The National Academies of Sciences, Engineering, and Medicine (NAS) has issued a report on assessing biodefense capabilities in view of new biotechnologies that could be used to reactive, alter, or design dangerous microorganisms or toxins. Specifically, the report scrutinizes synthetic biology (an umbrella term for a wide array of techniques available for the purpose of biological design). The report, Biodefense in the Age of Synthetic Biology, is publicly available here. This is an era where public health authorities must think beyond currently existing microorganisms (viruses, bacteria) and contemplate biological attacks or events resulting from novel biological agents.The U.S. Department of Defense asked the NAS to “develop a strategic framework to guide an assessment of potential security vulnerabilities related to advances in biology and biotechnology, with a particular emphasis on synthetic biology.”
In its endeavor, the study committee developed a framework to identify the relative level of concern that should attach to particular technological scenarios. In the event of an outbreak from a novel organism, or an attack with a novel toxin, how should public health officials determine the level of risk? The report's framework for assessing concern consists of four factors, along with descriptive elements within each factor. The factors are Usability of the Technology, Usability as a Weapon, Requirements of Actors, and Potential for Mitigation. Looking at these factors more simply, they assess the ease of using a technology, how feasible it is to use it as a weapon, the identification of what actors could achieve certain technical goals (having both knowledge and access to resources), and finally, the existence of measures to counteract a new biological threat. With that framework for guidance, the report ranks certain threats as warranting higher concern than others:
Of the potential capabilities assessed, three currently warrant the most concern: recreating known pathogenic viruses, making existing bacteria more dangerous, and making harmful biochemicals via in situ synthesis. The first two capabilities are of high concern due to usability of the technology. The third capability, which involves using microbes or synthetic pathways to produce harmful biochemicals or toxins to be used against humans, is of high concern because its novelty challenges potential mitigation options.
The report is a timely summary of how current genetic technologies recast and expand biosecurity threats. The framework that the NAS has provided for a methodical evaluation of a new biological organism or biochemical capability will allow public health and national security responders to more quickly determine risk and response during unanticipated events.
The recent release of 30 research papers, collectively describing the results from the ENCODE project, prompted headlines around the country characterizing this as a milestone in genetic research. As a follow-on project to the release of the original Human Genome Project sequence (which focused on the genes), the ENCODE project attempts to identify what the rest of the DNA in our genome – the so-called “junk” – is doing. It’s already known that only about 1% of the human genome actually contains the genes. What are the rest of the 3 billion bases for? “Junk DNA” was never an accurate or worthy title for it – it simply revealed the state of ignorance about the human genome. Now, the ENCODE consortium reports its further annotation of human DNA – they were able to “assign biochemical functions for 80% of the genome.” They “systematically mapped regions of transcription, transcription factor association, chromatin structure and histone modification.” This work begins to detail the underlying genetic switching mechanisms that operate behind the scenes in the genome and "regulate" how genes are expressed. No one doubts that nature is likely to have retained much of the DNA in our genome because of its usefulness. However, it's not clear that the ENCODE project, in cataloguing the signals from any biochemical event to describe "function," has yet produced a map with high genetic resolution. The Human Genome project - writ large - continues to unfold. However, the legal system has made use of DNA identification technologies that were developed from current genetic knowlege. If we look for any impact on the uses of DNA sequence information in the law, several consequences of the ENCODE research emerge. First, in using DNA as a forensic tool – relying on individuality in sequence to create a personally distinguishable identifier – criminal (and other) law has come to rely on a consensus use of a set of DNA sites (the STR (short tandem repeat) loci, retained in the FBI CODIS database) that were chosen because they offered variation but minimal biological information. Thus, these DNA sites could be used for forensic comparison while revealing very little about an actual person; that fact minimized any privacy invasion from use of these markers. In theory, they capture genotype differences without revealing critical phenotypic information. It was already known, for example, that some of the STR sites were found in an intragenic region – e.g., CSF1P0 maps to an intron – but these sites were not considered informative for any particular trait or condition. Could these STR sites now now be recharacterized as informative – and could they reveal more about the phenotype of an individual? My colleague, David Kaye, has more thoughts on ENCODE and "junk DNA." Practically, might the STR loci now be susceptible to a more critical look when a 4th Amendment-based privacy interest in invoked to challenge a government DNA database?
A key question emerges: does the ENCODE research meaningfully reclassify the STR loci for 4th Amendment purposes? This is not an abstract inquiry; already, the ENCODE data has been invoked in the rehearing of Haskell v. Harris by the 9th Circuit en banc in California. This case is a 4th Amendment challenge to the state’s practice of collecting DNA from arrestees (9th Circuit panel upheld; see earlier story). The Electronic Frontier Foundation, as amicus, has asked the court to reconsider the privacy interest advanced by the challenger in view of the ENCODE findings. A precise ENCODE-derived analysis of the STR loci is not available; the EFF letter simply states that “ENCODE has determined that “junk” DNA plays a critical role in determining a person’s susceptibility to disease and physical traits like height” (citing to the New York Times article on ENCODE) and that it is "highly likely that the genetic markers contained in each Appellant's DNA profile reveal much more information than just his or her identity." That's speculation which lacks any precision with respect to the STR loci that underlie the legal challenge. It’s true that ENCODE has certainly opened the door to a reunderstanding of purpose in much of the human genome (actually, that is its goal); it is not known whether the STR loci, however, are individually tracked to real phenotypic expression, and whether STR loci variations contribute to a more complex DNA profile for an individual than was previously thought. In general, the constitutional analysis now confronts an evolving scientific portrait of the human genome, but at this point, it has not been shown that the precise STR loci at the center of the DNA database challenges have been reconceptualized by ENCODE in a manner that undermines their genetically inert status. What's noteworthy about this period in DNA database litigation is that Haskell v Harris and the recent King v. Maryland (likely to be heard by the Supreme Court) are advancing the constitutional issues of DNA collection from arrestees in the nation's leading courts at a time where the underlying science is more in flux than usual.
The rapid development of new genetic tests and the expansion of commercial genetic testing reflect the years of molecular research that has uncovered genes, proteins, RNA and metabolites that are implicated in disease processes. Widely known are, for example, the single gene tests that allow a patient to be tested for a genetic predisposition to disease based on whether she has a mutation in the relevant gene (e.g., BRCA1 testing for breast/ovarian cancer). But these conceptually simply tests are rapidly being supplemented by the next wave of clinical molecular medicine. The collection of fields now known collectively as “omics” represent molecular science as it now seeks to explain biological phenomena through the collective behavior of multiple genes or proteins (essentially looking for patterns in large datasets). These become the fields of genomics, proteomics, metabolomics, etc. Because of the large data volumes, they lie at the intersection of biotechnology and computer science (i.e., bioinformatics). One commercially available example is the Mammaprint genetic test, which tests 70 genes in a breast cancer patient genes to develop a genetic “signature” that indicates the likelihood of recurrence (and the advisability of further treatment). Such a test requires the collection of accurate biological data accompanied by a computational model that integrates the profile information to arrive at a clinically relevant conclusion. Some of these tests to date have been regulated by the Food and Drug Administration as in vitro diagnostics (IVD), but there is no coherent regulatory scheme for genetic laboratory science that integrates the oversight of single-gene genetic tests with the more complex genetic signature tests. However, due to a recent case at Duke University where a researcher had engaged in the use of faulty gene pattern tests as the basis for decision-making in providing chemotherapy to cancer patients in clinical trials, the National Cancer Institute asked the Institute of Medicine (IOM) to review how “omics” tests are developed, validated and regulated. In the report recently issued, Evolution of Translational Omics: Lessons Learned and the Path Forward, the IOM has called for improvements in the way these "omics" tests are peer-evaluated, but also asks for the developers of such genetic tests to consult with the FDA prior to using such tests in clinical trials.These developments are very significant because we expect that the potential uses of molecular patterns to inform clinical decisions are numerous but potentially susceptible to irregular development and even (unintentional) misuse. The uneven involvement of the FDA in the regulation of genetic testing has contributed to the chaotic environment in which these technologies are entering clinical trials, and later, the marketplace. The real world consequence is that patients could be given faulty clinical information regarding their diagnosis, prognosis or treatment options. While the incentives to produce gene or protein signatures of clinical significance will increase, the consequences of misinformed clinical care are very real. A the IOM report concludes, all of this argues for a institutional culture – government, funders, universities, journals – that appreciate the novel, interdisciplinary nature of these laboratory tests and provide assessments that match their complexity.
The field of bioinformatics, generally described as the use of computer-based data processing applied to problems in the life/medical sciences has now delivered a promising set of results: expression profiles (a biochemical fingerprint of what is happening in the body) were compared between known diseases and known drugs in order to identify "cousins" - where both drug and disease have similar effects on the body. This was a random matching pool, looking to identify drugs which might have previously undiscovered relationship - i.e., potential use - for other medical conditions. In this report, scientists at Stanford reported some previously unrecognized connections between, for example, topiramate, an anticonvulsant drug, now shown to have a parallel biochemical signature with inflammatory bowel disease (IBD), suggesting a potential new application for the drug. Original papers here and here. The repackaging of old drugs for new uses is also referred to as repositioning (or repurposing). Legal issues? New uses for old drugs are potentially patentable, but the drugs themselves are likely to be either coming off patent or already in the public domain, so the focus turns to method patent claims, which recite a way of using a drug in a process (e.g., therapeutic application). From a regulatory perspective, the drugs may have already passed through the FDA-required clinical approval process, so safety profiles are established. But efficacy would still need to be studied (do they work?). From the view of R&D, here's an approach to drug discovery that doesn't require finding new chemicals, but simply (sort of) enlarging the profile of drugs that are already known.
