Spoiler alert: I give away the ending.
My husband and I took our kids to see Coyote vs. Acme this weekend. I went in expecting a new storyline from a cartoon I loved as a kid. I came out thinking about drug development, which I realize is not the normal reaction to ninety minutes of Looney Tunes.
We went to one of those theaters with reclining seats and allegedly fresh popcorn delivered to your chair. Our boys loved the movie. So did I. But somewhere along the way, my mind wandered.
The story behind the film is fairly well known at this point. Warner Bros. finished it in 2023 and then decided not to release it—not because it tested badly, but because the studio concluded that taking a tax write-off made more financial sense than putting a movie it had already spent tens of millions making into theaters. So the film went on a shelf. Three years later, another company bought it and released it.
The movie wasn’t the problem. The math was.
And that’s when I started thinking about healthcare.
We hear versions of this all the time: somewhere there are shelves full of treatments and cures being hidden from us. There aren’t. Drug companies are not warehouses stacked with cures they refuse to sell. Most experimental drugs that disappear from development disappear for the most mundane and important reason imaginable: they don’t work.
Biology is hard. Clinical trials fail. A molecule that looks brilliant in a lab can turn out to be useless in a human being. Anyone who tells you the cure for your disease is sitting in a freezer because a company doesn’t want you to have it is leaving a lot of the story on the table.
But healthcare does have shelves.
And some things land there even when the science hasn’t failed.
A therapy that was donated rather than discarded
Methylmalonic acidemia is a rare inherited metabolic disease. Children with it can’t properly break down certain proteins and fats, so toxic byproducts accumulate in the body. It brings metabolic crises, kidney disease, neurological damage, and for some children, eventually a liver or kidney transplant. There is no approved treatment that addresses the underlying cause.
For years, a company had been developing a gene therapy for it, designed to deliver a working copy of the defective gene. The program got orphan drug and rare pediatric disease designations. It cleared regulatory review. It reached the point of screening patients for a first-in-human trial.
Then, in 2023, the company stopped.
The therapy hadn’t failed. There were no disappointing results, because no patient had ever received it. The company discontinued the program for commercial reasons and withdrew its application to test the therapy in people.
That could easily have been the end of it. Instead, the company donated what it had — the manufactured therapy, the reagents, the regulatory materials — to the National Institutes of Health, where researchers had been partners on the work from the beginning. Investigators at the National Human Genome Research Institute and the National Center for Advancing Translational Sciences picked the program up, did the additional development work, redesigned the clinical plan, and went back to the FDA.
On May 8 of this year, the FDA cleared the therapy to begin a Phase I/II trial at the NIH Clinical Center.
It is not a cure that somebody buried. It is not even a proven treatment. It is an experimental therapy about to be tested in a small number of children for the first time, and it may well fail, as experimental therapies routinely do.
That’s the part I can’t get past. The scientific question almost never got asked because the commercial question had already been answered.
For one owner, continuing no longer made economic sense. For another, the unanswered question itself was reason enough to keep going.
The biology hadn’t changed. The math had.
Everybody can sell it. Nobody owns the evidence.
That story is one version of the problem. There’s a larger and less dramatic one.
Medicines that are already approved, already manufactured, already prescribed to thousands of people are sometimes found to have promise in diseases entirely unrelated to the ones they were developed for. This is called drug repurposing, and on paper it sounds almost absurdly sensible. If we already understand a drug’s chemistry, its manufacturing, its behavior in the human body, and a great deal about its safety, why wouldn’t we investigate every condition where credible evidence suggests it might help?
Sometimes we do. Often we don’t, and the reason is usually economic rather than scientific.
Once a medicine loses patent protection and inexpensive generics arrive, the question of who will spend millions of dollars proving it treats another disease becomes genuinely hard to answer. The original manufacturer may no longer have enough upside to justify the expense. Generic manufacturers compete on price and are not in the business of financing clinical development. Everybody can sell the medicine, which can mean nobody owns the evidence.
Nineteen years
Alkaptonuria is an extraordinarily rare inherited disorder — something on the order of one in a hundred thousand to one in a quarter million people. Patients accumulate a substance called homogentisic acid over the course of their lives, and it slowly damages connective tissue, producing severe arthritis, joint replacements, and complications in the heart and kidneys. For most of medical history there was nothing to offer beyond managing the damage.
But there was an existing medicine, already in use for a different rare metabolic disorder, that researchers found could dramatically reduce the substance that accumulates in alkaptonuria. The molecule wasn’t new. The mechanism wasn’t new. The hard part was proving what a known drug could do for a different group of patients.
The pivotal study was run inside the NIH’s intramural program. Forty patients. Three years of treatment against no treatment. It enrolled between April 2005 and March 2006, and it found improvements in pain, energy, and physical functioning alongside a greater than ninety-five percent reduction in the acid driving the disease.
It worked. And eventually a company willing to carry the application across the finish line took it up. The FDA approved it for alkaptonuria in June 2025.
But look at the timeline.
Nineteen years.
Academic researchers, patient advocates, publicly funded investigators and eventually a commercial sponsor handed the work from one to another because no single party had both the capability and the incentive to carry it alone.
And even the win came narrow. The approved indication covers the reduction of urinary homogentisic acid in adults. Children with the same disease, whose joints have not yet been destroyed and who arguably have the most to gain from starting early, remain outside the label.
We built a remarkable system for rewarding the person who invents a medicine. We never really built one for the person who figures out what else that medicine can do.
The space between what medicine knows and what the system recognizes
There’s a third shelf, and patients live on it in much larger numbers.
Some therapies are used routinely, off-label, for conditions their labels don’t mention — supported by published evidence, endorsed by specialists, reflected in professional guidelines. Off-label doesn’t mean experimental or unsupported. But an FDA-approved label, an insurer’s coverage policy, and a physician’s medical judgment are three different things, and they don’t always point the same direction. A patient whose doctor believes a treatment is appropriate, and for whom evidence exists, and whose diagnosis simply doesn’t satisfy a payer’s criteria, ends up in the gap between them.
Which is the same question again, wearing different clothes. If a therapy is already on the market from multiple manufacturers, who has the incentive to fund the studies that would formalize every use for which it might be appropriate? Sometimes someone does. Often nobody does. And the patient waits in the space between what clinical medicine has learned and what our regulatory and payment systems formally acknowledge.
We talk about medical innovation as though it follows a sequence: discover something, test it, get it approved, get it covered, give it to patients. Anyone who has spent time around sick people knows it rarely works that neatly.
Sometimes a program loses its sponsor before anyone even gets the chance to find out whether it works. Sometimes doctors learn that an old generic may do something useful and nobody can figure out who should pay to prove it. Sometimes the drug works, but the approved dose is miserable to live with. And sometimes physicians are already using a therapy because the evidence is good enough for them, while the insurer is still staring at the label.
These aren’t really the same problem. But they keep leaving patients in the same place: waiting for somebody to own the next step.
Somebody at the FDA is asking too
In May, the agency opened a public docket called Drug Repurposing for Unmet Medical Needs. It’s a request for information, which is bureaucratic language for the government admitting it doesn’t know something and asking the public to tell it. What the FDA wanted to know was which approved drugs might already have enough scientific evidence behind them to support a new use — a new indication, or a new population — that nobody is pursuing.
The notice is worth reading in the original, because of one phrase in it. The agency said it was focused on approved drugs for which “there appears to be no commercial interest in adding a new use.” Rare disease was named among the priority areas, along with metabolic and neurodegenerative conditions. Announcing it, the commissioner said that too many patients lack effective treatment options even when promising science already exists.
This highlights that it’s not always about a shortage of science, but rather, a shortage of anyone with a reason to act on the science we have.
The response tells you something too. The comment period was supposed to run thirty days and had to be extended another thirty. More than five hundred submissions came in — patient organizations, clinicians, researchers, and drug companies among them. People had answers. They had been waiting for someone to ask.
The docket closed on July 13.
The comments are public. Anyone can go read hundreds of specific, sourced accounts of medicines that might do more than their labels allow, submitted by people who know exactly what study would settle the question.
Asking is not the same as being able to do anything about the answer.
The FDA can approve a new use when someone brings it an application. What it cannot do is run the trial that produces the application. Somebody outside the agency has to design that study, fund it, enroll the patients, and file. And if the reason a use went unpursued in the first place is that no company had a commercial reason to pursue it, then collecting five hundred descriptions of the problem does not by itself change the economics that created it.
So we now have an unusually good inventory of what is sitting on the shelf.
What we don’t have is anyone whose job it is to take things off it.
Back to the coyote
Since 1949 we have treated Wile E. Coyote as the joke. He orders another product from Acme, it malfunctions spectacularly, a rocket sends him into a canyon or an anvil finds his head, and he gets up and orders again. Watch it long enough and you start to wonder whether the fool in the arrangement is really the customer.
The film gives him something he never had in the cartoons, which is recourse. Wile E. doesn’t sue Acme out of some civic awakening about product liability. He sues because he is still, after all this time, trying to catch the bird, and the proceeding is the one place he can be certain the Road Runner will turn up. The lawsuit is a tactic. The bird is the point.
And it doesn’t end in a verdict. It ends in front of a Senate committee that considers the whole matter and concludes it doesn’t have a role here. The two sides shake hands, call a truce, and go cheerfully back to their war. Nobody is found responsible. Nothing is ordered. The institution with the power to look decides that looking isn’t its job, and everyone goes home.
I laughed, and then I sat there a second longer than the joke required.
Because that’s the ending, and then there’s the other ending. Wile E. goes back to the desert. Not because he won anything, and not because he was denied and is starting over. He goes back because the chase is what he has, and the last thing on the screen is a question.
Maybe tomorrow?
He will chase that bird forever. Not out of stupidity. Because the pursuit is the thing he loves, and he has always known that, even when we thought we were laughing at him for not knowing it.
What happens across that film is that two cartoon characters are gradually overtaken by their humanity. The coyote turns out not to be a fool, and the grievance turns out not to be the real subject. The real subject is what you do with a pursuit that may never resolve, and whether you can love it anyway.
How many researchers and scientists are the coyote?
The one who spends a career on a disease that affects a few thousand people. The parent who learns genetics at two in the morning because nobody else is going to explain it to them. The NIH investigator who picks up a therapy a company walked away from and starts over. The advocate who files the comment, and files it again the next year, and the year after that.
None of them are promised a verdict. Most of them keep going anyway.
That is not a small thing, and I don’t think we should ask people to sustain it on love alone.
Which brings me to what I actually want.
We already built the beginning of it. The National Center for Advancing Translational Sciences (NCATS) was created in 2011 for precisely the gap I’ve been describing — the space between a scientific finding and something a patient can use. It is the reason a gene therapy for a rare metabolic disease didn’t disappear when its sponsor walked away. That happened because a public institution existed with a mandate to care about the unanswered question after the commercial question had been settled.
Here is how seriously we take it.
This year’s budget request proposed dissolving it and folding it into another institute. Congress declined, and in February funded it separately at roughly $942 million. But of that, more than $629 million is directed to a clinical trials network, leaving something in the neighborhood of $237 million for everything else — the abandoned programs, the rare disease therapeutics, the repurposing work, every unowned question across every condition.
And then there is the last $75 million, which is the part that actually undid me.
That line funds the Cures Acceleration Network. Congress created it in 2010 and gave it something unusual: the authority to make flexible awards outside the normal grant and contract rules. It is the same instrument DARPA uses, the one it used to fund mRNA vaccine technology back when the field thought that was a long shot. The tool for paying somebody to go answer a question nobody has a commercial reason to answer already exists, with real legal authority behind it, sitting inside NCATS.
It was authorized at $500 million. It is funded at $75 million.
We built the thing and never turned it on.
This is also, I think, the answer to a fair objection. When people hear “national institute,” they picture a slow federal laboratory doing the work itself, and they are right to be skeptical of that. But look at how we actually got commercial spaceflight. NASA did not build those rockets. It became a customer — it said what it wanted delivered, committed to pay on delivery, and let anyone compete to build it. The companies raised private money against a guaranteed buyer. Without someone willing to purchase the outcome, none of them exist.
That is precisely what is missing here. There is no shortage of people who would run these studies. University groups, small companies, disease foundations, and the investigators already doing this work would take it on tomorrow. What is missing is anyone who will pay for the answer, because the moment the evidence exists its value spreads to everyone at once and belongs to no one.
So I am not asking for a government laboratory. I am asking for a buyer. Post the unowned questions, pay on delivered evidence, and let whoever can answer them compete to do it.
What I would like to see is not a new agency layered onto the ones we have, and not a competitor to private drug development, which we need and should keep rewarding generously. Something simpler: take the center we already have, give it the standing of an institute, fund the authority we already granted it, and let it buy the answers nobody else has reason to produce. Call it the National Institute of Innovation, because that is what the work actually is.
A permanent home for the things no one owns. The second indication for the generic. The pediatric formulation. The dosing schedule that makes a hard drug livable. The combination spanning two companies’ patents. The abandoned program with the reagents still in the freezer and a scientist willing to pick it up.
A place whose entire job is finishing.
We know how to reward invention. We have never built a serious institution for completion, and completion is where an enormous number of patients are actually stuck.
How many other useful ideas are still waiting?
We should be extraordinarily good at inventing new medicines, and mostly we are. We need those breakthroughs, and we need a system that keeps rewarding the enormous risk of chasing them, because most of the time nobody finds anything and somebody has to be willing to pay for the trying anyway. I don’t want less of that. I want more of it.
But the next frontier may not only be what can we invent. It may also be what have we already invented that we haven’t finished figuring out. Because sometimes the science isn’t the thing sitting in the way. Sometimes, the way it was for a finished movie in a vault, it’s the math.
Wile E. Coyote gets to end on maybe tomorrow, and it works, because the chase is his and he loves it and the anvil never really kills him. Patients say the same two words and mean something entirely different by them. Maybe tomorrow somebody funds the study. Maybe tomorrow the label changes to include my kid. Maybe tomorrow is thirty more years of a life.
They shouldn’t have to spend that life falling off cliffs while we wait for someone to own the last mile.
One more thing, before I forget: go see it. It’s terrific, my kids loved it, and after everything it took to get onto a screen, it has more than earned the ticket.
