Amgen plows ahead with costly, highly toxic cancer drug dosing despite FDA challenge

When doctors began using the drug sotorasib in 2021 with high expectations for its innovative approach to attacking lung cancer, retired medical technician Don Crosslin was an early beneficiary. Crosslin started the drug that July. His tumors shrank, then stabilized.

But while the drug has helped keep him alive, its side effects have gradually narrowed the confines of his life, says Crosslin, 76, who lives in Ocala, Florida: “My appetite has been minimal. I’m very weak. I walk my dogs and get around a bit, but I haven’t been able to golf since last July.”

He wonders whether he’d do better on a lower dose, “but I do what my oncologist tells me to do,” he says. Every day, he takes eight of the 120-milligram pills, sold under Amgen’s brand name Lumakras.

Crosslin’s concern lies at the heart of a U.S. Food and Drug Administration effort to make cancer drugs less toxic and more effective. Cancer drug trials are structured to promote high doses, which then become routine patient care. In the face of evidence that thousands of patients become so ill that they skip doses or stop taking the drugs — thereby risking resurgence of their cancers — the FDA has begun requiring companies to pinpoint the right dosage before they reach patients.

The initiative, Project Optimus, launched in 2021 just as Amgen was seeking to market sotorasib. At the time, the FDA’s leading cancer drug regulator, Richard Pazdur, co-authored an editorial in the New England Journal of Medicine that said Amgen’s trials of the $20,000-a-month drug were “hampered by a lack of robust dose exploration.”

The FDA conditionally approved sotorasib but required Amgen to conduct a study comparing the labeled dosage of 960 mg with a dosage of 240 mg. The trial, published in November, showed that the 960-mg dose may have given patients a month more of life, on average, but caused more severe side effects than the lower dose.

Amgen is keeping the 960-mg dosage as it conducts further tests to get final approval for the drug, spokesperson Elissa Snook says, adding that the dose showed superiority in one study. Whether medically justified or not, the heavier dosage allows the company to protect 75% of its revenue from the drug, which brought in nearly $200 million in the United States last year.

And there appears to be nothing the FDA can do about it.

“There’s a gap in FDA’s authority that results in patients getting excess doses of a drug at excess costs,” says Dr. Mark Ratain, a University of Chicago oncologist who has pushed for more accurate cancer drug dosing. “We should do something about this.”

Deciding on dosage

It may be too late for the FDA to change the sotorasib dosage, although in principle it could demand a new regimen before granting final approval, perhaps in 2028. Under Project Optimus, however, the agency is doing something about dosage guidelines for future drugs. It is stressing dose optimization in its meetings with companies, particularly as they prepare to test drugs on patients for the first time, spokesperson Lauren-Jei McCarthy says.

“When you go in front of FDA with a plan to approve your drug now, they are going to address dosing studies,” says Julie Gralow, chief medical officer of the American Society of Clinical Oncology. “A lot of companies are struggling with this.”

That’s largely because the new requirements add six months to a year and millions in drug development costs, says Julie Bullock, a former FDA drug reviewer who advocated for more extensive dosing studies and is now senior vice president at Certara, a drug development consultancy.

In part, Project Optimus represents an effort to manage the faults of the FDA’s accelerated approval process, begun in 1992. While the process gets innovative drugs to patients more quickly, some medicines have proved lackluster or had unacceptable side effects.

That’s especially true of the newer pills to treat cancer, says Donald Harvey, an Emory University pharmacology professor, who has led or contributed to more than 100 early-phase cancer trials.

A study released last month in the Journal of the American Medical Association showed that 41% of the cancer drugs granted accelerated approval from 2013 to 2017 did not improve overall survival or quality of life after five years.

Many of these drugs flop because they must be given at toxic dosages to have any effect, Harvey says, adding that sotorasib might work better if the company had found an appropriate dosage earlier on.

“Sotorasib is a poster child for incredibly bad development,” Harvey says. The drug was the first to target the KRAS G12C mutation, which drives about 15% of lung cancers and was considered “undruggable” until University of California-San Francisco chemist Kevan Shokat figured out how to attack it in 2012.

Given the specificity of sotorasib’s target, Harvey says, Amgen could have found a lower dosage. “Instead, they followed the old model and said, ‘We’re going to push the dose up until we see a major side effect.’ They didn’t need to do that. They just needed more experience with a lower dose.”

The 960-mg dose “is really tough on patients,” says Yale University oncologist and assistant professor Dr. Michael Grant. “They get a lot of nausea and other GI side effects that are not pleasant. It hurts their quality of life.”

Higher doses may merely intensify side effects

The FDA noted in its review of sotorasib that in phase 1 studies tumors shrank when exposed to as little as a fifth of the 960-mg daily dose Amgen selected. At all doses tested in that early trial, the drug reached roughly the same concentrations in the blood, which suggested that at higher doses the drug was mostly just intensifying side effects like diarrhea, vomiting and mouth sores.

For most classes of drugs, companies spend considerable time in phases 1 and 2 of development, homing in on the right dosage. “No one would think of dosing a statin or antibiotic at the highest tolerable dose,” Ratain said.

Things are different in cancer drug creation, whose approach originated with chemotherapy, which damages as many cancer cells as possible, wrecking plenty of healthy tissue in the bargain. Typically, a company’s first series of cancer drug trials involve escalating doses in small groups of patients until something like a quarter of them get seriously ill. That “maximum tolerated dose” is then employed in more advanced clinical trials, and goes on the drug’s label. Once a drug is approved, a doctor can “go off-label” and alter the dosage, but most are leery of doing so.

Patients can find the experience rougher than advertised. During clinical trials, the side effects of the cancer drug osimertinib (Tagrisso) were listed as tolerable and manageable, says Jill Feldman, a lung cancer patient and advocate. “That killed me. After two months on that drug, I had lost 15 pounds, had sores in my mouth and down my throat, stomach stuff. It was horrible.”

Some practitioners, at least, have responded to the FDA’s cues on sotorasib. In the Kaiser Permanente health system, lung cancer specialists start with a lower dose of the drug, spokesperson Stephen Shivinsky says.

Smaller doses — and revenue

Amgen was clearly aware of the advantages of the 240-mg dosage before it sought FDA approval: It filed a provisional patent application on that dosage before the agency gave breakthrough approval for the drug at 960 mg. The company doesn’t appear to have disclosed the patent filing to investors or the FDA. McCarthy said the FDA was prohibited by law from discussing the particulars of its sotorasib regulation plans.

Switching to a 240-mg dosage could register a huge hit to Amgen’s revenue. The company markets the drug at more than $20,000 for a month of 960-mg daily doses. Each patient who could get by with a quarter of that would trim the company’s revenue by roughly $180,000 a year.

Amgen declined to comment on the patent issue or to make an official available to discuss the dosage and pricing issues.

Crosslin, who depends on Social Security for his income, couldn’t afford the $3,000 a month that Medicare required him to pay for sotorasib, but he has received assistance from Amgen and a charity that covers costs for patients below a certain income.

While the drug has worked well for Crosslin and other patients, its overall modest impact on lung cancer suggests that $5,000, rather than $20,000, might be a more appropriate price, Ratain says.

In the company’s phase 3 clinical trial for advanced lung cancer patients, sotorasib kept patients alive for about a month longer than docetaxel, the current, highly toxic standard of care. Docetaxel is a generic drug for which Medicare pays about $1 per injection. The trial was so unconvincing that the FDA sent Amgen back to do another.

U. of C. oncologist has run dosing trials on other drugs

Ratain, the U. of C. oncologist and a staunch critic of Amgen’s handling of sotorasib, told Centers for Medicare & Medicaid Services officials at a recent meeting that they should pay for sotorasib on a basis of 240 mg per day. But CMS would do that only “if there is a change in the drug’s FDA-approved dosage,” spokesperson Aaron Smith says.

Drug companies generally don’t want to spend money on trials like the one the FDA ordered on sotorasib. In 2018, Ratain and other researchers used their institutions’ funding to conduct a dosing trial on the prostate cancer drug abiraterone, marketed under the brand name Zytiga by Johnson & Johnson. They found that taking one 250-mg pill with food was just as effective as taking four on an empty stomach, as the label called for.

Although J&J hasn’t changed the Zytiga label, the evidence generated in that trial was strong enough for the standards-setting National Comprehensive Cancer Network to change its recommendations.

Post-marketing studies like that one are hard to conduct, Emory’s Harvey says. Patients are reluctant to join a trial in which they may have to take a lower dosage, since most people tend to believe “the more the better,” he says.

“It’s better for everyone to find the right dose before a drug is out on the market,” Harvey says. “Better for the patient, and better for the company, which can sell more of a good drug if the patients aren’t getting sick and no longer taking it.”

KFF Health News is a national newsroom that produces in-depth journalism on health issues.