Paxlovid (nirmatrelvir-ritonavir) does not reduce hospital admissions or deaths in vaccinated adults at higher risk of severe COVID-19, despite helping them recover faster, according to results from two national trials published today in the New England Journal of Medicine. With COVID-19 still causing significant illness and death, clinicians urgently need to know which patients benefit most from early antiviral treatment.
The findings, from 4,232 participants across the two trials, help clarify who should receive Paxlovid now that widespread vaccination has dramatically reduced the risk of severe outcomes.
The UK PANORAMIC trial, funded by the National Institute for Health and Care Research (NIHR) and led from the University of Oxford, enrolled 3,516 participants with COVID-19 across its Paxlovid arm (April 2022–March 2024). The Canadian CanTreatCOVID trial, led by Dr Andrew Pinto, director of Upstream Lab (Unity Health Toronto), enrolled 716 participants (January 2023–September 2024).
Both trials recruited adults aged 50 or over, or younger adults with additional conditions such as diabetes or asthma. Over 98% of all participants were vaccinated.
Paxlovid was approved in 2021 based on trials showing an 88% reduction in hospitalisation or death among unvaccinated high-risk adults. The two new trials tested the drug in mostly vaccinated higher-risk patients – a population very different from the one in which Paxlovid was originally studied. In May 2025, the National Institute for Health and Care Excellence (NICE) restricted routine NHS use to a narrower 'highest-risk' cohort (e.g. transplant recipients, severe liver disease), citing cost-effectiveness concerns.
"In today's highly vaccinated populations, the benefits of Paxlovid have fundamentally changed," said Professor Christopher Butler, Nuffield Department of Primary Care Health Sciences, University of Oxford, who led the UK trial. "While people feel better sooner from treatment with this important antiviral drug, we found no reduction in the already low rate of hospitalisations or deaths. This provides essential evidence for optimal, cost-effective targeting of this treatment."
The UK PANORAMIC trial was designed, coordinated and led by the University of Oxford's Primary Care Clinical Trials Unit, working with the NIHR Research Delivery Network across England and by Health and Care Research Wales, NHS Research Scotland, the Health and Social Care Board in Northern Ireland.
Across multiple treatments, the trial recruited over 29,000 participants. To complement recruitment through general practices, the Primary Care Clinical Trials Unit pioneered remote participation methods – online consent, medication dispatched through its in-house dispensing facility, and self-collected samples – removing many traditional barriers to taking part and thus 'democratising trials research.'
Participants taking Paxlovid recovered substantially sooner. In PANORAMIC, median recovery time was 14 days with Paxlovid compared to 21 days with usual care. CanTreatCOVID found recovery at 6 days versus 9 days, respectively, but measures recovery in a slightly different way. Paxlovid also significantly reduced viral load by day 5, which may have implications for reduced opportunity for spread.
The safety profile matched what is already known about Paxlovid. In PANORAMIC, 90.4% reported at least one side effect – predominantly altered taste (dysgeusia) and gastrointestinal symptoms typical for this drug – with around 8% discontinuing treatment because of side effects. Serious adverse events remained rare in both groups.
"These trials demonstrate precisely what evidence-based policy and medical care should look like – rigorously testing treatments as conditions change," said Professor Jonathan Van-Tam, former UK Deputy Chief Medical Officer and Co-investigator, University of Nottingham. "The UK led the world in COVID treatment trials because we have unified health systems, strong academic trial units, and standing research infrastructure through the NIHR – capabilities that are critical for future pandemic preparedness."
Very few participants in either trial were hospitalised within 28 days. In PANORAMIC, rates were 0.8% with Paxlovid versus 0.7% with usual care. In CanTreatCOVID, rates were 0.6% versus 1.2%. Neither study found a statistically significant difference. There were no deaths in either trial during the Paxlovid recruitment period.
This joint publication demonstrates how independent trials using similar protocols can deliver more robust findings than multiple smaller studies.
Oxford's Primary Care Clinical Trials Unit and Toronto's Upstream Lab are now applying these methods to trials of other respiratory infections through the ECRAID-Prime and TreatResp trials – carrying forward the research infrastructure developed during the pandemic to benefit broader public health and prepare for future outbreaks.
"These results demonstrate the value of rapid, large-scale evidence generation through NHS partnerships," said Professor Phil Evans, National Associate Director of Health and Care in the NIHR Research Delivery Network. "The UK's research infrastructure, built through the NIHR, and academic clinical trials units allowed us to generate this evidence quickly."
These findings apply specifically to higher-risk vaccinated adults (aged 50 or over, or aged 18 or over with comorbidities such as diabetes or chronic lung disease). They do not change recommendations for the highest-risk immunocompromised patients, for whom Paxlovid remains first-line treatment under current NICE guidance (TA878). Treatment should be started as soon as possible after symptom onset.
"Treatment decisions must evolve as the pandemic landscape changes," said Professor Butler. "The research infrastructure and methods developed through these trials provide a model for rapidly evaluating treatments during winter epidemics and future health emergencies."
