Illumina v. Roche
Case：19-1419
Decided on: March 17, 2020 by CAFC
【LRH评述】
中国的科学发现基本等同于美国的自然现象。在中美两国，二者都不属于可被授予专利权的主题。
科学发现和技术方案的相同点是，二者都是以客观世界为对象，尤其在涉及规律性方面的认识上，二者都需要人类思维的参与，是人“想”出来的。
但二者的不同点似乎更为明显。科学发现的对象原本就在那，与人类想到与否、行动与否没有关系；技术方案却是是人类创造产生的，是人类改造客观世界的成果通过类思维活动的表现。美国专利法第101条与中国专利法第2条其实并没有本质区别，而且似乎说得比中国专利法更为直白：专利法保护的是人类创造出来的东西。
LRH认为，科学发现或自然现象不是人类创造产生出来的，所以，不属于专利法保护的客体。就专利保护客体判断而言，其本质就是在可能涉及科学发现或自然现象的主题中，判断是否存在人类改造客观世界的成果。
不过，美国专利法排除自然现象的理由似乎不是从上述角度出发的，其目的是防止垄断作为科学工作基本工具的自然现象。
在美国，对于一个主题是否属于自然现象或法则，判断方法采用Alice/Mayo两步法：第一步是分析该主题是否“directed to”自然现象；第二步是分析权利要求中其他特征是否将该主题转变成适格的主题。本案中，CAFC特别强调两点内容：第一，在第一步中，主题中光含有自然现象是不够的，必须是以该自然现象为目标才行，即必须具有“directed to”要件；第二，有且只有当第一步得到满足的情况下，才进行第二步分析，即如果在第一步中，主题不是directed to自然现象，那么，就无需进行第二步分析。如何确定directed to要件呢？CAFC认为，如果权利要求要求保护的内容是自然现象本身，就可以认为满足该要件。例如，如果权利要求主题是如何发现自然现象或者如何检测自然现象的存在，那么，该权利要求就是在保护自然现象本身。
属于专利保护客体，仅仅是获得专利权的一个前提。要求专利保护的主题中存在人类改造客观世界的成果，仅仅说明该主题有资格申请专利保护。但是，是否最终被授予专利权，还要看该主题（以及相关文件）是否符合其他的授权条件，例如，新颖性、创造性、实用性、公开是否充分、是否以说明书为依据，等等。所以，不应该将保护客体判断与其他授权条件混为一谈。
在这个案件中，多数人意见应该是正确的。不同意见所提出的问题和阐述的道理是正确的，其不可取之处在于，将§101中的new and useful 直接等同于新颖性和创造性，实质上是说，判断保护客体是否适格也要以新颖性和创造性为依据。中国专利法对保护客体的定义中也有“新的”一词，但是，在审查指南中明确指出，这里的“新”无需进行新颖性分析。就像宪法规定了人身权和财产权，但是，具体的人身权和财产权保护由下位的法律部门统领，即便在不符合具体法律部门的规定时，也不需要指出还违反了宪法的相应规定。同样，保护客体的定义仅仅是个定义而言，在有具体规定对其相应内容进行限定的情况下，无需在考虑定义的对应内容；只有在没有具体条款对客体定义的某个内容，例如“技术方案”进行具体限定的情况下，才有必要适用该定义。
然而，在包括客体判断时不进行新颖性和创造性分析，并不是说人类改造客观世界的成果只是一个形式要件，而是应该是实质性要件。按照中国审查指南的规定，需要判断该成果是否解决了技术问题，是否利用了符合自然规律（而不是思维规律、社会规律或经济规律等）的技术手段，是否产生了符合自然规律的技术效果。
例如，针对利用公知计算机执行问答游戏过程控制的一个解决方案，审查指南分析说，该解决方案虽然通过游戏装置进入计算机游戏环境并通过执行计算机程序对游戏过程进行控制，但该游戏装置是公知的游戏装置，对游戏过程进行的控制既没有给游戏装置的内部性能例如数据传输、内部资源管理等带来改进，也没有给游戏装置的构成或功能带来任何技术上的改变。而该方案所要解决的问题是如何根据人的主观意志来兼顾两种游戏的特点，不构成技术问题，采用的手段是根据人为制定的活动规则将问答类游戏和成长类游戏结合，而不是技术手段，获得的效果仅仅是对问答类游戏和成长类游戏结合的过程进行管理和控制，该效果仍然只是对游戏过程或游戏规则的管理和控制，而不是技术效果。因此，该发明专利申请不属于专利法第二条第二款规定的技术方案，不属于专利保护的客体。
又例如，针对一种消费返利的方法，审查指南分析说，该方法是由计算机执行的，其处理对象是用户的消费数据，所要解决的是如何促进用户消费的问题，不构成技术问题，所采用的手段是通过计算机执行人为设定的返利规则，但对计算机的限定只是按照指定的规则根据用户消费金额确定返利额度，不受自然规律的约束，因而未利用技术手段，该方案获得的效果仅仅是促进用户消费，不是符合自然规律的技术效果。因此，该发明专利申请不属于专利法第二条第二款规定的技术方案，不属于专利保护的客体。
针对一种经济景气指数的分析和计算方法，审查指南分析说，该方法是由计算机执行的，其处理对象是各种经济指标、用电指标，解决的问题是对经济走势进行判断，不构成技术问题，所采用的手段是根据经济数据和用电数据对经济情况进行分析，仅是依照经济学规律采用经济管理手段，不受自然规律的约束，因而未利用技术手段，该方案最终可以获得用于评估经济的经济景气指数，不是符合自然规律的技术效果，因此该解决方案不属于专利法第二条第二款规定的技术方案，不属于专利保护的客体。
在中国，对于通过了客体认证，但在创造性分析过程中，通过与现有技术的对比，发现要求保护的主题实际的贡献（应该就是本案例中所说的claimed advance）并不在于技术方面时，可以直接以不符合创造性予以驳回，而不是归为不属于保护客体的情况。
例如，针对一种动态观点演变的可视化方法，审查指南分析说，发明专利申请的解决方案与对比文件1的区别在于步骤二中设定的情感的具体分类规则。从申请内容中可以看出，即使情感分类规则不同，对相应数据进行着色处理的技术手段也可以是相同的，不必作出改变，即上述情感分类规则与具体的可视化手段并非功能上彼此相互支持、存在相互作用关系。与对比文件1相比，发明专利申请只是提出了一种新的情感分类的规则，没有实际解决任何技术问题，也没有针对现有技术作出技术贡献。因此，要求保护的发明技术方案相对于对比文件1不具备创造性。
本案中，反对意见其实说的正是上面的情况。正如多数意见指出的，涉案发明基于所揭示的自然现象，通过客观存在的步骤（相当于利用了自然规律的技术手段），从自然状态的母体血液中得到了cff-DNA富集化的混合物，这也是一种符合自然规律的技术效果。因此，正如多数意见指出的，仅仅就保护客体而言，涉案发明没有问题。至于反对意见指出的各步骤均匀公知内容，确实还是应该放到创造性阶段解决为好，因为即便是公知，也还要判断公知技术组合是否显而易见，所以，不宜将复杂的技术方案——不是技术特征——是否具有创造性，放在客体适格阶段。
至于反对意见说的对科学发现的先占，其对技术进步的阻碍作用大于促进作用，确实是一个问题：涉案专利正如反对意见所说的那样，实质上是将科学发现与公知实验手段结合，而且是高度概括，其后果是实质上堵死了他人利用该科学发现在这个领域的研发。
但另一方面，也确实是涉案专利首先将这个科学发现应用于这个领域，并解决了存在的技术问题。正是因为这样，就更应该放涉案发明进行创造性审查程序：看看涉案发明的创造性高度到底怎样，如果确实是非显而易见，那么，也就意味着没有涉案发明，本领域技术人员恐怕在一段时期内仍然无法解决该领域存在的技术问题，因此，似乎也不能简单地说，因为使用了公知技术且高度概括，其促进作用就必然小于阻碍作用。毕竟，这只是增加了研发的难度，至于说堵死，似乎小看了人类智慧；再者，涉案发明并没有阻止该科学发现应用在其他领域呢。
【审查指南】
专利法所称的发明，是指对产品、方法或者其改进所提出的新的技术方案，这是对可申请专利保护的发明客体的一般性定义，不是判断新颖性、创造性的具体审查标准。
技术方案是对要解决的技术问题所采取的利用了自然规律的技术手段的集合。技术手段通常是由技术特征来体现的。未采用技术手段解决技术问题，以获得符合自然规律的技术效果的方案，不属于专利法第二条第二款规定的客体。
科学发现，是指对自然界中客观存在的物质、现象、变化过程及其特性和规律的揭示。科学理论是对自然界认识的总结，是更为广义的发现。它们都属于人们认识的延伸。这些被认识的物质、现象、过程、特性和规律不同于改造客观世界的技术方案，不是专利法意义上的发明创造，因此不能被授予专利权。例如，发现卤化银在光照下有感光特性，这种发现不能被授予专利权，但是根据这种发现制造出的感光胶片以及此感光胶片的制造方法则可以被授予专利权。又如，从自然界找到一种以前未知的以天然形态存在的物质，仅仅是一种发现，不能被授予专利权。
【判例摘抄】
Majority Opinion
35 U.S.C. §101 provides that “Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor......” Congress plainly contemplated that the patent laws would be given wide scope; the legislative history likewise indicated that Congress intended statutory subject matter to include anything under the sun that is made by man.
The Supreme Court has held that §101 contains an important implicit exception. Laws of nature, natural phenomena, and abstract ideas are not patentable. These exceptions exist because monopolizing the basic tools of scientific work might tend to impede innovation more than it would tend to improve it. However, the Supreme Court has advised that these exceptions must be applied cautiously, as too broad an interpretation of this exclusionary principle could eviscerate patent law.
Laws of nature and natural phenomena are not patentable, but applications and uses of such laws and phenomena may be patentable. A claim to otherwise statutory subject matter does not become ineligible by its use of a law of nature or natural phenomenon. On the other hand, adding conventional steps, specified at a high level of generality, to a law of nature or natural phenomenon does not make a claim to the law or phenomenon patentable.
To distinguish claims to patent-eligible applications of laws of nature and natural phenomena from claims that impermissibly tie up such laws and phenomena, we apply the two-part test set forth by the Supreme Court. First, we examine whether the claims are “directed to”a law of nature or natural phenomenon. If—and only if—they are, then we proceed to the second inquiry, where we examine whether the limitations of the claim apart from the law of nature or natural phenomenon, considered individually and as an ordered combination, transform the nature of the claim into a patent-eligible application.
At step one of the Alice/Mayo test, it is not enough to merely identify a patent-ineligible concept underlying the claim; we must determine whether that patent-ineligible concept is what the claim is “directed to”.
Under Mayo, we have consistently held diagnostic claims unpatentable as directed to ineligible subject matter. In contrast, we have held that method of treatment claims are patent-eligible.
The focus of the dispute in this case is whether the claims of the ‘751 and ‘931 patents are “directed to” the natural phenomenon, i.e., whether they claim the discovered natural phenomenon itself versus eligible subject matter that exploits the discovery of the natural phenomenon.
As an initial matter, [we have to determine] how to articulate the natural phenomenon that the inventors discovered. We adopt Illumina’s articulation of the natural phenomenon, i.e., that cell-free fetal DNA tends to be shorter that cell-free maternal DNA in a mother’s bloodstream.
We then turn to the crucial question on which this case depends: whether the claims are “directed to “that natural phenomenon. We conclude that the claims are not directed to that natural phenomenon but rather to a patent-eligible method that utilizes it.
……Those process steps change the composition of the mixture, resulting in a DNA fraction that is different from the naturally-occurring fraction in the mother’s blood. Thus, the process achieves more than simply observing the fetal DNA is shorter than maternal DNA or detecting the presence of the phenomenon.
Roche insists that the claims in this case are no more eligible than the claims at issue in Ariosa. We disagree. In Ariosa, the relevant independent claims are directed to a method “for detecting a paternally inherited nucleic acid” (claims 1 and 24) or a method “for performing a prenatal diagnosis” (claim 25). The only operative steps in the claims were “amplifying” (i.e., making more of) the cell-free fetal DNA and then “detecting it”, “subjecting it …to a test,” or “performing nucleic acid analysis on it to detect it.” We found those claims ineligible because, like the invalid diagnostic claims at issue in Mayo, Athena, and Cleveland Clinic, they are directed to detecting a natural phenomenon. In essence, the inventors in Ariosa discovered that cell-free fetal DNA exists, and then obtained patent claims that covered only the knowledge that it exists and a method to see that it exists. Here, in contrast, the claims are directed to more than just the correlation between a DNA fragment’s size and its tendency to be either fetal or maternal. And the claims do not merely cover a method for detecting whether a cell-free DNA fragment is fetal or maternal based on its size. Rather the claimed method removes some maternal DNA from the mother’s blood to prepare a fraction of cell-free DNA that is enriched in fetal DNA. Thus, the claims in this case are different from the claims that we held invalid in Ariosa.
Roche also argues, based on the Supreme Court’s decision in Myriad, that a naturally occurring DNA segment is a product of nature and not patent eligible merely because it has been isolated. But the claims here are not directed to the cell-free fetal DNA itself. The Supreme Court in Myriad expressly declined to extend its holding to method claims reciting a process used to isolate DNA. The Court stated:
It is important to note what is not implicated by this decision. First, there are no method claims before this Court. Had Myriad created an innovative method of manipulating genes while searching for the BRCA1 and BRCA2 genes, it could possibly have sought a method patent. But the processes used by Myriad to isolate DNA……are not at issue in this case.
Thus, in Myriad, the claims were ineligible because they covered a gene rather than a process for isolating it. Here we encounter the opposite situation, i.e., the claims do not cover cell-free fetal DNA itself but rather a process for selective removal of non-fetal DNA to enrich a mixture in fetal DNA. Thus, the Supreme Court’s decision in Myriad is not on point.
In our view, CellzDirect, while not directly on point, is instructive. In CellzDirect, the inventors discovered the natural phenomenon that some fraction of hepatocytes are capable of surviving multiple freeze-thaw cycles. Having made that discovery, they patented an improved process of preserving hepatocytes, that comprises freezing hepatocytes, thawing the hepatocytes, removing the non-viable hepatocytes, and refreezing the viable hepatocytes. We found that their claimed invention was patent-eligible because it was not simply an observation or detection of the ability of hepatocytes to survive multiple freeze-thaw cycles. Rather, the claims are directed to a new and useful method of preserving hepatocyte cells.
The inventors in CellzDirect did not invent hepatocytes or impart to hepatocytes an ability to survive cycles of freezing and thawing. Rather, they discovered that hepatocytes naturally have that ability, and they exploited that phenomenon in a patent-eligible method. So too here, the inventors of the ‘751 and ‘931 patents obviously did not invent cell-free fetal DNA or the relative size distribution of fetal and maternal cell-free DNA in maternal blood. And, like in CellsDirect, the inventors used their discovery to invent a method of preparing a fraction of DNA that includes physical process steps to selectively remove some maternal DNA in blood to produce a mixture enriched in fetal DNA.
Roche argues that the techniques for size discriminating and selectively removing DNA fragments that are used to practice the invention were well-known and conventional. And we recognize, of course, that the inventors of the ‘751 and ‘931 patents did not invent centrifugation, chromatography, electrophoresis, or nanotechnology. But while such considerations may be relevant to the inquiry under Alice/Mayo step two, or to other statutory considerations such as obviousness that are not at issue before us in this case, they do not impact the Alice/Mayo step one question whether the claims themselves are directed to a natural phenomenon. Again, CellzDirect is instructive, where we acknowledged that the inventors had not invented the well-known processes of “freezing” and “thawing”, but only in the context of the Alice/Mayo step two inquiry.
Rather than focusing on what the inventors of the ‘751 and ‘931 patents did not invent, we focus our Alice/Mayo step one analysis on what the inventor did purport to invent and what they claimed in their patents: methods for preparing a fraction of cell-free DNA by the physical process of size discriminating and selectively removing DNA fragments longer than a specified threshold. Those method are “directed to” more than merely the natural phenomenon that the inventor discovered. Accordingly, we conclude at step one of the Alice/Mayo test that the claims are not directed to a patent-ineligible concept, and we need not reach step two of the test.
III
In Ariosa, we recognized that the inventors had made a discovery with implications that would allow what had previously been discarded as medical waste to be used as a tool for determining fetal characteristics. We acknowledged the profound impact that the discovery had on the field of prenatal medicine, including that it created an alternative for prenatal diagnosis of fetal DNA that avoids the risks of widely-used techniques that took samples from the fetus or placenta. Nevertheless, under guidance from the Supreme Court, we determined that the discovery of that natural phenomenon, no matter how significant it was to the medical field, was not itself patentable, and neither was a method for detecting it.
The invention in this case is the product of further research on cell-free fetal DNA. This time, the inventors discovered that, not only does the fetal DNA exist in the bloodstream of a pregnant mother, but it has characteristics that make it distinguishable, and therefore separable, from the maternal DNA. Again, regardless how groundbreaking this additional discovery may have been, the inventors were not entitled to patent the natural phenomenon that cell-free fetal DNA tends to be shorter than cell-free maternal DNA. “Groundbreaking, innovative, or even brilliant discovery does not by itself satisfy the §101 inquiry”. Thus, they could not claim a method directed to the natural phenomenon, e.g., a method for determining whether a fragment of cell-free DNA is fetal or maternal based on its length. And they did not attempt to patent such a method.
The inventors here patented methods of preparing a DNA fraction. The claimed methods utilize the natural phenomenon that the inventors discovered by employing physical process steps to selectively remove larger fragments of cell-free DNA and thus enrich a mixture in cell-free fetal DNA. Though we make no comment on whether the claims at issue will pass muster under challenges based on any other portion of the patent statute, under §101 the claimed methods are patent-eligible subject matter.
Dissenting opinions
The Majority holds that the asserted patents are directed to patent-eligible subject matter. I respectively disagree and dissent. I conclude that the claims are directed to a natural phenomenon. The patents’ only claimed advance is the discovery of that natural phenomenon. The claims, the written description, and the legal precedent applicable to this case all support the conclusion that the patents are ineligible.
The Majority sidesteps well-established precedent by reasoning that the claims in this case belonging in a unique “bucket” reserved to patents that claim “a method of preparation.” By placing this case in that bucket and not in a “diagnostic case” bucket, the Majority summarily dismisses precedent like Athena, Roche Molecular, Cleveland Clinic, Genetic Techs., Ariosa, and others. Our precedent, however, does not support the Majority’s per se grouping of claims. A “method of preparation case” is treated no differently than any other process claim under our law.
The Majority disregards well-established precedent for conducting the Alice, step one, “directed to” inquiry by failing to consider the patent’s claimed advance.
In the three years following Alice, this court addressed numerous §101 cases without articulating a more definite “directed to” inquiry. Instead, we performed step one of the patent-eligibility inquiry by comparing the claims at issue to the claims held eligible or ineligible in earlier Supreme Court and Federal Circuit cases.
Since 2016, in a string of cases reciting process claims, we began conducting the “directed to” inquiry by asking whether the “claimed advance” of the patent “improves upon a technological process or is merely an ineligible concept.” Athena; Genetic Techs.
To determine a process’s “claimed advance,” we review the claims and the written description. If a written description highlights the discovery of a natural phenomenon—e.g., by descripting the natural phenomenon as the only “surprising” or “unexpected” aspect of the invention or that the invention is “based on discovery” of a natural law—the natural phenomenon likely constitutes the claimed advance. Ariosa; Athena; Cleveland Clinin.
In Ariosa, we concluded that the claims were directed to a natural phenomenon based in part on the patent’s disclosure that the natural phenomenon was a “surprising and unexpected finding.” In Athena, we concluded that the claimed invention was “only in the discovery of a natural law” based in part on the patent’s discovery that the inventors “surprisingly found” the natural law. In Cleveland Clinic, we concluded that the claims were directed to a natural law relying, in part, on the patent’s disclosure that “the invention are ‘based on the discovery’” of the natural law.
……The claimed advance is, therefore, the natural phenomenon. This conclusion is bolstered by the fact that the claimed method steps begin and end with a naturally occurring substance, as in Ariosa. …… The inventors did not create or alter any of the genetic information encoded in the cff-DNA (cell-free fetal DNA) in the claimed method steps. ……. The claimed method is therefore directed to a natural phenomenon.
The Majority fails to identify the claimed advance.
……
For method claims that encompass natural phenomena, the method steps are the additional features that must be new and useful. See Parker v. Flook (“The process itself, not merely the mathematical algorithm, must be new and useful.”). We must assess whether the additional features are new and useful within the field generally, not in the context of their particular application to the newly discovered phenomenon.
The method steps under review fail to transform the nature of the claims into patent-eligible applications. The three claimed method steps of……are not new, either alone or in combination.
For step two purposes, that the size discrimination and selective removal method steps were never before applied to the newly discovered natural phenomenon does not render those steps new and useful. …… They do not disclose an unconventional assay to the newly discovered natural phenomenon. Instead, they adapt commercially available DNA separation techniques to the natural phenomenon.
……
Simply appending routine, conventional steps to a natural phenomenon, specified at a high level generality, is not enough to supply an inventive concept. Thus, under step two, the claims of the patent in this appeal that are directed to patent ineligible subject matter are not transformed and made eligible under Alice step two.
Preemption
The Supreme Court has made clear that the principle of preemption is the basis for the judicial exceptions to patentability. Alice. As Mayo emphasized, “there is a danger that the grant of patents that tie up the use of laws of nature will inhibit future innovation premised upon them.”
Here, the claims are drafted in a manner that tie up future innovation premised upon the natural phenomenon because no skilled artisan would be entitle to rely on the natural phenomenon to isolate cff-DNA. That a skilled artisan could isolate or enrich cff-DNA using some unclaimed technique is not dispositive for preemption. See Athena v. Mayo (Chen, J., concurring with denial of the petition for rehearing en banc)(“That claims 7 and 9 do not preempt all ways of observing the law of nature isn’t decisive, as none of the steps recited therein add anything inventive to the claims.”) Like in Athena, the only claimed advance here is the discovery of the natural phenomenon, and as drafted, these claims significantly preempt use of that natural phenomenon.
