Free Medicine

Prescription drug spending in the U.S. is projected to nearly double by 2020. At the same time, new drug development is becoming even more expensive and less productive. The present regime dominated by IP and data exclusivity is failing.

The principles of free software map cleanly onto drug development because drugs are really information products. The molecular structure of a drug -- the sequence and spatial configuration of its components -- contains information about how to modify the drug's target, itself a specific molecular structure based on information contained in DNA. In another sense, the value of a drug is the information provided for by credible, government-sanctioned clinical trials, which show the safety and efficacy of the drug (see Eisenberg). Open source principles can promote free drug development in both respects. Networks of scientists with access to collaborative computational tools, like molecular libraries and free data sharing, can rapidly and effectively search for and refine potential targets and leads. Then distributing the clinical trials globally and sharing the data freely can mitigate their expense. Such an"open source" drug discovery network has been launched for tuberculosis drug development.

But, this open source version merely replicates current drug development. Conventional drugs are well suited for the treatment of simple or acute problems, such as short-term symptoms like acute pain, or most infectious diseases. Pharmaceutical companies are generally not interested in obtaining IP or data exclusivity for these kinds of drugs. Many problems are already solved by generics, and even needed new infectious disease treatments are not profitable. So, Big Pharma does this kind of drug development to solve other disease problems, but it can only lead to flawed solutions.

This is because wee are now going through the first stage of a genuine revolution in medicine. This revolution is the product of three big changes.

First, the underlying science, biology, has transformed. The most outwardly visible manifestation of this is the Human Genome Project, and the re-orientation of biology towards the primacy of DNA sequence and understanding the cell as a complex system of molecular interaction. But, the changes are broader than that. The techniques of molecular biology have changed to become more modular, kit-based, and reliant on instrumentation and software applications than it was before. There is now even the possibility of both computational and wetlab tinkering available outside the traditional laboratories, leading to the emergence of "DIY biology" and "biohacking" To be sure, kit-based biology is sloppy and requires substantial tuning, and computational methods are still limited. But, the trend is clear, and a substantial part of molecular biology is no longer a trial-and-error, basic discovery-oriented science.

Second, medicine is being transformed by the fundamental information and communications revolution that is changing the world as a whole. Twenty years ago, the molecular understanding of human health was limited to a few metabolites and protein levels, like glucose concentration and insulin levels. Now we are aware of the billions of bases in the human DNA sequence, and of the importance of the dynamic expression levels of 30,000 genes, hundreds of thousands of proteins, thousands of small molecule -- and even beyond that, the identities and dynamic function of the bacteria that live within us. And, the technologies to actually measure all of this information are becoming better and cheaper at an exponential rate. In addition to providing the means for storing and analyzing massive data sets for individuals, the global information revolution means that all of these data can be recorded and compared with data from other people in other conditions.

Third, the basic nature of the diseases that are important in medicine are changing. Medicine's future is dealing with chronic diseases, like diabetes, asthma, cardiovascular disease, and COPD, long-term infections like HIV/AIDS and hepatitis, and related long-term diseases like cancer. These are also not really "diseases"; in the way we understand infectious diseases -- they do not have a discrete causative moment, particular group of symptoms, specific range of outcomes, and most importantly of all, a definable "cure." Chronic diseases involve a complex of molecular pathways, and disease etiology and progression vary highly between individuals, which means that all of the information described in the previous paragraph is highly significant.

It should be clear why conventional drugs, as Big Pharma wants to develop, are only partially successful. There is a massive amount of information for a drug to interface with, and the information varies between individuals. So we have statins, like Lipitor, that have at best an uncertain impact, despite their ubiquity and cost. We have drugs like Vioxx that are apparently more specific than its predecessors, but which through that specificity somehow causes more serious side effects.

The future of medicine requires rethinking the relationship between information and health. Not a "ersonalized medicine," that is just about focusing the population for which to prescribe a drug. Rather, , the way forward is exemplified by successful first steps towards the use of social networks for conducting genome-wide association studies at one end of a new Free Medicine pipeline, and clinical trials at the other end of the pipeline, with biological "tinkering" in the middle. These developments will facilitate distributed innovation (or peer production) driven from the doctor-patient level. This is not itself new in medicine. Innovation once emerged from case studies rather than mass trials. In an era where we can measure individual variability, it makes sense to return to a more distributed and flexible form of medical development.

What will be required for this vision of Free Medicine to succeed? The main challenge is likely to be developing social networks through education and the spread of communication tools. But, as this challenge is solved, the infrastructure will exist to defeat the application of patents to genetic sequences, tissue banks, cell lines, and naturally occurring biomolecules and biochemical reactions. Ultimately, Free Medicine will not only reduce drug spending. It will also lead to better medicine, and thus reduce the costs of both medical care and poor health.

[Please feel free to comment, criticize, edit, etc. Thanks!]