Doctor with a syringe on the background of DNA.
Steve, a 60-year-old resident of the DC area who works in manufacturing, was always physically fit. In college, he played lacrosse in Division I, the highest level of intercollegiate athletics in the United States. Later, he stayed active by swimming, biking, and running--up until something strange happened around two years ago.
"It was hard for me to even get upstairs. I wasted away."
Steve, who requested that his last name be withheld to protect his privacy, started to notice weakness first in his toes, then his knees. On a trip to the zoo, he had trouble keeping up. Then some months later, the same thing happened on a family hike. What was supposed to be a four-mile trek up to see a waterfall ended for him at the quarter-mile mark. He turned around and struggled back to the start just as everyone else was returning from the excursion.
Alarmed, he sought out one doctor after the next, but none could diagnose him. The disabling weakness continued to creep up his legs, and by the time he got in to see a top neurologist at Johns Hopkins University last January, he was desperate for help.
"It was hard for me to even get upstairs," he recalls. "I wasted away and had lost about forty-five pounds."
The neurologist, Dr. Michael Polydefkis, finally made the correct diagnosis based on Steve's rapid progression of symptoms, a skin and nerve biopsy, and a genetic test. It turned out that Steve had a rare inherited disease called hereditary transthyretin amyloidosis. Transthyretin is a common blood protein whose normal function is to transport vitamins and hormones in the body. When patients possess certain genetic mutations in the transthyretin gene, the resulting protein can misfold, clump and produce amyloid, an aggregate of proteins, which then interferes with normal function. Many organs are affected in this disease, but most affected are the nervous system, the GI tract, and the heart.
Dr. Michael Polydefkis, Steve's neurologist at Johns Hopkins Bayview Medical Center in Baltimore, MD.
(Courtesy of Dr. Polydefkis)
For the 50,000 patients like Steve around the world, the only treatment historically has been a liver transplant—a major, risky operation. The liver makes most of the transthyretin in a person's body. So if a person who carries a genetic mutation for a disease-causing form of transthyretin has their liver transplanted, the new liver will stop making the mutant protein. A few drugs can slow, but do not stop the disease.
Since it is a genetic condition, a regular "drug" can't tackle the problem.
"For almost all of medicine from the 18th century to today, drugs have been small molecules, typically natural, some invented by humans, that bind to proteins and block their functions," explains Dr. Phillip Zamore, chair of the department of Biomedical Sciences at the University of Massachusetts Medical School. "But with most proteins (including this one), you can't imagine how that would ever happen. Because even if it stuck, there's no reason to think it would change anything. So people threw up their hands and said, 'Unless we can find a protein that is "druggable" in disease X, we can't treat it.'"
To draw a car analogy, treating a disease like Steve's with a small molecule would be like trying to shut down the entire car industry when all you can do is cut the power cord to one machine in one local factory. With few options, patients like Steve have been at a loss, facing continual deterioration and disability.
"It's more obvious how to be specific because we use the genetic code itself to design the drug."
A Radical New Approach
Luckily, Dr. Polydefkis knew of an experimental drug made by a biotech company that Dr. Zamore co-founded called Alnylam Pharmaceuticals. They were doing something completely different: silencing the chemical blueprint for protein, called RNA, rather than targeting the protein itself. In other words, shutting down all the bad factories across the whole car industry at once – without touching the good ones.
"It's more obvious how to be specific," says Dr. Zamore, "because we use the genetic code itself to design the drug."
For Steve's doctor, the new drug, called patisiran, is a game changer.
"It's the dawn of molecular medicine," says Dr. Polydefkis. "It's really a miraculous development. The ability to selectively knock down or reduce the amount of a specific protein is remarkable. I tell patients this is science fiction that is now becoming reality."
A (Very) Short History
The strategy of silencing RNA as a method of guiding drug development began in 1998. Basic research took six years before clinical testing in humans began in 2004. Just a few months ago, in November, the results of the first double-blind, placebo-controlled phase III trials were announced, testing patisiran in patients--and they surpassed expectations.
"The results were remarkably positive," says Dr. Polydefkis. "Every primary and secondary outcome measure target was met. It's the most positive trial I have ever been associated with and that I can remember in recent memory."
FDA approval is expected to come by summer, which will mark the first official sanction of a drug based on RNA inhibition (RNAi). Experts are confident that similar drugs will eventually follow for other diseases, like familial hypercholesterol, lipid disorders, and breathing disorders. Right now, these drugs must get into the liver to work, but otherwise the future treatment possibilities are wide open, according to Dr. Zamore.
"It doesn't have to be a genetic disease," he says. "In theory, it doesn't have to be just one gene, although I don't think anyone knows how many you could target at once. There is no precedent for targeting two."
Dr. Phillip Zamore, chair of the RNA Therapeutics Institute at the University of Massachusetts Medical School.
(Courtesy of Dr. Zamore)
Alnylam, the leading company in RNAi therapeutics, plans to strategically design other new drugs based on what they have learned from this first trial – "so with each successive experience, with designing and testing, you get better at making more drugs. In a way, that's never happened before...This is a lot more efficient of a way to make drugs in the future."
And unlike gene therapy, in which a patient's own genetic code is permanently altered, this approach does not cause permanent genetic changes. Patients can stop taking it like any other drug, and its effects will vanish.
How Is Steve?
Last February, Steve started on the drug. He was granted early access since it is not yet FDA-approved and is still considered experimental. Every 21 days, he has received an IV infusion that causes some minor side effects, like headaches and facial flushing.
"The good news is, since I started on the drug, I don't see any more deterioration other than my speech."
So far, it seems to be effective. He's gained back 20 pounds, and though his enunciation is still a bit slurred, he says that his neuropathy has stopped. He plans to continue the treatment for the rest of his life.
"The good news is, since I started on the drug, I don't see any more deterioration other than my speech," he says. "I think the drug is working, but would I have continued to deteriorate without the drug? I'm not really sure."
Dr. Polydefkis jumps in with a more confident response: "If you ask me, I would say 100 percent he would have kept progressing at a fairly rapid pace without the drug. When Steve says the neuropathy has stopped, that's music to my ears."
Giving robots self-awareness as they move through space - and maybe even providing them with gene-like methods for storing rules of behavior - could be important steps toward creating more intelligent machines.
Now roboticists are going a step further, training their creations to do even better: understand their own image in space and interact with humans like humans do. Today, there are already robot-teachers like KeeKo, robot-pets like Moffin, robot-babysitters like iPal, and robotic companions for the elderly like Pepper.
But even these reasonably intelligent creations still have huge limitations, some scientists think. “There are niche applications for the current generations of robots,” says professor Anthony Zador at Cold Spring Harbor Laboratory—but they are not “generalists” who can do varied tasks all on their own, as they mostly lack the abilities to improvise, make decisions based on a multitude of facts or emotions, and adjust to rapidly changing circumstances. “We don’t have general purpose robots that can interact with the world. We’re ages away from that.”
Robotic spatial self-awareness – the achievement by the team at Columbia – is an important step toward creating more intelligent machines. Hod Lipson, professor of mechanical engineering who runs the Columbia lab, says that future robots will need this ability to assist humans better. Knowing how you look and where in space your parts are, decreases the need for human oversight. It also helps the robot to detect and compensate for damage and keep up with its own wear-and-tear. And it allows robots to realize when something is wrong with them or their parts. “We want our robots to learn and continue to grow their minds and bodies on their own,” Chen says. That’s what Zador wants too—and on a much grander level. “I want a robot who can drive my car, take my dog for a walk and have a conversation with me.”
Columbia scientists have trained a robot to become aware of its own "body," so it can map the right path to touch a ball without running into an obstacle, in this case a square.
Jane Nisselson and Yinuo Qin/ Columbia Engineering
Today’s technological advances are making some of these leaps of progress possible. One of them is the so-called Deep Learning—a method that trains artificial intelligence systems to learn and use information similar to how humans do it. Described as a machine learning method based on neural network architectures with multiple layers of processing units, Deep Learning has been used to successfully teach machines to recognize images, understand speech and even write text.
Trained by Google, one of these language machine learning geniuses, BERT, can finish sentences. Another one called GPT3, designed by San Francisco-based company OpenAI, can write little stories. Yet, both of them still make funny mistakes in their linguistic exercises that even a child wouldn’t. According to a paper published by Stanford’s Center for Research on Foundational Models, BERT seems to not understand the word “not.” When asked to fill in the word after “A robin is a __” it correctly answers “bird.” But try inserting the word “not” into that sentence (“A robin is not a __”) and BERT still completes it the same way. Similarly, in one of its stories, GPT3 wrote that if you mix a spoonful of grape juice into your cranberry juice and drink the concoction, you die. It seems that robots, and artificial intelligence systems in general, are still missing some rudimentary facts of life that humans and animals grasp naturally and effortlessly.
How does one give robots a genome? Zador has an idea. We can’t really equip machines with real biological nucleotide-based genes, but we can mimic the neuronal blueprint those genes create.
It's not exactly the robots’ fault. Compared to humans, and all other organisms that have been around for thousands or millions of years, robots are very new. They are missing out on eons of evolutionary data-building. Animals and humans are born with the ability to do certain things because they are pre-wired in them. Flies know how to fly, fish knows how to swim, cats know how to meow, and babies know how to cry. Yet, flies don’t really learn to fly, fish doesn’t learn to swim, cats don’t learn to meow, and babies don’t learn to cry—they are born able to execute such behaviors because they’re preprogrammed to do so. All that happens thanks to the millions of years of evolutions wired into their respective genomes, which give rise to the brain’s neural networks responsible for these behaviors. Robots are the newbies, missing out on that trove of information, Zador argues.
A neuroscience professor who studies how brain circuitry generates various behaviors, Zador has a different approach to developing the robotic mind. Until their creators figure out a way to imbue the bots with that information, robots will remain quite limited in their abilities. Each model will only be able to do certain things it was programmed to do, but it will never go above and beyond its original code. So Zador argues that we have to start giving robots a genome.
How does one do that? Zador has an idea. We can’t really equip machines with real biological nucleotide-based genes, but we can mimic the neuronal blueprint those genes create. Genomes lay out rules for brain development. Specifically, the genome encodes blueprints for wiring up our nervous system—the details of which neurons are connected, the strength of those connections and other specs that will later hold the information learned throughout life. “Our genomes serve as blueprints for building our nervous system and these blueprints give rise to a human brain, which contains about 100 billion neurons,” Zador says.
If you think what a genome is, he explains, it is essentially a very compact and compressed form of information storage. Conceptually, genomes are similar to CliffsNotes and other study guides. When students read these short summaries, they know about what happened in a book, without actually reading that book. And that’s how we should be designing the next generation of robots if we ever want them to act like humans, Zador says. “We should give them a set of behavioral CliffsNotes, which they can then unwrap into brain-like structures.” Robots that have such brain-like structures will acquire a set of basic rules to generate basic behaviors and use them to learn more complex ones.
Currently Zador is in the process of developing algorithms that function like simple rules that generate such behaviors. “My algorithms would write these CliffsNotes, outlining how to solve a particular problem,” he explains. “And then, the neural networks will use these CliffsNotes to figure out which ones are useful and use them in their behaviors.” That’s how all living beings operate. They use the pre-programmed info from their genetics to adapt to their changing environments and learn what’s necessary to survive and thrive in these settings.
For example, a robot’s neural network could draw from CliffsNotes with “genetic” instructions for how to be aware of its own body or learn to adjust its movements. And other, different sets of CliffsNotes may imbue it with the basics of physical safety or the fundamentals of speech.
At the moment, Zador is working on algorithms that are trying to mimic neuronal blueprints for very simple organisms—such as earthworms, which have only 302 neurons and about 7000 synapses compared to the millions we have. That’s how evolution worked, too—expanding the brains from simple creatures to more complex to the Homo Sapiens. But if it took millions of years to arrive at modern humans, how long would it take scientists to forge a robot with human intelligence? That’s a billion-dollar question. Yet, Zador is optimistic. “My hypotheses is that if you can build simple organisms that can interact with the world, then the higher level functions will not be nearly as challenging as they currently are.”