Jerry Cangelosi — medRxiv user

The nonprofit preprint servers bioRxiv and medRxiv post scientific manuscripts within days of submission for anyone to read. This fast and open availability allows readers worldwide instant access to timely science. In this account, medRxiv user Jerry Cangelosi (University of Washington) shares his experience with medRxiv during and after the COVID-19 pandemic.

At the outset of the COVID-19 pandemic in the United States, testing for the virus required the collection of nasopharyngeal (NP) swabs. Also known as “the brain tickle”, this sampling method was invasive, deeply uncomfortable for patients, and hazardous for healthcare workers. It was a huge bottleneck in COVID screening and diagnosis.

In March 2020 I had the privilege of joining a consortium led by Dr. Yuan-Po Tu (Everett Clinic and UnitedHealth Group) that evaluated patient-collected nasal (anterior nares) swabs as easier, less invasive alternatives to NP. Dr. Tu’s equivalency study found that the nasal swabs were only slightly less sensitive than NP (94.0% [97.5% CI, 83.8 to 100.0]). Based on these findings, on March 24, 2020, the FDA updated its guidance to permit self-collected nasal swabs for COVID testing. By summer 2020 the first commercial at-home nasal swab tests for COVID received Emergency Use Authorization. Over time, nasal swabbing became universally accepted, although sensitivities were slightly lower than our initial report when asymptomatic and early-stage infections were included in studies.

On March 24, 2020, the same day as the FDA announcement, we submitted our results to the New England Journal of Medicine for peer review. At this point, the slow machinery of peer review took over.

The intervening weeks were interesting. Despite the FDA’s endorsement, many clinicians were understandably hesitant to adopt the new sampling method, especially given the hectic regulatory environment of early 2020. Our inboxes were flooded with requests for data, protocols, and inquiries that amounted to “is this real?” In turn, we pressed the journal for accelerated peer review (we were unlikely to be the only ones doing so during that period).

The solution was to post a preprint on medRxiv. Once it went online on April 6, 2020, clinicians, researchers, regulators, test developers, and people in the community could refer to it for protocols, data, results, and citation. Over 8,000 users did so before the peer-reviewed version was published on June 3, 2020, along with an audio editorial. Importantly, the preprint was open-access and the entire world could see it. Clinicians began to adopt the method, making COVID testing more tolerable and accessible for their patients. Importantly, commercial test developers could immediately move forward with the knowledge that at-home self-testing was feasible. During a period in which the global burden of COVID-19 was increasing exponentially, it is not far-fetched to believe that the availability of the preprint accelerated COVID control efforts and saved lives.

Over the same period, my research group at the University of Washington, working with numerous collaborators around the world, moved forward with a parallel effort to validate a novel non-invasive sampling method for tuberculosis testing (tongue swabs). This effort culminated in WHO endorsement in March, 2026, and global implementation is under way. It moved at a less hectic, more traditional pace than the COVID effort, but was nonetheless very rapid by tuberculosis standards. Once again, the use of medRxiv and other preprint servers facilitated the process by accelerating data sharing, minimizing redundancy, and promoting cooperation. Count me as a believer in the power of rapid, open information sharing as an accessory to peer review.