The KRAS drug story has a pretty specific shape to it so far.
Amgen’s Lumakras and Mirati’s Krazati got there first. They’re both covalent inhibitors. They form a permanent bond with a cysteine that only exists when KRAS has the G12C mutation.
That’s a brilliant design for that one mutation.
It’s also a hard ceiling: the other 80% of KRAS-mutant cancers — G12D, G12V, G13D, Q61 — don’t have that cysteine. The covalent strategy doesn’t work.
That’s why everyone is chasing pan-KRAS. Hit all the mutants. Open up colorectal, pancreatic, lung. The whole prize.
Today, BlossomHill Therapeutics dropped preclinical data at AACR 2026 on BH-501284, their pan-KRAS entry. Two numbers to know:
- Median IC50: ~0.83 nM
- Target residence time: >54 hours
Sub-nanomolar potency against pan-KRAS is rare. A residence time of more than two days is the real headline though.
The pan-KRAS race: binding mechanisms
BlossomHill — BH-501284
Non-covalent, pseudo-irreversible · >54 hr residence
AACR 2026 preclinical · IND-enabling next
BBOT — BBO-11818
Non-covalent · Active & inactive KRAS states · >500x selectivity over HRAS/NRAS
Phase 1 KONQUER-101
Kestrel — KST-6051
Oral small molecule · GTP + GDP states
IND cleared Q1 2026 · Phase 1 FALCON
Revolution Medicines — daraxonrasib
RAS(ON) multi-selective
Phase 3 RASolute 303 in PDAC
Okay so why does this matter….?
Covalent inhibitors get their long residence time from the fact that they form a chemical bond with the target. They’re stuck there. Non-covalent drugs usually come off quickly — they’re held in place by hydrogen bonds and weak interactions that break easily.
“Pseudo-irreversible non-covalent” means BlossomHill built a molecule that behaves kinetically like a covalent drug without actually forming a permanent chemical bond. You get the sustained target engagement — long residence time means the drug keeps KRAS shut down long after the plasma concentration drops — without needing a reactive warhead that’s limited to a specific cysteine.
And without a warhead, you’re not married to G12C anymore. You can go pan-mutant.
The Biotech Voyager
Early-stage biotech signals, personalized.
The signals that matter to you — contextualized and written directly to you — so you cut through the noise and immediately understand why it matters.
Also…BlossomHill isn’t a random startup.
The company was founded by Jean Cui, the drug designer behind three FDA-approved cancer medicines (crizotinib, lorlatinib, repotrectinib) and the co-founder of Turning Point Therapeutics, which Bristol Myers Squibb bought for $4.1 billion in 2022.
They raised a $100M Series B in Feb 2024 and added an $84M extension in December 2025. Total raised is somewhere around $260M.
Their lead program, BH-30643, is a mutant-selective EGFR inhibitor already in the SOLARA Phase 1/2 for EGFR-mutant NSCLC. It’s active against T790M and C797S resistance mutations.
They also have BH-30236, a CLK inhibitor targeting mRNA splicing, in Phase 1/1b for AML and MDS.
The pattern here: Cui’s team picks hard targets, builds around resistance from day one, and publishes clean chemistry. BH-501284 is pre-IND, so this is early preclinical.
But it’s the first glimpse of their KRAS program, and the residence time data is the kind of thing that separates “me-too” from “actually interesting.”
The pan-KRAS race is crowded. Revolution has daraxonrasib in Phase 3 already. Setidegrasib, a KRAS G12D degrader, just got into NEJM. BBOT has BBO-11818 in Phase 1. Kestrel’s KST-6051 starts dosing this year.
BlossomHill is late to the party, but they’re bringing a different tool. If the clinical data matches the preclinical profile — and that’s a big if — a >54-hour residence time could mean cleaner dosing, less frequent administration, and more durable target coverage than the competition.
IND-enabling studies next. Watch for a 2027 IND.
