The drugs killing dying patients are the drugs they can't get.

The FDA is regulating patients to death.

My husband Jake is still alive today because of clinical trials. Standard-of-care therapies for recurrent metastatic head and neck squamous cell carcinoma (R/M HNSCC) would have by now let his cancer kill him. But the process of keeping of him alive was and still is excruciating, and it requires a level of resourcefulness and agenticness most people don’t possess and can’t develop (I detail the Escher-meets-Kafka-like process in “Please be dying, but not too quickly”).

Though Jake and I have become inadvertent clinical-trial system experts, the skills we’ve learned, and that have kept him alive between July 2023 and today, will soon be obviated by his previous participation in the system, since the system prefers patients who have received less treatment, not more. Confused? I don’t blame you: so was I, when I learned in detail about how clinical trials work. Jake pretty much can’t keep hopping from clinical trial to clinical trial because, after his tumors learn how to outsmart SGN-PDL1V—the current experimental medication he’s on—he’ll graduate from his third to his fourth “line of therapy.”¹

The ”lines of treatment” issue is significant because most of the more promising clinical trials have restrictions on how many previous treatments a given patient can have. PDL1V, for example, requires that patients only have had two previous lines of treatment. Jake received pembrolizumab and chemotherapy in the summer of 2023, and then a clinical trial drug called Petosemtamab / MCLA-158 from Sept. 2023 – March 2024—which means he had two prior lines of treatment. If he had participated in some other trial first, he wouldn’t be eligible for the PDL1V that’s been so effective for him so far. The drugmakers who sponsor trials prefer to enroll trial patients early in the course of their disease—usually healthier patients, by metrics of the dying—because sicker patients have more problems, which confounds data and side effect profiles, and they’re statistically less likely to respond to the drug, since they’ve already failed so many others. Less of a response means worse data, which means a lower likelihood of a drug getting approved by the FDA. Therefore, drugmakers want patients who have had fewer treatments, not more. Each new treatment narrows the window of opportunity for Jake. I watch that window narrowing and know that, when it closes, his own light will be extinguished.

Remember that the main point of clinical trials isn’t to help the patients in the trial, but to get the FDA to approve a drug. The FDA can behave arbitrarily without recourse. Trial design deeply matters when it comes to what the drug will be approved for. Running a trial on patients receiving the drug as a first or second line therapy means drug companies can apply to the FDA for on-label use as a first or second line treatment, which increases the likelihood the drug may eventually replace the current standard-of-care regimen, which will lead to insurance companies paying for patients to get the drug earlier in their disease course, leading to significantly more patients getting the drug and more money for drug companies.

Research and development is expensive. Individual prescription drugs can be billion-dollar businesses. What do patients considered undesirable for clinical trials do? What do patients like Jake do?

Patients who are ineligible for trials (along with those who are waitlisted for trial slots, or don’t live near a relevant trial, or who don’t have the time, money, or capacity to travel continuously for care), just have to wait for these promising new drugs to become available on the market after a New Drug Application is passed through the FDA—a process which takes, on average, 10 – 14 years. Realistically, patients will die waiting. Promising therapies are announced, clinical trial data shared, drug company stock prices shoot up or fall down, hope feels like it’s always just around the next corner, and yet, with a drug approval time far, far, longer than the life expectancy of most terminal patients, these drugs are like Tantalus’s grapes, dangling forever just out of reach. 

The FDA is regulating emerging drugs to death, which means dying patients are also being regulated to death. They’re a cost of doing business, and one the FDA doesn’t consider when encouraging clinical trial restrictions and slowing down access.

Some clinical trial drugs are so promising that the FDA bestows them with the distinction of “breakthrough therapy,” and they’re placed on a “fast track” to approval. Drugs like Petosemtamab, which Jake was lucky² enough to receive, is one. And it’s clear why Petosemtamab is so good relative to a death sentence like recurrent / metastatic head and neck cancer:  37.2% of patients receiving Petosemtamab monotherapy (meaning it was not combined with chemo or other drugs) saw a 30% reduction in their tumor size and, most impressively, 72% of patients achieved “disease control,” which means the drug arrested the relentless and often rapid tumor growth for a median of six months.

Current standard-of-care for patients like Jake is, basically, to suffer nasty chemo side effects then die. New therapeutic regimens include Keytruda, to which only 19% of  patients respond³, or Keytruda plus a particularly gruesome combination of chemo (Cisplatin / Carboplatin + 5FU), which extends life a miserable 4.5 months for the 35% who respond. The median lifespan after a diagnosis of recurrence and/or metastases is just twelve months.

The 70%+ of dying patients who don’t respond to standard of care treatments deserve alternatives like Petosemtamab, which also comes with significantly fewer side effects than chemo or radiation. Some patients might even have their progression halted by standard of care, and then, when that fails, be able to move on to Petosemtamab. Sequential treatment might gift  an additional six months of life. By that time, Moderna’s mRNA cancer vaccine might finally be approved, or a new, even more promising drug could come to trial.I don’t care what the law says, it’s criminal, or at least deeply ethically questionable, to not approve Petosemtamab and withhold it from desperate patients as an additional or alternate option to current therapies. Research moves faster than approval.

Despite the outstanding response rates, “breakthrough status” and “fast track” designation (given in August of 2023), Petosemtamab won’t be available anytime soon to the almost 60,000 people with newly diagnosed cases of HNSCC in 2024—or the 12,000 expected to die. Instead, due to complex FDA requirements for New Drug Applications (NDA), and what is likely the drug company’s goal of trialing with the intent to eventually supplant current standard-of-care, Petosemtamab will remain accessible by only a small number of patients lucky enough to receive it in a phase 3 trial of Petosemtamab + Keytruda that won’t start until the end of 2024.

Don’t mistake me: Data is hugely important in determining if a drug is safe, effective, and effective in what combinations. But rapidly expanding access to terminal patients as those questions are being answered, instead of after, will save lives.

In “Please by dying, part three” I wrote:

“Safety” sounds great, but the FDA is providing safetyism. There’s a process by which the FDA can fast-track a drug based on promising early-phase data, allowing the drug to come to market contingent on confirmatory trials. In exchange for accelerated approval, pharmaceutical companies will perform follow-up studies collecting additional data.”

I propose creating a second, provisional market for terminal patients that allows partial approval of a drug while it’s still undergoing trials, making it available to trial-ineligible (or those unable to easily access a trial) patients for whom standard of care doesn’t provide a meaningful chance at remission. This partial approval would ideally allow physicians to prescribe the drug to this subset of patients as they see fit: be that monotherapy or a variety of personalized combination therapies, tailored to a patient’s needs, side-effect profile and goals. This wouldn’t just expand access to patients who are otherwise out of luck, but can give important data about real-world reaction and response.

As incentive, the FDA could require pharmaceutical companies to provide drug access to terminal patients as a condition of continuing forward in the trial process on the road to a New Drug Application. While this would be federally forced compassion, it would differ from “compassionate use.” To currently access a study drug via compassionate use, a physician has to petition both the drug company and the FDA on the patient’s behalf, which comes with onerous rules and requirements. Most drug companies with compassionate use programs won’t offer a drug until there’s already a large amount of compelling phase 2 data demonstrating efficacy, a patient must have “failed” standard of care and other available treatments, and the drug must (usually) be given as a monotherapy. Even if the drugmaker says yes, the FDA can still say no. Compassionate use is an option available to very small numbers, in limited instances, and with bureaucratic barriers to overcome. Not terribly compassionate, in my opinion.

The benefits of this “terminal patient” market can and should go both ways, much as lovers should benefit from each other instead of trying to create win-lose situations. Providing the drug to patients would come with reciprocal benefits to the pharmaceutical companies.  Any physician prescribing the drug to patients should be required to report data regarding how the drug was used, in what combination, and to what effect. This would create a large pool of observational, real-world data gathered from the patient who aren’t ideal candidates for trials, but better represent the large subset of patients who’ve exhausted multiple lines of therapies yet aren’t ready for the end. Promising combinations and unexpected effects might be identified from these observational data sets, and possibly used to design future trials.

I can guess that one of the pharmaceutical companies’ biggest concerns with this approach would be observational data confounding the trial data for New Drug Applications (NDA). The solution is simple. Just keep the “clinical trial track” and the “terminal patient track” separate, excluding data from the “terminal patient track” in a drug’s initial NDA. The FDA could therefore require faster access to drugs for a subset of terminal patients as a drug-company requirement to keep bringing a drug to the standard market, while also promising that these requirements won’t reflect negatively on a drug’s ability to ultimately get approval.

“But Bess,” you might be thinking, “Trials are vital! We don’t want to go back to the era of snake oil salesman. Won’t people get hurt if we start throwing drug at dying patients and in novel combinations that hasn’t been adequately tested?”

Define “adequate.”  It’s bad PR to roll out a drug that causes accidental deaths in people who would have otherwise lived a long and (relatively) healthy life, but it should be just as shameful to have drugs with the potential to treat terminal patients, yet withhold those drugs for a decade or more,  allowing patients to die. Watching someone drown when you have the capability to jump in and save their life may not be exactly ethically equivalent to pushing someone into the river who you know can’t swim, but it’s not that dissimilar. It’s ethical hair-splitting. 

Inaction is itself an action. If a patient comes to the ER struggling to breathe and I don’t intubate, showing the malpractice attorney data on the number of people who have ventilator-associated injuries as the reason I chose to let nature take its course, even though I could’ve intervened, won’t save me from a lawsuit, or bring the patient back.

The FDA is standing there watching patients drop dead and fill up the invisible graveyard. A federal law called EMTALA prevents ER docs from standing around, inactive, while someone dies. EMTALA requires that I provide stabilizing treatment to any patient that arrives at the emergency room regardless of their ability to pay, or their prior lines of therapy. Maybe we need a federal mandate for a federal department, making inaction on drug approval for fatal diseases more than just morally corrupt, but illegal. If withholding terminal patient’s access to promising drugs for extended periods of time came with a hefty fine that came out of higher-ups pockets, I bet the drug development pipeline would speed up a bit.

Consent is key. The history of medical experimentation on unwitting patients is dark and horrible. Plenty of quack cancer clinics promising miracle cures are out there. But consenting patients who can understand the risks of a drug and be shown existing, if incomplete, data? It’d be wonderful if they could have access to medications that might potentially prolong or improve their lives outside of a restrictive clinical trial system. Clinical trials are, effectively, already throwing first-in-human drugs at patients in novel combinations that haven’t been adequately tested in order to test them. Patients can provide their consent for trials. Therefore, they should be able to provide their consent for off-trial uses of a drug, provided that all the known risks and benefits are clearly and honestly presented.

The FDA should incentivize drug companies to provide early, expanded access to terminal patients, instead of allowing dying patients to become collateral on the slow road to drug approval; a future where paternalism and safetyism are rejected and instead agency is given to dying patients, allowing them to move scientific progress forward, potentially to their own gain, because they’re considered capable of consent; a future where real-world observational data can be a boon, not a potential burden to research and development; a future in which the invisible graveyard fills much more slowly.

The FDA’s torpor and lack of urgency in bringing drugs for terminal patients to market makes me wonder if, magically, no one working at the FDA with decision-making capacity has ever watched someone they love die. Perhaps they have drug access we plebs do not.

If you’ve gotten this far, consider the Go Fund Me that’s funding my husband Jake’s ongoing cancer treatment.

---

---

1. Most clinical trials restrict how many previous lines of therapy a patient can have—I don’t really understand the line-of-therapy limitation rationale—that aren’t first-in-human or very early phase one studies doing dose escalation—studies that are not exploring efficacy—will be closed to him. Eligibility criteria for trials past the “How high a dose can we give without killing someone? Oops, too high” phase prefer patients with one, maybe two prior lines of therapy, max.

2. “Luck” that involved an incredible amount of research, connection, systems experience, timing of imaging and, rumor has it, the sacrifice of a goat or two. (See Please be dying, but not too quickly part 1-3 for the saga). The study was also only available in the USA at UCSD in San Diego, with a limited number of slots. I believe 25 head and neck cancer patients were enrolled.

3. And this is of the subset of patients with a CPS score >1. A CPS score has to do with the expression of a protein called PD-1 which is used as an immunotherapy binder. Patients with a CPS score less than 1 aren’t eligible at all for Keytruda.