From Patents to Regulatory Exclusivities in Drug Development: A Comparative Transaction Cost Analysis

3. Trade Secrets

While pharmaceutical firms rely primarily on patents, they also leverage trade secrets as an additional form of protection. Secrets around manufacturing processes (the prototypical domain for trade secrecy) are more important for biologics than for small molecules given how easy it is to reverse engineer a small molecule. However, even proprietary small molecule development pro­grams strategically leverage trade secrecy over core intellectual property to im­pede the progress of competitors. Firms maximize secrecy over progress on core activities such as hit identification, lead compound optimization, formula­tion, and dose selection until patents are filed, and even then disclose as little as possible in their patent filings. Moreover, they hold underlying preclinical and clinical data, as well as CMC information, as close to the chest as possible for as long as possible.

Protecting trade secrets runs, however, into countervailing considerations for firms. First, to develop a drug, a firm typically must share at least some of its data and results with outside firms and institutions. But maintaining trade secrecy protection over this information entails significant costs. The firm must enter into confidentiality agreements with each external actor and must absorb the costs of implementing security measures, training employees, monitoring collaborators, keeping track of the secrets, and enforcing its rights. In the ag­gregate, these efforts can significantly slow progress. Moreover, the firm fore­goes access to potentially invaluable feedback on its research program from the broader scientific community. Finally, as drug regulators increasingly require public disclosure of clinical trial results, the scope of trade secrecy protection is likely to diminish somewhat. However, disclosure of the underlying data itself is currently only mandated in Europe, and it is only required after the regulator renders a marketing authorization decision, which can be years after study com­pletion. Underlying trial data often remain secret indefinitely for failed clinical programs in which no marketing authorization is submitted. This can harm com­petitors who might otherwise avoid making similar investments.

Overall, firms benefit from trade secrets for strategic reasons. These bene­fits come at a cost, however, not only to the firm, but also to social welfare—for example, by decreasing access to knowledge by researchers, regulators, and prescribers; encouraging duplicative research investments by competitors; and, in respect of clinical trials, by putting additional patients at risk. We return to this subject later.

II. Regulatory Exclusivities in Drug Development

While patents and trade secrets form the foundation of traditional pharma­ceutical sector intellectual property strategy, in the 1980s, governments began introducing a parallel system of incentives for drug development, administered by health regulators, such as the FDA, rather than by patent offices. Since then, the number and scope of these regulatory exclusivities have blossomed: they include exclusivities over new chemical entities, orphan drugs, new uses for ex­isting drugs, new antibiotics, and new biologics.

Regulatory exclusivities provide firms with broad protections that, in some circumstances, exceed those of patent rights because they are free and difficult to invalidate. Because they were introduced after the pharmaceutical sector structured itself around patents and trade secrets, these exclusivities have taken on a supplemental, rather than primary, role in a firm’s intellectual property management strategy. This misses a critical opportunity, especially as the in­dustry increasingly relies on collaboration with outside actors. As we argue in Part III, structuring those collaborations around regulatory exclusivities, as op­posed to patents and trade secrets, significantly lowers transaction costs, in­creasing productivity. This Part explores the nature of these exclusivities and how they operate.

Except for orphan drug exclusivity, which provides complete exclusivity over a disease indication, regulatory exclusivities are based on the data submit­ted to regulatory bodies by sponsors to gain approval for marketing a drug or biologic. These exclusivities provide that, from the date a drug or biologic is approved by the regulator, no other party may make reference to that drug to gain abbreviated approval of a follow-on drug with the “same” active ingredient (or “highly similar” active ingredient in the case of biologics). That is, a com­petitor firm cannot ask the regulator to rely on the original data submitted by the drug’s sponsor to obtain marketing authorization.

The period of exclusivity depends on the jurisdiction and whether the ap­proval is for a small molecule or a biologic. But in effect, during the period of exclusivity, a competitor wishing to sell a drug containing the same molecule or biologic would have to develop its own NDA, complete with full independent preclinical and clinical studies. At the same time, the competitor would not be entitled to any exclusivities of its own because its drug would not be considered “new.” As such, the enormous expense entailed would be extremely difficult to justify to the point that no firm is likely to do this except in extraordinary cir­cumstances.

Below we describe the core features of the primary exclusivity incentives offered by health regulators. Critically, none requires a drug sponsor to have filed patents covering the drug or to have withheld preclinical and clinical data from prior disclosure (though, for the reasons given above, sponsors are likely to have done both in most cases).

A. New Chemical Entity (NCE) Exclusivity

At the same time as it created pathways for the approval of generic drugs, the Drug Price Competition and Patent Term Restoration Act of 1984 (the Hatch-Waxman Act) implemented countervailing exclusivity provisions to en­sure continued incentives for drug development. Under the Hatch-Waman Act, if a sponsor registers a new drug containing a “new chemical entity” (an active ingredient that has not been previously approved), the FDA cannot accept a sub­mission from a competitor seeking to abbreviate its regulatory burden by “ref­erencing” the sponsor’s drug for a period of five years from the sponsor’s original approval. As it usually takes the FDA approximately a year to process an application, this provides a de facto exclusivity period of approximately six years. If Orange Book listed patents are involved, a would-be competitor can file a “paragraph IV certification” asserting invalidity or noninfringement of the listed patents to accelerate the FDA’s acceptance of its abbreviated submission to four years after initial approval. In practice, however, this shortened period is offset by the right of the sponsor of the original product to an automatic thirty-month stay of FDA approval upon initiation of infringement proceedings.

Other major pharmaceutical markets offer similar protections for drugs containing new chemical entities. The European Medicines Agency (EMA) can­not accept an application for a generic version of the original drug for a period of eight years (the data exclusivity period) and then cannot approve a generic application for an additional two years (the market exclusivity period), for a combined exclusivity period of ten years. Japan’s Ministry of Health, Labor, and Welfare provides eight years of data exclusivity. Meanwhile, Health Can­ada provides a six-year period of data exclusivity and an additional two years of market exclusivity for “innovative drug[s]” that contain new chemical entities, during which competitors may not make “a direct or indirect comparison” to the innovative drug to seek abbreviated approval.

These exclusivities effectively block follow-on submissions for drug prod­ucts containing the same active ingredient. This includes not only exact generic copies for use in the same indication—for example via the FDA’s abbreviated new drug application (ANDA) pathway—but also those attempting to refer to the pioneer product to seek partially abbreviated approval for additional indica­tions, dosage forms, or routes of administration. An example of this would be using the 505(b)(2) application pathway in the United States. For research in­volving a new active ingredient, new chemical entity exclusivity thus represents a powerful mechanism through which firms can protect their product markets.

B. Orphan Drug Exclusivity

To encourage the development of drugs to treat rare diseases or condi­tions, the U.S. Orphan Drug Act established a seven-year exclusivity period for new rare disease, or “orphan,” drugs during which the FDA cannot, without the consent of the drug sponsor, approve any application to market the same drug for the same disease indication. To avail itself of this exclusivity, a spon­sor must submit a description of the drug it is developing and the target popula­tion for the drug. If the FDA agrees that the indication qualifies as a rare disease or condition, the FDA will designate the drug under development as an orphan product. Giving broader protection than for ordinary drugs, the exclusivity not only bars subsequent abbreviated applications but also protects the drug from all competitors for the same indication, even those that would theoretically wish to conduct their own independent studies to prepare a full NDA. The orphan designation also qualifies a sponsor for various development incentives, includ­ing clinical research protocol assistance, grants to fund clinical studies, tax cred­its for clinical research costs, waivers of marketing application fees, and eligibility for accelerated review.

Similarly, the EMA provides a ten-year period of orphan drug exclusivity during which it cannot, without the consent of the holder of orphan drug status, either accept or grant an application for a similar medicinal product in the same therapeutic indication. Like the FDA procedure, a sponsor must submit an ap­plication to the EMA before seeking marketing authorization that describes the medicinal product and the proposed therapeutic indication. In addition to ex­clusivity, the EMA provides various drug development incentives, including protocol assistance, fee reductions, and the potential for grants from member states or the European Commission.

In both the United States and European Union, an application for orphan designation can be made at any stage of product development before the sub­mission of a marketing authorization application. In practice, however, the spon­sor generally waits until it has preclinical or preliminary clinical data to support its therapeutic plausibility. Other leading pharmaceutical markets including Singapore, Japan, Australia, Taiwan, and South Korea have followed suit with their own orphan drug regimes.

C. Pediatric Exclusivity

In response to concerns that many drugs are not tested in children, the Food and Drug Administration Modernization Act implemented pediatric exclusivity, which extends existing patents or regulatory exclusivities held by the sponsor for a period of six months in exchange for the completion of pediatric studies. Canada similarly provides a six-month extension of data exclusivity on “inno­vative drugs” in exchange for the completion of pediatric studies.

For its part, the European Union adopted a regulation governing medicinal products for pediatric use in 2006. The regulation requires the drug sponsor to submit a Pediatric Investigation Plan (PIP) during early clinical development of all medicinal products. Waivers are available if the drug is likely to be ineffective or unsafe in children, if it is intended for conditions that only occur in adult populations, or if the product does not represent a significant therapeutic benefit over existing treatments. Firms granted orphan status and complying with an approved PIP gain two years of additional orphan drug exclusivity, for a total of twelve years of market protection. Firms that have a patent and sat­isfy a PIP are also eligible for six months of patent extension.

D. Exclusivity for New Uses

A new application for a drug product that contains an active ingredient pre­viously approved by the FDA, though not eligible for new chemical entity ex­clusivity, is entitled to a distinct three-year period of exclusivity if the application contains reports of new clinical investigations (other than bioavail­ability studies). Three-year exclusivity is granted for various types of changes to a drug product supported by new clinical studies, including modified dosage forms, new indications, and switches from prescription to over-the-counter sale. This exclusivity bars subsequent competitor applications for the same conditions of use that rely on information supporting the NDA containing the new clinical investigations. Unlike NCE exclusivity, this exclusivity does not bar the submission of abbreviated follow-on applications, just their approval.

The European Union has a very different system to encourage firms to un­dertake new studies on previously approved medicinal products. There, a spon­sor is entitled to a one-year extension of its NCE exclusivity period if, during the first eight years of exclusivity, that sponsor obtains authorization for one or more new therapeutic indications that bring a significant clinical benefit in com­parison to existing therapies.

E. Exclusivity for New Antimicrobials

The Generating Antibiotic Incentives Now (GAIN) Act was passed by the U.S. Congress in 2011 to stimulate the development of new antibiotics and an­tifungal medicines. The Act provides incentives for the development of “qual­ified infectious disease products” (QIDPs), meaning drugs “intended to treat a serious or life-threatening infection.” The most significant incentive in the Act is that a sponsor of a QIDP is entitled to five years of additional regulatory exclusivity for the drug product, which is added to any NCE, orphan drug, or three-year new use exclusivities to which the product is otherwise entitled. Notably, scholars have argued that the QIDP criteria are far too broad in that they do not require the qualifying products to contain NCEs, leverage new mechanisms of action, or address unmet therapeutic needs. Owing to risk aver­sity, firms have largely been incentivized to pursue QIDP designations for new uses or incremental modifications to existing drugs rather than for new ones.

The European Union has yet to adopt an enhanced exclusivity regime for new antibiotics. However, the European Commission published a draft proposal in 2023 that, to address the emergence of antimicrobial resistance, would pro­vide a transferable exclusivity voucher to sponsors of new “priority antimicro­bials” that (1) represent a new class of antimicrobials, (2) have a mechanism of action distinct from other authorized antimicrobials in the European Union, or (3) contain an active substance not previously authorized in the European Un­ion that addresses a multidrug resistant infection or a serious or life threatening infection. A sponsor receiving the voucher would be able to use it to extend its data exclusivity on any one of its products, or sell it to another marketing authorization holder. While potentially politically feasible and more narrowly tailored to address unmet needs than the U.S. QIDP criteria, some have critiqued the proposal as inefficient from a cost-benefit analysis perspective and as un­fairly transferring the cost of new antimicrobial drug development onto patients requiring other drug products (who would continue to bear monopoly prices on an unrelated product during the period of extension).

F. Exclusivity for New Biologics

While principally beyond the scope of this Article, in part in response to concerns over the scope of allowable patents in relation to the lesser standard of “similarity” required in follow-on biologics (or “biosimilar”) applications, the U.S. Congress implemented an enhanced regulatory exclusivity scheme for new biologics as part of the Affordable Care Act in 2010. New biologics are enti­tled to twelve years of regulatory exclusivity, including four years before a bio­similar application can be submitted to the FDA and an additional eight years before the FDA may approve the application.

III. Comparative Costs Under Patents and Secrecy Vs. Regulatory Exclusivities Alone

Patents and regulatory exclusivities are parallel state-created entitlements designed to address underinvestment in research and development. The inno­vation that results from research and development is critical to economic growth. In most cases, firms do not have to choose between patent and exclu­sivity protections as they are not mutually exclusive. Nor must firms cede trade secrecy protections beyond what is required to obtain patents and satisfy limited regulatory requirements for public disclosure. This Part compares the relative private and social costs of conducting drug discovery and development under a paradigm that maximizes patenting and secrecy versus one that foregoes these practices and relies solely on regulatory exclusivities for eventual market protection. As backdrop, we first outline an important trend in the pharmaceuti­cal innovation paradigm towards increased collaboration that has significant im­plications for the relative social costs assessment.

A. The Rise of Drug Development Collaboration

Pharmaceutical firms have, for the last few decades, moved from a fully integrated in-house research and development process to one that is disaggre­gated, involving collaborations between many actors in both the public and pri­vate sectors who bring distinct skills, financing, and tools. Policymakers and funders are increasingly aware of the benefits of collaborative research models and have thus pushed researchers towards their adoption. For example, then Vice President Joe Biden urged researchers participating in the Cancer Moon­shot to “[b]reak down silos and bring all the fighters together—to work together, share information, and end cancer as we know it.” Major health research fun­ders including the Bill & Melinda Gates Foundation, the Wellcome Trust, the Medical Research Council (MRC), the National Institutes of Health (NIH), and the Canadian Institutes of Health Research (CIHR) have adopted policies and guidelines promoting open sharing of knowledge and data and open scientific collaboration by funding recipients.

The move from integrated pharmaceutical firms to drug discovery partner­ships represents a change from hierarchical to hybrid forms of organization. Two main features of drug discovery in recent decades have driven this change. The first of these is that the stock of ideas is increasing so that today’s scientists stand on the shoulders of yesterday’s giants. Benjamin Jones found that, as a result, researchers not only take longer to absorb foundational knowledge in their field but become more specialized so as not to be over­whelmed. Second, and relatedly, researchers compensate for this higher spe­cialization by conducting research in collaboration with others who bring different skills and knowledge.

Hybrid organizational forms emerge as a solution to coordinating diverse and specialized actors engaging in projects with high uncertainty, where it will likely be necessary to change directions during the execution of the project. Pharmaceutical research and development entails both significant specialization and high uncertainty, and the higher success of drugs developed in coordi­nated partnerships demonstrates the advantages of the hybrid form.

B. Comparative Effects of Patents and Regulatory Exclusivities on Drug Discovery

Studies highlight the importance of intellectual property to advance drug development, including within drug development collaborations, but often neglect to distinguish between the critical forms of intellectual property: patents, trade secrecy, and regulatory exclusivities. In collapsing the distinctions, these studies fail to investigate, or even acknowledge, the relative benefits and costs of each regime. We attempt to do so here, by first investigating the relative firm-level costs associated with patenting and secrecy in drug development compared to regulatory exclusivities alone, followed by a comparison of costs at the level of research partnerships and of society as a whole.

C. Comparative Firm-Level Costs

At the firm level, drug sponsors seek patents and trade secrecy protection to achieve product monopolies and stave off generic competition. The same benefit exists, however, with regulatory exclusivities. In fact, exclusivities pro­vide superior protection in several respects while entailing significantly lower firm-level costs.

Regulatory exclusivities are virtually costless to obtain (beyond the costs associated with seeking marketing authorization, which must be borne regard­less) and are automatically enforced by health regulators at no cost to the spon­sor. By contrast, firms relying on patents and secrecy must incur substantial costs in prosecuting patent applications across many jurisdictions, implement­ing and monitoring trade secrecy measures across the firm, and privately moni­toring and enforcing patents and trade secrets against infringers through litigation.

Moreover, unlike patents, regulatory exclusivities create incentives for firms to engage in clinical development and to pursue marketing authorization for products that are unpatentable. For example, a firm may not be able to obtain a patent due to prior disclosure of a compound that, while not novel or nonob­vious, has yet to be validated clinically. Regulatory exclusivities, in contrast, protect the significant investment required to generate preclinical and clinical study data for a new active ingredient regardless of whether that compound was previously disclosed in, or rendered obvious by, a prior art reference.

In addition, once granted, regulatory exclusivities provide firms with pro­tection directly tailored to their product markets and are practically immune from challenge. In contrast, patents, particularly secondary patents, face a high risk of being invalidated in litigation. They also risk being underinclu­sive in scope such that generic firms may be able to design around them through modifications to formulations, polymorphic forms, synthetic routes, manufac­turing methods, and chemical intermediates. Likewise, the validity and en­forceability of trade secrets in litigation is dependent on evidence of sufficiently robust efforts to protect the secrets from disclosure and proof that an alleged infringer did not independently develop the knowledge. These possibilities create substantial ex ante uncertainty for would-be product developers.

Finally, while the FDA reviews NDAs ever more rapidly, the total time from the authorization of clinical trials to approval to market a drug remains stubbornly at approximately eight years on average. Drug sponsors often must file patent applications early in a product life cycle to avoid inventions being rendered obvious by subsequent developments in the field, but this also results in a shorter effective patent term. Regulatory exclusivities, in contrast, do not take effect until the date of product approval, providing ex ante certainty with respect to the period of market protection, which is particularly relevant for products with longer clinical development horizons.

Firm-level advantages associated with regulatory exclusivities are not merely theoretical. The introduction of orphan drug exclusivities in both the United States and European Union led to a profound and sustained increase in both development efforts and product approvals for drugs to treat rare diseases compared to the immediately preceding period, despite the uniform availabil­ity of patents before and after introduction. Moreover, one of the main reasons that the United States implemented new chemical entity exclusivity was to en­courage clinical development and marketing of new chemical entities with little or no patent protection.

Lietzan reports that, between 2011 and 2014 alone, of the 105 new chemical entities approved by the FDA, 11 either lacked listed patents entirely or had patents expiring before the end of the exclusivity period. In a study covering 1998 to 2004, Junod found twenty-two new drug approvals protected by new chemical entity exclusivity had no corresponding patents listed. Lietzan pro­vides a number of other examples of new chemical entities approved with NCE exclusivity but without listed patents between 1984 and 2010.

Because of their lower costs, easier enforcement, and certainty, regulatory exclusivities may be particularly attractive to smaller firms with limited re­sources. As with programs protected by patents and secrecy, a sponsor’s regu­latory dossier and data package (and its rights to future regulatory exclusivities over them) can be transferred to larger firms later in the development pro­cess even if the firm producing the data does not itself file an application with the FDA, a topic to which we return later.

D. Comparative Partnership-Level Costs

As with firm-level costs, the core differences between patents, secrecy, and regulatory exclusivities manifest themselves in very different transaction costs in drug development partnerships. This Section explores two sets of transaction costs: those relating to the creation of collaborations and those involved in their ongoing governance.

In entering into a drug development partnership, collaborators will need to negotiate over various aspects relating to patents and secrecy, including rights to (and limits upon) use of preexisting patents and trade secrets, rights to patents developed through the partnership, and those derived from further (outside the partnership) research on partnership outputs, as well as revenue sharing from those patents. Collaborators will also need to contract over secrecy, at a mini­mum until patents are filed. Given the difficulty of valuing individual patents, they must either engage in extensive negotiations or postpone the decision, caus­ing uncertainty later.

The literature makes clear that negotiating an agreement with universities is difficult and often delays the start of projects by months. Some firms will circumvent the university by negotiating directly with university researchers, but this sets up uncertainty later over the retention of rights by the university. Often universities and academic researchers have nonmonetary motivations that diverge from those of their industry collaborators, adding complexity to the rela­tionship. Universities may acquire patents to enhance their reputations and rank­ings, or to deliver on promises to their boards and funders. Researchers may seek to acquire reputational gains, such as through publication in top journals or prizes, to attract top students to their labs, or to rank higher on grant applica­tions. These different motivations regarding patents and disclosure can com­plicate and slow down negotiations by six to twelve months. While standard form contracting, such as the Lambert toolkit in the United Kingdom, exists, these contracts still require significant follow-on negotiations.

Patents and secrecy also make governance of drug development collabora­tions more cumbersome in several ways. Research results must be reviewed pe­riodically and in advance of any publication to identify possible patentable inventions and to allow collaborators to remove information they deem confi­dential. If an invention emerges, the collaboration (or responsible party) must apply for patents and incur the costs of patent prosecution across jurisdictions. Meanwhile, applicants need to ensure that the invention and data supporting it are kept secret. Once a patent is issued, the collaboration (or responsible party) must monitor for any use of the invention by others and, if infringement is de­tected, must instruct lawyers to issue cease-and-desist letters, attempt to negoti­ate a settlement, or litigate. The allocation of rights and responsibilities for these activities often further complicates initial contract negotiations.

In addition, given the different skills needed along the drug development path, control over the drug will likely pass not only from one research team to another but from one organization to another. This entails costs of negotiating, documenting, and licensing or registering the assignment of applicable pa­tents.

In contrast, entering into and governing a collaboration that eschews patents and trade secrecy will generally enable significantly lower transaction costs and open new avenues of research. While negotiations over the sharing of data will still need to occur, given the absence of multiple preexisting and po­tentially future patent and secrecy rights, these should be simpler. These col­laborations are more compatible with the traditional academic incentive system as they require limited secrecy, permit open publications, and yet still increase reputation through networking with firms. Further, because of the lack of patents and stringent confidentiality obligations, standard form agreements are more plausible without the need for extensive follow-on negotiations. While this mode of collaboration precludes future market exclusivity based on patents, drug firms can still engage in collaborations without encumbering future regu­latory exclusivity rights they might wish to pursue through later development and submission of NDAs on new compounds. The lower transaction costs in­volved can also help mission-oriented nonprofits lead patent-free collabora­tions, particularly in areas of underserved therapeutic need, while regulatory exclusivities can help them attract commercial partners for marketing and dis­tribution of resulting products.

E. Comparative Social Costs

The pursuit of drug discovery and development in proprietary silos focused on patenting and secrecy gives rise to spillover costs on society that would not be present in a model that foregoes patenting and secrecy in favor of sole reli­ance on regulatory exclusivities. Patenting and secrecy, particularly at the dis­covery stages, can significantly impede the diffusion and use of new knowledge across the economy. These practices monopolize foundational scientific inputs and hinder downstream follow-on research that might otherwise derive signifi­cant utility if foundational discoveries were unencumbered and discovery data not kept secret. Moreover, secrecy gives rise to the phenomenon of duplicate drug development programs pursuing the same therapeutic hypothesis in paral­lel well into clinical development, multiplying by severalfold the aggregate so­cial cost when these programs fail. Secrecy around failed drug trials imposes substantial costs on researchers and vulnerable research subjects alike. This redundancy not only amplifies direct costs, but also gives rise to the opportunity costs of foregone breakthroughs that might have been achieved had duplicate resources instead been deployed to investigate other novel drug targets or mech­anisms. It further entails the significant ethical quandary of exposing large numbers of patients enrolled in clinical trials to drug candidates that would be known to fail but for secrecy across programs. Finally, delays caused by trans­actional complexity associated with patents and secrecy not only lead to in­creased firm-level and partnership-level costs, but in the aggregate decrease the number of drugs advanced.

In contrast, an approach to drug development centered around regulatory exclusivities that foregoes patenting and strict secrecy offers a number of social advantages over the siloed, proprietary model. It would enable the exploration of more therapeutic approaches in parallel by removing the informational asym­metries that cause large investments in redundant efforts on the same disease targets, thereby accelerating progress in the aggregate. It would help research­ers make decisions to stop ongoing projects early when competing groups openly invalidate relevant hypotheses. It would broaden the public domain and thereby enable the scientific community in both academia and industry to more quickly apply learnings from past successes and failures to generate new thera­peutic hypotheses or answer broader research questions through meta-analysis. It would improve research and development quality and outcomes by exposing drug research to secondary analysis, new ideas, and critiques from outside the firm or drug discovery partnership. Finally, it would provide more and earlier information to regulators, clinicians, drug purchasers, and patients to make bet­ter informed decisions around clinical trial approvals and participation, drug product marketing authorizations, health technology assessments, and prescrib­ing behavior.

A potential general critique, derived from the prospect theory of patent law first explicated by Edmund Kitsch, is that, absent early strong patent rights, the potential social advantages of open drug development could be undermined by rent-dissipating races by firms to be first to obtain marketing authorization with the same new chemical entity. Though a more detailed treatment of this issue has been advanced elsewhere, we suggest the following in response.

First, with respect to areas of market failure, the very need for the collabo­rative, risk-sharing model of drug development advocated here is in large part predicated on the existing lack of private financial incentives in the early stages of drug development. Firms do not invest in preclinical discovery and develop­ment in these areas even in the absence of competitors precisely because the economics do not offer the potential for future, risk-adjusted market returns. The calculus would be even less favorable with the added risk of losing a race with an open, collaborative public-private partnership.

Second, the practice of private firms discarding unpatentable compounds has been well documented even in potentially lucrative indications. A fortiori, it seems highly unlikely that private firms would invest in races to market yet to be clinically validated public domain compounds. Indeed, the highly promising yet unpatented lead compounds from the M4K Pharma initiative, described fur­ther below, were published in 2020. However, no private firm to date has sought to race these compounds to market. This is likely due to both the poor economics for the disease indication and the public domain status of the com­pounds. Perhaps more likely is that multiple firms would invest in advancing their own patented compounds for the same therapeutic hypothesis, but this is already common in the pharmaceutical industry and (as discussed above) a collaborative open drug development model offers the prospect of significantly reducing social loss from this form of redundancy. The risk-adjusted investment calculus would likely change significantly after clinical validation of openly de­veloped compounds like those from the M4K Pharma project; however, by that stage, other firms would be many years behind in development, again deterring investment in races to obtain marketing authorization.

Finally, in larger, more lucrative indications, firms may very well have stronger incentives to enter marketing authorization races in the absence of early patent protection. Policy proposals discussed in Part V are in part designed to curtail this form of rent-dissipating behavior.

For the foregoing reasons, a drug discovery and development paradigm centered on stacking patents, trade secrecy, and regulatory exclusivity entitle­ments entails significantly increased firm-level, partnership-level, and social costs than one that foregoes patents and secrecy and relies on regulatory exclu­sivities alone. However, despite their lower costs and significant advantages, drug development programs have not broadly switched to a regime of regulatory exclusivity from that of patents and secrecy. There are two principal reasons. The first is that, despite the costs to firms, patents and secrecy offer greater pro­tection, particularly at the preapproval stages and in extending a product’s life cycle, compensating for the higher transaction costs. Firms will not willingly give up patents and secrecy unless governments increase the incentives to forgo them and switch to regulatory exclusivities alone. The second reason is that firms have structured their affairs around patents and secrecy and thus are locked into them. Even in areas where this paradigm has been particularly unproductive—for example, in respect of unmet health needs—the process of getting firms to switch to other innovation modalities is slow and evolution­ary. There are, however, experiments involving a broad coalition of actors coming together to address these unmet needs where regulatory exclusivities combined with open science are making headway. We turn to this next.

IV. An Experiment in Open Drug Development

This Part explores the concept of a drug development partnership aimed at addressing unmet health needs that consists of firms, universities, philanthro­pies, governments, and patients and that seeks to leverage regulatory exclusivi­ties instead of patents to incentivize drug commercialization. While the firms in these partnerships may seek profits, the other partners may pursue a different set of rewards: reputation, prizes, attracting students, cost savings on healthcare, and longer and better quality of life. As discussed below, these partners seek to develop drugs in the most efficient way possible, selecting regulatory exclusiv­ities over patents, supplemented by open sharing of materials, data, and tools. Firms may be hesitant to participate in these partnerships, however, owing to entrenched path dependency around patents and secrecy in drug development. Governments may, therefore, need to initially take an active role in helping to form and fund these partnerships.

A. Open Science Drug Development Partnerships Aimed at Unmet Health Needs

Drug development today not only faces a productivity problem, seen in the ever-increasing costs of developing a new chemical entity, but also the inability to address some of society’s most pressing health needs. Alzheimer’s disease and other dementias will likely affect over 100 million people by 2050. Ten million are expected to die annually from antimicrobial resistance by the same year unless new antibiotics are developed. While rare diseases individually affect small populations, in total they affect between 263–446 million people at any one time. Yet neurodegeneration, antimicrobial resistant pathogens, and rare diseases, among others, are therapeutic areas that have been poorly served by traditional drug discovery modalities focused on patents and trade secrets.

To better address these unmet needs, mission-oriented partners in aca­demia, public institutions, philanthropy, and even the private sector are likely willing to take additional steps—beyond merely shifting their focus to regula­tory exclusivities as incentives—to reduce transaction costs. Below, this Ar­ticle explores drug development collaborations that have adopted the structure of an open science partnership. This form of collaboration aims to maximize the sharing of knowledge, data (to the degree possible within constraints of patient confidentiality), and tools both within the partnership and with those outside of it to accelerate research as much as possible.

Open science partnerships can reduce transaction costs by both streamlin­ing and speeding up project startup (by removing the most contentious compo­nents of negotiations: patents and secrecy) and by sharing data, tools, materials, and results quickly. With no delays due to secrecy, review of outputs for patent­ability, filing of patents, or patent enforcement, these partnerships can also sig­nificantly lower governance costs. The broad range of actors within these partnerships increases the diversity of expertise and the availability of data. This will be critical in constructing large, open datasets to advance development of new drugs—not only via improving the biological understanding of disease, but also with the identification of new targets and hits and in training artificial in­telligence models. This has been the experience with open databanks, such as the Protein Data Bank that led DeepMind to develop Alphafold, which predicts the three-dimensional structure of proteins, as well as biobanks, such as the Montreal Neurological Institute’s C-BIG repository. Sharing also enables parallel approaches to a problem based on the same molecule, data, or tools, thus increasing the chances that one of those branches is successful. Finally, as noted in Part III, open science partnerships can help firms to stop ongoing projects early by openly invalidating relevant hypotheses.

Because of the diversity of partners involved, these collaborations are po­sitioned to more efficiently distribute risk among participants. Partners also bring different resources: funding, molecular libraries, databases and biobanks, theories of disease, patient-derived data and materials, domains of expertise (bi­ology, medicinal chemistry, formulation, regulatory, clinical), policies, net­works, and reputation. As a result, no one entity bears the risk of failure. Further, the breadth of these partnerships also increases the likelihood that even if one part of the research results in failure, at least some of the partners can capture some benefit. For example, if a drug fails in clinical trials, the data gleaned from the trial can be used to better design another trial, or the data may suggest that the drug could be effective for another indication, or the data could provide training for artificial intelligence models. Under a proprietary partnership model, many of the key results of the research program and clinical trials would not be available to others: another firm would not develop a drug for another indication if patented by the first firm, and the data would not be available for training.

Central to this open science model is the creation of an organizational struc­ture to carry out several functions. First, such a structure is needed to manage the contractual, legal, and regulatory components of an open science drug de­velopment program and to act as the “applicant” or “sponsor” for regulatory filing purposes to obtain the regulatory exclusivities and other incentives avail­able to sponsors of new medicines. Second, the organizational form should be able to pursue and allocate diverse sources of funding for the project. Third, its social mission should position it to stimulate in-kind contributions from a range of experts and firms and its design should enable it to efficiently coordinate those contributions into a coherent scientific and regulatory plan. Fourth, the organization should act as a central body for project governance, resource man­agement, and decision-making.

B. Precedents in Open Science Drug Discovery

Open science partnership models view drug discovery as a problem of effi­ciently marshalling diverse sets of resources and skills. At base, these models focus on teams comprised of researchers from different fields, in different sec­tors, and in different locations. A central challenge is figuring out how to best coordinate inputs and manage resources during drug development. Outputs also vary and include not only drug candidates but also tools, databases, and bi­obanks. Early precedents are reviewed here.

1. Extending the Precompetitive Space to Chemistry

Most existing open science models focus on precompetitive research, al­though they have expanded the scope of precompetitive activities from biology and target identification to medicinal chemistry and even preclinical studies. Outputs are freely accessible, allowing downstream adopters to pursue either open or closed approaches to further translation and commercialization. A prominent example of this model is the Structural Genomics Consortium (SGC), a public-private partnership founded in 2003 and funded by many of the world’s leading pharmaceutical companies alongside philanthropic and public fun­ders. The SGC pools inputs and expertise from industry, philanthropy, and patient organizations and makes its research outputs openly available to the sci­entific community. The SGC’s publicly available outputs include the crystal structures of novel protein targets for which biological understanding is poor and clinical validation is lacking; potent and selective chemical probes that po­tently and selectively bind these novel proteins as tools for target validation and starting scaffolds for developing drug compounds; and the research data associ­ated with these structures and probes. The open availability of these tools and data allows a broad range of researchers to explore a protein target in multiple models of disease and report back through publication. In turn, this accelerates preclinical target validation and promotes rational structure-guided drug design, leading to lower clinical attrition rates and less duplication of ef­fort.

As an example, an open science partnership between the SGC and the On­tario Institute for Cancer Research (OICR) resulted in the identification and dis­semination of an unpatented chemical probe against the target protein WDR5, and enabled research publications by other groups linking the target to leukemia and other cancers. Spurred by this new knowledge of WDR5’s disease linkage, the OICR spun out a company that modified the probe compound into an opti­mized lead candidate, patented the modified structure, and performed the requi­site IND-enabling preclinical studies on it. The program was then licensed to Celgene for what was then the largest preclinical drug deal in Canadian his­tory.

The SGC was also at the center of the development of another chemical probe, JQ1, a first-in-class inhibitor of human bromodomains. The SGC’s in­volvement started in 2007 when its researchers discovered the three-dimensional structure of the human bromodomain BRD4, thought to be linked to cancer. The SGC deposited BRD4’s structure in the open Protein Data Bank. Next, the SGC, GlaxoSmithKline, and the Dana Farber Cancer Institute together cre­ated JQ1 as a chemical probe for BRD4, which was used to demonstrate the target’s potential in treating cancer. Not only did none of the partners seek a patent, they distributed JQ1 through simple open licenses and openly shared critical knowledge about the compound and the target with the scientific com­munity. As a result, a broad range of drugs targeting bromodomains have been studied by multiple firms in clinical trials over a wide range of indications.

A comparative analysis of JQ1 to other first-in-class compounds against other novel targets developed and patented through conventional closed propri­etary models showed that the public release of JQ1 led to a significant and sus­tained increase in both (1) downstream research and publication on the therapeu­tic applications of the drug target by a wider and more multidisciplinary research community and (2) follow-on innovation, as evidenced by the number of subse­quent patents filed involving bromodomains relative to the other targets. In­terestingly, these downstream patent filings came from a wider range of inven­tors and assignees in a greater number of therapeutic areas compared to the control target classes, suggesting that this open science partnership approach to drug discovery both broadens and accelerates downstream innovation and product development.

2. Efforts to Develop Drugs through Open Partnerships

Going beyond the precompetitive space, several partnerships have been early adopters of open science as a modality to discover and develop actual drug candidates. Some of these efforts have sought to leverage the principles of the open-source software movement to crowdsource drug discovery efforts. For example, Open Source Malaria (OSM) brings together researchers from aca­demia and industry to identify new compounds for anti-malarial drug develop­ment. GlaxoSmithKiline jump-started the project in 2010 by releasing a library of compounds into the public domain. Building on this library, the OSM community rapidly publishes all data online, including findings that po­tential leads do not work, helping other researchers to focus on other com­pounds. Another project, funded by both the World Health Organization and the Australian Government, focused on identifying low-cost processes to syn­thesize enantiopure praziquantel, an off-patent drug used to treat schistosomia­sis. The project placed all data in an open-source online elec­tronic lab notebook, ensuring that other scientists would be able to quickly ac­cess it. The project’s openness attracted significant unsolicited advice, experimenta­tion, and materials from both academia and industry, without any promise of reward.

These programs mobilized a broad set of incentives to advance their pro­jects, ranging from the selfless desire to improve public health to more self-interested goals associated with reputation-building, publication, and profit. For example, industrial participants have been willing to contribute to these pro­jects on a pro bono basis to promote employee morale, to enhance public rela­tions, and, in the case of contract research organizations (CROs), to publicly demonstrate their capabilities.

An important feature of these projects is that their openness contributed to growth. For example, the seed funding that supported initial research activity in OSM led to further contributions downstream. The open release of infor­mation and data in real time to the public attracted experts otherwise unknown to the project to participate, leading to accelerated research, improved transpar­ency, and peer evaluation.

As a follow-on to these early adopters, the SGC and several research part­ners, including the OICR and the CRO Charles River Laboratories, launched and executed a successful hit-to-lead program entirely in the open through a mission-oriented company called M4K Pharma (Medicines for Kids). This program leveraged an initial seed grant from OICR and open contributions from a large range of academic groups, public research organizations, clinicians, CROs, and pharmaceutical firms. Together, these collaborators synthesized, screened, and optimized several hundred new ACVR1 kinase (also called ALK2) inhibitor compounds against a stringent target product profile for the treatment of diffuse intrinsic pontine glioma (DIPG), a rare pediatric brain can­cer that is almost uniformly fatal yet underserved by traditional development models. With minimal transaction costs, the program generated several late-stage lead compounds that met the target product profile, with efforts currently focused on selecting a clinical lead for IND-enabling studies. It also placed all chemical structures and data in the public domain in close to real time. One compound called M4K2234, though not suitable for progression as a drug candidate for reasons related to pharmacokinetics, was also nominated and ap­proved as an SGC chemical probe and is broadly shared as a research tool for additional ACVR1 target studies.

Many of the same motivations for engaging in open science that have been reported for OSM and the open praziquantel project also animated the contrib­utors to M4K’s program. These include the altruistic desire to contribute to the public good and better patient outcomes, reputational goals related to academic publication, and commercial incentives relating to employee morale, corporate social responsibility, and showcasing technical capabilities.

As an important advance over previous open drug development programs, M4K Pharma adopted a corporate organizational form. M4K Pharma is an in­corporated company under the Canadian Business Corporations Act. It is wholly owned by Agora Open Science Trust, a charity whose mission is to ensure that any resulting products are made broadly accessible and affordable. This M4K Pharma entity plays the role of research and funding coordinator, contracting body, and of eventual drug product sponsor, providing a mechanism to register clinical trials, seek marketing authorization, and engage in licensing transactions with manufacturing and distribution partners. M4K Pharma explicitly es­chews patenting and secrecy to facilitate broad, open collaboration, and relies instead on regulatory exclusivities as incentives to attract commercial part­ners.

3. Open Clinical Trials

Preclinical biology and medicinal chemistry efforts are not the only domains where open science has the potential to transform drug development. Open approaches to clinical trials themselves may have an even greater impact, given the high failure rate of drugs and concomitant wasted investments. Of­ten, preclinical science points strongly to new therapeutic opportunities only for the therapeutic hypotheses to be later invalidated in clinical trials. This typi­cally occurs during Phase II trials where researchers first obtain evidence of whether the “mechanism[] of the selected disease target can be safely and use­fully modulated” in humans.

Given the competition between firms to develop a first-in-class drug, firms rely heavily on trade secrets at the clinical phase, resulting in extensive duplica­tion through parallel clinical studies. If at least the clinical proof-of-concept studies on new targets were carried out in the open, rapid dissemination of re­sults would reduce duplication while enhancing patient safety (by deterring re­dundant clinical studies on poor therapeutic hypotheses). Drug development efforts could then focus their investments on a greater number of de-risked and validated pioneer targets, enabling the pursuit of further indications or optimized follow-on medicinal chemistry efforts, ultimately with a higher likelihood of obtaining new medicines. Cost-avoidance models for negative clinical proof of concept studies could save the industry up to US$17.5 billion in 2024 dollars annually.

The benefits of clinical trial openness do not end with a more efficient route to clinical proof of concept for pioneer targets. In the traditional case, firms do not publish the results from many trials in a timely manner or at all, and if they do publish them, they leave large amounts of data unanalyzed. The Institute of Medicine’s Committee on Strategies for Responsible Sharing of Clinical Trial Data has called for more widespread and timely open sharing of clinical trial data. The Institute notes that doing so would allow other investigators to conduct secondary analyses and meta-analyses that strengthen the evidence base for regulatory and clinical decision-making, enable additional analyses of un­published data to generate new research hypotheses, and maximize the scientific knowledge and benefits gained from publicly funded research and from the con­tributions of clinical trial participants. Broad stakeholder involvement through openness can also help in the design of clinical trials by, for example, leveraging input from patients, care providers, and experts in the field. Clini­cal trial results are unlikely to achieve the same level of openness as preclinical studies because of concerns over trial integrity, protecting privacy, and respect­ing the informed consent of trial participants. Given this, one would expect some limitations on the rapidity and completeness of clinical data release.

To date we are not aware of any partnership that has sought to run spon­sored drug candidate trials using an open science paradigm that disseminates positive and negative trial data as quickly as possible, though calls for such part­nerships to more efficiently run clinical proof-of-concept studies have been made. Nonetheless, efforts to encourage more clinical trial transparency do exist. Trial registration on clinicaltrials.gov is now a widely mandated (though not universally enforced) requirement for funding, publication, and regulatory consideration. The Food and Drug Administration Amendments Act of 2007 requires summary results of clinical trials for FDA-approved drugs to be posted on clinicaltrials.gov regardless of whether the results have been published. The European Medicines Agency (EMA) moved a step further by requiring that all clinical study reports submitted to the Agency be made publicly available through a portal on its website once a marketing decision has been made. Several large pharmaceutical companies voluntarily make trial data available through an initiative called Clinical Study Data Request, although the process is cumbersome and restrictive, with requests being reviewed by an independent board and access granted only through a format that cannot be downloaded.

C. Open Science and the Commercialization Pathway

The previous Section canvassed the significant advantages of eliminating the transactional barriers and spillover costs associated with patenting and se­crecy, as well as the initial partnership efforts to move towards more open forms of drug development. This Section now explores the potential use of regulatory exclusivities as a firm’s (or partnership’s) sole commercialization asset to trans­late an openly discovered drug candidate to a marketed product. We argue that post-approval exclusivity protections on an openly developed drug product, combined with a few strategic approaches to control the competitive dynamics in the preapproval stages of development, can serve as the basis for a commer­cialization program sufficiently attractive to firms, particularly in less competi­tive therapeutic spaces. Because firms have an incentive to stack patent, secrecy, and regulatory exclusivity entitlements, however, Part V examines public poli­cies that could better align firms’ private incentives with a more socially optimal open drug development paradigm.

1. Preventing Competitor Use After Product Approval

Regulatory exclusivities protect sponsors of new drugs against competitive use of their data after regulatory approval, regardless of the public availability of the data. This means firms can engage in open partnerships and disclose their preclinical and clinical data during development, without foregoing exclusivity once on the market. This is most straightforward for rare disease products, whereas orphan exclusivities in the United States, European Union, and else­where act as complete bars to all competitor follow-on submissions in the same indication, regardless of the source of data.

The analysis is somewhat more complicated for NCE exclusivities in the absence of orphan status, but the conclusion is ultimately the same. Though a competitor may have access to the originator’s openly published preclinical and clinical data for an NCE, there is effectively no submission pathway by which the competitor could leverage these data to seek authorization for its own prod­uct. Neither the full NDA pathway, nor the available abbreviated pathways, would permit the inclusion of these data.

A full FDA NDA, which is filed pursuant to Section 505(b)(1) of the U.S. Food, Drug, and Cosmetics Act (FDCA), must contain full reports of adequate and well-controlled investigations, including clinical investigations demonstrat­ing safety and effectiveness for the use(s) indicated in the new drug’s proposed labeling. All of the investigations relied upon by the firm applying for ap­proval must either (1) have been conducted by or for the applicant or (2) be subject to a “right of reference” or use obtained by the applicant from the person by or for whom the investigations were conducted. In other words, the firm must either own, or have obtained an explicit license to use for regulatory pur­poses, all of the data submitted in its 505(b)(1) application. Thus, an origi­nator firm that published its regulatory data (without also granting a right of regulatory use) is protected from a would-be competitor seeking to file its own full 505(b)(1) application using the published data. This is true not only after approval of the originator product, but before approval as well.

A would-be competitor is also barred from using openly published data on an NCE to pursue abbreviated approval pathways during the period of NCE ex­clusivity. The FDCA permits two forms of abbreviated submission: (1) an application pursuant to section 505(b)(2), which contains full reports of safety and effectiveness but relies on one or more investigations not conducted by or for the applicant and not subject to a right of reference or use obtained by the applicant and (2) an ANDA pursuant to section 505(j) to market a generic version of a previously approved drug. An ANDA only requires information to show that the generic version has the identical active ingredient, dosage form, strength, route of administration, and intended use as the original, approved, drug (called the “reference listed drug” or “RLD”). In contrast to ANDAs, 505(b)(2) applications generally may be submitted in two scenarios: (1) for a new chemical entity, where the application seeks to rely in some part on pub­lished literature regarding studies for which the applicant has not obtained a right of reference to the underlying raw data (called a “literature-based 505(b)(2)”) and (2) for a change to a previously approved drug, such as a new indication, route of administration, dosage, or formulation, where the applica­tion seeks to rely in some part on the FDA’s finding of safety and effectiveness for the previously approved drug.

U.S. NCE exclusivity prohibits submission of all abbreviated applica­tions—including both types of 505(b)(2) applications as well as ANDAs—for any drug products containing the same active moiety as the originator product during the exclusivity period. Thus, not only would a competitor be unable to assemble a full 505(b)(1) NDA using published data without the sponsor’s ex­plicit license to do so, the competitor also could not rely on these data in any form of abbreviated submission, effectively foreclosing all potential routes to competitive use of published drug development data during the NCE exclusivity period.

While less regulatory guidance exists in the European Union, it is likely that a similar analysis to the above would apply in that jurisdiction. The EMA offers several legal bases for abbreviated applications for marketing authoriza­tion relying on pioneer data, including not only a generic pathway (analogous to that of the United States), but also a hybrid application (analogous to a U.S. 505(b)(2) application for changes to a previously approved drug), a mixed mar­keting authorization application (analogous to a U.S. literature-based 505(b)(2) but where, in addition to assembled bibliographical references, the applicant must submit its own preclinical or clinical studies), and well-established use ap­plications (applicable only to active substances that have been in well-established medicinal use in the European Union for at least ten years, with es­tablished safety and efficacy). During the period of European Union’s new chemical entity exclusivity, reference to data supporting approval of a pioneer product is not permitted in follow-on applications under any of these legal bases, regardless of whether the data is contained in the pioneer sponsor’s application dossier or in the literature. As such, competitor use of a firm’s public data should be completely barred during the period of E.U. NCE exclusivity.

In Canada, only two avenues are available to pursue abbreviated regulatory approval. The first is an abbreviated new drug submission for generics that is analogous to U.S. and E.U. generic pathways. Where the Canadian reference product is an innovative drug (i.e., contains a new chemical entity), a generic application is prohibited during the period of NCE exclusivity. The second abbreviated pathway in Canada is called a drug submission relying on third-party data (SRTD). The SRTD pathway enables an applicant to assemble a full new drug submission through substantial reliance on publicly available lit­erature and market experience instead of conducting its own testing to support safety and efficacy. However, Health Canada requires evidence of extensive current foreign market experience with the same medicinal ingredient (for a minimum of ten years) or reference to evidence on marketing of the same me­dicinal ingredient in Canada, neither of which would exist for a new product. Moreover, the SRTD pathway requires comparative pharmaceutical and bioa­vailability data to the previously marketed reference product, which would be prohibited by NCE exclusivity.

The above analysis demonstrates that a sponsor of an unpatented, openly developed drug product can rely on orphan drug exclusivity or new chemical entity exclusivity alone as an effective bar to competition from follow-on prod­ucts. Thus, in theory, a sponsor could both engage in partnerships that eschew patents, forego trade secrecy, and openly share drug development data to limit the transactional and social costs identified above, while still preserving robust commercial marketing incentives associated with regulatory exclusivities.

One drawback that firms would contemplate is that, outside of new antibi­otics and biologics, the effective period of market protection is likely to be shorter under regulatory exclusivities alone than under patents and regulatory exclusivities in combination. This is particularly true for NCE small molecules in the United States, where at five years (or seven years if the drug also treats an orphan indication), exclusivity is relatively shorter than in other jurisdictions (for example, compared to ten years in the European Union, or twelve years if the drug also treats an orphan indication). This is one key factor that would dis­suade firms from pursuing a regulatory exclusivity-only development trajectory in a broad range of scenarios. Part V discusses a solution to this problem.

Another drawback relevant for firms contemplating this pathway is that, unlike patents, regulatory exclusivities begin only upon marketing approval of the originator product, rendering it more difficult for firms to completely control the competitive dynamics during development, before regulatory approval. We turn to this issue next.

2. Preventing Competitor Use Before Regulatory Approval

None of the various regulatory exclusivities described earlier apply before regulatory approval. In theory, this opens the possibility that a competitor could seek to leverage an originator firm’s published drug development data to abbre­viate its own regulatory submission burden, and thereby race to beat the origi­nator to market. In practice, however, this would be difficult to implement, and a number of approaches could be taken by the originator firm to prevent it.

As noted above, a competitor could not submit a full 505(b)(1) NDA using the originator’s data in the preapproval period because the competitor would not itself have conducted the studies or obtained a “right of reference” to the study data. Additionally, the competitor could not submit a typical generic application, such as an ANDA under Section 505(j) of the FDCA or its analogs in Europe, Canada, and other jurisdictions, because those types of submissions require a previously approved reference drug, which would not exist in the preapproval period.

The only option available for competitive preapproval use of the origina­tor’s data would be via a literature-based 505(b)(2) application in the United States or an analogous mixed marketing authorization application in the Euro­pean Union. No such possibility exists in Canada, where the ten years of foreign market experience or domestic market experience criteria for SRTD applica­tions would not be satisfied.

Even in the United States and European Union, the submission of a litera­ture-based application relying on the originator’s published data is more theo­retical than practically feasible. To justify reliance on the published data, a competitor seeking approval on this basis would have to establish a “bridge” (usually through comparative pharmaceutical, bioavailability, and bioequiva­lence data) between its proposed drug product and the product used in the orig­inator’s studies. In the preapproval period, the competitor would not have access to the originator’s formulated product to conduct the requisite bridging studies.

Moreover, owing to their complexity, literature-only 505(b)(2) applications are extremely rare. Between 2003–2016, out of 451 total 505(b)(2) applications, only 14 were “literature-only” applications seeking FDA approval without re­ferring to an already approved drug, and only one of those was an oral formula­tion, where bioavailability considerations are complex compared to more direct routes of administration. All fourteen of those approvals were for pre-1962 drugs that had long histories of clinical use and had been grandfathered by the FDA or for drugs that would be unethical to test on humans prospectively (e.g., radioactive products). In the vast majority of cases, however, these fac­tors would not be present, rendering it highly unlikely that a would-be compet­itor could successfully leverage the literature-only 505(b)(2) pathway to bring a completely novel compound to market before the originator does.

To further deter this avenue, an originator firm that has foregone patenting and secrecy could choose to share its preclinical and clinical data during devel­opment under license terms that permit scientific use of the data but that prohibit competitive regulatory use. These terms would be directly akin to the terms un­der which the EMA releases clinical study reports after it makes a marketing authorization determination. While regulators themselves currently have no explicit mechanism for enforcing these contract terms, firms could enforce them privately in the courts, as they must do with patents and trade secrets. As another means to forestall competitive literature-only applications, an originator firm could strategically withhold certain elements of its regulatory data package until after approval, in particular datasets whose continued secrecy would entail lim­ited or no firm-level, partnership-level, or social spillover costs because of their limited value to collaborators and the research community. These might include, for example, data on manufacturing and scale-up processes, active ingredient source(s), and formulation. Given that withholding these data is not ideal in an open environment, Part V suggests regulatory changes that would obviate the need to withhold them beyond issuance of an IND.

For the above reasons, drug product developers that have foregone patents and secrecy have, in the vast majority of cases, relatively straightforward means to prevent competitor use of their data not only during periods of regulatory exclusivity after approval, but also in the preapproval period. What the existing policy landscape does not provide is a regulatory or legal means for a developer of an originator product, absent patents on the composition of matter or use, to exclude competitors from racing to conduct their own preclinical and clinical studies in an attempt to gain independent marketing authorization before the originator. In many cases, particularly in areas of underinvestment with lim­ited competitive dynamics, the originator’s development lead time should deter this phenomenon because of the ex ante risk that the competitor’s substantial independent development costs would be wasted playing catch-up in an ap­proval race that it subsequently loses. To augment lead time, originators with new drugs that address unmet needs or rare diseases can apply for regulatory programs that expedite development and approval phases. These programs in­clude fast track designation, breakthrough therapy designation, accelerated ap­proval, and priority review at the FDA. The EMA and Health Canada offer similar accelerated programs. Indeed, the greater the lead time, the less likely competitors are to incur those risks and the more lucrative the product market would need to be to justify it, for example in large indications. As previously noted, further limiting the likelihood of this behavior is the fact that most firms jettison drug candidates that are not patentable, let alone ones that are already in development by another firm. Nevertheless, this preapproval legal and regu­latory lacuna is one that firms contemplating a development pathway without patents and trade secrets would need to account for, likely deterring broader adoption outside areas of limited competition. We turn to solutions in Part V.

V. Increasing Incentives for Regulatory Exclusivities

Parts III and IV discussed the problem of path dependency with respect to patents and secrecy in drug development. These Parts also introduced the con­cept of open science drug development partnerships as proof of concept that developers can eschew costly and socially deleterious patenting and secrecy practices in favor of sole reliance on regulatory exclusivities. Though an open developer’s data would be protected against competitive regulatory use in both the pre- and post-approval periods without patents and secrecy, some limitations in the current policy landscape exist that deter broader adoption of this approach. Specifically, we identified (1) the relatively shorter period of market protection offered by regulatory exclusivities compared to patents, particularly with re­spect to NCE small molecules in the United States and (2) the lack of a legal or regulatory means to prevent competitors from racing to be first to market with an openly developed NCE through the competitor’s own independent studies, which may be a risk in more lucrative indications. Given the positive social spillovers of reducing reliance on patents and secrecy within drug development partnerships, we sketch out policy modifications that governments could intro­duce to address these limitations and nudge more programs in this direction. In light of the political economy dynamics at play, however, this Part advocates for the introduction of voluntary incentives, rather than a full regime change that would disrupt the existing proprietary framework as a robust option for product developers.

A. Preapproval Incentives

With respect to the preapproval period, our goal is to offer solutions that mimic the added exclusionary benefits of patents and trade secrets to firms with­out the deleterious implications noted previously. First, to eliminate the risk of competitors seeking to race the originator to market by conducting independent studies on the same NCE, the FDA and other health regulators could grant a time-limited exclusivity over the NCE to a product developer that first files an IND application for clinical trials on the NCE. The exclusivity would pre­clude competitors from filing any IND or NDA for the same or similar NCE, akin to how orphan drug exclusivities preclude competitor NDAs. To obtain such an IND exclusivity, the product sponsor would need to establish that it (and affiliates and partners) has not patented the NCE and commit to open disclosure of its clinical data. Once granted, the FDA would publish the contents of the sponsor’s IND application for research use, and would monitor ongoing com­pliance with data disclosure requirements, absent which the exclusivity would be invalidated. The period of this exclusivity could be calibrated to bal­ance the need to adequately incentivize open product developers with the im­perative of encouraging others to advance the product if the originator fails for reasons un­related to the product’s safety and efficacy. In a similar vein, the IND exclusivity could also be made subject to a diligent implementation requirement, such that a sponsor that failed to progress its drug candidate through clinical development in a timely manner would surrender its exclusivity over the NCE.

Alternatively, or concomitantly, to vitiate the relative advantages of trade secrecy over preclinical and clinical data, the FDA and other drug regulators could offer developers a reasonable period of exclusivity over their openly dis­closed trial data during the preapproval period. This would address the theoret­ical concern of a competitor using the originator’s own published data in a “literature-only” submission to leapfrog the originator to market. Such an initi­ative would create less monitoring burden on the FDA than full IND exclusivity, as there would be no need to verify ongoing data disclosure or diligent imple­mentation. The entitlement could be granted for a fixed period and only to data that have already been published in an accepted open repository, for example. In fact, the FDA, perhaps in conjunction with other regulators, could create an open drug development data repository that enables sponsors to voluntarily reg­ister and deposit their preclinical and clinical data during the development pro­cess. Datasets that are registered and deposited in this repository would be made publicly available for research use while simultaneously affording the depositor exclusive regulatory use. The depositor could assemble its entire regulatory dos­sier in the repository and submit it to the FDA as an NDA once complete. As with the IND exclusivity proposed above, the period of preapproval data exclu­sivity should be calibrated to balance incentives to the developer with the need to ensure the data do not sit idle.

B. Post-Approval Incentives

With respect to the post-approval period, regulators could eliminate the dis­crepancy in the period and strength of monopoly protection for a drug product covered by patents versus one covered only by regulatory exclusivities.

First, the FDA and other drug regulators could convert the period of NCE exclusivity (and any extension to it that a sponsor might be entitled to) from a regime of data exclusivity to one of complete protection over the NCE, akin to orphan drug exclusivity but without being limited to specific indications. This change would be designed to eliminate the theoretical possibility that a compet­itor may wish to complete its own preclinical and clinical studies to register a competing product. To obtain this protection, a sponsor would need to establish that its drug product is not covered by patents and that it had openly published its preclinical and clinical data in an accepted repository.

Next, to address the discrepancy between existing regulatory exclusivities and the periods of monopoly protection offered under patents for most new drugs, the FDA and other drug regulators could offer an exclusivity extension for drugs that are developed without patents and where data are openly shared. The extension would be similar in design to those offered in the United States for pediatric studies and new antimicrobials, but to qualify a sponsor could be asked, for example, to (1) demonstrate upon submission of its marketing appli­cation that it has diligently made its preclinical and clinical data publicly avail­able via an accepted data repository and (2) provide a certification to the regula­tor that the drug developer has not sought patents on the drug. Having done so, the sponsor would be ineligible to list any patents against its drug in the FDA Orange Book or foreign equivalents. The extension period would likely need to be substantial to induce more product developers to pursue this pathway. It could either be a uniform fixed period, or subject to determination on a case-by-case basis, with longer extensions being offered for products that address certain priority health needs, which are first-in-class against a novel target, or for spon­sors that commit to a price ceiling for the drug product to help ensure access and affordability. The latter feature would impose an additional monitoring burden but could help control escalating drug costs on a voluntary basis. Econometric studies could help determine the length of the extension under appropriate con­ditions.

C. International Harmonization

To further reduce regulatory complexity and align incentives for using reg­ulatory exclusivities instead of patents and secrecy, drug regulators should seek to harmonize regulatory exclusivity incentives for drug developers who forgo patents and secrecy internationally. Data protection provisions in bilateral and multilateral trade agreements have often been derided as a ploy to limit patent law flexibilities that could otherwise be used to generate access to affordable generic medicines in lower-income markets. Nevertheless, if introduced to encourage open, patent-free drug discovery to address unmet local medical needs, regulatory exclusivities may represent a positive use of trade flexibilities.

Conclusion

There is no single cause of declining productivity levels in drug develop­ment, and each factor will require different solutions. It is crucial, however, to address existing inefficiencies in the production and use of knowledge in devel­oping drugs from target to market.

While the pharmaceutical sector relies on time-limited product monopolies to justify the significant costs and risks of developing new drugs, these monop­olies need not be based on patents and secrecy, whose widespread use drives many of the existing inefficiencies. Regulatory exclusivities provide strong, and in many cases more certain, monopoly protection against competition. At the same time, they require significantly lower firm-level costs to obtain and en­force, and they are not invalidated by prior disclosure, enabling early data pub­lication without foregoing product monopolies. Because of this, regulatory exclusivities give public research funders more latitude to require data sharing without undermining translational goals.

In addition to firm-level advantages, drugs developed within collaborations between firms, university researchers, patient organizations, philanthropies, and governments are increasingly more likely to be successful than siloed programs. These collaborations leverage far more and different kinds of expertise, re­sources, funding, and risk tolerances than possible within a single firm. How­ever, partnership arrangements come with significant transaction costs, most notably around patenting and trade secrecy, which increase with the size and complexity of the project. Eschewing patents and secrecy in these partnerships makes them easier to initiate and govern and eliminates the need for stringent measures to protect confidentiality within them.

Finally, the broader adoption by firms and partnerships of a strategy relying on regulatory exclusivities without patents and secrecy would not only lower firm-level and partnership-level transaction costs but would also result in posi­tive spillovers in terms of the number and diversity of drugs brought forward. External research groups could leverage resulting increases in public domain knowledge to more quickly generate and test novel therapeutic hypotheses and to divert resources away from redundant avenues of investigation towards more productive uses. This would help spur an overall acceleration of drug develop­ment.

Because of strategic path dependency, however, firms may be uncomforta­ble experimenting with a new model of drug discovery. We therefore suggest that governments encourage and help fund private-public partnerships among firms, academics, public research institutions, and philanthropies to develop drugs for unmet health needs. We propose that these efforts take the form of open science partnerships that provide open access to results, data, tools, and materials, and that they rely on regulatory exclusivities instead of patents and secrecy to incentivize commercial translation.

We also identify policy options that governments and regulators could adopt to encourage firms and partnerships to adopt open forms of drug develop­ment more broadly. These options primarily involve expanding the scope and period of regulatory exclusivities for programs that voluntarily forego patenting and secrecy. States could hone these policy recommendations to the particulars of their respective pharmaceutical sector strengths and weaknesses.