a consultation with a doctor.

Sources

• F2G and Shionogi. Data from the global phase 3 OASIS study of olorofim versus AmBisome followed by standard of care in invasive aspergillosis, press release, 21 October 2026. shionogi.com

• OASIS trial registration, ClinicalTrials.gov. NCT05101187

• Maertens JA et al. Olorofim for the treatment of invasive fungal diseases in patients with few or no therapeutic options: a single-arm, open-label, phase 2b study. Lancet Infect Dis. 2025. PMID 40541222

• Oliver JD et al. F901318 represents a novel class of antifungal drug that inhibits dihydroorotate dehydrogenase. Proc Natl Acad Sci USA. 2016. PMID 27791100