PRINCIPLE trial finds antibiotics azithromycin and doxycycline not generally effective treatments for COVID-19

Azithromycin and doxycycline are two commonly used antibiotics investigated as separate treatments in the trial. Both drugs are being used by some doctors in the hope of treating COVID-19 in the early stages of the illness.

After reviewing interim analyses of both the azithromycin and doxycycline arms of the PRINCIPLE trial, the independent Trial Steering Committee advised the trial investigators, who concluded that there is no beneficial effect in patients aged over 50 who are treated with either antibiotic at home in the early stages of COVID-19. This is because interim data analyses show that neither treatment reduces the time taken for people to first report that they feel recovered sufficiently to achieve meaningful clinical benefit. The PRINCIPLE trial did not look at the effects of these drugs in patients who were already admitted to hospital with COVID-19.

No further people are therefore being randomised to receive azithromycin or doxycycline in the PRINCIPLE trial.

Summary of the key clinical findings:

1. Azithromycin: A total of 526 eligible participants were randomised to azithromycin (500 mg once daily for 3 days) within the first 14-days of onset of COVID-19 and compared with 862 participants randomised to usual care. Per protocol, randomisation into the azithromycin arm was stopped for futility due to a small probability (0.024) of clinically meaningful benefit compared to usual care at an interim analysis with incomplete data. After 28-days of follow-up on the randomised participants, the results showed the estimated median time to self-reported recovery for azithromycin was 0.94 days shorter compared to usual care (95% Bayesian credible interval -0.56 to 2.43 days), with a low probability (0.23) of being a meaningful benefit. Also, there was no evidence that azithromycin reduced hospitalisations or deaths compared with usual care (model-based estimated increase in hospitalization rate of 0.3%, Bayesian credible interval: -1.7% to 2.1%).
2. Doxycycline: A total of 798 eligible participants were randomised to doxycycline (200 mg on the first day followed by 100 mg a day for 6 days) within the first 14-days of onset of COVID-19 and compared with 994 participants randomised to usual care. Per protocol, randomisation into the doxycycline arm was stopped for futility due to small probability (0.044) of a clinically meaningful benefit compared to usual care at an interim analysis. Based on the interim incomplete data, both the estimated clinical benefit in the time to recovery (less than 1 day benefit) and hospitalisation rate (less than 2% benefit) is small for doxycycline. Final results for the doxycycline arm of the PRINCIPLE trial will be published after the full 28-day follow-up on all randomised participants are available. 

Additional details and analyses on both the azithromycin and doxycycline arms of the PRINCIPLE trial will be shared in the near future in peer-reviewed journals.

Recruitment into the budesonide arm of the PRINCIPLE trial continues as planned.

Clinical guidance in England currently recommends treatment with oral doxycycline for suspected pneumonia in people with COVID-19 in the community if the cause is bacterial, or if it is unclear whether the cause is bacterial or viral and the symptoms are particularly concerning. The PRINCIPLE trial shows that doxycycline is not effective as a treatment for suspected COVID-19 in the absence of bacterial pneumonia, and should not be used in this way.

Professor Chris Butler from the University of Oxford’s Nuffield Department of Primary Care Health Sciences and Co-Lead of the PRINCIPLE trial, said: ‘Azithromycin and doxycycline have anti-inflammatory, antibacterial and possibly antiviral effects, and so were considered as potential treatments for COVID-19 in the community. While we are completing the analysis of the full range of study outcomes, and in different patient groups, our findings show that a three-day course of azithromycin or a seven-day course of doxycycline has no important clinical benefit in terms of the time taken to feeling recovered, and so will not help most patients with COVID-19 in the early stages their illness. These are two important findings, as both azithromycin and doxycycline have been used for treating COVID-19 in the community even in the absence of suspected bacterial pneumonia, so this practice should now be re-considered - particularly because overuse of antibiotics in the community can fuel the emergence of antimicrobial resistance. PRINCIPLE is one of the first trials to report about doxycycline for COVID-19 worldwide, and as this drug is in common use for this condition, this should help guide prescribing decisions for COVID-19.

‘The PRINCIPLE trial has grown into a UK-wide community collaboration to find effective treatments for COVID-19 that can be used in the community. Treatments that can speed recovery and prevent hospitalisations are urgently needed worldwide. The trial is ongoing and will continue to evaluate COVID-19 treatments suitable for use in the community that might speed-up recovery and potentially reduce the need for hospital admission. Many more people are still needed to join the study because there are several other treatments that need to be tested.’

Professor Richard Hobbs, Head of the University of Oxford’s Nuffield Department of Primary Care Health Sciences and Co-Lead of the PRINCIPLE trial, said: ‘While it is disappointing that neither azithromycin nor doxycycline speed-up recovery for those with COVID-19 in the community, these are both important findings which will reduce the use of ineffective antibiotics for this illness. This finding shows the importance of doing rigorous clinical trials in real-world settings before treatments are rolled out on a wide scale. Widespread use of treatment should not be based on laboratory studies and opinion alone. We remain incredibly grateful for the huge efforts from many patients, GP practices and other organisations in delivering this national, flagship primary care study in these challenging times.’

Azithromycin and doxycycline are widely used in primary care to treat respiratory tract infections. Data from the Oxford-Royal College of General Practitioners Research (RCGP) and Surveillance Centre (RSC) shows that during the COVID-19 pandemic General Practitioners have been more likely to prescribe antibiotics to people with lower respiratory infections, and the use of azithromycin has increased compared with 2019, and doxycycline use has remained static despite a reduction in new cases of respiratory disease.

Professor Simon de Lusignan, Director of RCGP RSC and Co-Principal Investigator of the PRINCIPLE trial, said: ‘Whilst I fully understand why colleagues would want to do everything for patients who might have COVID-19, PRINCIPLE usefully provides evidence that prescribing these antibiotics for most people with possible COVID-19 is not helpful.’

The PRINCIPLE trial is primarily evaluating whether treatment prescribed in the first 14 days of COVID-19 illness can speed up recovery and prevent the need for hospital admission. It is open across the UK to people aged over 50 with certain underlying health conditions, or anyone aged over 65. Those with coronavirus symptoms, or a confirmed SARS-CoV-2 infection, can join easily from home online, over the telephone or via their GP practice from anywhere in the UK, without needing face-to-face visits with the trial team in Oxford. Any health and social care professional is able to introduce potential participants to PRINCIPLE and more than 3,700 people have joined to date.

Participants are randomised to receive either usual care, or to be treated with the addition of a trial treatment. Follow-up information is collected online or by telephone, and from medical records.

PRINCIPLE is funded by UK Research and Innovation and the Department of Health and Social Care through the National Institute for Health Research as part of the UK Government’s rapid research response fund.

Further information is available at www.principletrial.org