Francis Collins was fresh off one of the most important scientific advances of all time—decoding a rough draft of the entire human genome—when he first met 4-year-old Sam Berns.

Sam did not look like a typical preschooler. His hair was falling out. He wasn't gaining weight. He was beginning to look like an old man. But as the two played catch in Collins's backyard, "it was clear that this was a wonderfully precocious little boy," Collins says.

The symptoms had shown up in the first few months of Sam's life, when his growth essentially stopped and his bones began to deteriorate. His parents tried doctor after doctor, test after test. Finally, they hit on Sam's diagnosis: progeria, a disease so rare that few of his doctors had seen it outside of a textbook. It affects only one in every 4 million to 8 million children.

The diagnosis gave Sam's parents an answer but not much hope. There was no effective treatment for progeria, never mind a cure. No one knew what caused it. Hardly anyone had spent much time studying the disease. Those who had couldn't crack it. The only certainty was a grim prognosis: Children with progeria usually died around age 12 or 13, often from a stroke or heart disease.

Collins had done some research into progeria early in his career but had been frustrated by his inability to make much progress. "I thought to myself, 'Maybe someday, something will be learned about this' ... but I didn't know where to go with it," he recalls.

More than 15 years later, Collins met Sam's father, Scott Berns, at a cocktail party in Washington, where Berns had landed a prestigious White House fellowship. Berns was surprised that Collins had even heard of his son's disease. Berns and Collins started talking about what they could do to jump-start progeria research. A few months later, Sam's parents brought the boy along when they visited Collins's house to talk about setting up a scientific workshop. And the closer Collins grew to the energetic, determined child, the more invested he became in the search for a cure.

So invested, in fact, that when a new researcher came to work in Collins's lab, in need of a project and willing to try something risky, they decided to take on the disease. "With full knowledge that this probably was not going to work," Collins says, they would give it a year and see whether they could come up with something to help Sam.

They found it: the single genetic mutation that causes progeria. "In those 3 billion letters in the human DNA instruction book, this is one letter that's misspelled," Collins explains. The breakthrough was staggering and, relatively speaking, it happened virtually overnight. And just six years after Collins and his researcher identified the progeria gene, they had a proven therapy that adds about five years to patients' lives. Another drug is in development now.

None of that softened the blow when, this past January, Sam died. He was 17. "I would not imagine being able to stand at his graveside without crying," says Collins, now director of the National Institutes of Health. "It's such a sense of personal loss."

The doctors and scientists hunting for new cures and treatments work in a constant state of tension. They operate in a tremendously high-stakes environment, pouring years of their lives into research as the people who inspire them continue to suffer and even die. Drug hunters face failure after failure, almost never followed by success. Decades of work flame out. Promising ideas turn into dead ends. For every 10,000 compounds they explore, scientists wind up with just one drug approved by the Food and Drug Administration. Even when medical science moves as fast as it can—and today, it's moving faster than ever before—it's still an agonizingly slow process.

"As much as we say that failure is part of what we do—if you're not failing, you're probably not doing science that's very interesting—it still hurts," Collins says. "It is frustrating, because you want to come up with the answer. You want to save lives. That's what we all get into this medical research area to try to achieve, and yet the challenges are immense. And we make progress, oftentimes, in very small baby steps, even though what we're hoping for are big leaps."

When incremental advances do happen, they're almost never predictable. Collins's success with progeria would not have been possible without his 13 years of slow, steady work to decode the human genome. It also might not have happened if he had skipped that cocktail party. Collins was in a position to help Sam largely because he had worked on progeria earlier in his career. But once he discovered the underlying cause of the disease, he realized that his earlier patient—the one whose case familiarized Collins with the disease—hadn't had progeria after all. And the drug that ultimately added years to Sam's life was a product that had previously been written off as ineffective. It had been developed as a cancer drug but, Collins says, "didn't work worth a darn for cancer."

All of those contradictions, surprises, twists, and turns add up to a quest that can be maddening but also deeply affirming and rewarding for the people who pursue it. "You have to be a special person to be a drug hunter," says Stevin Zorn, who leads drug development at the Lundbeck pharmaceutical company. "The percentage of ideas that actually make it through to success are small, but if there is no idea, there's no cure."

There are some 7,000 known diseases in the world. Only about 500 have treatments, and even fewer have cures. There are viruses like Ebola, which are devastating and difficult to manage even though scientists understand perfectly well what causes them. There are ticking time bombs like cancer and Alzheimer's, the causes of which are just beginning to be understood. And there are scores of rare diseases, such as progeria—which science has largely ignored.

Scientists are moving as fast as they can to close those gaps. Buoyed by unprecedented technological advances over the past two decades, today's researchers are not satisfied with simply treating symptoms. They want to find the root cause of each disease and fix it. They want cures.

* * *

Zorn is among the scientists taking on the biggest and most daunting challenge of them all: the brain.

Zorn initially came to medical research simply because he was good at science, but personal tragedies pushed him toward the brain. His best friend from high school suffered a psychotic break during his second year of college and was diagnosed with schizophrenia. Medications didn't work; neither did a treatment center. His friend's depression and delusions ended in suicide.

Then, shortly after Zorn became engaged, his fiancée's mother was diagnosed with Alzheimer's. "She deteriorated quickly," Zorn says, from forgetting small things to forgetting her daughter. He and his wife cared for her as long as they could, through late nights where she would empty their closets or barge into their bedroom. After she began wandering away from their house, assisted living was the only option. "It was a heart-wrenching period, for close to 10 years, where we watched my mother-in-law go from being a lovely person to an empty shell."

Now Zorn is among the small army of researchers trying to unlock cures, or at least decent treatments, for Alzheimer's and other brain disorders.

No one knows exactly what Alzheimer's is, or why, precisely, some people's brains betray them. The best hypothesis revolves around a protein called amyloid, which shows up in large plaques on the brains of people affected by Alzheimer's, similar to the way plaques form on teeth and in arteries.

Nearly fifteen years ago, researchers at pharmaceutical giant Eli Lilly started working on a drug that would attach itself to amyloid particles and prevent them from binding together into plaques. After screening thousands of compounds, they found one that worked in a petri dish. So they engineered some mice to develop amyloid plaques. The drug worked in the mice, too, and at that point, Lilly's scientists let themselves get their hopes up.

"Those were some amazingly exciting times," says Ron DeMattos, Lilly's chief scientific officer for neurobiologics. "You never could come up with a test that proved us wrong."

They were able to develop a drug that wasn't toxic to humans and could actually be absorbed into the brain—a difficult barrier to overcome, because our brains are excellent at keeping out foreign material.

Along the way, a new genetic mutation was discovered in a small group of people in Iceland who lived long lives and seemed to be shielded from Alzheimer's. The finding made researchers even more confident that amyloid was the right target. To know whether their drug was reducing the amount of amyloid in the brain, though, they had to invent a way to measure the protein. Previously, they could only measure amyloid during autopsies. By 2012, Lilly's scientists had cleared all of those hurdles and were wrapping up Phase 3 clinical trials—the last phase before asking the FDA for approval.

The drug failed. Most people in the trial, which had taken more than two years to conduct, showed no significant improvement.

"I was sitting there listening through the phone, and just hearing the response, and the sheer emotion that came out of what a lot of people term a failure," DeMattos remembers. But he cites that moment as among the best of his career. "In reality, this was one of the very first times that we, and myself as a scientist, thought, 'My gosh, we're really gonna get this beast.' That was just so exhilarating, I can't wait to feel that again."

Although failure is frustrating and expensive, drug hunters say the key to progress is to fail well. A well-designed study that's conducted properly should teach you something, even if the outcome isn't what you wanted. "Some of the greatest insights I've ever had were failures—and some pretty big failures," DeMattos says. "We're going to fail thousands of times."

In this case, the lesson was to start earlier. When Lilly's scientists dug a little deeper into their data, they realized that their drug had produced a decent benefit for trial participants who had "mild" Alzheimer's. Every time they cut the data a different way, they remained convinced they were seeing real progress earlier in the disease.

But failure is still disheartening, especially when it comes so late in the game. Another Alzheimer's drug also failed in Phase 3 trials in 2012, around the same time as Lilly's, and the dual setbacks were a blow both to advocacy groups and to the scientists themselves. "The first thing is real disappointment," says Husseini Manji, the global head of neuroscience at Janssen Research & Development, which conducted the second Alzheimer's trial. "You can imagine, people working on this project, they've given their life, so to speak, working on it. And of course it was going to be disappointing. You want to ... make sure that they weren't overly demoralized. To some degree it's completely understandable."

"This is hardly a 9-to-5 job," Manji says. "I want people when they're in the shower to be thinking about novel ways—all day and night, to be thinking about novel strategies."

* * *

When he's not working on a vaccine for malaria or Ebola, Rip Ballou is the guitarist (recently "forced into a vocal role") in a band called Dokter Rokter.

The bandmates are connected to Glaxo-SmithKline, where Ballou leads clinical development from a lab in Brussels. They practice once a week and play gigs around the city, all in an attempt to find some space away from the emotional, all-consuming nature of their current effort—to find the product that can halt Ebola's rampage.

"It's not everybody that can do vaccine development. And there are a lot of people who get out of it because it takes too long, or it's too frustrating, or they don't have the 20 years of patience," Ballou says. "I've come close to it at times."

When Ballou first decided to focus on infectious diseases, he was working with soldiers at the Walter Reed Army Medical Center who had fallen prey to an especially nasty parasite while training in Panama for jungle warfare. The work required him, for the first time, to move from the bedside into a laboratory. And he hated it.

"After about a year and a half of killing mice, I went to my boss and said, 'I just can't do this anymore. I spent eight years learning how to be a very good doctor. I want to take care of patients, I just cannot kill any more mice. I'm going back to the hospital,' " Ballou recalls.

His boss offered him another option: Switch over to malaria, where vaccine development was still the priority but testing happened in the field, with real people. He agreed, and the ability to stay close to his patients has kept Ballou in vaccine development for 30 years. "That was a big turning point in my career. I could have easily walked out the door, and today I'd be a cardiologist or something," he says.

Plenty of young drug researchers have been in Ballou's shoes. One success, even an incremental one, can carry veteran researchers through strings of failures. But new scientists have to jump right into the letdowns, learning the hard way what it feels like to fail over and over while they wait years for that first breakthrough.

"After a while, after you've been through that experience a dozen times, you can kind of see your way past it. I gotta say, the first few times, when you're a junior researcher, you feel like you have so much riding on whether your hypothesis is right," Collins says. "You get this all connected up with your own sense of self-worth. I remember those times where I just thought, I don't know how I'm going to pick myself up from this. It was just like such a personal sense of failure."

Most people who get into drug discovery are able to get over that hump sooner or later. These researchers say good management helps—things like starting a company band to help preserve (or enforce) work-life balance, or ensuring that scientists receive recognition for their efforts after a failed trial. Staying close to patients also helps. And to some degree, people with the right temperament "self-select" for drug-development careers, Ballou says. Patience is the most important virtue.

Ballou worked through roughly 15 years, and 12 failed drugs, before he hit on a malaria vaccine that showed promise. He even infected himself with the disease, becoming one of his own research subjects so he could feel the pain of the patients he's trying to help.

His experience with malaria carried over naturally, he says, when NIH came calling, seeking a large pharmaceutical company to help it quickly develop an Ebola vaccine.

"The pace of this is just extraordinary," Ballou says. "Six months ago, nobody was prepared for this, and so there weren't resources allocated to it. ... If a vaccine is going to be part of the solution, we literally have months to be able to bring it to the field. And there are so many moving pieces on this."

The first round of clinical trials should wrap up soon for the drug that GlaxoSmithKline is helping to develop, and the initial animal tests were encouraging. But the animal tests were also encouraging during NIH's first try at an Ebola vaccine, which began human trials in 2003 but ultimately didn't work. "The concept was beautiful, but when you actually put it in humans, there was a problem," Ballou says.

The pressure on drug hunters is generally high, but it really mounts in the middle of a deadly epidemic, when the public demands answers that science simply doesn't have.

* * *

Anthony Fauci still remembers the day—June 5, 1981—when he saw the first reports about five gay men in Los Angeles who had a virus doctors couldn't understand. A month later, there were 26 more, and the cases had spread to San Francisco and New York City.

"It became clear to me that we were dealing with a new disease. I didn't know what it was. I didn't know what caused it. But it certainly was something that I, as an infectious-disease person and as a researcher, had never seen anything like it before," says Fauci, who took over NIH's National Institute of Allergy and Infectious Diseases in 1984, while the HIV/AIDS epidemic was still mounting.

On top of the scientific challenges he faced, Fauci and officials from the FDA and the Centers for Disease Control became the public face of the government response. He was a personal target of activists, who were angry at how quickly the disease was spreading and how slowly the government seemed to be moving. In 1990, roughly 1,000 protesters stormed the agency, carrying coffins and demanding faster access to new treatments. Larry Kramer, a leading AIDS activist, wrote an open letter to "the incompetent idiot Anthony Fauci" in the San Francisco Examiner.

"From 1981 through literally all of the '80s, almost all of my patients died," Fauci recalls. "Literally, it was the most dark era of my life, professionally and personally."

Yet, in retrospect, it's stunning how quickly researchers achieved progress. A typical patient who walked into a clinic in 1981 would have had six to eight months to live. Today, the same patient could have another 50 years. At the beginning, though, the challenge was among the most daunting modern medicine had encountered.

What Fauci calls the first "glimmer of hope" came in 1987, when patients began to get better with a dose of the drug AZT. But the results didn't last. The virus was becoming immune to AZT. "They did well for a limited period of time, and then they started to do poorly again, which gave us the big insight that you probably are going to need not a single drug against HIV," Fauci says.

He and his team, in partnership with the pharmaceutical industry, spent roughly the next decade trying to find a combination of drugs that would work. They scoured the virus for what Fauci calls "vulnerabilities"—physical properties in the virus itself or moments in its life cycle where it would be especially susceptible to drugs. They took on one vulnerability after another, and by 1996 they had developed a combination of three drugs that showed a "phenomenal, transformational effect," Fauci says. The next year, the number of AIDS deaths in the United States declined for the first time. Now, the disease's growth is slowing worldwide, and in the U.S., it has become a manageable chronic condition, rather than a death sentence.

The fight isn't over, of course. The current cocktail of drugs can suppress the virus to undetectable levels, but it can't eliminate the infection. As soon as patients quit taking their medicine, the virus comes roaring back. " 'Cure' is still problematic. I'm not so sure we're going to be able to cure HIV," Fauci says. Work on a vaccine has also hit a wall.

Even so, the HIV/AIDS experience has not only helped AIDS patients live full lives, it has also transformed drug development. And it has changed the way the government rolls out potential remedies in the middle of a crisis, too. It led, specifically, to the use of a "parallel track" in drug development that allows patients to take experimental medicines while clinical trials are going on.

That was, in the 1980s and 1990s, a radical idea, and one that scared many scientists, particularly FDA researchers whose mission had always been to rigorously evaluate new drugs in tightly controlled studies. Now, though, the parallel track is thoroughly ingrained in medicine. It may well be saving Ebola patients' lives: Almost all of the Ebola patients treated in the U.S. have received one experimental drug or another, and lived. That doesn't mean the drugs work, as Fauci is careful to emphasize, but NIH has not hesitated to let dying patients take drugs that haven't been fully tested, and it has even asked drug companies to jump in and help scale up production.

* * *

Science is still losing its fight with Ebola, which is killing more than half the people it infects in West Africa. Researchers haven't landed on the solution to problems like Alzheimer's, despite spending billions of dollars a year trying. There are certainly no cures for the host of diseases such as progeria that are so rare that the cost of drug development might never be offset by pharmaceutical sales.

And the challenges standing in the way of the next breakthrough are no longer just the difficulty of the scientific questions at play and the stresses of the job. Researchers must now look for funding and opportunities in a low-budget environment.

Harold Varmus, the director of NIH's National Cancer Institute, was awarded a Nobel Prize for research that helped prove that cancer is caused by genetic mutations rather than a foreign molecule. He was also NIH director in the 1990s, when the agency's budget doubled. That increase in funding had ripple effects: NIH gave out more grants, so research universities grew, and more students got doctoral degrees. Thanks to that boost, new technologies were developed, which opened up new opportunities for scientists to learn even more.

Today, Varmus says, the foundational advances in our understanding of cancer are coming faster than science can translate them into therapies—and that's largely because of a lack of money. Budget cuts and flat funding have reduced NIH's purchasing power by about 20 percent over the past decade, and the agency took a $1.5 billion hit from sequestration.

"Now, we certainly have a situation where we have too many people trying to do work with all these new ideas with less money than was available over a decade ago—substantially less. If you look at adjusted dollars, the loss is about a quarter of our resources," Varmus says. "We're at a point at which cancer research is going very well. It's going very well in this country. But there are danger signals on the horizon here."

Those investments, public or private, are important because progress builds on progress. A tool developed in the 1970s unlocked an effective AIDS treatment, which unlocked a new way to create safer drugs for other infections. The discovery of the structure of DNA, in 1953, paved the way for decoding the human genome, which has now paved the way for an ambitious new effort to map the human brain.

"It really is sort of a blend of engineering and art, where creative things have to flow out, and you have to try it out on the canvas. And then, like, 'Oh, God, that's ugly. Try another thing,' " says Sasha Kamb, senior vice president of research at the biopharmaceutical company Amgen. "We don't have a blueprint for what we're trying to make."

That was true of the genome project, which launched with a 15-year time line that Collins now says was "basically made up." Most of the genome map's potential hasn't yet been tapped, but Collins says it has already helped scientists identify the underlying causes of about 5,000 of the 7,000 known diseases. In most cases, they're still a long way from turning that knowledge into treatments or cures, but they've taken the first step.

The next challenge is mapping the brain. Ultimately, it will cost billions of dollars, and Collins says it's impossible right now to know what being finished might even look like. But in the quest for new cures, he says, the most important thing is to keep going—to keep trying, failing, recalibrating, inching forward, and sometimes being surprised.

"I tell my postdocs, you know, in my lab, if you're doing a really good job, it's possible that 1 percent of your experiments might actually contribute new knowledge. If you can boost that to 2 percent, you're going to be a hero," Collins says. "Those moments, those you file away, and when you're having a tough experience with the latest failure, you kind of remember, 'Oh, yeah, it is possible, once in a while, to make progress.'"