Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

# Life sciences intellectual property licensing at the Massachusetts Institute of Technology

## Abstract

Academic institutions play a central role in the biotech industry through technology licensing and the creation of startups, but few data are available on their performance and the magnitude of their impact. Here we present a systematic study of technology licensing by one such institution, the Massachusetts Institute of Technology (MIT). Using data on the 76 therapeutics-focused life sciences companies formed through MIT’s Technology Licensing Office from 1983 to 2017, we construct several measures of impact, including MIT patents cited in the Orange Book, capital raised, outcomes from mergers and acquisitions, patents granted to MIT intellectual property licensees, drug candidates discovered and US drug approvals—a key benchmark of innovation in the biopharmaceutical industry. As of December 2017, Orange Book listings for four approved small-molecule drugs cite MIT patents, but another 31 FDA-approved drugs (excluding candidates acquired after phase 3) had some involvement of MIT licensees. Fifty-five percent of the latter were either a new molecular entity or a new biological entity, and 55% were granted priority review, an indication that they address an unmet medical need. The methodology described here may be a useful framework for other academic institutions to track outcomes of intellectual property in the therapeutics domain.

## Main

The process of drug development in the pharmaceutical industry is undergoing a profound shift in its industrial organization. Increasingly, smaller biotech firms apply recent academic research in the life sciences to develop new drugs, which are then acquired by pharmaceutical giants using their financial war chests and access to low-cost capital to purchase expertise1.

The number of biotech startup companies formed as a result of technology licensing has increased from 145 in 1994 to 278 in 2000 to 1,024 in 20162,3, and this trend has had a remarkable impact on drug development (Supplementary Tables 1 and 2). Of the 30 top-selling drugs worldwide in 2000, only 5 were traceable to biotech universities, and only 2 of these were in the top 10; the remaining 25 were developed by big pharma. By 2018, more than one-half of the top 30 drugs were sourced from academia, including 6 of the top 10.

Despite the growing importance of technology licensing to the biomedical ecosystem, surprisingly few data have been collected on the impact of technology transfer by academia. We address this deficit by providing a detailed analysis of the portfolio of life sciences intellectual property (IP) of MIT from 1983 to 2017. Although other studies have been published on patenting activity at specific universities4, this study is a systematic analysis of a portfolio of therapeutics companies that have licensed the IP of a specific academic institution.

For this study, we limit our scope to a subset of the life sciences sector, one that includes biotechnology, pharmaceuticals and other therapeutics, defined as firms in the business of developing new drugs, either small molecules or biologics. We focus on therapeutics for two reasons. First, the ecosystem physically surrounding MIT has its primary focus on drug discovery and development. Second, this focus allows us to use the number of drug approvals as a core metric of innovation, as is done every year by journals, such as Nature Reviews Drug Discovery5.

We develop a framework for tracking innovation originating in academia that builds on previous work measuring innovation in the pharmaceutical industry1,6,7. Using the Approved Drug Products with Therapeutic Equivalence Evaluations report (Orange Book) by the Food and Drug Administration (FDA; http://www.fda.gov/drugs/drug-approvals-and-databases/approved-drug-products-therapeutic-equivalence-evaluations-orange-book), we attempt to link MIT IP to drugs approved by the FDA8. However, explicit lines of causality are rare owing to the complexity of the journey from IP licensing to FDA approvals, which complicates the determination of the degree of contribution from the IP versus the degree of contribution from the company.

To provide a clearer picture of the impact of MIT IP, we separately track the origin of each drug in our dataset to weight the contributions of MIT licensees—that is, companies that have licensed MIT IP—to their respective approved drugs. To measure different types of innovation, these drugs are also labeled as new molecular entities (NMEs) or new biological entities (NBEs), and as drugs granted priority review (PR)1,6,7. Ultimately, this framework enables us to compare the innovation contributed by MIT licensees against the benchmark of historical pharmaceutical industry averages. In addition, we analyze the following company outcomes: capital raised in initial public offerings (IPOs), merger and acquisition (M&A) volumes, total drug candidates developed and number of patents granted. Finally, we examine several case studies that demonstrate the broader contribution of MIT to the growth of the biotech industry beyond IP licensing (for example, the role of MIT faculty as co-founders and advisors to biotech companies that have gone on to be very successful).

## Company screening

From an initial list of 225 life sciences MIT licensees, we identify 76 therapeutics companies on the basis of company business descriptions and financial filings. We further narrow the bulk of our analysis to 33 companies that were or are currently publicly listed on an exchange, owing to the availability of detailed information in their financial filings (see Supplementary Tables 3 and 4 for overview information on companies).

With the exception of Aprecia Pharmaceuticals, none of the private licensees has brought a drug to market single-handedly. This is not surprising, given the capital-intensive process of commercializing therapeutic candidates. To address these capital needs, biotech companies tend to tap the public markets or engage in strategic business development or M&A transactions with larger biopharmaceutical companies. Thus, we focus primarily on public companies, but also summarize private M&A outcomes.

## Orange Book citations

Following the approach in Stevens et al.8, we use the Orange Book (http://www.fda.gov/drugs/drug-approvals-and-databases/approved-drug-products-therapeutic-equivalence-evaluations-orange-book, https://data.nber.org/fda/orange-book/historical/1986-2016/ and https://data.nber.org/data/fda-orange-book-data.html) and the US Patent and Trademark Office databases (http://patft.uspto.gov/netahtml/PTO/index.html and https://assignment.uspto.gov/patent/index.html#/patent/search) to link MIT IP and FDA-approved drugs (cutoff at December 2017). To identify drugs that owe their origin, at least in part, to MIT IP, we search the publication for New Drug Applications (NDAs) that cite patents assigned to MIT (Supplementary Table 6).

We find entries for four small-molecule drugs that cited MIT patents (Table 1). (See Supplementary Note, section D, for citations that fall outside the scope of our analysis or occurred after our cutoff.) Redux was brought to approval by Indevus in partnership with Wyeth for obesity. However, its approval was later withdrawn owing to safety concerns9. Gliadel, Sarafem and Spritam are examples of a direct link between MIT IP and approval. The technology behind Gliadel, an implantable wafer loaded with the chemotherapy agent carmustine for use in glioblastoma treatment, was invented by Robert Langer’s laboratory at MIT, and developed by Guilford Pharmaceuticals, a spinout of Scios Nova. Scios acquired Nova Pharmaceutical, which had licensed the IP from MIT. Scios Nova was uninterested in developing the product, so MIT facilitated the creation of Guilford, a spinout started specifically to develop and commercialize the wafer and related technologies10.

In the development of Sarafem, the MIT professor Richard Wurtman patented his discovery that low serotonin levels in the brain contributed to premenstrual dysphoric disorder. He then founded Indevus to license the patent to Eli Lilly, which already marketed Prozac, a selective serotonin reuptake inhibitor that increased serotonin levels in the brain for the treatment of depression. Eli Lilly subsequently developed the compound for premenstrual dysphoric disorder and launched a newly branded version called Sarafem11.

The use of Orange Book citations as a measure of impact of MIT IP has several limitations. First, the absence of a medical school and hospital predisposes research at MIT to platform-based technology that may be applicable to many different drugs/diseases, over optimization of specific drug compounds. Therefore, most MIT patents do not cover composition of matter, which is most pertinent for market exclusivity protection and inclusion in the Orange Book; they usually focus on explication of mechanisms. This also means that MIT licensees are likely to spend a large part of the patent term seeking the optimal target for the platform (for example, Sangamo Therapeutics and its zinc finger nuclease (ZFN) gene-editing platform). Given that the time to market for new drugs can be as long as 20 years, it is not uncommon for MIT patents to expire before a drug is approved.

To further complicate the link between academic IP and approvals, companies often license additional IP from other institutions and build on existing technology to file new patent applications during the development process. In addition, investigational drugs typically undergo many iterations of formulation studies. Consequently, it is not surprising if the initial MIT IP is ultimately displaced from the Orange Book by more recent patents that can afford greater protection.

Clearly, the impact of MIT IP extends beyond citations in the Orange Book. MIT IP also plays an important role in catalyzing a company’s financing by attracting interest from venture capitalists, and serving as a foundation for future research and development (R&D). To provide a clearer picture of the innovation contributed by MIT IP, we analyze the financials and the R&D portfolios of MIT licensees as well as provide several other measures of impact, including capital raised, M&A outcomes, drug candidates discovered, drug approvals and patents granted.

## IPOs

Of the 33 public biotech MIT licensees, 26 completed the standard IPO process, 4 reverse-merged into publicly listed companies, and 3 listed through alternative pathways. We use the form S-1 financial filings—registration forms submitted to the Securities and Exchange Commission for new securities—to produce Table 2, which summarizes the results of the IPOs that have sufficient available data. (The financial filings for firms that conducted an IPO before 1996 were not available.) It includes the proposed share price range, the final share price, the proceeds net of underwriting fees, the shares issued and outstanding, and the dilution due to financing.

On average, the IPOs of MIT licensees raised about US$41 million in net proceeds, diluted the existing shareholders 26% in the offering, and achieved a post-IPO valuation of US$218 million. Given that therapeutics companies typically conduct an IPO to finance R&D and clinical trials when their assets are under development, the average valuation post-IPO is in line with expectations. The capital generated by the IPOs and the resulting shareholder dilution were also plotted over time (Supplementary Note, section E). Over the studied period, MIT licensees raised more capital, although this also came with increased dilution to shareholders. After adjusting for inflation, net proceeds experienced a high level of growth, an indication of the dramatic expansion of the biotech capital markets over the past two decades.

## M&As

Among the 76 biotech MIT licensees, 23 companies were acquired in 11 private and 12 public M&A transactions, with a total value of US$30.7 billion (Table 3 and Supplementary Note, section F). This large volume was driven primarily by three public company deals totaling US$23.5 billion. Millennium Pharmaceuticals (acquired by Takeda), ARIAD Pharmaceuticals (acquired by Takeda) and Cubist Pharmaceuticals (acquired by Merck) were all fully integrated biopharmaceutical companies with potential blockbuster assets at the time of transaction. Millennium marketed Velcade for multiple myeloma and mantle cell lymphoma, but also had a diverse pipeline of small-molecule inhibitors and monoclonal antibodies. ARIAD marketed Iclusig for select hematological malignancies, and had the potential blockbuster small-molecule drug brigatinib for ALK+ non-small cell lung cancer on the cusp of FDA approval. Cubist marketed a variety of antibiotics, including the blockbuster Cubicin for Staphylococcus aureus and complicated skin and skin structure infections, and had a late-stage clinical pipeline of antibiotics.

A substantial difference in size between private and public M&A is expected owing to the crucial role that public markets play in funding the growth of development-stage biotech companies12. However, one particularly notable private deal was Merck’s acquisition of SmartCells, founded by Todd Zion, an MIT chemical engineer. In 2010, SmartCells was sold for more than US$500 million, including upfront and downstream milestones, based on a preclinical asset called MK-2640 for type 1 diabetes13,14. The value created for SmartCells shareholders is unknown as the breakdown of upfront cash and contingent value rights was not disclosed. Another large private deal involved Civitas, which was on the cusp of an IPO when it sold to Acorda Therapeutics for US$525 million. Civitas was developing a drug for Parkinson’s disease ready for phase 3, and had filed an S-1 just 1 month before Acorda stepped in15.

## R&D pipeline

The average drug undergoes at least a decade of translational research and several clinical studies before it is approved by the FDA. Each phase of clinical testing costs millions of US dollars, typically leading to multiple publications and a better understanding of the targeted disease and its pathway. As a result of this complexity, we believe that examining the full R&D pipelines of MIT licensees—that is, all drugs developed over a company’s lifespan—can provide valuable insights into their economic and biomedical contributions to the biopharma industry.

To track their pipelines, we went through each company’s annual financial filings to the Securities and Exchange Commission (form 10-K) and manually extracted all investigational compounds in the clinical phase. We focus on the number of unique pipeline drug candidates (quantity), the highest stage of development for any indication (depth) and the therapeutic areas involved (breadth), with less emphasis given to the specific number of indications targeted. This is to maintain consistency with Mullard5, which excludes label expansions, but it is also due to the difficulties in quantifying the scope of a candidate during its early clinical phases (Supplementary Note, section G).

The final dataset consists of 281 drug candidates from 33 public MIT licensees over 23 years of development, spanning from 1994 to 2017 (Figs. 1 and 2; see Supplementary Note, section G, for breakdown by company). We find that oncological, neurological and genitourinary-related drugs are the most popular areas of development in MIT biotech companies—more than 50% of the companies were involved in at least one of these therapeutic areas. Of the 281 candidates, 51 are FDA-approved products. Among the companies, Alkermes outperforms the rest in both quantity and depth, leading the group with 22 approved products and an aggregate of 46 pipeline drugs developed or marketed in its portfolio. Other notable companies include Millennium with 26 compounds, and Indevus with 21 drug candidates. In the next section, we categorize the 51 approved drugs in the dataset to better reflect their origins and the contributions of MIT licensees.

## Drug approvals

The number of drug approvals by the FDA is the key benchmark of success in the biopharma industry. However, the number of approvals of NMEs/NBEs and of drugs with a PR designation can help further differentiate novel drugs and drugs that address large unmet medical needs or provide substantial clinical benefits1,5,16,17. To capture both types of innovation, we screen the financial filings of the 33 public companies, and use the Drugs@FDA database to generate a list of approved drugs with their targeted indications, years of approval, properties, NME/NBE statuses, and PR designations (Supplementary Note, section H).

Although label expansions for already approved drugs were excluded earlier to maintain consistency with the prior literature5, they are accounted for separately in Supplementary Note, section H. For inclusion in our dataset, we define a label expansion as the FDA approval of a drug for a different disease than prior approvals, and omit expansions into various lines of therapy or patient populations within a disease. For example, in 2003, Velcade received initial approval to treat third-line multiple myeloma. We exclude the 2005 expansion of Velcade into second-line multiple myeloma, but we do include the 2006 addition of mantle cell lymphoma to Velcade’s label.

The journey from MIT-licensee company formation and IP licensing to a successful drug approval is rarely straightforward. The high cost, high risk and lengthy timeline of drug development often result in tortuous paths, and can involve split ownership in the form of licensing deals and collaboration agreements, or the exchange of ownership over time, often through M&A transactions1. Therefore, we classify drug approvals in four ways to more accurately identify the origins of the assets and clarify the contributions of the MIT licensees to their respective FDA-approved drugs—Partners, Originators, Originators* and Acquirers—which are described below (Table 4).

Companies classified as Partners led the clinical development of drugs that were initially discovered by MIT licensees or were enabled by technologies from MIT licensees. Partners were common in the dataset, as biotech firms often license their technology to validate their science, generate non-dilutive capital or share risk. For example, Curis discovered a Hedgehog pathway inhibitor for basal cell carcinoma that eventually became Erivedge under a collaboration agreement granting Genentech (now Roche) an exclusive, royalty-bearing license. Although Curis played a key role in the discovery of the compound, Genentech had complete responsibility over all clinical development and commercialization, ultimately spending substantially more resources and time on the asset. In contrast, ARIAD discovered and advanced Iclusig through clinical trials on its own without any partners.

Companies classified as Originators had ownership of assets at varying stages of development that were ultimately acquired and developed by MIT licensees. For example, Millennium acquired LeukoSite in 1999, which came with Campath and LDP-341. LDP-341 later became Velcade, a first-in-class proteasome inhibitor for multiple myeloma18. In contrast to the case of ARIAD, where the approved molecules originated within company laboratories, Millennium did not contribute to the discovery of Velcade. The Originator classifier highlights the cases in which any MIT IP was likely far removed from the discovery of the assets.

The classification of Originator* company differentiates firms with drugs that were acquired after phase 3, which includes compounds that were NDA-ready, NDA-filed or already marketed. When Millennium acquired LeukoSite, an NDA for Campath in chronic lymphocytic leukemia had already been filed, whereas LDP-341 was only in phase 118. Thus, Millennium contributed meaningfully to the clinical development of Velcade, but not to Campath. The MIT licensee objectively had no role in the development of Originator* drugs, let alone any contribution from MIT IP. These drugs are excluded from our analyses.

Many successful MIT licensees were bought by Acquirers. For example, ARIAD, Abgenix and Millennium had pipeline assets that were eventually approved by the FDA following their acquisition. In these cases, we still attribute credit to the MIT licensees for their involvement. For example, Takeda acquired Millennium in 2008 and retained the business as its dedicated US oncology subsidiary. In 2015, Millennium and Takeda brought Ninlaro, a second-generation proteasome inhibitor, to market19. We consider Ninlaro as developed by Millennium with Takeda as a Partner to attribute credit appropriately, instead of excluding the asset entirely because the approval occurred after Millennium was no longer a standalone entity. Similarly, Abgenix’s XenoMouse monoclonal antibody technology continued to produce drug candidates such as Repatha20 after the company was acquired by Amgen. We list such cases in Supplementary Note, section H, and include them in our subsequent analysis.

Including 7 drugs approved post-acquisition, a total of 58 unique drugs have been on the market and under the ownership of MIT licensees (Table 5). However, these 58 include 27 Originator* drugs that were acquired by MIT licensees, which dilute the impact of MIT IP on compounds for which MIT licensees played a role in their discovery or clinical development. Excluding these assets, 22 out of 31 (71%) were an NME/NBE or had PR, key indicators of innovation and impact on unmet needs. Further excluding assets classified as Partners narrows the group to those drugs for which MIT licensees led their development to market. This subset had 16 approvals, of which 12 (75%) were an NME/NBE or had PR. These final 16 include a number of well-known drugs, including Gliadel, Vivitrol, Aristada, Cubicin, Onivyde and Velcade.

## IP

In the context of evaluating the impact of MIT IP, the number of patents granted to MIT licensees could be used as a proxy for the contribution of MIT’s licensees to biopharma innovation. However, it is important to note that this metric is not ideal. The mission of the biopharma industry is ultimately to improve the health outcomes of patients. It is more appropriate to measure innovation by assessing the approval of drugs, rather than granting of patents, whose technology may never reach patients.

The US Patent and Trademark Office database was used to collect data for the 33 public MIT licensees (cutoff at December 2017; Supplementary Table 11). The 33 companies licensed 258 unique patents from MIT in initial startup agreements totaling US$39.9 million in royalties. These companies were additionally granted 2,512 patents between 1985 and 2017, clearly showing that MIT licensees continue to innovate beyond an initial license from MIT. The number of patents filed by a therapeutics company can be very different depending on the nature of the innovation involved. Although certain types of innovation lend themselves to large patent estates to protect an investigational compound, others might rely on just a few patents based on core biological insights. We find that companies that developed novel technology platforms (for example, Millennium, Alnylam and Sangamo) were the most productive in terms of number of patents granted per year. In contrast, companies focused on developing in-licensed assets, such as Conatus Pharmaceuticals, tend to be focused on clinical development rather than on innovating new technology. In general, we do not find a strong correlation between the number of drugs approvals and the number of patents granted for MIT licensees (Supplementary Fig. 7). Although patent data do not measure the direct impact on patients, they do reveal one aspect of MIT’s contributions not reflected in drug approvals. Companies such as Alnylam and Sangamo are pioneering new technologies and had not yet achieved FDA approval before the cutoff of our dataset. However, as suggested by their extensive patent portfolios, both firms contributed substantial advances to the scientific community’s understanding of short interfering RNA (siRNA) and ZFN technologies as therapeutics during this time. (In August 2018, Alnylam received FDA approval for patisiran, a first-in-class siRNA therapeutic21.) ## Discussion An analysis of the 2,529 new drugs approved by the FDA from 1991 to 2017 shows that 31% were NMEs, and 24% had PR (Supplementary Table 12). In contrast, MIT licensees played a traceable role in the approval of 31 drugs over the same time period, of which 55% were NMEs and 55% had PR. This comparison is limited because of its small sample size and because the preference for NMEs and PR candidates by smaller biotech companies is unknown. Nevertheless, this suggests that MIT licensees may have been more innovative than the industry average. As shown by the small number of Orange Book citations, the convoluted link between academic IP and FDA approval makes an analysis that tracks the outcomes of academic IP difficult. This limitation arises owing to multiple factors: IP licensing is only an early step in a lengthy drug development timeline, companies may license IP from several institutions, and asset ownership is frequently multi-partied and variable over time. However, we believe that our approach—examining Orange Book citations, IPOs, M&As, R&D pipelines, drug approvals and IP of MIT licensees—provides a fair and comprehensive framework to simultaneously acknowledge MIT origins and recognize the complex, multi-party contributions that extend beyond MIT and are required to commercialize drugs. In some cases, MIT licensees found success after pivoting away from their initial IP and strategy. For example, Cubist Pharmaceuticals purchased daptomycin, a discontinued compound for Gram-positive infections in phase 2, from Eli Lilly in 199722. Cubist pushed daptomycin across the finish line as Cubicin in 2003. The drug subsequently became a blockbuster, and enabled Cubist to further acquire Adolor, Optimer, Calixa and Trius, with each acquisition resulting in an FDA-approved drug. Other examples include ARIAD and Millennium. However, even without clear links between drug approvals and academic patents, academic IP still contributes to three critical aspects of a company’s success: bringing together unique people and talent; catalyzing a company’s financing; and serving as a foundation for future R&D. The case of Millennium is a prime example: its technology platform enabled the firm to sign multiple partnership deals with large pharmaceutical companies and raise substantial capital from the public markets. By 1998, Millennium had struck deals totaling US$1 billion with the pharmaceutical giants Eli Lilly, Roche and Bayer23. These alliances provided Millennium with substantial funding and access to key downstream technologies for drug development24. In 2000, Millennium raised more than US1 billion from a follow-on public offering and a convertible debt sale. Its lucrative partnerships validated its platform and attracted substantial capital from investors. This combination of business development and financing activity ultimately allowed Millennium to make its transformative acquisitions of LeukoSite and COR Therapeutics. Clearly, the initial licensing of MIT IP had a broad impact on the organization. Our analysis does not capture all aspects of MIT’s contribution to therapeutic innovation. Within our dataset, biotech companies such as Alkermes, Abgenix and Millennium have licensed their technologies to large pharmaceutical companies. Although drugs disclosed within the financial filings of MIT licensees were tracked, it is possible that others predominantly owned by outside partners have been missed. Specifics of early-stage drug development activity are not necessarily disclosed in a large pharmaceutical company’s financial filings, and a biotech company may be contractually limited in its own disclosures of such programs. Similarly, access to public company financial filings before 1994 was limited, and preclinical assets or drug discovery technology of an acquired private company may remain undisclosed. One example that falls outside our framework is the case of Idun. Idun was founded in 1993 and licensed multiple MIT patents based on the work of Robert Horvitz. Idun and Abbott Laboratories (now AbbVie) collaborated to develop inhibitors of Bcl-2, a regulator of apoptosis. Although the initial molecule produced by this collaboration did not progress owing to poor oral bioavailability, Abbott developed the follow-on compounds navitoclax and venetoclax. The latter has been approved for the treatment of various hematological malignancies including chronic lymphocytic leukemia. It has also been granted multiple breakthrough therapy designations based on its profound therapeutic impact particularly in diseases of high unmet need such as acute myeloid leukemia. Further beyond our dataset, four biotech giants, Genzyme, Biogen, Amgen and Genentech, were co-founded by MIT faculty or people with connections to MIT. However, these companies did not license MIT IP and consequently were not included in our analysis (https://be.mit.edu/directory/harvey-f-lodish and http://web.mit.edu/sharplab/shortbio.html)25,26. It is not uncommon for MIT faculty and affiliates to launch therapeutics companies without licensing MIT IP, such as in the case of Verastem, co-founded by Robert Weinberg and Eric Lander27. Finally, MIT research publications that are not translated into patents also advance the therapeutics landscape. Our analysis does not capture these sources of impact because they do not involve an explicit licensing transaction with the MIT technology licensing office. On the other hand, as shown by the analysis of IP generated by the MIT licensees, entirely new classes of potentially transformative drugs are on the horizon whose development uses MIT IP. A brief survey of drugs in the development phase underscores MIT’s contribution to the current creation of innovative therapeutics. Alnylam, bluebird bio, Editas Medicine and Sangamo are just four examples of firms that have licensed MIT IP to develop platform technologies capable of discovering new drugs. Alnylam was founded in 2002 by a group of MIT faculty, including Langer and Phillip Sharp, to develop RNA-mediated interference (RNAi) therapeutics based on siRNA discoveries. The firm licensed patents from MIT on the formulation and delivery of siRNAs, and also engaged in a 5-year research collaboration to improve delivery to target tissues. Alnylam has a broad clinical development pipeline of seven assets that includes patisiran, an RNAi therapeutic for hereditary ATTR amyloidosis, a severe neuropathy, which is under PR by the FDA as of the cutoff of our dataset28. (In August 2018, Alnylam received FDA approval for patisiran as a first-of-its-kind targeted RNA-based therapy21; see Supplementary Note, section K, for other examples.) It must be noted that other parties contributed to these innovative platforms in addition to MIT, such as Tekmira, which licensed its siRNA patents to Alnylam, and the Scripps Research Institute, which licensed its zinc finger proteins IP to Sangamo. In fact, Sangamo’s 2017 annual report references the ZFN IP licensed from the California Institute of Technology and the University of Utah, but not the 1996 patent license agreement between Sangamo and MIT. Although the original patent from 1996 may not be scientifically important relative to the more recent ZFN patents, perhaps it was critical to nucleate the company and enable Sangamo to become the leader of ZFN technology. This highlights the critical point that developing a drug starting from an initial academic patent can require more than 20 years, at which time the foundational patent may be expired and no longer commercially relevant. Most MIT licenses are not composition of matter patents, which are the most commercially valuable Orange Book listable patents. Unlike many other institutions active in academic-stage life sciences research, MIT does not have a medical school. This precludes large animal or early clinical testing before academic licensing and may bias the typical MIT license towards technology that underpins a new platform rather than towards an identifiable drug that would have a clearer, shorter path to market. A company that licenses technology may spend a decade developing the platform and identifying the optimal targets/diseases to pursue. Thus, it is not surprising if MIT patents have expired by the time a drug is approved. Nevertheless, MIT IP has played an important role in the formation of innovative companies and development of novel therapies. Numerous drug candidates produced by these companies are likely to reach the market in the coming years. ## Perspective As a result of the complexity of academic IP transactions and the difficulty in determining causation, the systematic tracking of the outcomes of academic IP transactions has been neglected, despite its obvious benefit to all players and its potential to increase the rate of biomedical innovation. The use of the framework developed in this paper to analyze university IP would not only allow cross-university comparisons, but could also provide convincing data in favor of increasing funding to bridge the financing gap for preclinical assets and drug discovery technology. As shown by the recent US130 million collaboration between Deerfield Management and Duke University29, there is both precedent and desire for such funding to accelerate innovation at the academic level.

More speculatively, with such data in hand, one could conceive of an academic translational medicine fund that raises money from limited partners to invest in therapeutics companies that license IP among a consortium of universities, each of which contributes IP to a commingled pool using a standardized IP agreement, managed by an external third-party portfolio manager dedicated to selecting and developing assets on behalf of investors in the fund, similar to the CRISPR patent pool initiative by MPEG LA30. If this fund were large enough, it could also provide additional value-added services such as animal studies, medicinal chemistry, toxicology and even clinical-trial designs, not unlike the support offered by the US National Institutes of Health’s National Center for Advancing Translational Sciences.

Some seasoned biotech venture capital firms may be skeptical about the prospect of multiple universities, each with its own unique group of stakeholders, goals and constraints, collaborating productively in commercializing their IP. However, the downward pressure by funding organizations on overhead rates at a time of tremendous expansion in life sciences research has created the potential for a more conducive environment for cooperation between multiple universities, especially if motivated by the prospect of major funding. One example is the research consortium between Celgene, the University of Pennsylvania, Columbia University, Johns Hopkins and Mount Sinai Medical School, established to accelerate the discovery of new cancer treatments31.

Given the assumption that innovation in biopharma is primarily constrained by a lack of capital rather than talent or worthwhile ideas, an academic translational medicine fund could greatly accelerate the commercialization of new drug discovery technologies and the discovery of new and innovative medicines. As suggested by our data, such an investment could prove beneficial not only for the researcher and the investor, but also, and more importantly, for patients.

## References

1. 1.

Munos, B. Lessons from 60 years of pharmaceutical innovation. Nat. Rev. Drug Discov. 8, 959–968 (2009).

2. 2.

AUTM U.S. Licensing Activity Survey FY 2016 (Association of University Technology Managers, 2016).

3. 3.

Campbell, E. G., Powers, J. B., Blumenthal, D. & Biles, B. Inside the triple helix: technology transfer and commercialization in the life sciences. Health Aff. 23, 64–76 (2004).

4. 4.

Mowery, D. C., Nelson, R. R., Sampat, B. N. & Ziedonis, A. A. in Industrializing Knowledge: University-Industry Linkages in Japan and the United States (eds. Branscomb, L. M., Kodama, F. & Florida, R.) 269–306 (MIT Press, 1999).

5. 5.

Mullard, A. 2016 FDA drug approvals. Nat. Rev. Drug Discov. 16, 73–76 (2017).

6. 6.

Cohen, F. J. Macro trends in pharmaceutical innovation. Nat. Rev. Drug Discov. 4, 78–84 (2005).

7. 7.

Lanthier, M., Miller, K. L., Nardinelli, C. & Woodcock, J. An improved approach to measuring drug innovation finds steady rates of first-in-class pharmaceuticals, 1987–2011. Health Aff. 32, 1433–1439 (2013).

8. 8.

Stevens, A. J. et al. The role of public-sector research in the discovery of drugs and vaccines. N. Engl. J. Med. 364, 535–541 (2011).

9. 9.

Food and Drug Administration. Indevus Pharmaceuticals, Inc.; withdrawal of approval of a new drug application. Federal Register https://www.federalregister.gov/documents/2007/01/30/E7-1414/indevus-pharmaceuticals-inc-withdrawal-of-approval-of-a-new-drug-application (30 January 2007).

10. 10.

Prokesch, S. The Edison of Medicine. Harv. Bus. Rev. March−April, 134–143 (2017).

11. 11.

Herper, M. Lilly gives away Prozac, makes money. Forbes (16 November 2001); https://www.forbes.com/2001/11/16/1116lilly.html#3ecdbf4635ee

12. 12.

Morrison, C. & Lähteenmäki, R. Public biotech in 2014—the numbers. Nat. Biotechnol. 33, 703–709 (2015).

13. 13.

Merck to acquire SmartCells, Inc. Merck https://www.merck.com/news/merck-to-acquire-smartcells-inc/ (2 December 2010).

14. 14.

Matheson, R. The half-billion-dollar idea. MIT News http://news.mit.edu/2013/todd-zion-smartcells-0408 (8 April 2013).

15. 15.

Fidler, B. Alkermes spinoff Civitas, poised for IPO, sells to Acorda for $525M. Xconomy https://xconomy.com/boston/2014/09/24/alkermes-spinoff-civitas-poised-for-ipo-sells-to-acorda-for-525m/ (24 September 2014). 16. 16. New drugs at FDA: CDER’s new molecular entities and new therapeutic biological products. Food and Drug Administration http://www.fda.gov/drugs/development-approval-process-drugs/new-drugs-fda-cders-new-molecular-entities-and-new-therapeutic-biological-products (2020). 17. 17. Priority Review. Food and Drug Administration http://www.fda.gov/patients/fast-track-breakthrough-therapy-accelerated-approval-priority-review/priority-review (2018). 18. 18. Millennium and Leukosite agree to merge to accelerate Millennium’s move toward commercialization. Millennium Pharmaceuticals https://www.bamsec.com/filing/91205799001797?cik=915320 (14 October 1999). 19. 19. Weisman, R. Takeda gets FDA approval for multiple myeloma pill. The Boston Globe (20 November 2015); https://www.bostonglobe.com/business/2015/11/20/takeda-gets-fda-approval-for-multiple-myeloma-pill-developed-cambridge/YRxwHUrcljJ3pY5iS6VvtJ/story.html 20. 20. Raal, F. J. et al. PCSK9 inhibition-mediated reduction in Lp(a) with evolocumab: an analysis of 10 clinical trials and the LDL receptor’s role. J. Lipid Res. 57, 1086–1096 (2016). 21. 21. Alnylam announces first-ever FDA approval of an RNAi therapeutic, ONPATTROTM (patisiran) for the treatment of polyneuropathy of hereditary transthyretin-mediated amyloidosis in adults. Alnylam Pharmaceuticals https://investors.alnylam.com/press-release?id=22946 (10 August 2018). 22. 22. Tally, F. P. & DeBruin, M. F. Development of daptomycin for Gram-positive infections. J. Antimicrob. Chemother. 46, 523–526 (2000). 23. 23. Herper, M. Biotech battle scars. Forbes (14 September 2007); https://www.forbes.com/2007/09/14/biotechnology-venture-capital-biz-sci-cx_mh_0914venture.html#20de6b2b681a 24. 24. Thomson StreetEvents. Q1 2001 Millennium Pharmaceuticals earnings conference call transcript. BamSEC https://www.bamsec.com/fulltext-search/1002637?q=aventis&cik=1002637 (2001). 25. 25. Tung, R. MIT related to top biotech companies. The Tech http://tech.mit.edu/V115/N67/biotech.67n.html (1996). 26. 26. Study traces impact of government-funded MIT research on biotechnology and drugs. MIT News Office http://news.mit.edu/1996/biotechmitgov (22 January 1996). 27. 27. Verastem. Verastem secures$16 million in series A financing for novel cancer stem cell drugs. Biospace https://www.biospace.com/article/releases/-b-verastem-b-secures-16-million-in-series-a-financing-for-novel-cancer-stem-cell-drugs-/ (16 November 2010).

28. 28.

Alnylam Pharmaceuticals. Alnylam announces FDA acceptance of new drug application (NDA) and priority review status for patisiran, an investigational RNAi therapeutic for the treatment of hereditary ATTR (hATTR) amyloidosis. Business Wire https://www.businesswire.com/news/home/20180201006407/en/Alnylam-Announces-FDA-Acceptance-New-Drug-Application (11 August 2018).

29. 29.

Duke University. Duke University & Deerfield Management announce four points innovation. Duke University Office of Licensing & Ventures https://olv.duke.edu/news/duke-deerfield-announce-four-points-innovation/ (18 December 2019).

30. 30.

31. 31.

Lawrence, S. Celgene backs four NCI Cancer Centers with \$50M to create research consortium. FierceBiotech https://www.fiercebiotech.com/biotech/celgene-backs-four-nci-cancer-centers-50m-to-create-research-consortium (22 June 2016).

## Acknowledgements

We thank MIT’s Technology Licensing Office for agreeing to participate in this study and providing access to their historical data, in particular, G. Lindsay, J. Lura and L. Millar-Nicholson. We thank D. Gifford, R. Horvitz, R. Langer, H. Lodish, E. Roberts, P. Sharp, P. Szolovits, C. Westphal and H. Williams for helpful comments and discussion, and J. Cummings for editorial assistance. Research support from the MIT Laboratory for Financial Engineering is gratefully acknowledged. The views and opinions expressed in this article are those of the authors alone, and do not necessarily represent the views and opinions of any institution or agency, any of their affiliates or employees, or any of the individuals acknowledged above.

## Author information

Authors

### Corresponding author

Correspondence to Andrew W. Lo.

## Ethics declarations

### Competing interests

All authors were affiliated with MIT during the course of this project. L.N. was head of MIT’s Technology Licensing Office at the start of the project, and currently consults for biopharma companies and investors. A.W.L. reports personal investments in private and public biotech companies, biotech venture capital funds, and life sciences mutual funds, and is a cofounder and partner of QLS Advisors, a healthcare analytics company; an advisor to BrightEdge Ventures; a director of Roivant Sciences Ltd, BridgeBio Pharma and Annual Reviews; chairman emeritus and senior advisor to the AlphaSimplex Group; and a member of the Board of Overseers at Beth Israel Deaconess Medical Center and the NIH’s National Center for Advancing Translational Sciences Advisory Council and Cures Acceleration Network Review Board.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

## Supplementary information

### Supplementary Information

Supplementary Note, Figs. 1–7 and Tables 1–12.

## Rights and permissions

Reprints and Permissions

Huang, S., Siah, K.W., Vasileva, D. et al. Life sciences intellectual property licensing at the Massachusetts Institute of Technology. Nat Biotechnol 39, 293–301 (2021). https://doi.org/10.1038/s41587-021-00843-5

• Accepted:

• Published:

• Issue Date: