Patent

- Are trade secrets patentable?

     The media reported early last week that an ex-employee of Goldman Sachs had been arrested by the federal authorities for allegedly stealing the codes of the company's proprietary trading system. (See here & here.) According to the reports, US prosecutors told the court that the revelation of the secret codes would cost the company "millions upon millions of dollars". What is more, the prosecutors said that the stolen codes could be used to "manipulate markets in unfair ways". It also appeared that the ex-employee left GS in June this year to work for another company also engaging in the business of proprietary trading.

     This piece of news did not raise too many eyebrows in the markets. GS's stock prices were virtually unchanged through the week. It does raise a few interesting related questions, though. For example, if the codes can indeed be used to unfairly manipulate markets when fallen in the wrong hands, did GS take advantage of that malign capacity while it had the secret possession? Also, did the ex-employee sign any non-compete agreement that should have prohibited him from working in a similar field?

     I am not going to talk about these questions in this posting. Rather, I would like to focus on another question related to the intertwined paths of trade secret and patent. -- Now that the cat is out of the bag, can GS seek to patent the previously secret strategies to prevent others from using them? Or, more generally, even if the secret codes had not been stolen, can GS patent the strategies after profiting (supposedly) from them secretly for some time?

     On a first look, there seems to be little reason for GS to pursue patents for the stolen strategies. Proprietary trading firms normally prefer to keep their strategies confidential, because if too many people trade the same strategy at the same time, any profit opportunities will quickly disappear. Even though theoretically a patent may allow the owner to stop others from using the strategies when they are no longer proprietary, figuring out who is/are using the strategies and enforcing the patents can be challenging. Moreover, it may take years to obtain a patent, when the strategies will most surely have outlived their usefulness.

     There is, however, no guarantee that others will not obtain patents and seek to prevent GS from using the strategies. Also, it is not entirely impossible to find out whether the major players (maybe not all the players) are using the same strategies. Patents therefore may be effective and useful in preventing these major players from using the same strategies. In addition, even if the strategies have lost their values in proprietary trading, they may still have significant values in hedging, corporate financing, preventing "unfair market manipulations", non-proprietary trading for clients, risk management, and other uses.

     Whether the secret strategies are patentable will depend generally on the normal criteria, principally whether the strategies are new and non-obvious. A more specifically related issue here is whether the past secret use of the strategies by GS would preclude the strategies from the patent grant.

     This issue arises because the main goal of the patent grant is to encourage public disclosure of inventions. It would contradict this goal, if an inventor is permitted to keep an invention secret and profit from it for a long period of time, and then continue to be eligible for the patent monopoly when trade secret is no longer a desirable or viable strategy.

     The patent law does not explicitly address this issue. The Patent Act requires an inventor to submit a patent application within one year only if the invention is "in public use". (Patent Act §102(b)) The courts, however, have broadly construed the term "public use" in this context to include most commercial activities, excluding only strictly confidential uses. For example, the courts have held that giving corset steels to a close friend to wear intimately without restrictions was public use (Egbert v. Lippmann, US Supreme Court, 1881); selling to the public a product manufactured by a secret method was public use for the secret method (Metallizing Engineering v. Kenyon Bearing, Second Circuit Court, 1946); limited market testing of a product manufactured by a secret method was public use for the secret method (Dippin' Dots v. Mosey, Federal Circuit Court, 2007).

     Giving these previous cases, does the "secret" use of the strategies by GS in the public markets to profit for itself (and perhaps for its clients) constitute "public use"? (Assume that GS invented the strategies and has used the strategies for more than one year since the invention.)

     I have not found any case that directly addresses this fact pattern. And, it can be argued that because GS used the strategies strictly in confidence, not giving up the strategies to the public domain, the uses were not "public use", even though GS did profit from them. The previous lines of cases, however, were not concerned about whether the public had learned about the invention. Rather, the core of the holdings was that the inventor had given up the patent rights by keeping the invention secret for too long.

     Therefore, it is most likely that GS will not be able to obtain a US patent on the secret strategies. Other countries, however, may still allow trade secrets to be patentable.

-- In the context of greenhouse gas emission and climate change

     It is well known that air is a type of "public goods", which normally cost little to use. A public good generally carries the characteristic that everyone can use it without significantly affecting the use by others, and therefore free of costs. (For example, see here.) Water is another "public good" natural resource. In some occasions, lands and soils can be another type of "public good".

     The "public good" nature of air is the fundamental root behind the climate change issues supposedly caused by human activities. Because it costs little to emit greenhouse gases into the air, there is little financial incentive to stop. Unless, that is, people voluntarily choose to stop, or are involuntarily required to stop by government regulations.

     Government regulations, therefore, play an important role in managing the air quality, because they create financial incentives, and thus markets, for people to change behaviors. Availability of financial incentives and markets, on the other hand, is critical in stimulating capital investments and innovations. (Many people have studied this area. See, eg, here.)

     This causal relationship between regulations, markets, and innovations is clearly illustrated by a paper I recently read. The paper was written in 2001 by Taylor, Rubin, & Hounshell of Carnegie Mellon U, and titled "The effect of government actions on technological innovation for SO2 control". The authors analyzed the correlations between the Clean Air Act and innovative activities in the reduction of SO₂ emissions among power utilities. One of the approaches they used was to collect the numbers of patents issued each year in technologies related to the reduction of SO₂ emissions from 1887 to 1997. They discovered that US patents related to SO₂ control technology surged during a 5 year period encompassing the enactment of the 1970 Clean Air Act, rising from under 5 patents issued each year before 1967 to more than 75 patents issued in 1971. And the level of issued patents remained above the 1971 level for the rest of the study period that ended in 1997. (See Fig 4 of the paper here.)

     The authors noted that prior to the Clean Air Act f 1970, when EPA was established, the patenting activities in SO₂ control technologies were nearly non-existent, even though the federal governments had generously granted research funds in the areas. Today, the federal government has been debating whether to directly regulate greenhouse gas emissions (by taxes or caps) or to avoid direct regulations by providing research funding. At lease in terms of patenting activity, the Taylor paper should shed some light on the debate.

     Google rolled out its brand new "Patent Search" tool yesterday. (See here.) I quickly tested the tool. Looks like it includes only issued US patents, not published patent applications. (I am guessing non-US patents and patent applications should be coming soon.)  At this point, the Google Patent Search tool does not offer much more than the search features provided in USPTO web site. (Here.) I did a quick search for the phrase "spinal cord injury" on both the USPTO web site and Google. USPTO returned 1875 results, but Google returned 672. Not sure why yet. Also, for each individual patent, Google Patent Search displays the "Abstract", "Claims", "Citations", and "Referenced By" sections as texts. The "Description" section, however, is only available as images. The USPTO database is all text-based, except older patents.

     I am sure Google will be improving this patent search tool. But what's next beyond patents? How about court records, or other public records? Will Google's free-use business model threaten Westlaw and Lexis? Will consumers like myself be the true beneficiaries of Google's ambition to catalogue whatever there is to be catalogued?

     Good time to be a writer!

     The Supreme Court has shown sudden interests on patent disputes in recent years. Just this year alone, the Court has determined/heard at least 5 patent cases. See table below.

     Each of these cases can have potentially far-reaching impacts on our patent system, and can mean lots of $ gained or lost for many companies. Many industrial giants have therefore taken advantage of this once-in-a-life-time opportunity and sought to advocate their own interests by submitting amicus briefs to the high court. Many of these amicus briefs expressly supported one particular side. For me, this is an interesting opportunity to find out who supports what.

     In the table below, I have listed some companies who have filed amicus briefs for stronger or weaker patent rights in the 3 cases: eBay v. MercExchange; MedImmune v. Genentech; & KSR v. Teleflex. The info is gathered from Dennis Crouch's blogs, Patently-O. (See here, here & here.)

     In these high-stake cases, traditional manufacturing-based conglomerates (i.e., GE, 3M, du Pont, J&J, Qualcomm) have consistently advocated for stronger patent rights. On the other hand, non-traditional, information-based giants (i.e., Intel, Microsoft, Yahoo, Micron, Cisco, Time Warner) have consistently advocated for weaker patent rights. Such is the bifurcated state of our world of patents!

     My guess is that such a state of bifurcation is a result of how these companies come up with their inventions. More particularly, it hinges on whether the inventions incorporate many public domain, unpatented, components. For the information industry, most of their innovations are intangible contents, software, or processes that embody numerous existent and commonly used technologies. On the other hand, for the traditional manufacturing industries, their inventions are less likely to incorporate unpatented, commonly used machines or processes.

     Given this consistent bifurcation of patent positions in the private sector, the question is whether the Supreme Court will take notice and reconsider the patentability of some subject matters such as business methods and software. In LabCorp, the Court decided not to deal with the question and let a patent on scientific relationships stand. Three Justices (Js. Stevens, Breyer, & Souter), however, dissented and would have invalidated the patent. The same three Justices also joined in eBay to question the validity of business method patents. I think if the Patent Office is unable to develop a more satisfactory examination system for business method or software patent applications, the Supreme Court will soon pick up where it left off.

     Suppose that composition S is an important medicine. It was found, however, that the composition degrades rapidly when it is exposed to the air, generating corrosive acids and causing difficulties in its storage and transportation. Subsequently, Company A discovered that the degradation was due to reactions between the composition and a group of acids commonly present in the air. Because of this finding, Company A was able to further discover that by adding composition W to composition S, the mixture would no longer degrade in the presence of the acids. Company A wants to apply for a patent on the mixture composition. There is, however, one problem. Before Company A's discovery, Company B had already synthesized the mixture of W & S, although at that time the mixture was only used for the purification of composition S, and Company B (or anyone else) did not know the mixture could prevent the reaction between S and the acids in the air. Can Company A obtain a patent for the mixture composition?

     This question doesn't seem too difficult and the law in this area is well settled. -- An inventor cannot obtain a patent on a machine or composition simply because she/he discovered a new way of using it, although the inventor can obtain a patent on the new method.

     As an example, consider Viagra, the wonder drug for men. The chemical compounds contained in Viagra were originally synthesized by Pfizer to treat cardiac diseases, not male impotence. (Pfizer obtained a patent in 1993 for the compounds and for methods of using the compounds to treat various cardiac diseases.) Later on, having discovered Viagra's effectiveness in treating male impotence, Pfizer obtained a new patent on the method of using the compounds to treating male impotence, but not on the compounds themselves. Of course, Pfizer had already held patent on the compounds. If a different inventor discovers another new use of the same Viagra compounds, however, the inventor will still not be able to obtain a patent on the compounds for the newly discovered use, only a patent on the new method.

     This rule seems reasonably clear. However, in a recent case appealed to the Court of Appeals of the Federal Circuit (Abbott Lab v. Baxter) it was not so clear to the district court judge who presided over the trial.

     In that case, what happened was essentially the same story of Company A & Company B that I described at the beginning paragraph, where Abbott was Company A and Baxter was Company B. The main differences were that Abbott did obtain a patent on the mixture composition, and that despite the well-settled law the district court held that the patent was valid even though Baxter held an earlier patent where the mixture had been synthesized. (See here & here for more info.)

     What confused the trial judge was the difference between a new use of a method and a new use of a machine/composition. -- A new use of an existing method can be patentable, if the new use achieves a different purpose. But an existing machine/composition is not patentable simply because a new use of it achieves a different purpose. The judge applies the rule for a new use of method to a new use of machine/composition. Consequently, CAFC reversed the district court's decision and invalidated Abbott's patent.

     Credits for CAFC!

     I have posted several write-ups on spinal cord injury related patents. (See here in the patent archive.) These patents involve technologies in the fields of cellular, molecular, and genetic biology, or other related fields. Even with these discoveries, we understand the search for an effective treatment for spinal cord injury is still an ongoing effort. However, the pace of advancement in these fields during the past 2 decades has been amazing.

     As a testament to the exploding rate of advancement, this year's Nobel Prize in medicine was awarded to 2 American scientists for their works in a new field of genetic biology. -- RNA interference. (See here.) I am not going to touch on the technical details of their works. Suffice it to say that RNA interference allows scientists and doctors to stop the activities of a specific gene. This sounds simple, but the scope of potential scientific and medical applications is immense. For example, if a disease is known to be caused by a specific gene, RNA interference may allow doctors to silence the gene and thus cure the disease. Because of the importance of the discovery of RNA interference, Science Magazine named it in 2002 as the Breakthrough of the Year. And, of course, the Nobel Prize.

     It is also possible to apply RNA interference to treat spinal cord injury. For example, we know some "inhibitor" proteins released after a spinal cord injury prevent the spinal cord neuron cells from re-growing. RNA interference can be used to silence the genes that control the production of these "inhibitor" proteins, which reduces or prevents the production of the proteins and allows the neuron cells to re-grow. (See here.) Some companies have already begun commercial developments in this area. (See here.)

     A lot to look forward to!

     A final note. It may be debatable whether patents are necessary to stimulate innovation, but there is little doubt that patents have contributed significantly to the advancements in medicine. Here, the 2 Nobel Prize winners indeed obtained a patent for their discoveries.

Patent number: US6,506,559

Title: Genetic inhibition by double-stranded RNA

Inventor: A. Fire; S. Kostas; M. Montgomery; L. Timmons; S. Xu; H. Tabara; S. Driver; C. Mello

Assignee: The Carnegie Institution of Washington, Washington, DC; The University of Massachusetts, Boston, MA

Filing date: 12/18/1998 (provisional application filed on 12/23/1997)

Issue date: 01/14/2003

     Deferred compensation plans have become very common in corporate America. Generally, an employer sets up a deferred compensation plan to allow its employees to postpone the payment of a portion of the compensations they have earned. The deferred compensations normally are re-invested and not taxable until when they are actually paid to the employees. The more widely know varieties of deferred compensation plans include pension plans, 401(k)'s, stock-bonus plans, and profit-sharing plans.

     Although the number of deferred compensation plans has grown rapidly, these plans are not without risks for the employees. There is the market risk associated with investment. And there is the credit risk of the employer, who must re-pay the employees. The workers of GM, Delta Airline, and United Airlines recently learned the hard way about this credit risk. (See: "G.M. Freezes Pension Plan of Its Salaried Workers", NYTimes.com; "Delta to dump pension plans", CNN.com; "Airlines lobby for relief on pensions", USAToday.com.)

     In light of these recent sagas related to deferred compensation plans, I find the following patent interesting.

Patent number: US 6,766,303

Title: Method for hedging one or more liabilities associated with a deferred compensation plan

Inventor: D. J. Marshall

Assignee: Goldman Sachs & Co.

Filing date: 10/15/2001

Issue date: 07/20/2004

     The most straightforward way for an employer to hedge its liability associated with a deferred compensation plan is to actually invest the compensations that have been deferred. For example, if the employees participating in a plan have chosen to invest a total of, say, $1m in an S&P500 index fund, then the company can simply buy $1m of the S&P index fund under the plan. This approach, however, ties up the funds, and most employers want to use the deferred compensations for other business purposes. Patent US6,766,303 offers an alternative method to hedge this liability without tying up the funds. -- Total return swaps.

     Basically, the employer would enter into a swap agreement with another party. For a set period of time, the employer agrees to pay the other party periodic interests (normally a floating rate), and in exchange the other party agrees to pay the employer an amount equivalent to the total return of the specified investments in the deferred compensation plan. For example, let's say a plan offers only one investment option, an S&P500 index fund. The employer enters into a total return swap, where the employer pays a floating rate and receives the total return of the S&P500 index. The S&P500 index return that the employer receives from the swap would hedge against its obligations under the deferred compensation plan. With this approach, the employer is not required to tie up funds because no money is exchanged upfront. Additionally, other financial instruments can also be used, such as forward contracts, options, or synthetic forwards.

     The economics of this swap hedging approach is not very different from that of the cash hedging method I mentioned earlier. In cash hedging, an employer can borrow money at a floating rate and then invest the money according to the employees' choices under the plan. In other words, just like in a total return swap, the employer pays a floating rate and receives the total return of the specified investment. If an employer does not have easy access to credits, however, the swap markets may be a more attractive hedging alternative.

     It remains to be seen whether this patent can improve the funding shortfalls in many retirement-type plans. It should be noted, however, that many investment choices do not lend themselves easily to swap-type hedging. For example, many hedge fund companies allow their employees to invest their deferred compensations in some proprietary funds developed within the companies. Because the company may be reluctant to disclose the total returns of these proprietary funds, total return swaps are not good hedging tools for these plans. So, either the companies actually invest the deferred compensation in these funds, or the plans will most likely be unhedged.

[This posting is part of a series on spinal cord injury related patents. Previous postings of the series can be found in the Patent archive here.]

Patent number: 6,033,660

Title: Method of treating a nervous system injury with cultured Schwann cells

Inventors:     J. P. Mather; R. Li; & J. Chen

Filing date:     05/10/1995

Issue date:     03/07/2000

     Let me begin with something not directly related to this patent. Dr. Wise Young is a prominent expert in spinal cord injury, and treated Superman, Mr. Christopher Reed, when he was paralyzed from a horse-riding accident. Dr. Young once formulated a multi-step plan for spinal cord injury treatment. The plan consisted of the following steps:

1. Give methylprednisolone (a steroid)
2. Remove any bone that is compressing the spinal cord
3. Stabilize the spine
4. Consider Schwann cell implants as experimental therapy
5. Begin rehabilitation as soon as possible
   

     (See here.)

The 4th step above involves "Schwann cells", which is the focus of patent US6,033,660.

     Schwann cells are found in our peripheral nervous system (PNS), but not in the central nervous system (CNS, which includes the brain and the spinal cord). The neurons in our PNS differ from those in our CNS in one very important respect. -- PNS neurons can re-grow after injury, but not CNS neurons. This observation has stimulated volumes of inquiry and research. It is believed that Schwann cells play a pivotal role in the regenerative capacity of PNS neuron cells. After an injury to the PNS, Schwann cells begin to reproduce in a fast pace. The increased population of Schwann cells secretes large amounts of substances (neurotropic factors) that help guide or facilitate the re-growth of the injured neurons. And then, at the end of this process, Schwann cells form a shield that covers and protects the newly generated neuron cells. This understanding leads to the exciting possibility that Schwann cells may also be used to stimulate re-growth of injured CNS neurons, if they are made available at the vicinity of CNS injuries. That's why Dr. Young suggested that Schwann cell therapy be considered an essential part of CNS injury treatments.

     Many types of therapy for CNS injuries utilizing Schwann cells have been proposed, such as peripheral nerve graft and implantation of guidance channels. These therapeutic methods, however, require abundant supplies of isolated pure Schwann cells. This patent describes a method of isolating and purifying Schwann cells derived from the materials obtained in a small biopsy.

     I will not delve into the technical details of the patent here. Just a quick note on Dr. Young's treatment plan. It seems there have been controversies over the effectiveness of methylprednisolone (the first step) in treating spinal cord injuries. (See here.) To me, however, that is a sign of progress!

     In eBay v. MercExchange, the Supreme Court rejected the "general rule" established by the Court of Appeals for the Federal Circuit (CAFC) in issuing a permanent injunction. (See here for more info.) The eBay decision might have rendered another CAFC product, & sibling of the "general rule", an endangered species. -- The presumption of irreparable harm in issuing a preliminary injunction. (See here for more info.)

     Both the "general rule" and the presumption of irreparable harm arose from CAFC's reading of a patent holder's statutory right to "exclude" others from accessing the patented invention. Seems to me, the "general rule" is simply a corollary of the presumption of irreparable harm when applied to permanent injunctions. Consider the 4-factor test returned by the Supreme Court in eBay for permanent injunctions. With the presumption of irreparable harm, the first factor is turned in favor of the patentee. The public interest, the 4th factor, favors promoting the progress of science, and thus also favors the patentee generally. As to the 2nd factor, adequacy of monetary compensation, if the patent holder's right is to "exclude", then monetary compensations generally will not be adequate. That's 3 out of 4. And, thus the "general rule" of issuing permanent injunction once the patent is determined to be good and infringed.

     However, the Supreme Court rejected this "general rule". That threw into question CAFC's reading of the extent of the statutory patent exclusive right, and thus raised doubts over the legitimacy of the presumption of irreparable harm. Moreover, because the 4 factors include irreparable harm, it won't make sense if irreparable harm is presumed. (In a recent post-eBay case, z4 Technology v. Microsoft Corporation, the court wrote, "[the] suggestion, that the right to exclude creates a presumption of irreparable harm, is not in line with the Supreme Court's holding" in eBay.) Granted, eBay and z4 involved permanent injunction, not preliminary injunction. However, the Supreme Court noted in Amoco Production v. Gambell (1987) that "[t]he standard for a preliminary injunction is essentially the same as for a permanent injunction with the exception that the plaintiff must show a likelihood of success on the merits rather than actual success".

     There is a good argument to be made after eBay that in preliminary injunction proceedings NO irreparable harm for the patentee should be presumed by the court, eve if the patentee clearly establishes patent validity and infringement. Patentee must separately show irreparable harm. -- One more followthrough of the mere 6-page eBay opinion.

[This posting is part of a series on spinal cord injury related patents. Previous postings of the series are here, here & here.]

Patent number: 5, 762,926

Title: Method of grafting genetically modified cells to treat defects, disease or damage
          of the central nervous system

Inventors: F. H. Gage; M. Schinstine; J. Ray; T. Friedmann; M. D. Kawaja;
                   M. B. Rosenberg; & J. A. Wolff

Filing date:06/05/1995

Issue date: 06/09/1998

     This patent is a sibling of another patent (US5,082,670) that I summarized previously (here). Both patents actually have the identical title. Procedurally, after the application for the 5,082,670 patent was filed at the Patent Office in 1988, the inventors added some new materials to that application, and submitted a separate application in 1991. The 1991 application was then further modified, eventually becoming a new application filed in 1995, which issued as the 5,762,926 patent.

     [Note: Following the general practice, I will denote the two patents by the last 3 digits of their respective patent numbers, preceded by an apostrophe: patents '670 and '926.]

     Both the '670 and '926 patents present methods of implanting genetically engineered cells directly to the Central Nervous System (CNS) to treat diseases or injuries. Implanting cells directly to the CNS provides an alternative approach to treating CNS diseases, as opposed to administering treatment indirectly through medicines. Of course, there are pros and cons for each approach, which I will not discuss. (See also my previous post here.)

     In essence, the later '926 patent offered several improvements over the earlier '670 patent. Patent '670 described the broad framework of selecting, preparing, transfecting (transferring a gene to a cell), and implanting donor cells. In particular, the inventors presented 3 examples where donor cells were genetically engineered so they could produce certain beneficial materials (for curing a kidney disease, promoting neuron growth, and curing Parkinson disease, separately). These genetically engineered cells were then implanted to the CNS of some rats. The examples showed that the genetically engineered cells indeed survived the implantation and continued to produce the desirable materials in sufficient amounts. Moreover, the materials that these donor cells produced did result in the desired therapeutic benefits after the implantation.

     Built on top of this framework, patent '926 presented several approaches to increase the survival rates of implanted cells, to regulate the production of the desired materials, and to reduce the immune response. It discussed 10 examples to illustrate how these newly improved methods could be practiced.

     With respect to spinal cord injuries, example 10 discussed in the patent is particularly interesting. In that example, the implanted cells were genetically modified to produce Nerve Growth Factors (NGF), which, as its name implies, are capable of stimulating / promoting the growth of neuron cells. (See here for info.) The inventors implanted these cells into one side of the brain of a rat, and at the same time implanted other un-modified (thus non-active) cells into the other side of the brain. They discovered that within one week after the implantation, minute neuron networks began to develop around the modified cells (that were capable of secreting NGF), and continued to develop into the eighth week. No such neuron development, on the other hand, was observed around the un-modified / non-active cells, indicating that the neuron development was most likely attributable to the secreted NGF.

     Patents '670 and '923, of course, did not claim the invention of NGF. Rather, they claimed methods of making NGF work by implantation. The discoveries of NGF and its role in neural development were amazingly huge breakthroughs in our understanding of the human neural system. So was the discovery of neurological implantation. So was the discovery of genetic engineering. Now came these patented inventions to combine them together. It took so many giant steps in science and technology to grow just a tiny bit of neurons in the spinal cord. And, it will take many more giant steps to come up with an effective treatment for spinal cord injuries. So fast (to scientists), and yet so slow (to patients)!!!