‘Nano Scale [is] Magic Scale’: On EPR,Unicorns,and Enchantment in Nanomedicine

Abstract

Nanotechnology is a scientific field where STS scholars have observed the influence of magical thinking. Here we contribute to that literature by describing a phenomenon in the field of drug delivery called the EPR (Enhanced Permeability and Retention) Effect. When the ‘nano’ boom of the 2000s brought intense interest from outside pharmacology, EPR spawned so many efforts and enterprises in nano-based drug delivery that it became the leading basis for nanomedicine’s promise. Following the 2016 failure of a prominent nano-biotech firm exploiting EPR, the collapse of the field offers insights into the role such theories or narratives played in the formation of nanomedicine and, more broadly, how contemporary technoscience may incorporate elements of enchantment—particularly where exposed to financialization.

Introduction

Here we recount how an observation in the 1980s by pharmacologists, that particles of a certain size tended to accumulate preferentially in certain tumours—that is, certain tumours transplanted into certain mice, a common experimental system for cancer research—inspired a generation of research into drug delivery systems. Dubbed and reified as the EPR Effect (for Enhanced Permeability and Retention; see below), this research program was recast during the 2000s in terms of then-ascendant ‘nanotechnology’ and was often held up as a prime example of the promise of ‘nanomedicine’. We show that the EPR Effect, its investigation, and the attempted exploitation of its potential for cancer therapy, played a keystone role in organizing a research community within the pharmacological discipline of drug delivery, in securing funding for its participants by enabling related research projects to be labeled ‘nanomedicine’, and in underwriting the medical promise and glamour of nanotechnology more broadly. That is, despite its relatively weak empirical basis and no claim to demonstrable utility, the EPR Effect played a central, nucleating role in the crystallization of nanomedicine both internally, structuring the professional activity of scientists, and externally, where it attracted and shaped investments both public and private. This central role became especially apparent when, in 2016, the Effect’s credibility suddenly evaporated after a prominent EPR-based drug candidate failed in clinical trials. These powerful repercussions reflect the usually unseen work that theories or stories like EPR can do in organizing and supporting scientific fields, and raise questions about the way the storytellers accomplish such effects. Further, we will argue that EPR can be considered an enchantment—what Max Weber called a ‘mysterious incalculable force’—and can be interpreted as the product of the collective narrative labour that turns tenuous scientific narratives into bankable certainties. We suggest that alertness to those special stories that we call enchantments can enrich the analysis of technoscience more generally.

We read EPR’s arc alongside that of Japan’s Protein 3000 project which, despite meeting its technical goals, was deemed a failure even in its early stages precisely because it lost the ‘magic’ that bound disparate participants and patrons together (Fukushima, 2016, p. 10; Star & Griesemer, 1989; Stemerding & Hilgartner, 1998). The situation with EPR was the inverse: Despite failing its technical goals (of leading to cancer drugs that worked outside the contrived mouse system where it originated), it had too much magic—enchantment—to be abandoned. We use ‘enchantment’ in Bennett’s (2001) sense of secular moments of wonder that animate ethical and political attachments, and interpret it as the product of collective narrative labour. In the case we treat here, obvious signs of enchantment will be found in actor’s categories such as ‘wizard’, referring to the entrepreneurial MIT nanoscientist Robert Langer, and ‘magic bullet’, referring to a highly selective nanomedical drug delivery platform. The latter metaphor, also discussed by other scholars of nanomedicine (Bensaude-Vincent & Loeve, 2014), was introduced over a century ago by Paul Ehrlich—and, like Ehrlich’s famous drug Salvarsan, ultimately traces its roots to Paracelsian alchemy (Bosch & Rosich, 2008; Galdston, 1954; Lenoir, 1988). What matters for us is not its historical pedigree but the enchantment work it performs today (Rip, 2006).

Promissory expectations, long recognized in STS as performative engines of innovation (Borup et al., 2006), become even more potent analytic tools when cast as stories whose empirical truth is secondary to their capacity to co-ordinate action under radical uncertainty (Beckert, 2013; see Shiller, 2017 on ‘narrative economics’ and Arnaldi et al., 2014 on nano and uncertainty). Yet fictionality alone cannot explain why some narratives hold the collective imagination for decades while others evaporate. Here we draw on anthropological analyses of modern enchantment for an answer (Appadurai, 2020; Josephson-Storm, 2017). Enchantment is a mechanism that translates probabilistic futures into organized collective action, science fiction into social reality. It is what allows venture capitalists to treat an experimental tumour-targeting nanoparticle that sometimes works in mice as a near-inevitable blockbuster medicine, or to interpret Langer’s portfolio as evidence of infallible entrepreneurial prowess. The Weberian lens adds depth to an analysis framed in terms of fictional expectations––it shows how charismatic figures (Langer) or artefacts (EPR graphs) come to embody that future in ways spreadsheets alone can’t, and it explains why investors stay committed despite feeble empirical support (such as BIND’s unimpressive early trials). The spell isn’t just cognitive probability but also is social adhesion.

We treat the EPR Effect not as a settled fact (which it was, to many scientists) but as an epistemic fiction—a charismatic ‘unicorn’ (Erden & Levy, 2020) that mobilized billions despite questionable evidence. Viewing its career through the lens of enchantment highlights the charisma, ritual, and narrative stickiness that pure expectation analyses miss. We do not consider other areas of nanomedicine such as (i) nano-diagnostics, (ii) regenerative/implantable nanomaterials, and (iii) non-EPR drug carriers. In money terms at least, all of these were minor parts of nanomedicine’s total promise compared to EPR-based drug delivery. Instead, we focus on how an EPR-based cancer nanotherapy promise ran from mouse cages to venture capitalist press releases and back again in a self-reinforcing cycle, effectively driving the whole field of nanomedicine for well over a decade. Like the 1720s South Sea Bubble, MacKay’s (1869) emblem of speculative ‘madness’, EPR was a creature not just of fictional expectations (Beckert, 2013) but also of what we are calling enchantment. Where coffee-house pamphlets once spun flimsy seafarer rumours into great fortunes, today Science editorials and slide decks pitching to venture capitalists turn selected mouse data into tomorrow’s cancer cure, animating technoscience with the charismatic storytelling of finance.

The Emergence of the EPR Effect

Long before nanomedicine became a keyword, the Enhanced Permeability and Retention (EPR) Effect captured a following within pharmacology. By suggesting that the very chaos of leaky tumour vasculature could be turned into a privileged gateway for macromolecular cancer drugs, the EPR story bound bench scientists, clinicians and early industrial partners into a shared imaginary of precision without molecular targeting (see Noury & López, 2017). That early coalition, documented in animal experiments and Phase I trials well before the 2000 launch of the US National Nanotechnology Initiative, shows how enchantment can precede hype while acting as the epistemic glue that holds a venture together until louder narratives arrive.

In 1992 the rising young British pharmacologist Len Seymour published a review on activity in his field, drug delivery systems. His focus was a recently described phenomenon where (it was held) tumours have leaky blood vessels (i.e. enhanced permeability), which, in combination with their typically poor lymphatic drainage (i.e. enhanced retention), tends to make them accumulate macromolecules and other blood-borne particles of a certain size range (hence EPR, Enhanced Permeability and Retention). EPR was essentially a theory—a hypothetical mechanism of general applicability—to explain evidence from a set of mouse experiments suggesting that, simply put, big molecules and certain-sized particles could get easily from the bloodstream into a tumour, but they had trouble getting out. The thrust of Seymour’s (1992) review was that the Effect offered an important way forward for research aiming to deliver cancer chemotherapy agents selectively so as to achieve better clinical outcomes.

The background to Seymour’s review was research during the 1980s showing that the capillaries in some solid tumours in mice tended to have more permeable linings, enabling macromolecules and particles, normally too large to escape the bloodstream, to flow into the space in and around the tumours. As Seymour (1992, p. 151) put it, a ‘ripple of excitement’ had swept through the world of drug delivery when these observations first were mooted, because they suggested an obvious way to concentrate anti-cancer agents where they were needed most. Two specific avenues of research had been stimulated.

Immediately, researchers working in the hot and richly-funded field of monoclonal antibodies had perceived a way of delivering their imagined ‘smart bombs’, consisting of a toxic drug linked to a custom-made antibody specifically attaching to cancer cells, directly to the tumour target zone (Bensaude-Vincent & Loeve, 2014; Cambrosio & Keating, 1988, 1995; Poste & Kirsh, 1983). These large, ‘actively targeted’ antibody-based drugs (molecular weight about 150,000 Mr and 10-15 nm in size) would be able to get to the cancer tissue because of the leaky tumour blood vessels, they hoped.

The other avenue stimulated was ‘passive’ drug targeting, so-called because it involved no specific molecular interaction between tumour cells and drug or drug vehicle. The concept was that particles loaded with anticancer drugs would leak out of the circulation into the vicinity of the tumours based only on size and other physical properties affecting diffusion (shape, charge), otherwise remaining in circulation and sparing normal tissue because they would be too big to escape normal vessels. The much higher concentrations of toxic drug around the tumours would cure the cancer.

However, said Seymour in framing his review, by 1992 the drug delivery field had come to regard these hopeful avenues as a ‘fading prospect’ (Seymour, 1992, p. 151). First (as emphasized by one of Seymour’s key sources but not Seymour himself), not all the tumours used in the animal experiments exploring passive targeting had especially, or predictably, leaky vessels. Second, much of the work attempting to exploit passive targeting had employed liposomes, particles consisting of an oily bubble containing a watery core, and this work had found that only the smallest producible liposomes—about 30 nm according to a key authority—could escape the circulation into tumours. And liposomes that small could not transport much drug. Sadly, complained Seymour (1992), ‘the influence of particle-based drug delivery’ advocates—here, liposome researchers—was such that their discouragement had suppressed passive drug targeting research in general (p. 151).

Seymour’s goal was to revitalize efforts at passive targeting for cancer drug delivery—with a different kind of delivery vehicle. A 1986 study by Japanese pharmacologists Matsumura and Maeda exploring the leaky tumour vessel phenomenon had first named the EPR Effect. Where it differed from previous work along these lines was in not using liposomes as a vehicle but instead a protein drug linked to polymers, which expanded its size and changed other physical properties. Specifically, the Japanese team showed that the cytotoxic drug NCS, a small protein derived from fungi, had superior anticancer properties against mouse tumours when they linked it covalently to a styrene-maleic acid copolymer (SMA). SMANCS, the polymer-linked version, exhibited a much longer serum half-life, staying longer in the bloodstream, than original NCS, and it accumulated dramatically in the tumour compared with bloodstream concentrations (Matsumura & Maeda, 1986). Exploring the idea that this accumulation was largely due to the increased size of the drug molecules, Matsumura and Maeda found that the bigger the protein-polymer conjugate, the better it tended to concentrate in tumours. They explained this desirable accumulation not just in terms of the tumours’ enhanced permeability (due to capillary leakiness, letting more of the big drug molecules out), but also retention, in that tumours tended to have poor drainage by lymphatic vessels that especially retarded clearance of larger particles. In this and related studies soon after, they suggested the effect could be exploited using drugs or drug vehicles above 85,000 Mr in molecular mass (Kobayashi et al., 1988). For Seymour the implications were that more drug delivery systems in the size range immediately below 30 nm should be explored via drug modification with polymers.

In 1992, Seymour, together with his postdoctoral mentor the pharmacologist Ruth Duncan, were already collaborating with Maeda—and in fact were EPR’s leading champions in Europe. Working at Keele University and then University of London, Duncan had around 1989 formed a collaboration with Prague polymer chemist Karel Ulbrich to target cancer drugs to tumour tissue by attaching polymers to them, following Maeda except using hydroxypropylmethacrylamide (HPMA) rather than SMA, via a chemical link that would be cleaved to release the drug once inside a target cell. The drugs, typical small molecule chemotherapy agents like daunomycin and doxorubicin, would thus concentrate their prodigious toxicity on the cancer cells and spare the rest of the body—provided they could be delivered preferentially to tumours (Cassidy et al., 1989; Seymour et al., 1990, 1991). This was where the EPR Effect was crucial for Duncan. With radioactively labelled HPMA polymers of various sizes from about 20,000 Mr to about 780,000 Mr, Duncan and her collaborators, together with Maeda’s group, confirmed that the largest accumulated best in the mouse tumours Maeda had been using—particularly drug carriers from 78,000 Mr upward (Seymour et al., 1995). The finding came a little late to influence the design of the drug candidate that the Duncan and Ulbrich team had already selected for clinical testing in the mid-1990s, an HPMA-doxorubicin conjugate of only about 28,000 Mr dubbed PK1. PK1 was abandoned after Phase I clinical testing (Duncan et al., 1998; Reynolds, 1995; Vasey et al., 1999).

During the late 1990s, Duncan and collaborators extended their HPMA-based anticancer platforms by developing more biodegradable polymer carriers (Duncan, 1999; Hreczuk-Hirst et al., 1999, 2001). In a landmark 1999 review, she reflected on ten years of modest gains and, echoing Seymour’s 1992 insight, positioned the EPR Effect as the linchpin for polymer therapeutics: protecting large protein drugs targeting angiogenesis and apoptosis, safeguarding gene therapy polynucleotides in circulation, and of course concentrating polymer-cytotoxic conjugates at solid tumours. Although liposomes and antibody drug conjugates also aimed to exploit the EPR Effect, Duncan argued that polymer modification affords unmatched control over retention and tumour penetration. This argument forged a decisive link between the EPR Effect and polymer delivery platforms, an innovation that would later be subsumed under the banner of ‘nanomedicine’. At no point in her 1999 review was any drug class referred to with the ‘nano’ prefix, even though most of those discussed fell into the size range above 1 nanometer and below 1 micron (Duncan, 1999). But by 2006 Duncan was working hard to make polymer therapeutics and ‘nanomedicines’ synonymous, going so far as to define a drug ‘nanoparticle’ as ‘formed from natural or synthetic polymers’, thus excluding liposomes and antibodies (Duncan, 2006, p. 689).

NNI and ‘Nanomedicine’ Research Involving EPR

This was just one of many strange effects worked by the US National Nanotechnology Initiative (NNI). Through the NNI, launched in 2000, state policy institutionalized the enchantment already latent in laboratory EPR work. Nano-evangelist and NNI administrator Mihail Roco’s slogan that the nanoscale was a ‘magic scale’ appeared verbatim in early NSF solicitations, and President Clinton’s East Room announcement promised breakthroughs that were ‘only imagination away’. By embedding that language in official funding calls, with magic sitting alongside technical aims in the very first NNI programme description, the government converted a charismatic promise into a bureaucratically sustained future, effectively codifying enchantment (M. Kearnes & Macnaghten, 2006; Mody, 2004; Roco, 2023).

The magnitude of that commitment helps explain why expectations around the EPR Effect became so inflated and remained buoyant for so long (Crabu, 2014). Coordinated across US agencies including NSF, NIH, DOE and NASA, NNI spending climbed from roughly $500 million in 2001 to nearly $2 billion annually by 2010, passing a cumulative $40 billion by 2023 (Roco, 2023). Similar ‘nano-roadmaps’ in France, the UK, Japan, China and elsewhere reinforced a transnational funding ecology that treated the EPR Effect as a ready-made portal to futuristic cancer cures, years before robust human data existed (M. B. Kearnes & Rip, 2009; Miller, 2017).

Within this performative coil of funding and rhetoric, persuasive narratives around ‘magic bullets’ and nanotech ‘magic’ didn’t simply follow investment trends but actively amplified them, each round of solicitations and grand-challenge framing serving as its own ritual spell to conjure belief and draw fresh capital. In this sense, the NNI’s rebranding of polymer platforms as ‘nanomedicine’ functioned like financial hype cycles (Shiller, 2017) but on a broader stage, not because the science itself changed overnight, but because of state-sponsored storytelling. Among US agencies, the NIH has been the biggest distributor of NNI research funding, ¹ reflecting the especially high hopes of medical benefits from nanotechnology noted by many scholars (e.g. Crabu, 2014; Noury & López, 2017). Through its Common Fund mechanism for interdisciplinary projects, from 2010 to 2015 the NIH supported translational research in eight Nanomedicine Development Centers as well as many investigator-driven grants. ² The NIH’s National Cancer Institute (NCI) spawned a fleet of Centers of Cancer Nanotechnology Excellence (CCNE), linked to its existing Cancer Centers—elite cancer hospitals like Stanford, Sloan-Kettering, Johns Hopkins, Massachusetts General Hospital (MGH), and Emory. ³ The NIH’s Heart, Lung and Blood, General Medical Sciences, Biomedical Imaging and Bioengineering, and Environmental Health Sciences Institutes each established dedicated nanomedicine programs as well. ⁴ The National Science Foundation, the second-largest cumulative contributor to the NNI, has also invested hundreds of millions of dollars in projects labeled ‘nanomedicine’. ⁵

The high hopes and flood of funding that followed on the heels of the NNI had at least two specific effects on the drug delivery field. For nanotechnology program directors and advocates, they induced a search for successful examples of health care innovation that could be called ‘nanomedicine’, especially as the years of NNI expenditure wore on. The cancer drug Doxil, a liposomal preparation of the cytotoxic chemotherapy agent doxorubicin invented in the 1980s and first approved in 1995, was frequently cited as an example of how nanotechnology benefits health despite the fact that it was developed long before the NNI (Barenholz, 2012). ⁶ Thus, NNI put drug delivery in a transnational spotlight and demanded new breakthroughs. For drug delivery pharmacologists (like many other biomedical researchers), it motivated them to repackage existing research programs as ‘nanomedicine’, to access the new funding streams and capture other patronage affected by nano-enthusiasm. The way Duncan rebranded her existing polymer research program as ‘nanoparticle’ drug delivery has already exemplified this. The same would be true for much of the cancer drug delivery research that relied on the EPR Effect, because EPR offered a rationale for focusing on the physical properties of drug delivery vehicles in the nanometer size range. That is, EPR gave drug delivery researchers and other scientists studying the biological properties of particles a convenient way to tap into nanomedicine money and enthusiasm. Fed by NNI funding, entrepreneurial researchers across the United States, like Chad Mirkin, Ulrich Wiesner, Joseph DeSimone, and Robert Langer—all of whom relied significantly on the EPR Effect in their cancer-related work on the biological properties of particles—expanded their already large laboratory capacity, trained more students, and patented prolifically. ⁷ With the state positioning itself as a handmaiden to high tech industry, ⁸ they also spun off ‘nanomedical’ companies with vigour. These included firms such as Elucida, Liquidia, and Nanosphere, promising to enhance the bioavailability of drugs, decrease toxicity and thus increase effective dosage, or to simultaneously identify and target diseased tissue ‘theranostically’. ⁹ But amongst the many formidable academic groups across the US to share in the multibillion dollar nano-bonanza, Langer’s MIT lab stands out. Viewed by the finance world as a ‘wizard’ of biotech (and often compared with Thomas Edison, the ‘Wizard of Menlo Park’), in the 2010s Langer held over a thousand patents, and these were said to have spawned hundreds of companies, more than 30 of which he co-founded himself. ¹⁰ Langer led the CCNE at MGH during its 2005-2015 lifespan, and was credited as a key figure in the development of the aforementioned Doxil—now recast not just as a success in ‘nanomedicine’ but one that worked because of the EPR Effect (Maruyama, 2011). As we will argue, the label of ‘wizard’ or enchanter for Langer deserves some serious consideration.

The impetus behind nano-oncology was never just routine grant-seeking. It mobilized a ‘promissory economy’ (Brown, 2003) amplifying expectations to a pitch that dwarfed ordinary academic boosterism and, at times, even the exuberance typical of recent biotech (Fortun, 2008; Gaillard et al., 2023). Plotted on the familiar Gartner hype-cycle, ¹¹ the field raced from ‘technology trigger’ to ‘peak of inflated expectations’ with exceptional speed (see also Loesch, 2006). Yet those curves capture only the volume, not the tenor, of the claims: Advocates invoked something like a tumour-sponge effect that would make chemotherapeutics seek out cancer cells as unerringly as iron filings find a magnet. Beyond merely the peak decibel level, nanomedicine’s enchantment lay in the narrative architecture that channelled resources and interest. Langer’s pronouncements illustrate a weaving of ideas and institutions to reinforce certain activity patterns within and expectations around nanomedicine, and much of that work (which we are calling enchantment) involved EPR in particular. One prominent 2007 review article by Langer and colleagues on the prospects of nanomedicine concludes with the demure protestation that ‘although we are still far from Nobel Prize winner Paul Ehrlich’s ‘magic bullet’ … we will soon enter an era in which nanocarrier-based approaches will represent an important modality within therapeutic and diagnostic oncology’ (Peer et al., 2007, p. 758). In another prominent review Langer and two former postdocs, Andrew Z. Wang and Omid Farokhzhad (the latter soon to return in our account), went further in specifying the nanoparticles that would soon be these not-quite-magic bullets – those that ‘through the enhanced permeability and retention (EPR) effect, preferentially accumulate in tumor cells’ (Wang et al., 2012; similarly, Tong & Langer, 2015). In such dual-audience platforms as Nature Nanotechnology Langer artfully sold the EPR Effect to business readers while still dispassionately reviewing the science, therefore conferring value on enterprises exploiting it—not just by projecting cancer cures, but also addressing specific business concerns such as the superior intellectual property protection afforded by nanoparticle formulations (Burgess et al., 2010; R. Langer, 2013). He also worked to reshape the public policy environment to favor his enterprises, for example in a 2011 special Policy Forum in Science magazine (the most influential US scientific journal) arguing for more NIH money and special interdisciplinary peer review procedures for research on ‘nanoparticle therapeutics’ and similar biomedical innovation involving the ‘convergence’ of physical and life sciences. Such interventions were performative in the sense that they not only ‘mobilized the future’ to secure resources (Brown, 2003, p. 5)—public and private investment in Langer’s own projects—but also constituted ‘nanomedicine’ more broadly by defining expectations about what kind of activity it was and what it could accomplish (Birch, 2017). Through such performances Langer (with perhaps a few others) made EPR-based magic bullets against cancer central to the sociotechnical reconfigurations supporting ‘nanomedicine’—in fact, as we shall see, its greatest promise.

EPR in Business: The Effect as Value-giver

As the NNI created the ‘nanomedicine’ brand, and entrepreneurs like Langer made EPR-based cancer drug delivery central to it, the mounting expectations spawned a small boom in speculative cancer-therapy enterprises. By 2003 Seattle-based Cell Therapeutics was trialling a polymer-paclitaxel conjugate said to exploit the ‘enhanced permeability and retention’ of tumours (C. J. Langer, 2004), while Solutia’s pharma arm built a polymer-drug facility for the same purpose (Scott, 2003). In 2006 Debiopharm licensed NanoCarrier’s Medicelle micelle technology to package platinum drugs ‘for superior EPR-mediated accumulation’. ¹² The emblematic venture was BIND BioScience (later BIND Therapeutics), founded in 2006 by Robert Langer and Omid Farokhzad. To this enterprise Langer applied his standard method of priming the investors with publications in high profile scientific journals because ‘the peer review process of a highly selective and competitive journal validates that the idea may be a significant breakthrough’—i.e. helps impart charisma (R. Langer, 2013). As with all biotech startups, patents were fundamental to the prospects of the firm and therefore to its valuation by venture capitalists and other investors (Rasmussen, 2014; Robbins-Roth, 2000). By the time BIND was ready to float its IPO on the stock market in 2013, its drug delivery ‘platform [was] protected by 16 US patents and 50 US patent applications’, according to an impressed analyst (Neuman & Chandhok, 2016). ¹³ And many of these foundational Langer-Farokhzad nanoparticle patents depended explicitly on EPR; that is, EPR underwrote the value of BIND. ¹⁴ The financial potency of the EPR enchantment, together with Langer’s reputation, was evident in the nearly $70 million in private capital BIND had already secured by the end of 2007, and the deals with Pfizer and others that dangled milestone payments worth up to a billion dollars (Maine & Thomas, 2017). The prospectus for BIND’s IPO invoked EPR as the way its Accurin^® particles would reach tumours, and pictured them atop a trajectory that began with Doxil, asserting that ‘Accurins represent the next stage in the evolution of targeted therapies and nanomedicine.’ ¹⁵

To many investors, the BIND’s EPR-based ‘story’ made for a very compelling ‘value proposition’, according to an investor newsletter after the fact. ¹⁶ BIND’s narrative was all about the firm’s ability to ‘program’ Accurin particles for delivery of many different ‘payloads’ or drugs, tuning their size and properties to target specific tissues and cancer cell types and also to design them for optimal pharmacokinetics. While active targeting—the placement of cancer-cell-binding ligands on the Accurin surface—was part of the Accurin story, the particles depended entirely on EPR to get out of the circulation and into the target tumour vicinity. BIND was not alone in that EPR dependency. As an investor newsletter in 2015 observed, the ‘global cancer nanomedicine market’ was ‘driven by nanoparticle drug-delivery systems that depend heavily on the EPR effect’. ¹⁷ EPR-dependence was not then framed as a liability for nanomedicine as a whole—even though cancer therapeutics made up the most valuable slice of nanopharma, and nanopharma itself represented about 90% of projected nanomedicine value. Indeed, EPR underpinned more than a third of nanomedicine’s overall value. It was the most important daemon animating the field. ¹⁸

BIND’s showcase drug candidate, BIND-014, used a conservative polylactide-polyethylene glycol shell, a well-known prostate-cancer ligand, and the cytotoxic agent docetaxel (Hrkach et al., 2012), thereby combining passive EPR targeting with active receptor binding. Enchantment logic is visible at every step of its development: It starts in the NNI funding programs (citing ‘magic’) presumably behind some of the Langer lab work on which it was based. It continues through the patents and prospectuses that not only cited EPR but built it in as an assumption that drug would self-navigate chaotic tumours. And it is present in the venture valuations, pharma partnerships and, ultimately, the stock’s market price, all reflecting Langer’s wizardry. Together with his charisma, EPR served as the venture’s value-giver, translating BIND’s story, an epistemic fiction, into bankable security.

The Unravelling of BIND

With BIND representing the promise of the cancer nanomedicine field so conspicuously, the clinical trials of BIND-014 attracted great attention from both scientific and financial audiences. For example, short-term positive responses of two out of three patients in the drug’s earliest clinical trials attracted prominent mention (Fonseca et al., 2014), even though their official purpose (like all Phase 1 trials) was only to measure pharmacodynamics so as to fine-tune dosing, and to watch for signs of toxicity. BIND had initiated two Phase 2 (safety and preliminary efficacy) clinical trials of BIND-014 in 2013, one in 2014, and one in 2015, all involving dozens of patients. ¹⁹ While the firm reported promising results with one of the first Phase 2 trials, treating prostate cancer, in January 2016, it also announced that it would not pursue this indication further because ‘of the evolving treatment landscape’—i.e. other emerging treatments looked better. ²⁰ Hopes for BIND-014 were riding on the remaining trials, one for a common form of advanced lung cancer, and one for metastatic cancers of several original varieties including cervical and head and neck. In April 2016 these results were announced: In the lung cancer trial the drug showed some marginal efficacy, and in the other almost no sign of efficacy at all. ²¹ Simultaneously with the release of those trial results, BIND announced a drastic cost-cutting restructure to continue core activities with reduced staff. ²² That would not be enough; within a month BIND had filed for bankruptcy, and two months later Pfizer acquired the company’s remaining assets for $40 million—down from a peak firm valuation of over $230 million. ²³ The bankruptcy of BIND effectively meant, both to the business world and concerned sciences (including pharmacology, oncology, etc), the bankruptcy of the EPR Effect.

The work that the EPR enchantment had been doing was evident once the spell was broken (and timing suggests that it was the market correction that broke the spell among the scientists, the financiers awakening them to their two decades of unjustified faith). Among drug delivery scientists, the implications were perceived as enormous. As one scientist put it, ‘the verdict has been handed down: the EPR effect works in rodents but not in humans’—raising the question, ‘What is the future of nanomedicine?’ (Danhier, 2016, p. 109). That concluding question confirms the point already discussed, that EPR-based cancer drugs had been the most promising and valuable fruit the nanomedicine field could offer; the mention of rodents, however raises doubts about a particular, traditional way in which essentially all cancer drug research was conducted. The same idea—that if EPR was wrong, then maybe the problem was mouse models in general—was aired by others too. One review analyzing hundreds of articles in cancer nanomedicine, by a prominent group of drug delivery scientists, argued that ‘the current approach’, based largely on the EPR Effect as characterized in mice, ‘was broken’ and would never translate promising drug candidates into effective new cancer therapies. They proposed, among other major changes, an obligatory testing stage for delivery systems in large animals whose vasculature might differ from rodents (Wilhelm et al., 2016). Similarly, for another leading scientist, the failure of BIND-014 attested to the general failure of the mouse models used for test drug delivery systems to predict clinical value over the past four decades. The way forward in drug delivery was for senior scientists and ‘stars’ in the field to step aside so as to enable entirely new and fresh approaches from younger people (Park, 2017). Clearly the perceived problem with the traditional order went beyond mice. In the eyes of some, the EPR fiasco reflected a political problem with the field, the distribution of resources as well as the pattern of practice that had become traditional.

Unsurprisingly, leaders of the old regime resisted questions about how work in their field was organized, or how it related ultimately to health care. For example, Langer argued that the EPR Effect must actually exist in human patients with solid tumours, just as in mice, and that the reasons the Effect had not yet led to clinical benefit must have to do with minor differences between species that could be straightforwardly overcome (Mitchell et al., 2017). For NCI researcher McNeil (2016), the lesson was not that nanomedicines which worked in small animals did not work in the clinic, but that the sponsors of clinical trials were too timid, testing valuable new nanotechnology vehicles with old anticancer drugs from which nothing but incremental improvement could ever be expected. But there were not many such defenders of the status quo; the clamour was at the other end of the spectrum. Some even disagreed about whether BIND-014 failed solely through its EPR-based passive targeting, or through failure of the active targeting strategy too (Wolfram & Ferrari, 2019). That cast doubt on the entire ‘magic bullet’ idea. Indeed, some went so far as to suggest that cancer nanotherapeutics as a whole should aim at making drugs with lower systemic toxicity, rather than targeting them at all (Youn & Bae, 2018). This is not the place to delve deeply into the clashing factions of scientists and their differing investments and values. Suffice it to note that loss of faith in the EPR Effect led to proposals of radical reform in the practices, and even the social order, of both oncology and drug delivery: abandonment of screening in mouse models, reform of drug translation pathways, even abandonment of targeted cytotoxic agents altogether (i.e. the magic bullet). The vision was dissipating, the sociotechnical configurations around nanomedicine that Langer had worked so hard to fix—or better, to bind, and through EPR’s promise to endow with life—were unraveling.

Thus, by implication of all this questioning, the work the EPR enchantment had been doing among drug delivery scientists included helping to coordinate their efforts, to rationalize their experimental approaches, to reinforce their field’s power structure and institutions, and of course to win them resources. Because it played no comparable constitutive role in the biotech investment world, loss of faith in the EPR Effect did not have such far-reaching effects there, no fundamental challenges to finance methods or business models or to cancer medicine as a market. The questioning remained more localized: Because EPR was the main basis of the nanomedicine’s social promise and commercial value, its failure brought a general devaluation of the field in favor of alternative investment opportunities. Thus BIND served as a nice example of how nano- and biotech stocks ‘served up only heartbreak’ for investors in 2016, and how even the most promising IPO bets can be flops (Anon, 2016a, 2016b). Such lessons in heartbreak reflect the perspectives of investors and traders on the public market. In contrast, for the venture investors and the companies themselves, BIND’s IPO remained an example of brilliant monetization (Maine & Thomas, 2017): The stock fetched an impressive premium when floated, irrespective of subsequent performance. Still, many of the investors that had been enriching the nanomedicine field now forsook it, provoking efforts by analysts and entrepreneurial scientists over the next few years to lure them back. Their argument was that despite high failure rates and the common impression that nano-delivery systems had been ‘oversold’, there was still great potential in them especially for biologics (proteins or nucleic acids)—even though the only successes that could be cited there were still older liposome-based chemotherapies (van der Meel et al., 2017; also Norouzi et al., 2020). As for mention of the EPR Effect, after early 2016 (when suggestions were still afloat that its discovery might win Maeda a Nobel Prize) its presence in the investment literature dwindled quickly. ²⁴

In 2019, the US National Cancer Institute discontinued funding to its flagship CCNEs (so that their final year would be 2020), both signaling and amplifying nanomedicine’s fading prospects. At what turned out to be the field’s nadir, in October that year eminent French nanomedicine researcher Patrick Couvreur blamed overinvestment in EPR for the field’s parlous state, pleading that his fellow drug delivery scientists maintain faith in the ‘magic bullet’ vision even without it (Couvreur, 2019, pp. 319–320; see also Hare et al., 2017). Barely a month after Couvreur’s piece, though, a dreadful new infectious disease from Wuhan, China, changed the context, providing nanomedicine—still largely centred on particle-based drug delivery—a chance to reinvent itself. The possibility of rehabilitating their field with a fresh story about fighting Covid occurred to nanomedicine’s leaders as early as March 2020, when a prominent editorial called for seizing the opportunity ‘to deploy nanotechnology advances as frontline tools’ against the new coronavirus, citing the vaccine developed by Moderna—co-founded by Langer in 2010—which had just begun Phase 1 testing (Chan, 2020). A later editorial triumphantly framed the impending Pfizer-BioNTech and Moderna vaccine approvals as ‘a milestone for nanomedicine’, capable of restoring faith in the field lost in the EPR debacle (Anon, 2020). And indeed, the vaccines’ spectacular efficacy rekindled enchantment in nanomedicine—albeit with some reservation. Kisby et al. (2021) warned against ‘shallow glorification’, urging the field to present itself as normal science, built on decades of incremental work, rather than (by implication) a supernatural breakthrough machine. In this clash between scientists declaring a total vindication of nanomedical drug delivery and those urging more cautious public statements, one might detect the kind of tensions observed by Loesch (2006)—based on differing ‘folk theories’ about science-society relations as described by Rip (2006).

Conclusion: Enchantments and Unicorns

Modern rationality, and science especially, ‘means that there are no mysterious incalculable forces that come into play, but rather that one can, in principle, master all things by calculation […thus] the world is disenchanted’, as Max Weber put it in 1917 (Weber, 1991, p. 139). Yet the EPR saga shows how technoscience keeps summoning such forces under new names. Turning away from nanomedicine’s astonishing, perfectly timed revival by the coronavirus pandemic of 2020-2022, let us reconsider what the trajectory of EPR and BIND can teach us about the way technoscientific projects can capture collective imaginations so as to organize work and resources, and how attention to what we are calling enchantment can add to the analysis of modern science and technology.

When BIND Therapeutics went public in 2013, investor demand did not approach the US $1 billion firm valuation mark that confers ‘unicorn’ status. Five years later another Langer venture, Moderna, leapt onto NASDAQ with a valuation of almost US $8 billion— well before the company had any product on the market. ²⁵ The contrast was not about revenues (neither firm had any) but about how compelling their ‘stories’ were, their narrative horsepower, and also context. As veteran biotech executive-turned-venture-capitalist Tim Harris argues, today’s blockbuster listings are driven by ‘group-think’, typically around ‘preclinical stories with nothing in the clinic’. Indeed, the past decade has seen the emergence of a unicorn ‘herd’ or rash of overvaluation in biotech IPOs, which he attributes to a new investment model: ‘dropping huge amounts of money on the table [just] to create a unicorn’ (Harris, 2024, pp. 394–395), that is, performatively displaying venture capitalist commitment to a vision, and thus stimulating and focusing investor group-think. Clearly, the generation of what Weber called ‘prophetic pneuma, which in former times swept through the great communities like a firebrand, welding them together’ (Weber, 1991, p. 155), is hardly the dead art he supposed. Rather, that increasingly ‘spectral’ financialization which has been engulfing all realms of productivity since the 1990s (Comaroff & Comaroff, 2000) is fully assimilating the previously resistant realm of medicine, perhaps by longterm neoliberal design (Nik-Khah, 2014), bringing the occult art of conjuring animal spirts to new levels of refinement.

A kindred form of unicorn has been observed inside nanomedicine laboratories. Erden and Levy (2020) use the term for entities or phenomena that, although never securely demonstrated, nevertheless legitimize and organize research quests. Unifying the two senses—financial and epistemic—we can define the technoscientific unicorn as a fiction that mobilizes extraordinary labour and resources regardless of supporting evidence. The Enhanced Permeation and Retention (EPR) Effect on which BIND was built fits the definition perfectly: Its empirical footing was questionable from the outset and, as a general phenomenon (not restricted to certain tumours in certain mice) never got much stronger. Yet for two decades EPR bound together polymer chemists, venture capitalists, and federal agencies under the banner of tumour‑selective delivery. We prefer to call it an enchantment.

To review, BIND’s clinical testing of its lead product did not merely fail to meet endpoints; it staged a mini-drama of enchantment and disenchantment. Investors had found its Accurin ‘story’ highly compelling, its ‘value proposition’ irresistible; hopes, both financial and medical, were riding high from BIND’s 2013 IPO. Buoyed by the headline that two of three Phase I patients (really, pharmacology experimental subjects) showed improvement, the company launched four Phase II trials between 2013 and 2015 amid breathless promissory talk of ‘programmable’ EPR-enabled tumour-hunting drugs. January 2016 brought the first jolt to the dream, when BIND quietly dropped its pursuit of the prostate cancer market; the April data-dump (showing marginal benefit in lung cancer, none in a multi-tumour bundle) dispelled it. Layoffs were announced the same day, bankruptcy followed within a month, and Pfizer scavenged remaining assets for a pittance just two months later.

Our enchantment framing helps us understand why EPR, and the BIND Accurins embodying it, attracted, among countless claimants as smart investments and promising cancer cures, so much enduring commitment and affective engagement from investors and nanomedicine experts in the first place. The object endowed with charisma outshines other objects of anticipated gain and/or hope, in a manner not fully explicable in framings foregrounding mere promise (Anderson, 2007; Borup et al., 2006; Brown, 2003; Fortun, 2008) or fictional expectations (Beckert, 2013; Shiller, 2017). Furthermore it offers additional insights into dramatic boom-bust cycles, as exemplified by the quick resurgence of targeted nanoparticles for vaccine delivery, ²⁶ and the flourishing of beguilments like ‘Trojan horse’ (Ibarra, 2021; Stenzel, 2021) that represent alternatives to the ‘magic bullet’ metaphor. In many respects, we feel we are building on Latour’s (2012) insights that the modern is distinguished from the ‘primitive’ chiefly in its denial of the magical mingling of human and nonhuman agency that remains constitutive to both. Enchantment is thus no vestigial remnant of a pre-modern past; rather, it serves as a binding agent, holding heterogeneous elements (data, dollars, dreams) in a single working composite much as it always did. We also feel that the rediscovery of enchantment in current science resonates with recent historical research revealing science as originating in the magic arts and the spiritual, rather than revolting against them (Gal, 2021, Ch. 6; Grafton, 2023; Newman, 2019).

For science and technology studies the EPR story crystallizes three intertwined mechanisms. First, enchantment glues alliances: In Beckert’s terms, a shared fiction of tumour‑selective drug uptake bound pharmacologists, clinicians, investors, and regulators until commercially crucial contradictory data made the gap between narrative and evidence impossible to ignore. Second, disenchantment reveals infrastructure, the ordering work enacted by the now-extinguished enchantment, and helps explain why a fresh unicorn narrative often quickly takes the place of an older one. A vacuum is created where the spell is broken—a vacuum where re-enchantment occurs, as with lipid-nanoparticle vaccines in Covid’s wake. Third, charisma migrates with capital: As classic sociology of charisma predicts, authority vested in individuals can be redeployed in new arenas, so figures like Robert Langer could carry reputational capital from polymer anti-tumour vehicles into the vaccine space, smoothing the transfer of belief, talent and funding. Speculative finance’s search for rapid, liquid exits makes that charisma fungible, letting promises be cashed out long before realization can test belief.

Students of science and technology should treat spectacular failures and the vacuums they leave not as dead ends but as rare, advantageous observation windows on the diverse powers holding modern enterprises together. This methodological insight carries different imperatives for different publics. Governance bodies and public funders should install reflexive checkpoints that question a platform’s founding metaphors, whether EPR, endosomal escape, Trojan horse, or the next bewitching formulation, before that metaphor ossifies into policy and infrastructure. STS scholars should seize vacua as privileged ethnographic intervals in which submerged sociotechnical imaginaries surface and rival futures openly contend. Entrepreneurial actors ought to recognize (as most already must) that phrases such as ‘Trojan horse’ and ‘magic bullet’ are not innocent descriptors but strategic assets that canalize capital, labour, and imagination. Reading them critically can widen the repertoire of possible innovations.