Preprint Mandates Are Real: What Clinical Authors Need to Know About medRxiv and openRxiv in 2026

For most of its history, posting a preprint was something researchers did by choice. If you wanted to share your work early, stake a priority claim, or gather feedback before peer review, you posted to bioRxiv or medRxiv. If you preferred to wait for the journal version, no one could stop you. That era is closing. As of 2026, a growing number of major research funders require preprints, not as a suggestion but as a condition of funding. Two of the most consequential mandates are now in force, and the platform that hosts the world's largest biomedical preprint server has itself gone through a significant transition. Understanding what has changed is not optional for researchers funded by HHMI or the Gates Foundation, and it is increasingly worth knowing for everyone else.

The changes stack: a new nonprofit governance structure for bioRxiv and medRxiv, a new DOI system that took effect in December 2025, a rebuilt commenting platform launched in April 2026, and funder mandates that now specify not just whether you post but when, under what license, and whether a revised preprint is required after peer review. This post works through each of those changes in the order that matters to a clinical or medical researcher preparing to submit a manuscript.

The openRxiv Transition: What Changed at bioRxiv and medRxiv

Since bioRxiv launched in 2013, both it and its medical counterpart medRxiv (launched in 2019) had been operated by Cold Spring Harbor Laboratory (CSHL) in New York. That arrangement made sense when these were small, experimental infrastructure projects, but the platforms grew far beyond what a single research institution could comfortably absorb as a side project. By early 2025, bioRxiv was receiving hundreds of thousands of submissions per year, and medRxiv had become the primary venue for early clinical and epidemiological research.

In March 2025, Cold Spring Harbor formally transferred stewardship to openRxiv, a newly created independent nonprofit backed by a $16 million grant from the Chan Zuckerberg Initiative. Richard Sever, one of the original co-founders of bioRxiv, moved from his role at CSHL Press to become chief scientific strategy officer of the new organization. The practical effect for authors submitting to medRxiv or bioRxiv was minimal at first. The same URLs, the same submission systems, and the same screening processes carried forward. What changed was governance and long-term financial stability. An independent nonprofit can raise funds and plan roadmap features without being subject to a parent institution's competing priorities.

For clinical authors, the most visible consequence of the transition has come later. In December 2025, openRxiv switched all new preprints to a new DOI prefix, 10.64898, replacing the older CSHL-associated prefix. If you posted a preprint before December 1, 2025, your existing DOI is unaffected. If you post now, your preprint will carry the new prefix. Both formats remain resolvable and citable. The change reflects the organizational separation from CSHL and also reinforces one of openRxiv's stated principles: a single DOI per article, across all versions, always resolving to the most recent version. This is different from how some publishers handle versioned documents, and it has practical implications for citation.

Hypothesis Commenting: The April 2026 Upgrade

Preprint commenting has always been one of the mechanisms by which the scientific community is supposed to improve work before formal peer review. In practice, commenting on medRxiv under the old Disqus system was modest. More than 29,000 comments had accumulated across both platforms since launch, which sounds substantial until you consider the volume of papers posted. Engagement was uneven, with high-profile preprints attracting attention and most others receiving very little.

On April 7, 2026, openRxiv replaced the Disqus commenting system on medRxiv and the Neuroscience category of bioRxiv with a new system powered by Hypothesis, an open-source annotation platform. The near-term plan is to expand Hypothesis to all bioRxiv categories. The new system brings author email notifications, meaning you will now receive an alert when someone leaves a public comment on your preprint. This sounds small, but it matters because one of the consistent criticisms of preprint commenting was that authors rarely knew when comments existed. Notification closes that loop.

The change is also significant for researchers who follow others' work. Hypothesis supports annotation by outside parties on any web page, not just on the server itself, and its integration with academic platforms is broader than Disqus ever was. Whether this actually increases substantive feedback on clinical preprints remains to be seen, but the infrastructure barrier is lower than it was six months ago.

The HHMI Mandate: Preprints Before Journal Submission

Howard Hughes Medical Institute announced its Immediate Access to Research policy in 2025. Enforcement began on January 1, 2026, and it applies to articles submitted for publication on or after that date. The mandate covers HHMI Investigators, Freeman Hrabowski Scholars, and scientists at the Janelia Research Campus, specifically when a member of the lab is listed as first, last, or corresponding author.

The core requirement is straightforward but represents a genuine change in workflow: an initial preprint must be posted before the manuscript is submitted to any journal for the first time. This is a timing condition, not a retrospective one. You cannot satisfy it by posting a preprint after the journal sends the manuscript out for review. The preprint must precede first submission.

The policy also requires a revised preprint when peer review feedback, new results, or new analysis lead to substantial revisions of the original work. HHMI defines a revised preprint as an additional version posted when revisions are substantive, not for minor language edits or formatting changes. What counts as substantive will require judgment, but if reviewers requested new experiments, significant reanalysis, or major rewriting, a revised preprint is expected.

Two further requirements apply to license and hosting. Preprints must be posted under a Creative Commons Attribution 4.0 International license (CC BY 4.0), which allows anyone to reuse and redistribute with attribution. HHMI designates bioRxiv, medRxiv, ChemRxiv, and arXiv as preferred servers depending on discipline, and for clinical and biomedical research, medRxiv is the appropriate venue. The policy also signals a shift in how HHMI itself evaluates its researchers: preprints, not just published papers, will form part of the basis for internal assessments of scientists' work.

The Gates Foundation Mandate: Running Since January 2025

The Bill & Melinda Gates Foundation's preprint requirement is slightly older than HHMI's. It took effect on January 1, 2025, and applies to research manuscripts funded by the foundation. Under the updated policy, all Gates Foundation grantees are required to post funded manuscripts as preprints on a recognized preprint server before or at the time of submission to a journal. Unlike HHMI, the Gates policy does not require a revised preprint, but the initial posting obligation is clear.

The Gates Foundation also made a parallel decision on Article Processing Charges. As of January 2025, the foundation no longer pays APCs for hybrid journals, the subscription-access journals that charge per-article fees to make individual papers open access. This is a meaningful shift. For grantees whose preferred journals are hybrid models, it changes the financial calculus of journal selection. Work funded by Gates now has to find its open-access route through fully gold open-access journals, institutional agreements, or the preprint server itself. In 2026, the foundation has also entered into a partnership with openRxiv, reinforcing medRxiv and bioRxiv as the preferred venues for Gates-funded life and health science research.

Researchers who receive Gates funding and publish in global health, infectious disease, maternal health, or agricultural science are the most directly affected. If you are unsure whether a specific grant triggers the mandate, check your award agreement and the foundation's open access compliance page directly, as the detailed conditions can vary by program area.

What medRxiv Actually Accepts and How Screening Works

For researchers posting to medRxiv for the first time, it is worth understanding what the screening process does and does not do. medRxiv screens submitted preprints for basic quality standards before they go live, but this is not peer review. Screeners check whether the submission is recognizable as a scientific manuscript, whether it contains potentially dangerous clinical information (such as claims about unproven treatments for serious conditions), and whether it falls within medRxiv's scope of health sciences. Routine screening typically takes one to two days.

Clinical case reports with identifiable patient information, promotional material, and non-scientific content are excluded. Preprints describing clinical trials should carry a trial registration number. Papers that would ordinarily be posted to bioRxiv (basic biology without direct clinical application) are redirected there. You cannot post the same paper to both bioRxiv and medRxiv; attempting to do so results in withdrawal.

One practical point that catches authors off guard: medRxiv is not anonymous. Author names, affiliations, and contact information are part of the public record from the moment of posting. If your co-authors are concerned about public exposure before journal acceptance, that conversation needs to happen before the preprint goes up, not after. This is especially relevant for clinical research where the corresponding author's identity may attract media attention.

How Major Journals Handle Prior Preprint Posting

Most major biomedical journals now accept manuscripts that have been posted as preprints. NEJM, The Lancet, JAMA, BMJ, and PLOS Medicine all explicitly permit prior preprint posting, and many require or encourage authors to disclose the preprint DOI at submission. If you are posting a preprint in order to comply with an HHMI or Gates mandate, the journal almost certainly allows it, but the specific expectations vary.

Some journals ask you to indicate the preprint DOI in the cover letter or submission form. Some have explicit fields in their submission systems labeled something like Prior Preprint or Posted Preprint. Failing to disclose a preprint at submission is not treated as a minor clerical omission at journals that ask for this information. It can be viewed as an incompleteness in the submission record, similar to not disclosing a conflict of interest. The correct practice is to include the medRxiv DOI proactively, state what changed between the preprint and the submitted manuscript if revisions were made, and note the CC BY 4.0 license if applicable.

One area where journal policies differ is on whether the preprint can be updated during peer review. Most journals do not prohibit updates, but some ask that you not post a revised version while the paper is under active peer review at their journal. If the HHMI mandate requires you to post a revised preprint after receiving substantive reviewer comments, there may be a brief tension with such policies. In practice, the right approach is to contact the editorial office if you are unsure. Journals are increasingly used to this situation, and a short, specific inquiry is more useful than making an assumption.

Why Researchers Post Preprints, and Why It Matters Beyond Compliance

For researchers who are posting under mandate rather than by choice, the benefits of preprinting can feel secondary to the compliance obligation. But the 2023 openRxiv community survey, which drew more than 7,000 respondents, shows that even authors who are not mandated to post cite real strategic reasons for doing so. The most common motivation was increasing awareness of the research. The next most common were controlling the timing of dissemination, staking a priority claim, and having citeable work available for grant applications before the peer review process completes.

For clinical researchers, the priority claim function deserves attention. In fields where multiple groups are working on similar problems, a preprint with a clear DOI and a timestamp establishes precedence in a way that journal submission does not. The journal's online publication date can be months or even more than a year after submission. A preprint posted the day you submit is public record immediately.

There is also a signal value in how the field responds to early posting. medRxiv preprints that receive substantive comments or media attention before peer review carry more information to reviewers and editors than papers that arrive cold. That is not always an advantage, particularly if early feedback exposes a flaw you had not noticed. But for strong work, early public visibility is generally a net positive.

Managing Versions and Long-Term Citability

One of the practical anxieties around preprints is the question of versions. If you post an initial preprint, then update it after peer review, and then the published paper differs from both earlier versions, readers and citers face a choice. Which version should be cited? How do readers know which version someone who cited the paper actually read?

The openRxiv approach, where all versions share a single DOI that resolves to the most current version, is designed partly to address this problem. A reader who follows your DOI always lands on the current version. The version history is still accessible, and version-specific links exist for those who need to cite or compare a specific iteration. In practice, when the final published paper exists, that should be the preferred citation for most purposes, with the preprint DOI noted where the manuscript was publicly available before peer review. Some journals automatically deposit a link from the preprint to the final paper when they accept a manuscript; medRxiv supports this through its journal transfer tool and established links with major publishers.

For papers posted under HHMI's CC BY 4.0 license, there is also the question of downstream reuse. CC BY 4.0 allows anyone to copy, distribute, and adapt the work for any purpose, including commercially, as long as attribution is provided. This is the same license used by PLOS Medicine and similar fully open-access journals, and it is more permissive than the default copyright retained by many paywalled journals. If this concerns you, check your institutional copyright policy and discuss it with your co-authors before you post.

What NIH-Funded Researchers Need to Know

NIH has not issued a preprint mandate equivalent to HHMI or Gates Foundation. What it has done, as of July 1, 2025, is eliminate the previous twelve-month embargo on public access to NIH-funded papers. Under the current NIH Public Access Policy, any peer-reviewed manuscript arising from NIH funding must be deposited in PubMed Central and made freely available immediately upon publication. There is no delay permitted. This is a significant change from the prior policy, which allowed a twelve-month window between journal publication and required open access.

What this means in practice is that if you publish in a subscription journal using NIH funding, the journal must either provide immediate open access under an acceptable license, or you must deposit the accepted manuscript in PubMed Central immediately. Many journals that previously relied on the twelve-month embargo period have had to revisit their NIH compliance approaches. If your target journal's open-access and NIH compliance information is unclear, checking directly with the editorial office or with your institution's library is worthwhile before you submit.

Separate from the access policy, NIH has been working through a proposal to cap allowable article-processing charges that grantees can charge to grants. Various options have been under discussion, ranging from a flat cap of around $2,000 to conditional caps depending on journal tier. No final rule has been issued as of this writing in May 2026, but if you are planning a submission to an expensive fully open-access journal using NIH funding, monitoring whether an APC cap takes effect before your publication date is prudent.

A Practical Workflow for Compliant Preprint Posting

Pulling these pieces together, the question for a clinical author in 2026 is how to build preprint posting into the normal manuscript preparation workflow rather than treating it as an afterthought. The simplest approach is to treat the medRxiv submission as a parallel process to journal selection, not a sequential one that comes after.

The sequence that works well: once co-authors have approved a draft for external review, prepare the medRxiv submission in parallel with identifying your target journal. Confirm the license to use (CC BY 4.0 for HHMI and Gates-funded work, other CC licenses or standard copyright for others). Submit to medRxiv and obtain the DOI, usually within one to two working days. Then submit to the journal, disclosing the preprint DOI in the appropriate field or cover letter. If substantial revisions are later required and you are under HHMI's mandate, plan the revised preprint as a deliverable in the same revision cycle as the revised manuscript, not a separate task to handle later.

One administrative detail that is easy to miss: when you submit to medRxiv, you are asked to confirm that co-authors have approved the submission. For compliance purposes, document this approval. Some institutions are beginning to ask for records of preprint posting as part of grant reporting, and the co-author approval record protects the corresponding author if a dispute arises later.

Where This Is Heading

The funder mandates in place as of 2026 are likely early examples of a broader trend. ASAPbio, the organization that has been advocating for preprints in biology and medicine for years, has been tracking a growing number of research funders adopting preprint-posting requirements. HHMI and Gates Foundation are large and visible enough that their policies carry practical weight, but smaller funders in global health, European research councils, and public health agencies are watching the same landscape.

The infrastructure to support mandated preprints is also improving faster than it was two years ago. The openRxiv transition gave medRxiv a governance structure capable of responding to policy demands and building new features. The Hypothesis commenting launch in April 2026 is one early example of that. There are also ongoing conversations about structured peer review being associated with preprints, meaning that external reviewer comments might eventually be linked directly to medRxiv records. That would change the relationship between preprints and formal journals in ways that are still unpredictable, but the direction is clear: preprints are becoming a more permanent part of the record, not just a waiting room before publication.

For clinical researchers, the practical implication is to stop treating preprints as a separate activity reserved for certain types of papers or certain career stages. If you receive funding from any major funder, check whether a preprint mandate applies. If it does, build the posting step into your standard submission checklist. The overhead is genuinely small, and the cost of non-compliance, which can range from funder reporting issues to reputational friction, is not.

Further Reading