The Petrie-Flom Center for Health Law Policy, Biotechnology, and Bioethics at Harvard Law School and the Center for Advanced Studies in Biomedical Innovation Law (CeBIL) at the University of Copenhagen recently announced a new collaboration, the Project on Precision Medicine, Artificial Intelligence, and the Law (PMAIL). PMAIL will provide a comparative analysis of the law and ethics of black-box personalized medicine, explaining the shortcomings of the current innovation policy landscape in Europe and the US, and providing a comprehensive examination of various policy options to better harness the potential of black-box medicine.
Professor I. Glenn Cohen, faculty director of the Petrie-Flom Center, will lead PMAIL. In a Q&A with Harvard Law Today, he discussed the center’s latest project.
Can you give us a little background on what precision medicine is?
“Precision Medicine” is an approach to disease treatment and prevention that takes into account individual variability in genes, environment, and lifestyle for each person treated. Increasingly, physicians understand that optimal treatment is not “one size fits all.” For example, a treatment that works for one person with breast cancer may not work for another person.
In this project we are particularly interested in approaches to medicine that use large health care data sets to better match treatments to patients. In particular, we are interested the use of machine-learning algorithms and artificial intelligence to examine newly available troves of health data, including genomic sequences, patient clinical care records, and the results of diagnostic tests to make predictions and recommendations about care. In some ways, the goal of precision medicine is for medicine to work a little more like Netflix, the movie-watching site that tries to provide individually tailored recommendations to its users. Precision medicine has tremendous potential to improve patient care and outcome, as well as reduce wasteful medical spending, but also has some risks.
The Project has a particular interest in “black-box” medicine, what does that mean?
The algorithms used in precision medicine are sometimes called “black-box” either because these algorithms are based on unknowable machine-learning techniques or because the relationships they draw are too complex for explicit understanding. These algorithms, and artificial intelligence techniques more broadly, represent a point where health technology may outstrip human understanding.
The risk in using complicated algorithms and cutting edge artificial intelligence to predict and guide health care is how can we then evaluate these technologies. Some of these new techniques will help us innovate and push the boundaries of medicine but others may be based on errors and lead to sub-optimal care for patients. Certainly, careful oversight will be needed to make sure precision medicine achieves its potential while minimizing risks to patients.
What are some other challenges facing this field?
The benefits we get out of precision medicine are limited by the sophistication of the algorithms used and the richness of the data sets they mine. Currently, intellectual property law does not properly incentivize companies to share their big data sets and this is a problem because these artificial intelligence techniques rely on very large data sets. Additionally, several recent big intellectual property cases, notably Mayo v. Prometheus Labs and Association of Molecular Pathology v. Myriad, have undercut the protections an inventor can expect for algorithms which, in turn, may reduce the incentives of developers to refine the algorithms.
These intellectual property issues reflect the core challenge: our legal and regulatory frameworks may not be perfectly suited for this exciting new discipline. Artificial intelligence and machine learning in medicine create tremendous possibilities of transforming health care for the better—but these techniques are so different from traditional approaches to medicine that we need new tools to understand how precision medicine should be developed, regulated, and deployed in care settings. We need to make sure that we are not losing out on the benefits of artificial intelligence in health care while also making sure that there are appropriate patient protections.
How does PMAIL fit into this landscape?
The purpose of PMAIL is to create a roadmap for “catching up” our legal and regulatory systems to the reality that precision medicine is here and will only get more popular. There has been a lot of work done in the scientific and medical communities to develop precision medicine techniques and apply artificial intelligence to health care treatment decisions, but relatively little legal scholarship done to understand how our legal system should respond. PMAIL will seek the best legal and regulatory frameworks for harnessing the potential of these techniques while also protecting patients.
PMAIL will sit squarely at the intersection of health care delivery, cutting edge data science, and the law. PMAIL seeks to answer important health technology questions, such as: What are the concrete legal hurdles to the further development of black-box medicine? How can we incentivize the development of quality health care algorithms that improve patient care? How does the legal framework differ in the US and Europe and could we learn from each other? How much control should patients have over how their data is used to produce personalized medicine models and the use of these models in their health care?
How does PMAIL fit into the rest of the Petrie-Flom Center’s work?
The Petrie-Flom Center has for over a decade worked to be the “lawyer in a room of scientists and doctors.” Harvard University is such a leader in the biomedical sciences. Being embedded in this fantastic environment helps us to be on the legal and ethical frontline of new technologies, such as CRISPR, Mitochondrial Replacement Therapy, etc. One of the Petrie-Flom Center’s area of expertise is to provide frameworks for how our legal, regulatory, and ethical structures can best respond to medical innovations.
PMAIL reflects our commitment to understanding how legal systems can promote medical and scientific innovations. We are proud that CeBIL at the University of Copenhagen and the Novo Nordisk Foundation approached us to be part of their bigger project on the challenges facing biomedical innovation globally. PMAIL is the result of their interest in our participation. It is a great example of our Center’s commitment to cutting edge bioethical and health care law questions.
What impact do you hope PMAIL will have?
I anticipate that PMAIL will help influence intellectual property and other regulatory policies, both in the United States and Europe. In the last several decades we’ve often thought of medical innovation in terms of new drugs and devices, but we now are living in the age of data. Health care is no exception, especially with the rise of precision medicine.
The Petrie-Flom Center is excited to host a project that will help update our legal systems to promote the effective use of big data and artificial intelligence to revolutionize health care delivery.
For more information on PMAIL and the Center’s other initiatives, visit the Petrie-Flom Center website.