Dr. Ruxandra Teslo, a genomics researcher, identifies “[T]he Biggest Obstacle to New Cancer Cures” in an op-ed for NYT.

The cost of running a clinical trial in the United States climbed steeply in the past few decades. For Phase 1 trials specifically, the cost for generating the evidence needed just to open one has doubled, on average, since 2017, and for some complex therapies, the cost can reach several million dollars per patient.

In response, drug companies are opening fewer trials than they otherwise might have, and they are increasingly taking the ones they do conduct abroad, to countries such as Australia and China. That leaves American patients with fewer chances to enroll in a clinical trial. The timing makes this especially bitter: A.I. and the genomics revolution are turning long-promised ideas into potentially curative therapies.

It is no longer the science, Dr. [David] Hong [an oncology professor at The University of Texas and an oncologist at MD Anderson Cancer Center] believes, that is holding back potential cures but our capacity to try new medicines in clinical trials. His fear, echoed by the many oncologists and medical experts I interviewed, is that the most important science may happen elsewhere or not at all. “For the first time, patients in America will no longer be among the first to benefit from the latest biomedical discoveries,” he told me.

While the deep cuts to science and education funding since the Trump 47 administration took office last January surely haven’t helped, the problem has been building for a very long time.

The damage is hard to assess: the trials that were never run, the breakthroughs that never happened. What we do know is that the search for new cures is broken. We can’t always ask for or expect medical miracles, but we can demand a system that gives doctors the best chance to save lives.

Much of the slowdown in drug development has been caused by the accumulated weight of decades of regulatory requirements, piled one atop another over years. The paperwork burden alone can be enormous: Novo Nordisk’s former chief scientific officer once noted that the documentation required to request approval for two insulin products, if printed and stacked, would rise higher than the Empire State Building.

That accumulation follows from a medical framework that treats any potential risk, however negligible, as something that must be mitigated and a reason to delay while drastically underweighing the cost of delay itself. Regulatory agencies, hospitals and drug companies all cover themselves against litigation or blame. But responsibility for mitigating the second, often much greater risk — that the disease afflicting the patient takes advantage of the delay, with potentially deadly results — belongs to no one. “We often forget that the most toxic thing for the patient is the cancer itself,” Dr. Hong said. In many cases, he can recount, down to the month, how much longer his patients are expected to live, which is why weeks lost to paperwork are not abstract to him.

The trade-off is obvious, but where one draws the line between ensuring the trials do no further harm to patients and postponing advances that might save tens of thousands is not at all clear. But surely the deadliness of the disease being studied should factor in: the former risk is less pressing if the patients are likely to die soon from the underlying condition the trial is attempting to cure.

Regardless, it seems obvious that we have put too many obstacles in the way of progress:

The shadowy nature of the drug development system helps explain, at least in part, how we arrived here. Most of what happens between a drug developer and the F.D.A. is invisible: Communications are largely confidential, and the requirements attached to clinical trials are rarely published or compared across cases. But I have spoken with dozens of the researchers who work inside this system, and their accounts point in the same direction.

The frustrations they describe stretch across every level of the system. One is redundancy. Often, the same trial plan will be reviewed by separate ethics boards (also known as institutional review boards) at different hospitals, on top of a review cycle at the F.D.A., even though the National Institutes of Health has for years urged the opposite. These reviews can sometimes drag on for months, often for purely procedural reasons, even though there’s no evidence that such long reviews increase patient safety.

And then there are the F.D.A.’s manufacturing standards, the set of rules for how a new drug must be produced. Academic clinicians repeatedly told me that these have become the largest bottleneck to getting novel cell and gene therapy treatments to patients with advanced cancers. To meet these standards, developers typically must make their material in specialized facilities, certified to commercial-scale standards.

The problem is that doing so makes little sense for early-phase trials that need just a tiny batch of medicine for what might be a handful of patients. And by substantially increasing manufacturing costs, the rules can result in clinicians offering these experimental therapies to only a small fraction of potentially eligible patients. (The F.D.A. claims to offer exemptions from full manufacturing requirements for Phase 1 trials, but the exemptions are so vague that medical researchers tend not to take advantage of them.)

The results of all this are what one might expect:

Slower and costlier trials are one reason the number of new drugs approved per billion dollars of research and development spending has roughly halved every nine years since the 1950s, with an apparent plateau in the past 10 years. Researchers have a name for this phenomenon: Eroom’s Law. It’s a deliberate inversion of Moore’s Law, the semiconductor industry’s principle of exponential progress. Whereas Moore’s Law tracks a world of compounding gains, Eroom’s Law shows the opposite: The more we spend on drug discovery, the less we get.

There’s a whole lot more to the essay, but that’s the gist.