In December 2021, nirmatrelvir plus ritonavir (Paxlovid) was approved for emergency use in the treatment of COVID-19. The EPIC-HR clinical trial reported in the New England Journal of Medicine that Paxlovid led to an 89% reduction in hospitalization or death among unvaccinated patients with early COVID-19. However, the effect of Paxlovid among vaccinated populations was not studied and clinicians have been uncertain about potential contraindications.

In a population-based cohort study published in the Annals of Internal Medicine, Mass General Brigham researchers report Paxlovid is associated with a 44% reduction in hospitalization or death among highly vaccinated adults aged 50 years or older. Scott Dryden-Peterson, MD, MSc, an infectious disease specialist at Brigham and Women’s Hospital and medical director of COVID outpatient therapy at Mass General Brigham, and Ann Woolley, MD, MPH, associate clinical director of Transplant Infectious Diseases at the Brigham, led the study.

In this Q&A, they discuss the impact of their results, what clinicians should know when prescribing Paxlovid, and future COVID-19 research.

Q: Can you describe the goal of your study and the methodology to accomplish it?

We wanted to inform the community about the benefits of Paxlovid among vaccinated populations, as well as determine the best method for prescribing it. To answer this question, we created an emulated clinical trial using observational data from the Mass General Brigham Enclave, an electronic medical record database with information on patients who test positive for COVID-19 within our system. We estimated the effect of Paxlovid in preventing severe illness, which we defined as being hospitalized or dying from COVID-19, among vaccinated adults aged 50 years or older.

Throughout the country, there is a large difference in who gets prescribed Paxlovid and who doesn’t. We used the data to identify patients who were potentially eligible to receive Paxlovid. For the patients who didn’t receive a prescription, we used inverse probability-weighted models to account for things that could predict whether someone would be prescribed. This allowed us to estimate the effect of Paxlovid among a “pseudo-population,” without potential treatment bias based on factors like age, comorbidities, race and ethnicity, and vaccination status.

Q: What are the key takeaways from this study that clinicians should know?

There was a 44% reduction in the risk of being hospitalized or dying from COVID-19, after accounting for differences in those who were or were not prescribed Paxlovid. It had a stronger effect in preventing death than hospitalization. Both of those findings suggest that even in highly vaccinated people, and often in those who had prior COVID-19, Paxlovid reduced risk substantially.

The risk of severe outcomes of COVID-19 have decreased due to protection conferred by vaccination and prior infection. However, COVID-19 hospitalization and death remain common among patients with multiple medical problems. Early use of Paxlovid can prevent these outcomes.

Q: What were the benefits of utilizing a real-world cohort from Mass General Brigham?

The Mass General Brigham Enclave was instrumental in this study. It allowed us to emulate a clinical trial with a real-life cohort representing our Mass General Brigham patient population. So the outcomes we saw resonated with what we’re seeing clinically. Additionally, the analysis could be done in the midst of a strong wave to directly provide advice to patients and clinicians affected at that moment.

Especially with a large cohort, it’s helpful to have local data for safety reasons so clinicians can account for potential drug interactions and feel comfortable prescribing this. It also allows our institution to have a more strategic direction in how we prescribe Paxlovid. We need to ensure everyone can prescribe this and it becomes the standard of care for high-risk patients across the system.

Q: How can primary care providers determine if their patients should be prescribed Paxlovid?

Clinicians need to consider potential contraindications to prescribing Paxlovid. Many patients who can benefit from this treatment are receiving a medication that they may need to pause to safely receive Paxlovid. They need to weigh the risk of holding an existing medication with the benefits of Paxlovid. Most of the severe adverse reactions to Paxlovid reported to the FDA involved drug interactions, particularly the transplant medication tacrolimus.

If Paxlovid is not in your patient’s best interest due to contraindications, there may be an alternative agent such as remdesivir or molnupiravir.

Antiviral treatments work against the virus when it’s actively replicating, so they are only effective if given within the first five days of a positive test. We should encourage prompt testing by patients and ensure that they let their provider know even if symptoms are mild.

Q: Are there any ongoing or upcoming COVID-19-related studies from your team?

We’re fortunate to have treatments in our repertoire to help protect patients when they get COVID-19, but we want to really understand who is at high risk of severe illness. This will help us determine who should be prioritized for antiviral therapy. We also see marked differences by geographic area, race, ethnicity, and income status in terms of who can access treatment in a timely way. Further understanding those disparities could lower barriers for communities that have been historically disadvantaged in their care.