How To Prevent Pitfalls When Pursuing Patent Protection

When B. Stanley Pons and Martin Fleischmann announced that they had achieved cold fusion last March, the scientific community understandably clamored for details. But information was not forthcoming, at least in part because patents were rapidly being filed on the work. Even though scientific secrecy can be maddening in cases such as this, patents are very much a part of science. In today's environment, when even organisms and equations are being patented, filing a patent application can be complicated—especially the first time around. Here, for Scientists faced with the formidable and confusing process of applying for a patent, are some frequently asked questions-along with the corresponding answers. What is patentable? According to law, a candidate for patent protection must be "novel, nonobvious, and useful." Patents cover "compositions of matter," such as new superconducting compounds; systems, such as Genentech's specification of the host cell, vector, gene, and genetic controls used to produce recombinant human insulin; and processes, such as a new synthetic route to produce polystyrene. In recent years, the scope of what can be patented has broadened appreciably. Since 1982, for example, algorithms have been patentable, reversing the long-standing consideration of these mathematical expressions as scientific truths, therefore exempt from patent protection. A molecule, cell, or organism is patentable if it does not exist in nature. This might be a change as minor as an amino acid substitution in a protein, or as inventive as artificial seeds. Merely determining the sequence of a protein or gene isn't patentable, unless a new way is devised to do so. But a natural product is patentable if a novel, nonobvious use is found for it. "Many antibiotics, for example, are compounds that exist in nature," explains Jim Magee, a patent attorney at General Electric's Research and Development Center in Schenectady, N.Y. "But since they exist in nature in an impure state, they are not useful. [In such a case] the patent is directed to the method for making the compound by fermenting a particular organism." What makes an invention "non-obvious" or an "obvious extension" of existing knowledge? A patentable invention must not be obvious to a "scientist skilled in the art" who knows everything on the topic that is publicly available. It is not so much the particular alteration that is proposed, but the result, that is crucial. "Take an analog that is a simple hydrocarbon," Magee says. "If you merely extend the chain, or replace a methyl group with an alkoxy group, it may or may not be obvious. What's important in determining nonobviousness is if the properties are changed in a useful way." Determining non-obviousness is especially confusing in a fledgling field such as biotechnology, where inventions can be quite unusual. "Biotechnology patenting is an area of the law that is still embryonic. But we are getting to the point that certain kinds of changes are more often seen, and in the future these will be considered to be obvious," says Lisa Raines, an attorney for the Industrial Biotechnology Association. Many of these types of changes are being worked out on the clotbuster tissue plasminogen activator (tPA). Nineteen companies have filed patents on their particular tPA recipes. The first patent, issued a year ago to Oxford University for tPA derived from cultured human colon fibroblasts, specifies two "glycoforms," which are different placements of sugar side chains that affect how quickly the molecule works. Other tPA patent applications use different culture systems, different amino acid substitutions that alter activity, and even a form made in the milk of transgenic mice. What is a nonobvious change in pioneering tPA may in the future be considered an obvious extension in the manufacture of biopharmaceuticals. Is a new use for an existing invention patentable? You can't reinvent the wheel, but you can indeed take out a patent for a new use of an existing invention. Says Magee: "A good example is Minoxidil, an old compound [used] to treat hypertension. Suddenly, Upjohn discovered that it promotes hair growth under certain conditions. They couldn't patent the compound, but they filed a patent application on a method to promote hair growth—by administering a certain amount of Minoxidil. If similar compounds were known to increase hair growth, then there would have been nothing startling about Minoxidil. It has to be a truly surprising event:" New uses can be found for existing devices, too. Vaser Inc. of Indianapolis has patented a laser instrument to clear clogged arteries. It involves a balloon, but not in the conventional way that one is inflated in balloon angioplasty to press plaque against artery walls. "It's a positioning balloon used at much lower pressure, used to position a catheter, which is rotated to ablate circles of plaque," explains president Ned Kivett. A subtle difference, perhaps, but one deemed different enough to merit a patent. What goes into a successful patent application? Most of the details of writing the application are best left to a patent attorney-such as assigning the work to one of 400 classes and one of 115,000 subclasses set up by the Patent and Trademark Office. The attorney actually writes the application, based on information provided by the inventor. And this is why the scientist must effectively communicate what was done. The process begins by keeping careful records. At General Electric Corp., for example, researchers must have a witness sign their notebooks each day, so there can be no future questions of when and where something was discovered' "Next, the inventor writes the equivalent of a technical paper, describing the work done. Mention the who, what, why, how, when, and where of the project. Include a section on 'background prior art,' which the patent attorney can read to help him understand it," says Magee. Don't do what one chemist did-hand the attorney reams of jargon and lecture him in technical gibberish. John Marsella, owner of five patents and senior principal research chemist at Air Products in Allentown, Pa., says that, to work with a patent attorney effectively, it helps to understand how he or she thinks. "There's a pattern to how they look at things, and it takes a while to figure out. One thing you can do to speed the process is to start a literature search not as a scientist, but as a patent examiner would. It will be less complete from the scientific standpoint, less specific." Of course, it is essential to find out if the invention has already been patented. Some 89 computer databases are available to search among awarded U.S. patents, but patent applications are not public knowledge, nor are trade secrets. Consulting foreign patents (listed by Derwent Publishing Ltd. in London) may provide clues to patents filed in the United States, because these patents are generally processed faster than domestic ones. Inventing around an existing patent requires both great creativity and expert legal counsel. What limits should be claimed in a patent application? How far the limits of a patent extend often depends on the judgment of a particular patent examiner, and can run counter to what the scientist considers reasonable. "In biotechnology, in most cases the patent claims manufacturing the substance in any microorganism, or yeast, or species of bacteria," says Raines. "But even on this there is variation within the patent office. One examiner may feel that just because it is produced in bacteria doesn't mean it is produced in yeast, and may ask for evidence that it is indeed produced in yeast. Another examiner may accept production in another organism as being an obvious extension, without proving it." Marsella finds the same to be true in chemistry. The best patents are broad as well as specific, and this is most apparent in "first of its kind” patents. Genencor of San Francisco holds the first patent for an engineered enzyme with "improved end-use"; the patent covers any amino acid substitution in subtilisin, a protease. On another front, AT&T's Bell Laboratories' patent on an algorithm that is used to allocate resources is applicable to a wide range of activities—including making investments, routing phone calls, and scheduling planes. Clashes can develop when companies hold patents on different aspects of the same substance. In 1982, Rorer Group Inc. of Fort Washington, Pa., received a patent for factor VIII (the clotting factor needed by hemophiliacs) extracted from human blood. But when the company sought to broaden its protection in 1985, it was too late. Genetics Institute of Cambridge, Mass., had already filed to cover the gene for factor VIII, techniques to express it, and host cells. Like science itself, the patenting process is unpredictable and ever-changing. And the skills needed to master it-inquiry, creativity, organization, and communication-are the very same tools that make for good science. Ricki Lewis teaches biology at the State University of New York, Albany. The Patent Attorney: An Essential Source Of Advice Patent attorneys are part and parcel of industrial science, and researchers in large companies need not go to any effort at all to find them. "Here at the research center, a patent attorney checks in with each scientist often," says staff chemist Larry Lewis of General Electric, which is issued at least one patent every day. But even smaller companies hire their own full-time patent attorneys. Genentech has a team of six patent attorneys who work on the dozen patent infringement lawsuits the firm is currently embroiled in. For the academic scientist, a few questions to the appropriate department head or research office are likely to reveal the name of a patent attorney or firm that works for the university. "Most universities are quite interested in patent successes, became it brings in revenues. They recruit them from outside law firms," says Jim Magee, a patent attorney for General Electric. "Otherwise, simply look in the phone book. Many patent law firms are located in New York City; Washington, D.C.; Chicago; and California, places where there is a high concentration of technically oriented business and industry. Call up, make an appointment, and just sit down at a face-to-face meeting," he adds. But be certain not to take matters into your own hands if this is not your institution's policy. Once in the presence of the patent attorney, will the scientist have to deliver a freshman-level lecture on his or her field? Probably not, because a patent attorney must have some technical background—which usually means an undergraduate degree—in the area in question. "Maybe he won't have as in-depth a knowledge as a working scientist, but he knows how to ask questions. Show the attorney, your notebooks and explain what you did and why you did it," says Magee. However, John Marsella, a research chemist at Air Products Inc., has found that sometimes a bachelor's degree isn't enough. "They understand the basics, but not the research process," he says. But if this is indeed a problem, it shouldn't be for long, as more people cap an advanced degree in a science with a law degree. Already patent attorneys can be found with Ph.D.'s in a variety of scientific disciplines. Even more effective may be a collaboration between the Patent and Trademark Office and the Industrial Biotechnology Association. Their new "Biotechnology Institute" will tap academic, industrial, and government expertise to provide a continuing education program for patent examiners-a necessity in this rapidly evolving field. —R.L.

How To Prevent Pitfalls When Pursuing Patent Protection

What is patentable?

Interested in reading more?

Become a Member of