How are COVID-19 vaccine developers and regulators responding to variants?

By Lisa Larrimore Ouellette, Nicholson Price, Rachel Sachs, and Jacob S. Sherkow

The remarkable news of record-breaking COVID-19 vaccine development has been clouded by the increasing emergence of new variants of the SARS-CoV-2 virus. Like other viruses, SARS-CoV-2 mutates over time, due to random errors in copying its genetic sequence. When one of these mutations helps the virus survive and reproduce—such as by making the virus more transmissible—that variant will spread more rapidly than the original virus through natural selection. The global effort to control the pandemic has thus been framed as a race between the vaccines and the variants: can the world be vaccinated before the virus evolves to evade the vaccines? In this post, we examine how COVID-19 vaccine developers are responding to the spread of variants, how the FDA plans to regulate updates to the vaccines, and other innovation policies governments should consider to combat the variant spread.

How are COVID-19 vaccine developers responding to variants?

As SARS-CoV-2 variants sweep the globe, a critical question is whether the currently authorized vaccines provide protection to individuals who come in contact with those variants. Answering that question in the field is difficult, but three vaccines (including Johnson & Johnson’s) showed lower efficacy in clinical trials conducted in South Africa (origin of the more transmissible 501Y.V2 variant) than in other locations. Clinical trials for the two mRNA vaccines authorized in the US (from Pfizer–BioNTech and Moderna) were completed before the more recent variants became widespread, so researchers have leaned on in vitro analyses instead—measuring how well the blood serum of vaccinated individuals performs in reducing live viral “plaques.” The most detailed summary to date of this in vitro evidence reports that the Pfizer–BioNTech and Moderna vaccines show markedly decreased plaque reduction activity against the three most common variants: B.1.1.7 (first reported in the UK), 501Y.V2 (first reported in South Africa), and P.1 (first reported in Brazil). (There’s not enough data yet on J&J’s vaccine.) Translating these decreases in activity to measures of actual protective ability are difficult, for a variety of practical and immunologically complex reasons. Nonetheless, the in vitro data is not good news. 

If existing or novel SARS-CoV-2 variants escape the immunity conferred by current vaccines, hope may lie in modifying vaccines to target the variants. All of the currently authorized vaccines in the US and Europe prime the immune system to respond to one specific protein on the outside of the SARS-CoV-2 capsid, the “spike” protein, named so for the crown-like spikes sticking out from the surface of the virus that give coronaviruses their name. All of the variants thus far seem to modify this spike protein. Moderna and Pfizer–BioNTech’s vaccines provide an mRNA copy of the spike protein to be produced by the body’s immune cells, thus generating a response. J&J and AstraZeneca do the same, but with DNA.

Adjusting these mRNA or DNA sequences is relatively easy, and so developers can—in theory—adjust their vaccines to target these variants by simply encoding for the variant sought to be targeted, or producing a blended version that contains mRNA or DNA copies of both the “old” SARS-CoV-2 variant and newer ones (a “multivalent” vaccine). The way the currently authorized vaccines are made, these changes would be slightly easier for Moderna and Pfizer–BioNTech. For J&J and AstraZeneca, the science is similarly straightforward but takes about twice as long (3 months). And this doesn’t account for the additional challenge that individuals could, over time, develop immunity to the particular vector used by the two manufacturers (a modified adenovirus).

Developers are aware of these possibilities and most have announced plans to adjust their vaccines to incorporate these variants. Moderna, the first firm to have a vaccine candidate ready for testing in early 2020, was also the first to announce a candidate for variants on January 25 of this year. On February 24, it announced it was ready to begin its variant-based clinical study as soon as regulators approve the trial. BioNTech, meanwhile, said it could adjust its vaccine in about six weeks. Both modified vaccines would need additional time for testing and manufacturing before they could reach the public. In the meantime, on February 25, Pfizer and BioNTech began to study the effects of giving already vaccinated individuals a third booster dose of their existing vaccine. J&J is also working on updated versions of its vaccine. And researchers are considering ways to make vaccines more resilient in the first place rather than requiring constant redesign. Despite the global spread of variants, there may be a silver lining in that this move toward innovative vaccine “updating” may “not only tackle viral variants but also provide solutions across the globe at a fraction of the cost.”

How does the FDA plan to regulate vaccines for variants?

The typical regulatory process of approving a new vaccine requires three phases of randomized controlled clinical trials, lasting several years. The length of this process cannot compete with the speed of virus mutation, and even the compressed timeline within which COVID-19 vaccines initially came to market is likely to be too slow. As a result, there is a need to consider whether additional regulatory approaches can be used to speed vaccines for variants to market. Particularly where a given vaccine approach has been rigorously demonstrated to be safe and effective against one form of a virus through this randomized trial process, regulators may use other types of clinical evidence to evaluate the efficacy of a slightly modified version of that vaccine.

On February 22, the FDA released new guidance for medical products related to SARS-CoV-2 variants, including an updated version of their existing vaccine guidance. The guidance describes the required nonclinical and clinical data the FDA expects to use in evaluating the efficacy of vaccines targeted against emerging variants, including clinical studies that compare an immune response to variants induced by the modified vaccine against an immune response to the authorized vaccine. The agency directs sponsors to perform studies on individuals who have not yet received a COVID-19 vaccine as well as on those who received the authorized vaccine (for whom the new vaccine would be a booster shot).

The FDA’s guidance document for vaccines targeted at SARS-COV-2 variants is not exhaustive, though, and questions remain to be answered. In particular, the guidance notes that it “does not address how it will be determined that a vaccine based on a new viral sequence is needed.” FDA’s Acting Commissioner, Dr. Janet Woodcock, did not give specific criteria as for how the agency would determine when such an update was needed. At a news briefing, she emphasized the need to “to anticipate this and work on it so that we have something in our back pocket before the threshold is upon us.”

Using immunogenicity assessments (rather than full-scale clinical trials) to evaluate vaccines modified to address virus variants is not a new endeavor for the FDA. The seasonal flu vaccine is regulated in much the same way—vaccine manufacturers use already-proven methods and platforms and adapt them for the flu virus strains projected to be circulating in the upcoming season. The agency has specifically issued guidance for pandemic flu situations, instructing manufacturers of existing licensed seasonal flu vaccines as to the immunogenicity studies they would need to complete before manufacturing a pandemic influenza vaccine using the same process. Professor Amy Kapczynski has written about the ways in which much of the information regarding flu virus strains moves through a global virus-sharing network, using open science rather than traditional exclusivity models. Learning from these models may help reduce the cost of developing—and distributing—modular vaccines, hopefully decreasing some of the inequities currently present in global vaccine distribution.

What other policies should innovation policymakers consider to address variants?

Authorizing developed variant vaccines is not the only task for policymakers, who should consider several other routes to keeping variants at bay. Not all the action is at the FDA, even though it undoubtedly plays a central role. Other health-related policy interventions would be useful in understanding variants and limiting their spread, better understanding different vaccine options for variants, and improving vaccine performance once vaccines are developed.

The best way to deal with new and troubling variants is to avoid their development in the first place. While policymakers can’t control how the virus evolves, they can help limit the chances it has to do so—and that means ramping up production and distribution of vaccines globally. Uncontrolled spread results in more variants, including those which can evade existing vaccines or reinfect the already infected. Pushing for rapid, complete vaccination across the globe is thus not merely the right moral and ethical approach; it’s also important from a purely self-interested perspective of countries with vaccine supply. Interventions we have suggested in prior work, including sharing manufacturing know-how with other manufacturers, would help these efforts and, consequently, decrease the chance of future variant development.

Once variants exist, it is important to know what, where, and how widespread they are. This means genetic surveillance—sequencing infected patients to know which variant they have contracted. Genetic surveillance is a problem of nonexcludable innovation, which suggests that private actors are unlikely to spend enough on it, leaving governments to fill the gap. Some governments do genetic surveillance very well; the United States doesn’t, at least nationally. Such surveillance should also help scientists understand how effective existing vaccines are in managing the spread of variants. 

Better understanding the performance of vaccines suggests its own potential policy intervention. Each vaccine was tested in a different time frame, with different populations, in different places, and—crucially—with different variants at different levels circulating in the population. As a result, despite headlines trumpeting different vaccine performances (and some policy reactions), no one really knows whether Johnson & Johnson’s or AstraZeneca’s vaccines are actually less effective than Moderna’s and Pfizer–BioNTech’s. Even less is known about how a vaccine modified to address a variant would compare with any of them. Running comparative randomized controlled trials between multiple different vaccines seems unlikely—indeed, comparative trials are underfunded generally—but at the very least policymakers should consider robust systems for collecting observational data to understand how different vaccines perform. 

Once vaccines are approved, getting them made remains a challenge—and with variant vaccines, that challenge will continue. Flexible manufacturing platforms should make it easier to get variant vaccines into scaled-up production more rapidly, and further investment in such platforms could help with that challenge. As we’ve noted, older processes to manufacture vaccines were materials intensive—some involve incubating viruses in chicken eggs. mRNA vaccines, on the other hand, are less resource intensive and seem especially well suited to variants, since the encapsulated mRNA sequence can be easily altered without changing other production parameters. Government efforts to improve manufacturing generally, and especially around mRNA or other flexible manufacturing platforms, could be essential to producing variant vaccines if COVID becomes endemic and continues to change. 

Finally, getting vaccines into arms is always the last step, without which nothing matters; investing in better vaccine roll-out is essential if variant boosters will become routine. An example for this exists, of course: the annual flu vaccine. Someday we all may get annual COVID vaccines, tailored to that year’s new variants. But for that to work, the health system would need to do a better job keeping track of who gets what, whether first and second vaccine shots can be mixed and matched, and how to insure that ongoing distribution is efficient, affordable, and equitable. 

This post is part of a series on COVID-19 innovation law and policy. Author order is rotated with each post.

Advice about the patent bar for current and prospective law students

Guest post by Professor Eric E. Johnson (ericejohnson.com)

I recently asked fellow intellectual property professors and others about advice for law students interested in taking the patent bar. The IP community generously responded, and I have synthesized their wisdom and opinions here, with some of my own advice sprinkled in. Of course, opinions differ and things change, so students should consider this post as a jumping-off point for doing their own research and asking their own questions.

Most important points:

• You don’t need to be admitted to the patent bar to do most things involving intellectual property law. For practice concerning copyrights, trademarks, trade secrets, and publicity rights, the patent bar is irrelevant. Even litigating patent infringement in federal court does not require the patent bar.
• The patent bar is necessary for those who prepare patent applications and those who represent clients before the U.S. Patent and Trademark Office (USPTO) on patent matters.
• You don’t have to be a licensed attorney to be admitted to patent practice before the USPTO. But you do need to meet certain pre-requisites (such as having an engineering degree, a certain kind of hard science degree, or some alternative qualification spelled out by the USPTO). And you have to pass an exam and meet the moral character requirements. 
• If you’re a non-lawyer and are admitted after passing the exam and meeting the moral character requirements, then you’re a “patent agent” (just not a “patent lawyer”). In so far as that goes, “patent bar” is a somewhat inaccurate nickname for admission to practice before the USPTO in patent matters.
• The USPTO’s main source of information for becoming registered as a patent agent/lawyer (including requirements to be eligible to register for the exam) is here. Note that it may take quite a while for the USPTO to process your paperwork to register for the exam. So plan ahead.

Prep options for the exam:

• Practising Law Institute (PLI) Patent Office Exam Course. This seems to be the most popular and well-known prep course. It is said to be thorough but pricey. They may offer a student discount, for which you might need an e-mail address with .edu.
• Patent Education Series (PES) Patent Bar Review Course. This is said to be good in particular with helping students learn how to search the Manual of Patent Examining Procedure. It is known to be less pricey than PLI. 
• (The Patent Resources Group course got a good review, but the website says the course has been retired.)
• There are self-study books, apparently. But they didn’t seem to be a very popular option with those in the know. A self-study book might be more plausible in combination with a law school course focused on patent prosecution practice, if one is offered. (But note that is not the same thing as the more commonly offered law school course labelled “Patent Law” or “Patents,” which is much more general in coverage.) At any rate, the self-study option is probably best for those who don’t mind the prospect of taking the exam a second time if necessary. If you’re a one-and-done type, the prep-course option is probably your best bet.

Tips for studying that various people offered:

• Expect to spend a minimum of one month of focused study, or maybe more like 2-3 months. It’s a tough test; only about half pass.
• Consider taking both the PLI and PES courses, if you can afford it.
• Knowing 35 U.S.C. § 102 under the America Invents Act seems particularly important.
• Complete as many practice questions as possible. 
• Taking a regular patent law course in law school generally does little to prepare you for the patent bar exam, although it may help provide a broad framework and may increase your interest in the subject. (And vice-versa: Having passed the patent bar exam is not that much of a leg up on a general patent law class.)

Career thoughts that various people offered:

• The patent bar requires intensive study. And there is wide agreement that it is a bad idea to be distracted during your 1L year by any other time commitments. Thus, received wisdom strongly counsels against studying for the patent bar while simultaneously taking 1L classes. Even studying during 2L or 3L year is potentially troubling. (In other words, don’t sacrifice grades for the patent bar.)
• A number of people agreed that getting the patent bar out of the way before law school or during a summer of law school can be a great idea. 
• A registration to practice before the USPTO can distinguish your résumé from the multitude of others that firms may receive.  
• Résumés of law students who are registered patent agents are much more likely to receive consideration for patent-prosecution-related summer clerkships/jobs/internships and permanent positions.
• It can be particularly advantageous to have your patent-registration number in hand before going to the Loyola Patent Law Interview Program.
• And some expressed the sense that having a patent registration number could open up opportunities for jobs at smaller firms where the work would embrace a variety of IP matters, including copyright and trademark.
• There were reports that some patent litigation firms want their patent litigators to pass the patent bar. This is despite the fact that infringement litigation in federal court only requires being a licensed attorney, not being licensed to practice before the USPTO. On the other hand, there are litigation-like patent proceedings that take place before the USPTO (inter partes review and post grant review). These have been increasing in volume, and in many firms, these are handled by the patent litigation group.
• But a number of people perceived downsides of hurrying toward the patent bar.
• Before law school, it’s hard to know what kind of work you’ll eventually end up wanting to do. Law school is a time of being exposed to many things—some of which may surprise you as a great fit for you. And patent prosecution is far from the only lawyer work where a science/engineering background is helpful.
• Along these lines, there was a concern that having a patent agent registration number may unhelpfully pigeon-hole a student as being a patent-prosecution-only person. Even within some IP firms, there was a sense that a patent-registration number might limit a junior lawyer’s exposure to other types of IP-related work. 
• On the other hand, pigeon-holing is a generic concern for any stand-out line on a résumé that is associated with a particular field. So it’s a concern, but you can’t spend your lifetime running from achievement out of a desire to keep your options open. 
• If you find yourself pigeon-holed, you may just have to push harder to get the opportunities you want in order to become more well-rounded. That could be within the firm. Or it might mean leaving the firm. There were stories of lawyers lateraling from their first job to a firm where they could have a broader practice—or leaving to start their own, very fulfilling, practice.
• Even for a person committed to patent prosecution as a career, there may be a mentoring advantage in waiting to take the exam until after law school. If one feels confident about being able to get a patent-prosecution job without the registration number in hand (perhaps because of having sought-after engineering credentials or an advanced degree), then that person is likely to get very helpful mentoring in the initial phase of their career. This is because a supervising attorney will closely evaluate the new associate’s work when it’s being done under the supervising attorney’s name and number. On the other hand, once the associate has their own registration, they may be left to get the work done on their own. 
• And if you get the job first, your employer may pay for your prep course.
• Make sure you consider the possibility of pursuing a career with the USPTO itself. Being a patent examiner can mean the opportunity for valuable on-the-job training, and it can make you very marketable for private-sector jobs later on. Plus, the federal government tends to be a good employer—with good benefits and lots of stability. And USPTO jobs are no longer just in the D.C. area.
• Because of the availability of online research tools, some patent prosecutors are heading for the hills—working remotely so they can live near hiking, skiing, etc. Having a remote practice likely takes experience, a stable client base or cooperating employer, and some intrepidity. But it can make for a great life.
• Patent prosecution work isn’t for everyone, but for those who love it, it can lead to very happy law careers. In at least one person’s opinion, admittedly anecdotal in basis, patent prosecutors seem to have some of the highest career satisfaction in the legal profession.

© 2021 Eric E. Johnson. Licensed under the Creative Commons Attribution-ShareAlike 4.0 International License (CC BY-SA 4.0). Konomark—requests for gratis permissions beyond the scope of the Creative Commons license are welcomed.

Waiting for Google

Google v. Oracle was argued (after being reset last term for additional briefing) on October 7, during the first week of this term. We still don't have an opinion, and the time delay makes me worry that the opinion(s) will be more Guffman than Godot.

While we wait (and it could be any time), I wanted to point to a recent case that illustrates the concern at issue in this case if Oracle wins. The facts of this case are remarkably similar to my remote control analogy, which I continue to think is the best analogy to date (and which I hope some Supreme Court clerk happened to read).

The case is Pyrotechnics Management, Inc. v. XFX Pyrotechnics LLC and FireTEK, a W.D. Pa. case that just issued a preliminary injunction. The facts for our purposes are not complicated. Both parties make a device used to control fireworks displays. The device issues and responds to commands, which are basically numbers sent along the wire. It's called a protocol by the plaintff (and anyone else familiar with this kind of technology). It's functionally no different than an API like that in Google v. Oracle - it's a set of commands that tells devices to do something.

Here, the copyrighted work isn't even software. The registration deposit materials is a manual that describes the protocol - send this command (number) with these optional parameters (more numbers) and things will happen. Here is the briefest of examples:

In other words, send a 12-byte message containing (in this case) 3 actual bytes of information. 0x23, 0x23,0x46, plus a CRC check (a common error check practice). Like the remote control in my analogy, it's sending 1s and 0s down the wire, in a particular order.

The plaintiff's expert said the copyrighted work must have been copied because how else could you make a device that sent these same numbers? The defendant's device sent and received these same numbers down the wire. There's not even a claim that the numbers were generated in the same way.

And the court agreed. It said the selection of numbers required originality, and there was no reason for a competitor to use the same numbers. Furthermore, the fact that the deposit wasn't even software didn't mean that it couldn't be copied in software if the same numbers were implemented. In short, the plaintiff now owns the set of remote control codes and nobody else can make a compatible remote control.

If you don't see why this is problematic, then nothing else I write will convince you otherwise - you are basically OK using copyright to obtain a 100 year backdoor patent on any device that sends and receives an arbitrary set of commands. But this is not a good result. Copyright policy - let alone the statute - simply should not allow for this outcome that uses copyright to limit functional information sharing in computer programs and electronic devices. 

As I advocated in my amicus brief and many other blog posts, I've got no quarrel with the argument that the set of numbers collected by the plaintiff may be copyrighted.  But that copyright cannot extend to enforcement against the method of using those numbers to operate devices. So, you can't copy the description of the protocol verbatim, perhaps, but it is not infringing to send numbers down a wire. The statute couldn't be more plain on this - methods of operation cannot be protected, no matter the form in which they are expressed. What else can the statute mean, if not this narrow meaning?

I'll end by noting that the district court here did not even mention Baker v. Selden, which explicitly endorses this rule - that the expression of a method is not infringed when others use the method:

The fact that the art described in the book by illustrations of lines and figures which are reproduced in practice in the application of the art, makes no difference. Those illustrations are the mere language employed by the author to convey his ideas more clearly. Had he used words of description instead of diagrams (which merely stand in the place of words), there could not be the slightest doubt that others, applying the art to practical use, might lawfully draw the lines and diagrams which were in the author's mind, and which he thus described by words in his book.

The copyright of a work on mathematical science cannot give to the author an exclusive right to the methods of operation which he propounds, or to the diagrams which he employs to explain them, so as to prevent an engineer from using them whenever occasion requires.

Furthermore, we will accept some copying of expression if that's the only way to use the idea:

And where the art it teaches cannot be used without employing the methods and diagrams used to illustrate the book, or such as are similar to them, such methods and diagrams are to be considered as necessary incidents to the art, and given therewith to the public; not given for the purpose of publication in other works explanatory of the art, but for the purpose of practical application.

Baker v. Selden also provides a direct analogy on the facts - there, the bookkeeping system was described in a document, but the plaintiff could not stop others from practicing the system. That rings true here, where the copyright is in a document explaining the system of numbers; from Selden: "The use of the art is a totally different thing from a publication of the book explaining it." Here, the commands were described in a document, but the plaintiff should not be able to stop others from practicing the commands: 

As an author, Selden explained the system in a particular way. It may be conceded that Baker makes and uses account-books arranged on substantially the same system; but the proof fails to show that he has violated the copyright of Selden's book, regarding the latter merely as an explanatory work; or that he has infringed Selden's right in any way, unless the latter became entitled to an exclusive right in the system.

The court's failure here to mention Baker v. Selden is problematic. I hope the same is not true in Google v. Oracle.

 

Charles Tait Graves: Idea Submission Cases, Desny Claims, and Trade Secret Law

I thoroughly enjoyed Charles Tait Graves new article: Should California’s Film Script Cases Be Merged into Trade Secret Law?, which was recently published in The Columbia Journal of Law & the Arts.  Graves is a partner at Wilson Sonsini and teaches trade secret law at UC Hastings Law.   

The article deals with so-called "idea submission" cases. The fact pattern is as follows. Plaintiff, who is sometimes called the "idea man" in older cases, shares an idea with Defendant, hoping for monetary compensation even though there's no express contract stating terms of payment. Defendant subsequently takes the idea and commercializes it without paying Plaintiff. (There's an excellent discussion of the idea submission cases in Chapter 4 of Elizabeth Rowe and Sharon Sandeen's Trade Secret Law casebook). 

At least in California, the Plaintiff-idea person will likely have two distinct types of legal claims in this scenario: (1) a claim for breach of an implied-in-fact contract, which in California is called a Desny claim; and (2) a claim for civil trade secret misappropriation, which since 2016 can be brought under both state law (e.g. under the California Uniform Trade Secret Act) and federal law via the Defend Trade Secrets Act (DTSA). Graves recounts in tremendous detail how these two different legal regimes developed on separate ends of the map of California, in Southern and Northern California, respectively. Graves' thesis is that, even though these two areas of law have been historically addressed separately, they have a lot in common and can learn a lot from one another.   

I interviewed Graves about the article, transcribed below.

CAH: What is an idea submissions case?

CHARLES TAIT GRAVES: We need to be very careful in defining this.  The term "idea submission case" has very different meanings in different states. In California, we typically mean a Desny claim, and that is what I talk about in the article. The Desny version of the idea submission claim, which arises most often in the Hollywood/entertainment context, is a very tightly controlled implied-in-fact contract claim that can arise when someone submits an idea (typically a film script, a screen play, or a fleshed-out idea for a film), and there is offer for payment and an acceptance, followed by use without payment.

I will also note that I define idea submission cases in the article to include only contract-based claims like Desny claims, because I am intentionally not including the penumbra of tort and other common law claims that are likely preempted by trade secret or copyright law.  

CAH: So you might have told a national tale. Trade secret law is obviously now federal even if state claims are still available. But you choose to limit this to California. Why?

CHARLES TAIT GRAVES: The states differ in how they treat idea submissions. For example, in California, the "idea man" (now we would say "idea person") can have an implied-in-fact contract claim based on submission of an idea conditioned on payment, even if the idea was not novel. However, in New York, for example, the law is different: novelty would be required.  So the idea submission law is very different state to state.  This is why this is a state-specific story.

CAH: OK, so in the universe of your article we have two different legal claims to address the idea submission situation in California. Readers know a lot about trade secret claims and what is required for a trade secret claim. But we are less familiar with this other kind of claim you discuss: the Desny claim. What is a Desny claim and how did it come about? How do the Desny claim and the trade secret claim differ? 

CHARLES TAIT GRAVES: The Desny claim is an implied-in-fact contract claim. It takes its name from the famous case, Desny v. Wilder (Cal. Sup. Ct. 1956), in which the California Supreme Court created the elements that plaintiffs must satisfy to win on this kind of claim. 

Desny emerged out of many years of bewildering California case law. At that time, in the 1950s, courts were struggling to figure out what to do with the idea submission cases coming out of Hollywood. These usually involved screen plays or concepts for films, and usually the plaintiff believed they were entitled to compensation after the film was made based on their script or ideas. There was a lot of confusion and multiple sorts of approaches in the common law. The situation was even more confusing because, prior to Desny, there used to be "common law copyright" protection for ideas in California.  Under that regime, ideas could be claimed as property rights and litigated under a "plagiarism" approach without involving contract law at all.  But in 1947, California’s legislature amended, and significantly narrowed, the state statute governing common law copyright protection, Civil Code Section 980. The 1947 amendment to California law left a gap in the law.

So then in 1956, the California Supreme Court confronted this fact pattern in the Desny case, and had to decide how to fill the gap and address confusion in the case law. The plaintiff in Desny (Victor Desny) came up with a story for a film based on a media frenzy that had happened in the real world over a man trapped in a cave, and Desny was trying to get in touch with the well-known director Billy Wilder through Wilder's assistant, so that Wilder could make the film and (Desny hoped) compensate him. Desny claimed he told Wilder's assistant that use of his film idea was conditioned on payment, and that the assistant told him that if Wilder used the submission, he would be paid. [For a somewhat light-hearted take on the facts of this case, check out this piece in Variety].

So with these facts in the background, the court had to decide what kind of claim could be brought to get compensation for idea submissions, and what would be the elements of such a claim? The court put together a test that said: (1) if you submit an idea (like a screenplay or a synopsis) for sale, and (2) its use is conditioned on payment, and (3) the defendant knew or should have known there was a payment condition, yet (4) still voluntarily accepted the idea, and (5) the defendant actually used it, and (6) it had some value (even if it was not novel)...there can be a claim for breach of implied-in-fact contract. 

But it is a very complex, difficult claim to bring. There are a lot of elements to prove. It can be hard to bring these claims, and there are situations where plaintiff can't win them. For example, courts have held if the idea was previously published, or even if it was shown earlier to the defendant without the same condition of payment for use, then there is no viable Desny claim. Also, these claims cannot be based on merely inchoate discussions with no plausible promise of payment. There has to be a tight nexus between disclosure of the idea or screen play to the defendant and conditioning payment. 

CAH: If the director, Billy Wilder, had done what he did today, and Desny had brought a trade secret claim instead of breach of (implied) contract, how would this have gone?

CHARLES TAIT GRAVES: I suspect that if this came up in 2021, the courts would just see this as a trade secret case and apply trade secret law in the usual way.  They would decide whether this met the requirements of a trade secret, including whether it was sufficiently secret (this could be doubted, since the film idea centered around the public facts of the real-life cave rescue effort); whether it derived sufficient independent economic value from secrecy; and whether reasonable secrecy efforts were taken. 

But at the time of Desny, California trade secret law was very primitive. Almost all the cases were in the employee mobility/customer list context, not necessarily the high tech sphere like today. Hollywood was more the powerhouse at that time, not Silicon Valley. And these film script cases had previously been seen in a statutory context that was now gone post-1947.  So in a sense, I think it is a historical accident that the California Supreme Court came up with this strange, non-trade-secret, California-specific claim to deal with what might now look like a trade secret claim. Today, I think the courts would have applied the elements of trade secret law that we are familiar with, they wouldn't struggle to craft this new kind of implied-in-fact contract claim.  

But nonetheless, the Desny claim is how the Supreme Court of California decided to deal with the situation. And very interestingly, the courts have stuck with the Desny claim and its elements all the way from the 1950s until today. The Ninth Circuit and the California courts use this very same test in dealing with idea submission scenarios like Desny. It took them a while to get there, but it stuck. 

CAH: So let's talk about independent derivation. A big part of your article focuses on the independent derivation "defense" to misappropriation. The trade secret statutes say that misappropriation of a trade secret does not include, among other things, "...independent derivation..."  How do the Desny implied-in-fact contract cases, and trade secret cases, deal with independent derivation, and what can they learn from each other?

CHARLES TAIT GRAVES: Independent derivation was the original impetus for the article. I noticed that courts on the Desny idea submission side had been publishing these very interesting opinions finding that certain claims were not actionable because there was an independent source for the film script or idea, other than the plaintiff.  Courts would find that, even though plaintiff gave defendant an idea for a film, the evidence showed defendant already had the same idea before defendant even met plaintiff, or even that third parties, maybe even other studios, gave the idea to the defendant, and the people who worked on the production never even saw the plaintiff's materials. There are very detailed discussions of these issues in the Desny line of cases coming out of Hollywood, and decades of published cases that deeply analyze the independent development situation.  

In contrast, on the trade secret side there is little development of independent derivation, and not much of a structure for assessing it, at least in California. There is case law on who bears the burden to prove independent derivation, but not much on how to prove it, like what evidence is needed and what factors really matter. So one of the insights for this article was that we have all this law coming out of Southern California, involving movies and tv, and can't we look at this law on the trade secret side, where we are trying to get at the same question: Did the defendant independently derive the information that the plaintiff says was misappropriated? Why not look at these cases about tv shows and movies, and see how they assess independent derivation, and use that for trade secret cases?

CAH: What are some key lessons you see in the Desny line of cases' approach to independent derivation that trade secret law could use?  And would these apply anywhere, or only in California trade secret cases?

CHARLES TAIT GRAVES:  I think there are lessons that can work anywhere. Even though the Desny claim is California-specific, the way the Desny idea submission courts adjudicate the issue and assess the evidence is broad, and would work in the trade secret context, even outside California. I am somewhat surprised others have not connected the two lines of cases before on this point.

Going back to the 1950s, courts in Desny idea submission cases look at two things when assessing an independent derivation argument. First: the temporal structure. That is, when did defendant claim to have conceived of the information? Was it before or after defendant got any information from the plaintiff?  If it's after, the burden is higher for defendant to satisfy the court that this didn't actually come from the plaintiff. Second: who was the source of the independent derivation? Was it the direct recipient of the idea, so for example, Billy Wilder himself or his assistant? Or was it others within the same large studio, or even people at a different company? If it's the same individual who received the trade secret, or someone closely connected to them, it will be tougher to prove independent derivation. On the other hand, when defendant can point to an independent source like another movie studio or a company in another country, that is a better defense. 

It is easier to swallow an independent development argument when the defendant has evidence that they were already in development with a third party when the plaintiff came along.  For example imagine a company hires a new employee in a three-hundred person company, and the former employer sues for trade secret misappropriation. Imagine the defendant argues, "we were developing this first, before the employee was hired," and second, "it was our team over in Singapore that gave us this idea, it wasn't even anyone in the new hire's group."  Thinking about things this way, in terms of when did the independent development occur in relation to when the idea was received, and who was the independent source of the idea, can help courts tackle the under-developed issue of independent derivation in trade secret law.  

CAH: So you go into a lot of detail in the paper about preemption. But you come out concluding that, for a variety of reasons, Desny claims aren't preempted by the California Uniform Trade Secrets Act, just as they aren't preempted by copyright law. So as a matter of policy, do you think there's anything wrong with having multiple legal regimes to address this kind of a wrong? Or is there a room for both of them?

CHARLES TAIT GRAVES:  I don't think these claims are necessarily in conflict. There is a policy sameness here. On the one hand, the Desny claims were clearly created by the courts to protect the idea person, the individual who goes to a powerful studio to submit their idea, and just gets steam-rolled.  California courts created the Desny claim out of whole cloth precisely to protect the idea person, the weaker party, in the idea submissions context. Meanwhile, trade secret law is also guided, at least in California, by a strong policy of favoring the interests of the departing employee who is accused of taking their employer's trade secrets. The interests of the departing employee are paramount. So in this sense, there is a policy sameness, not a policy conflict. Both claims have an underlying policy of respecting the interests of the weaker party, whether it's the idea person-as-plaintiff or the departing employee-as-defendant. That said, you could certainly imagine scenarios where there would be a collision, for instance if a plaintiff manages to squeeze a Desny claim out of a situation for which it wasn't designed in order to get a windfall.

CAH: In concluding, do you have any observations on what these legal claims say about the difference between Southern California, the source of entertainment law and the Desny claim, and Northern California, where trade secret law tends to take center stage?

CHARLES TAIT GRAVES: One thing I can say is that you will notice the two sides almost never meet. There are almost never conferences on the intersection between trade secret law and entertainment law, even though you see conferences on the intersection between patent and trade secret law, things that are a bit more obvious.  I don't know why we have these two separate legal cultures. I can only speculate, but it's possible that it has to do with the fact that Southern California/Hollywood arose as a separate economy, and then Northern California/Silicon Valley arose as a separate economy.  Only now in the past twenty years are they sort of joining together. So my guess is that the people who do entertainment law and the people who do trade secret law just grew up at different law firms, worked in different practice areas, went to different symposia at law schools, read different treatises... It's like two ships passing in the night. So maybe this is why we don't often think about the fact that, actually, these two areas of law are really very similar. 

CAH: It is in a sense the fact pattern that joins them. Someone is stealing an idea without payment. It doesn't strictly matter if it's a technical prototype for a backpack or an idea for a film script.  When we have legal regimes addressing that kind of scenario, it's virtually inevitable they would have similarities.  

***

We also talked at length about the intricacies of preemption. For that, you'll have to read the article!    Again, Charles Tait Graves' article, Should California’s Film Script Cases Be Merged into Trade Secret Law? is now published and available online at The Columbia Journal of Law & the Arts.