The future of medicine

In the future things will be ... uh, better ... mostly

ast week I was reading one of my favorite books: Diseases of Trees and Shrubs by Sinclair, Lyon and Johnson. It's a magnificent book that lists diseases from more than 350 biological agents, many of which have evocative names like Nectria cinnabarina, which might sound like a tasty fruit drink but is actually a fungus that infects mountain ash and many other kinds of trees.

It's amazing how many different ways your tree can get sick. But what struck me was how well it describes the current state of medical science. We know a lot about all the different things that can make you sick, and we'd like to think that doctors can cure most of them, but it's just not true. And the idea that we need medical care to stay alive, and we're somehow deprived if we don't get it, is unfortunately to a large extent wishful thinking.

Three dimensional ribbon structure of a protein molecule

In the book the tree diseases are well described, but what about the treatment? For that you must turn to Tree Maintenance by the legendary tree doctor P.P. Pirone, who says there are basically two things you can do: (1) cut off the diseased branch or (2) cut down the whole tree and plant a new one. Fungicides, insecticides, and so forth sometimes work but the general thinking is: why bother? Let's not get sentimental: it's just a damn tree.

We don't like to believe that kind of thinking might be going in inside the heads of our medical doctors, but consider: we have basically four kinds of treatments: surgery, drugs, radiation, and lifestyle changes.

Oh sure, sometimes surgeons reattach parts that fall off you for one reason or another. But mostly their job is to chop things off of you. Most of what drugs and radiation do is kill things that are alive or screw up some biochemical process in the body. Everything else, from exercise to diet, is a lifestyle choice, which means some politician is just aching to pass a law prohibiting you from doing it.

Changing our diet is not easy: the body demands nutrients, and if all that's available is the nutrient-poor food we're accustomed to, the body will keep us eating until we get the quantity of nutrients it needs. If we're so superstitious that we cut off the fat, as the cargo cult style of nutrition (“if it looks like fat it makes you fat”) dictates, the body rejects it: it becomes tasteless. So we pour spices on it and sear the edges of our beef to try to give it some taste, thereby creating heterocyclic amines and polycyclic aromatic hydrocarbons, which are mutagenic and, they're saying, may cause cancer.

There's a growing realization that many of our diseases are actually caused by bacteria. Some intestinal bacteria are essential, making essential things like vitamin K, while others are evil little bastards that act like tiny chemical factories, turning perfectly harmless food into a soup of carcinogens that the body can absorb. Bacteria are suspected of causing everything from migraines to multiple sclerosis. Yet there is little understanding of what all the different species of bacteria do, or how to convince them to stop doing it.

Don't get me wrong: doctors deserve praise for reducing the incidence of cancer, but our treatments still consist of surgery and radiation—slash & burn—and chemical poisons, because that's all we have. We can only think in terms of the tools that are available. Just as we had no way of understanding the function of the heart until engineers had invented the pump, it will take radical new technology before that mindset gives way.

Pharmaceutical companies know this as well. They're having a crisis because the age of cheap small molecule drugs, like acetaminophen, chlorpromazine, penicillin and zolmitriptan, is coming to an end. Nowadays, many of the new drugs are antibodies, protein molecules which are hideously expensive to produce. And the customers are not happy about that.

Nature has always used proteins to carry out the business of life. Enzymes are small molecule-sized protein-based machines that carry out various chemical reactions in the body. Antibodies are simpler: they seek out and attach themselves to some other molecule, making it nonfunctional. They work well, but they have to be laboriously grown in cultured cells, then purified and tested, all under stringent GMP conditions. And that takes plenty of dough.

But what if we could engineer our own proteins? We could build programmable protein nanomachines that could tweak your biochemistry only where it's needed. If your c-ras protein in one cell has mutated, nanomachines could come along and fix it, curing cancer one cell at a time. If they find a prion, which is a normal protein that has somehow gotten bent out of shape, they'd fold it back up properly. If your insulin receptors are screwed up, they'd fix that.

The way nature does it—by programming the peptide sequence into our DNA—is impractical. The folding problem, as it's called, is an NP-complete problem. It's so computationally difficult that today's powerful computers can only predict the structure of the smallest of proteins.

Nature solves these computations the hard way: by trial and error over millions of years. That's way too slow for us. But there's no reason we can't build proteins the way we build everything else: by attaching components together like Lego bricks. The computer industry already has a pattern we can follow: the field-programmable gate array, or FPGA.

An FPGA is a fixed matrix of configurable logic blocks. A single FPGA design can be programmed to perform virtually any function, without changing its structure, simply by changing the interconnections among the logic blocks. We can use a similar principle to design custom enzymes to do things that Nature could never have conceived of.

In the future our bodies will be full of field-programmable superenzymes, built around tiny computers. Those computers will get instructions from a network of larger diagnostic computers implanted in every organ of your body. They'll monitor your blood glucose levels, your electrolyte balance, your C-reactive protein levels, and hundreds of other parameters, make a diagnosis, and then build the requisite tools to do whatever is needed.

Like real enzymes, which often contain metal ions to help catalyze various reactions, these programmable superenzymes will have metal components. Nature is limited to abundant materials in the biosphere. But with artificial enzymes, we can use exotic materials like gallium arsenide, titanium, carbon fiber, and even carbide tools. Nature has already shown us how: bacteria swim using small rotating molecular motors. With the fusion of computer technology and biology, our artificial superenzymes will be virtual machine shops, floating around in the body waiting for a signal that something needs to be repaired.

But there's a down side. A computer that keeps you alive can just as easily kill you. To keep evolution going, your health computer will have to plant you, maybe while you are having the most fun, say in the middle of your 150th birthday party while somebody named Bambi is sitting on your lap.

That means that little computer will have to have really tight security to make sure you don't hack into its programming to make yourself immortal. If it loses its secure network link to the manufacturer, or if detects another health computer attacking it, it will start screaming to the government on its dedicated terahertz microwave band, and shortly thereafter two guys in a black SUV will show up at your front door to ask you some awkward questions.

But all your doctor will have to do is plug you in to their diagnostic computer once every six months, and do basically what they do now: type a bunch of stuff into their computer that nobody will ever read. Then they'll slap an inspection sticker on your forehead, tell you to eat more fruits and vegetables, and shove you out the door. So not much change there.

They'll tell you their job is to make sure your health computer is still working, but what they're really doing is making sure you haven't replaced it with a cheap Chinese counterfeit or reprogrammed it to tell it your schlong is supposed to be ten inches long. Their most important job is to make sure you haven't tampered with it.

So maybe what I'm really saying is that in the future, computers really will kill us all and the job of doctors will be to make sure that happens. And, ironically, we'll be happy about it.