An oral peptide blocked cancer's most-wanted protein across many different mutant forms, not just the one version the approved drugs can hit. The catch, as always with this kind of result, is that the work is still in mice and dishes.

The drug is LUNA18, also called paluratide, from Chugai Pharmaceutical in Japan (part of the Roche group). In work published online July 30 in Molecular Cancer Therapeutics ↗, a team led by Hitoshi Sase and Hiroshi Tanaka describes it as a first-in-class pan-RAS inhibitor. Pan-RAS means it goes after the whole RAS family rather than a single version of it. That is the entire point.

Why one letter matters

RAS is the protein that mutates in roughly a quarter to a third of all human cancers, which for decades made it the most sought-after and most frustrating target in the field. It sat in a group scientists called undruggable, because its surface gave small-molecule drugs nothing to grab.

The breakthrough came with drugs that hit one specific version, KRAS G12C, where a single amino acid swap (glycine to cysteine at position 12) leaves a reactive handle a drug can latch onto. Sotorasib from Amgen and adagrasib from Bristol Myers Squibb both work that way and are approved for lung cancer. But G12C is a minority of RAS mutations. The much larger share sit at other positions, G12D and G12V and G13D among them. Others come from the gene being amplified into extra copies. G12D is the single most common RAS mutation, frequent in pancreatic and colorectal cancer, and none of these forms present the reactive handle a G12C drug needs. That drug does nothing for them.

LUNA18 takes a different route. It is a cyclic peptide, a small ring of amino acids rather than a conventional small molecule. It binds the inactive, switched-off form of RAS and keeps it from being flipped back on. Because that off-state pocket is shared across mutants, one drug reaches many of them. In the paper, LUNA18 acted against cells carrying G12 and G13 mutations and against tumors driven by extra KRAS copies.

The escape-hatch problem

The more interesting result is about durability. Tumors treated with a G12C-only drug tend to route around it. Under long-term treatment and growth-factor stimulation in the lab, LUNA18 held signaling down more completely and for longer than a G12C inhibitor did. The authors trace that to normal, unmutated RAS. Even in a cancer driven by one mutant copy, the wild-type version of the protein still helps run the MAPK growth pathway. A drug that only silences the mutant leaves that back channel open. LUNA18 quiets the wild-type protein too, closing it.

That points to combinations rather than a solo act. LUNA18 worked together with drugs that hit receptor tyrosine kinases and the MAPK pathway, the two circuits tumors lean on when RAS is pressured. In xenograft models where a G12C drug alone ran out of room, adding LUNA18 produced more durable shrinkage than either drug by itself.

Reaching an intracellular target with an orally dosed peptide is its own trick, since peptides are usually too big and too fragile to survive the gut and slip inside cells. Chugai built LUNA18 through an mRNA-display screen of an enormous peptide library, then optimized the hit into something that can be swallowed. An early-stage human trial in patients with RAS-altered solid tumors is already underway.

This is a peptide reaching for a target that defeated small molecules for forty years, the same story peptidemodel covered when a designed peptide jammed FOXM1, another cancer driver drugs have struggled to hit ↗. RAS is the harder case and the bigger prize. LUNA18 does not yet have a card on peptidemodel; it belongs to the anticancer ↗ class of peptide candidates the platform tracks.

None of this is a clinical result. Cell lines and mouse tumors are where most promising cancer drugs look their best, and RAS in particular has a long record of preclinical wins that thinned out in people. What the paper establishes is a mechanism with a clear logic: hit the whole family, close the wild-type back channel, and the tumor has fewer doors to escape through. Whether that holds in a patient is the question the trial exists to answer.