Back in the old days, black swans used to take their tea with purple zebras, play hopscotch with orange kangaroos, and ride the unicorn skiff down the Cadbury-cream river with yellow porcupines. That is to say, black swans didn’t exist. At least Europeans were certain that they didn’t until a seventeenth century Dutch explorer actually discovered one on a trip to Australia. It turns out that in the land down under, black swans are not only possible, they’re common. So if you happen to see an inverted swan waddling by outside of your window, don’t panic. First off, you're probably in Australia; and secondly, you've just seen a reminder of how a single unexpected observation can topple hundreds of years worth of dogma.
I first heard the story of black swans in a talk by the Chief Scientific Officer at TEVA Pharmaceuticals, Michael Hayden, a few months ago in Jerusalem. Hayden’s topic was not actually swans, though, but rather drug discovery. He summed up his philosophy on finding new drugs by saying that people who display rare physiological attributes, either beneficial or harmful, may be the key to our next blockbuster medicines. He was referring to people who experience pain excessively and also people who don’t experience pain at all, who are extremely tall, who are resistant to deadly diseases like AIDS, or people who don’t age. Hayden called these types of people black swans because, like their namesake, they exhibit traits that we never would have expected to exist. That is, until the day we actually encountered people who have them.
Perhaps the idea of developing drugs by studying black swans seems like no big deal to you. Of course we discover new drugs by studying people with rare conditions, there’s nothing new about it, right? The answer is somewhere between “sort of” and “no.” Actually, new drugs are typically developed either by searching for molecules that can cause a beneficial-seeming effect in cells growing in a petri dish, or by testing large collections of drug-like molecules for their ability to target a protein that is thought to be important to a disease. These approaches are problematic because they often lead to new drugs that are dangerous in humans, have major unintended effects, or simply don’t work. Even a drug that perfectly targets a protein we think is important can be a failure, because targeting that protein in a human may lead to effects not encountered in the cells assayed. It is because of such an unintended effect, for example, that the hypertension drug Sildenafil became the erection drug, Viagra.
A black swan approach would be different. The key to Hayden’s philosophy is that if we are able to identify the cause of a black swan trait and to perfectly target it with a new drug, then we already know what the side effects of the drug will be from the get-go. The side effects, if there are any, should be readily apparent from observing the black swans after whom the drug was fashioned. This idea isn’t just science fiction. I mentioned before that there are certain otherwise healthy people who simply do not experience pain. A black swan approach to developing a new pain medicine would involve identifying the mutated protein that causes this rare trait (if such can be found), and then developing a drug that makes this protein in normal people act like the mutated one. A well-targeted drug would have a very low chance of being harmful, because we already have examples of people who have this condition permanently, yet who display no seemingly adverse side effects. In fact this example is real. Drug makers are now developing a new class of highly effective, non-addictive painkillers based on this no-pain trait, which may soon be changing the lives of sufferers of chronic pain.
There is another black swan named Brooke Greenberg, who may hold the secret to aging.
Brooke enjoys being swung around in the air, playing with her sisters, crawling on the floor, giggling, and getting attention. Do these activities seem strange for a twenty year old? That’s because despite being twenty, Brooke stopped developing around the age of five, and still has the mentality and physical appearance of a toddler. Her case has baffled doctor after doctor, and has led to stupefying interviews and media appearances by her family. It’s too early to tell if Brooke Greenberg actually holds any keys to immortality, but it certainly appears that in some aspects of aging she has either stalled or has almost done so, which provides hope. Taking a forever-young pill may be possible in the not-so-far future.
The key then lies in being able to figure out what’s causing Brook’s condition. If we can determine its cause, maybe we can develop drugs that can mimic it. But the thing about black swan traits is that up until recently, we haven’t had a reliable way to figure out what was causing them. We could do a battery of physiological tests, take blood samples, search for abnormal proteins or tissue functions, and write case studies, but these often just led to more head scratching or a long process of biological discovery, rather than a straightforward path to a cure.
This whole process may be on the verge of an overhaul.
In January, 2013, in an interview alongside Brooke’s family on the prime time show “Katie,” a geneticist at Mount Sinai Hospital named Eric Schadt explained that his group has sequenced Brooke’s genome, and is in the process of analyzing it for clues about aging.
Until very recently, sequencing an individual’s genome would have been either impossible or prohibitively expensive. Not so anymore. A full human genome now can be sequenced with reasonable accuracy for something around 5 thousand dollars, down from 50 thousand dollars just in 2009, and hundreds of thousands or more just a few years before that. This opens enormous new avenues for understanding black swans like Brooke Greenberg. Many of the differences between her genome and a “normal” one will be readily apparent, and as more people’s genomes are sequenced as references, the differences will stand out more and more clearly. Knowing these differences is the key to then developing cures.
Of course, knowing what’s in the genome is only a part of the whole picture. But, as Michael Hayden pointed out in his talk, there is a hidden advantage to the black swan approach over other drug discovery methods. This is that when dealing with a black swan trait, we have a very good hunch before we even start drug development what a well-targeted drug would have as its side effects.
The importance of knowing the side effects of a drug from the get-go cannot be overstated. Developing drugs is almost inconceivably expensive -- in the neighborhood of four to eleven billion dollars per drug. The majority of this cost is from the huge number of drugs that fail during clinical trials because they’re not safe, which in turn pump up the cost of the drugs that do make it. Although it may not seem to matter how much drug companies pay to develop drugs, it matters hugely -- lowered cost could be a game changer, making drugs cheaper for consumers, and enabling drug companies to focus on many diseases that are not now on their radars because they simply aren’t cost effective.
Black swan-based drug discovery may be the ticket to reducing drug failures because of side effects, but black swans can sometimes portend some pretty serious side effects, too. Brooke Greenberg is no exception. First off, it may not be so simple to isolate Brooke’s anti-aging trait from fundamental and crippling developmental abnormalities, which have kept her not in the prime of her life, but as a toddler. Secondly, Brooke has had a smorgasbord of medical complications, including stomach ulcers, seizures, a stroke, a brain tumor, and non-uniform tissue aging, which led to her esophagus closing and her needing to eat through a feeding tube. These are discouraging. Yet Brooke has also recovered from many of her complications astoundingly quickly, lest we forget that her body is harboring a great many secrets. To some of these secrets, her genome sequence may yet reveal answers.
Brooke Greenberg was born into an exciting age, in which a confluence of technology and knowledge has given a condition like hers a tremendous potential to help others. It’s yet unclear if her condition will enable us to attack aging, but genome sequencing gives us a leg up on any other time in history, and we have reason for optimism. Sequencing a person’s genome now costs on the order of Lasik eye surgery, and it’ll soon be much cheaper. The possibilities are as numerous as the traits we can see in people around us. We’ve really only just cracked open the doorway revealed by genomics, and the next leaps in medicine, in biology, in our very culture, may be incubating even now, somewhere, in the nest of a black swan.
Although I like the idea of drawing research inspiration from people who possess black swan traits, I do not expect a black swan approach to drug discovery to succeed more frequently than conventional approaches. I have doubts about the assertion that we can know the side effects of a black swan drug (a drug that mimics the biological cause of a black swan trait) by studying people who naturally possess the black swan trait.
ReplyDeleteFirst, the apparent health of people who naturally possess the black swan trait cannot be taken as assurance that the trait will manifest itself in others without harm. People who naturally possess the black swan trait will be few in number. What statistical confidence could we have about side effects? Regardless of number, people who naturally possess the black swan trait cannot constitute a random sample of a more diverse population.
Second, apparent health is not true health. We can only know about the side effects that we try to find and can observe in the time we spend looking for them. Why would a black swan approach to drug discovery be any less susceptible to the error of overlooked side effects than any other approach? We make this error when we have much stronger evidence in favor of a drug than the evidence that can be gleaned from observational studies; some drugs pass large clinical trials and make it to market before harmful side effects are discovered. Only time and mountains of empirical data can reveal the risks and rewards of a drug treatment. (That is, until we solve biology.)
Your discussion of Brooke Greenberg and a forever-young pill makes me wonder, is the ultimate goal of medicine to heal or to extend life indefinitely? Is there really any distinction? We die when a critical organ or multiple critical organs fail to function properly. If we repair a 10-year-old’s failing organ, are we healing or extending life indefinitely? What if we repair a 90-year-old’s failing organ? Can we draw a distinction by what we consider normal for a person’s age? Normal seems to change every generation. Can we draw a distinction by the cause of the organ failure? If it is caused by an infection or trauma, then repair the organ, but if the failure is caused by calcified tissue, then let the organ fail. Murky waters… I suspect that you have read and thought about this already. No need to respond, though, unless you are self-motivated to do so.
Hope all is well in TA!
Hey Andrew,
ReplyDeleteThanks for your thoughts. Actually, I agree with you that searching for black swans won’t be a magic bullet -- there could still be serious side effects, and you bring up a really good point about many drugs failing once they get tested in large enough populations. However, the game of making drug development cheaper is a game of statistics of attrition. How much can you save at each stage of the drug development pipeline? If you cut out 50% more drug candidates that are bound to fail before they get into clinical trials, you’ve saved billions. Even if black swan drugs only have a 50% higher chance of succeeding than traditional drugs, that’s literally billions of dollars saved.
The special thing about black swan drugs is that we’ve seen how the whole human organism reacts to a perfect therapy (I’m calling the actual black swan mutation the ‘perfect therapy’ here), so at least in some cases, it’s fairly clear what the side effects will be -- at least, as you aptly pointed out, for those ‘black swan’ individuals. Still, I’d rather know the full side effects of a drug over decades of a few patients’ lives than the side effects of a drug in a very large population of immortalized cells, or mice. The benefit gained from looking at an entire human, even if it’s only one or a few, is probably greater than that gained from looking across a wide variety of non-human models.
Your final point is a good one, and thought provoking. I don’t know the answer. A friend of mine asked me if I’d like to live to 200, in good health, given the option -- I said probably yes. Then she asked if I’d want to live forever. I didn’t know how to answer…
matto