This week's Root Room: Debby Learns to Listen First in the Waiting Room — [Link]. Teaser at the end of the Plain Talk section.

What happens when a drug stops working, and why that is not the end of the story.

Branch from Issue 007

In Issue 007, we followed the phase 3 IRIS trial from its June 2000 launch to the moment the Data Safety Monitoring Board (DSMB) called a stop. 76.2% of patients on imatinib (Gleevec) achieved a complete cytogenetic response while the comparison arm was at 14.5%.

The trial didn't stop because it failed, rather because the evidence was so clear that continuing to randomize patients to the inferior arm was unethical. So, according to the guidelines of the trial defined by the DSMB Charter or a similar plan defined early in the planning stages of the study, course shifted to seek approval early rather than delaying this new treatment for patients longer than necessary.

The FDA approved Gleevec 71 days after Novartis submitted its application. Not because the standard dropped, but because the evidence was unusually clear.

Most patients on imatinib did extraordinarily well. Key word being "most". Most people today still do well on imatinib, Gleevec.

In 2008, a 60-year-old man in Los Angeles was told he had cancer. In his first public statement about that moment, he feared that he might be dead in a few months.

He is 78 years old today, and he goes by the name Kareem Abdul-Jabbar. He has CML, and the drug that's kept him alive for nearly 2 decades is the same drug approved in 71 days, from a trial that stopped early, from basic research that Novartis initially declined to fund. Thankfully for the world at large, he has been a strong advocate for the targeted therapies such as Gleevec ever since having gotten the drug 7 years after it was on the market.

However, if you thought we had gone the full Root to Rx, farm to table story, well, the Gleevec story is not finished.

For most patients, Gleevec changed everything. Before Gleevec, the 5-year survival rate for CML was roughly 30%. After, it became roughly 90%. Most patients today are treated with a daily pill and live for decades. CML no longer was a death sentence. Now it's a chronic condition.

Now for the "but". But, for a subset of patients, Gleevec stopped working. It wasn't immediate, and it wasn't everyone. Over time, a single mutation in the BCR-ABL gene changed the shape of the protein just enough that the drug could no longer bind, because that's at the core of what makes cures or treatments for types of cancer difficult. Cancers are constantly changing due to the nature of their biology.

The most common and most stubborn of these is called T315I. Researchers named it the gatekeeper mutation.

One letter change in a 3 billion-letter book. A single amino acid, threonine, replaced by another, isoleucine, and it meant Gleevec no longer fits as an inhibitor.

Think of Gleevec as a key, and BCR-ABL as a lock. Gleevec was designed to fit that lock precisely, to block the always-on signal that drives CML. T315I changes the shape of the lock so the key no longer fits inside.

What did not happen: researchers did not abandon BCR-ABL as a target.

Second-generation inhibitors, nilotinib and dasatinib, were designed to bind more tightly and to cover most of the resistance mutations that Gleevec could not address.

For many patients with imatinib resistance, they worked.

T315I was harder. It defeated the second generation too. It became the specific problem the field spent years working to solve.

Then came asciminib. Asciminib doesn’t compete at the same binding site as Gleevec or any second-generation inhibitor. Rather than targeting the ATP-binding site in the kinase domain, asciminib targets the myristoyl pocket, a completely different regulatory region of the BCR-ABL protein. So it can bypass the mutation entirely.

Gleevec did not end the story. It started one. When resistance emerged, the pipeline did not abandon BCR-ABL as a target. It got smarter about how to hit it. Second-generation inhibitors addressed the mutations Gleevec could not. And asciminib arrived with a fundamentally different approach, targeting a different part of the protein entirely. Not a better version of the same key. A different door.

That pattern is not unique to CML and so, throughout much of the clinical research pipeline, the battle looks like this:

Identify the molecular driver, build a targeted therapy, iterate (perform repeatedly) when resistance emerges.

Right now, in 2026, it's playing out with a disease that has been one of medicine's most resistant problems.

Pancreatic cancer has the lowest 5-year survival rate of any major cancer. Approximately 13%. The molecular driver in roughly 90% of cases: a mutated RAS protein stuck in the on position. The same kind of always-on signal as BCR-ABL in CML.

For decades, RAS was considered undruggable. No suitable binding pocket. Extremely high affinity for its substrate (meaning it's like a golden retriever catching a ball out of the air. More likely than not, that ball is getting caught and little Billy trying to steal the ball from the dog will lose out). Selective inhibitors were widely thought to be chemically impossible.

Those results were simultaneously published in the New England Journal of Medicine on May 31, 2026, the same day they were presented at the ASCO plenary session. The FDA has also opened an expanded access program for daraxonrasib, allowing eligible patients outside of clinical trials to access it while the formal NDA review proceeds.

Daraxonrasib has cleared Phase 3 and the evidence has now passed peer review. The FDA review floor is still ahead. That is the next step.

One more thing to know about this story. Former Senator Ben Sasse of Nebraska is enrolled in a daraxonrasib trial. He was diagnosed with stage 4 metastatic pancreatic cancer in December 2025, given 3 to 4 months to live. In a 60 Minutes interview, he reported a 76% reduction in his own tumor volume over 4 months on the drug.

That number, 76%, describes one patient's individual response. It is not the trial average. The trial average is the 13.2 versus 6.7 months above. Both numbers matter. They describe different things.

The cost of dismissing a drug that actually works is not a bad investment. It is a person, a father, a sister, a grandparent, a friend. And it would not exist without years of publicly funded research that enabled the mechanism that Revolution Medicines pursued in clinical trials.

These are the party test moments from this issue to share with friends, family, and I won't be mad if you share them with random strangers on the street, although they might be.

Kareem Abdul-Jabbar was diagnosed with CML in 2008. He is 78 today. The drug keeping him alive was approved in 71 days, not because the bar was low, but because the evidence was that clear.

When Gleevec stopped working for some patients, medicine did not stop. It designed new keys. Then it found a different door.

In April 2026, a drug called daraxonrasib doubled median survival in pancreatic cancer per pre-FDA review press releases. The same pattern as Gleevec, but a different disease tackled by the same machinery. You may have seen by now the standing ovation, mid-speech, where this data was presented at ASCO 2026. It is incredibly uncommon for a mid-speech standing ovation there, I might add.

The Gleevec arc started with a drug no one wanted to fund. Issue 009 follows the money, not to confirm what you already think, but to find out what the evidence says when you run it through the Toolkit.

These 3 issues give you the foundation for everything in this arc.

Issue 003b: How to evaluate any health claim using the Skeptic's Toolkit. Start here.

Issue 001: Why medical distrust is rational, and what to do with it anyway.

Issue 004: How a drug gets from a lab to your pharmacy. The pipeline in plain language.

Maria got her results back. 3 weeks in, and Gleevec is already working.

Debby came prepared. She had the IRIS data. She knew the DSMB. She had every answer Maria might need.

She was the least helpful person in the room.

In this week's Root Room, a character named Cora does something with 5 letters that Rooty and Sam kept forgetting. Debby learns what the Toolkit cannot teach by itself.

Then her friend Maria was diagnosed with CML.

In this week's Web Companion, Debby goes somewhere she's never been. She brings all the right information, but she is the least helpful person in the room until someone named Cora front loads that small card with 5 letters on it by doing something Rooty and Sam kept forgetting to do.

Read The Root Room: Debby Learns to Listen First in the Waiting Room — [Link]

For a deeper look at the science behind this issue, the T315I gatekeeper mutation, asciminib's mechanism, and the full citation tier breakdown for the daraxonrasib data — read The Informed at [Link].

Know someone being told vaccines were rushed, that pharma only cares about profits, or that natural is always safer? This is the toolkit they did not know they needed.

Share Root to Rx with one person this week. When they subscribe, you both unlock the full Skeptic's Toolkit as a downloadable reference.

228 Park Ave S, #29976, New York, NY 10003, United States