For over thirty years, oncologists had a name for certain cancer targets: undruggable. Proteins so smooth, so featureless, so stubbornly resistant to pharmaceutical intervention that the scientific consensus was simply to move on and focus on something else. In 2026, that word is becoming obsolete. Two landmark advances — RAS inhibitors for solid tumors and menin inhibitors for leukemia — represent the crumbling of walls that cancer researchers spent decades beating their heads against. The story of how these impossible drugs came to exist is one of the most exciting chapters in modern medicine.
The Undruggable Target: KRAS
RAS is a family of proteins — KRAS, NRAS, HRAS — that act as molecular switches inside your cells. When a growth signal arrives from outside the cell, RAS flips to the “on” position, tells the cell to divide, and then flips back off. Clean, controlled, essential for normal biology.
Except when it’s not. When KRAS gets mutated, the switch gets stuck permanently in the “on” position. The cell receives a relentless signal to divide, and it does — endlessly. KRAS mutations appear in approximately 30% of all human cancers, including lung, colon, and especially pancreatic cancer, where roughly 90% of tumors carry a KRAS mutation.
Scientists discovered the KRAS oncogene in the early 1980s — over forty years ago. Almost immediately, drug developers wanted to build a molecule that could block it. The problem was maddening: KRAS is tiny and almost perfectly smooth, a small globular protein with no obvious pockets or grooves where a drug molecule could grab hold. Trying to design a drug for KRAS was like trying to stick a Post-it note on a marble.
Researchers spent decades trying and failing. By the late 1990s and early 2000s, the scientific consensus was blunt: KRAS is undruggable. Move on.
The Hidden Pocket: Shokat’s Discovery
Not everyone accepted defeat. In 2013, Kevan Shokat at the University of California, San Francisco made a discovery that nobody expected. One specific KRAS mutation — called G12C — actually creates a tiny hidden pocket in the protein, located under what’s called the “Switch II” region. The pocket only appears when the protein is in its inactive state, but it exists.
Shokat’s lab demonstrated that you could design a small molecule to slip into that pocket and essentially lock KRAS in the off position permanently. That single paper blew the field wide open.
By 2021, the FDA approved sotorasib (brand name Lumakras, manufactured by Amgen) as the first-ever direct KRAS inhibitor, targeting G12C-mutated non-small cell lung cancer. It was a historic moment — the first successful drug against a target that the entire field had written off as impossible.
Cracking G12D: The Pancreatic Cancer Problem
But here’s the catch. The G12C mutation is relatively common in lung cancer, but the most prevalent KRAS mutation in pancreatic cancer is a different one: G12D. Approximately a third of pancreatic cancers carry it specifically. And G12D is even harder to target than G12C because it doesn’t have the same accessible pocket.
If G12C was the secret door in the marble, G12D was the vault with no keyhole.
Multiple research groups attacked the problem from different angles. MRTX1133, a non-covalent inhibitor capable of binding both the active and inactive forms of the KRAS G12D protein, provided the first proof-of-concept that direct targeting was possible. HRS-4642, another G12D inhibitor, showed early Phase I data with tumor shrinkage in lung cancer patients. Verastem Oncology reported preliminary results from their VS-7375 program in late 2025, designed to block KRAS G12D in both its active “on” and inactive “off” conformations — a dual mechanism that locks the switch from both sides.
The biggest news from Dana-Farber Cancer Institute’s 2026 breakthroughs list is that a novel RAS inhibitor for pancreatic cancer has entered a Phase III clinical trial, led by Brian Wolpin, director of the Hale Family Center for Pancreatic Cancer Research. Phase III is the last major hurdle before seeking regulatory approval — a head-to-head comparison against standard treatments in a large patient population.
Andrew Aguirre, co-director of Dana-Farber’s Center for RAS Therapeutics, called the revolution in targeting RAS “one of the biggest therapeutic advances in the history of clinical care for pancreatic cancer patients.”
That’s not a small statement for a disease where the standard of care barely budged for decades. Gemcitabine — a chemotherapy drug approved in the 1990s — remained the backbone of pancreatic cancer treatment for over twenty years. RAS inhibitors represent the first time researchers can directly go after the primary genetic driver of the disease.
Pancreatic cancer has a five-year survival rate of approximately 13%, dropping below 5% for metastatic cases. It’s projected to become the second leading cause of cancer death in the United States by 2040. The arrival of targeted therapy for its dominant mutation is genuinely paradigm-shifting.
Menin Inhibitors: Breaking Leukemia’s Deadly Partnership
The second major breakthrough of 2026 targets a completely different cancer through equally innovative science. Acute myeloid leukemia (AML) is one of the most aggressive blood cancers — fast-moving, difficult to treat, and devastating for patients who relapse.
Menin is a protein in the nucleus of cells that normally helps regulate which genes get turned on and off during blood cell development. It works closely with another protein complex involving KMT2A (also known as MLL). Under normal circumstances, this partnership is tightly controlled.
Two types of genetic damage weaponize this partnership. In some AML patients, the KMT2A gene gets rearranged — broken and fused to another gene, creating a mutant protein that hijacks menin to permanently activate genes that drive leukemia. In another subset, a gene called NPM1 gets mutated and similarly co-opts the menin-MLL machinery. Menin isn’t the villain itself — it’s being used by the villain.
Scott Armstrong at Dana-Farber spent over twenty years studying this biology. His lab worked out exactly how the mutant MLL protein and menin conspire to activate the leukemia program, then set out to find a drug that could pry them apart — literally breaking up the molecular partnership.
Working with Syndax Pharmaceuticals, Armstrong’s group described a class of menin inhibitors in 2019 that could eradicate leukemia in preclinical models. The compound that emerged from this work is revumenib (brand name Revuforj), an oral pill taken by mouth.
The FDA approvals came in rapid succession. In November 2024, revumenib was approved for relapsed or refractory acute leukemia with KMT2A translocations — in both adults and children. In October 2025, a second approval followed for AML with NPM1 mutations, making revumenib the first and only menin inhibitor approved for multiple subtypes of acute leukemia.
The pediatric approval is particularly significant. KMT2A-rearranged leukemia disproportionately affects young patients, and having an option beyond intensive chemotherapy is meaningful for families facing this diagnosis.
Together, KMT2A rearrangements and NPM1 mutations account for roughly 40% of all AML cases in both children and adults. That’s an enormous fraction of patients who now have a fundamentally new treatment option that didn’t exist two years ago.
Jacqueline Garcia at Dana-Farber called it “a monumental step forward.” The next frontier is combination therapy — pairing revumenib with chemotherapy or other targeted agents to produce deeper, more durable remissions. Those trials are already underway.
Why “Undruggable” Keeps Falling
The KRAS and menin stories share a deeper lesson. Both targets were considered impossible for decades. Both required researchers who refused to accept the consensus. And both were ultimately cracked by scientists who developed new ways of seeing — discovering hidden pockets, understanding dynamic protein conformations, and exploiting vulnerabilities that previous generations of tools simply couldn’t detect.
This pattern is accelerating across oncology. Dana-Farber’s 2026 list also highlights protein degraders — drugs that don’t just block a problematic protein but tag it for destruction by the cell’s own recycling machinery. Eric Fischer at Dana-Farber predicted that this decade could be “the clinical decade for protein degradation.” Instead of blocking the lock, you shred the entire door.
Other breakthroughs on the horizon include personalized cancer vaccines designed to train a patient’s immune system against their unique tumor’s specific mutations. Liquid biopsies capable of detecting cancer relapse from a simple blood draw. Radioligand therapies that deliver targeted radiation directly to tumors while sparing healthy tissue.
The toolkit has exploded. For decades, cancer treatment was essentially three tools: surgery, radiation, and chemotherapy — a blunt approach. Immunotherapy arrived and transformed outcomes for some cancers. Now we’re entering an era of precision oncology, where understanding the molecular wiring of individual cancers enables drugs designed to go after specific broken components.
The Technologies Behind the Breakthroughs
Several enabling technologies made these advances possible. Cryo-electron microscopy allows scientists to visualize protein structures at near-atomic resolution, revealing the hidden pockets and dynamic conformations that earlier tools missed. AI-driven drug design can screen millions of potential molecular interactions computationally before a single compound is synthesized. And our deepening understanding of protein dynamics — the fact that proteins are not rigid objects but constantly flexing, breathing, and shifting between states — has opened targeting strategies that static structural analysis could never reveal.
The lesson from KRAS is that “undruggable” often just meant “we haven’t figured it out yet.” Shokat found the hidden pocket in 2013. Armstrong cracked the menin-MLL interaction after two decades of work. Each breakthrough built on tools and insights that didn’t exist when the targets were first declared impossible.
What This Means for Patients
For pancreatic cancer patients — facing a disease where the standard of care has been essentially unchanged for twenty years — directly targeting the primary genetic driver represents genuine hope. These are still clinical trials, and not every patient will respond. But the trajectory is historic.
For leukemia patients — particularly children with KMT2A-rearranged disease and adults with relapsed AML — revumenib offers a fundamentally different mechanism of action beyond conventional chemotherapy. The speed from lab discovery to two FDA approvals in approximately five years is remarkable.
And for the broader oncology field, the crumbling of “undruggable” targets signals that the remaining supposedly impossible proteins are living on borrowed time. The combination of structural biology, computational chemistry, and biological insight is systematically dismantling the barriers that defined cancer treatment’s limitations for a generation.
Cancer had a very bad year. The science says it’s about to get worse — for cancer.