Skip to content

AACR Annual Meeting 2026

April 17–22, 2026 · San Diego · 2418 items

After the meeting★★★

KRAS / RAS targeting: from platform to Phase 3

199 people · 233 posts15 topics

Start here

KRAS is one of the most common cancer drivers, and for years it was undruggable: the first inhibitors only caught one subtype (G12C) in its off state, then resistance arrived. AACR 2026 is where that story changed — one chemistry can now hit several KRAS forms, and there is a Phase 3 survival readout.

Three facts carry this chapter: daraxonrasib in second-line pancreatic cancer moved median overall survival from 6.7 to 13.2 months; G12D lung cancer showed response rates above 50%; resistance is no longer just a mutation list — the cell changes identity.

First-generation G12C drugs (sotorasib, adagrasib) bind KRAS in the off state. After treatment, RTK feedback reloads KRAS into the on state. Single-agent response sits around 30–40% and progression-free survival around six months — well below EGFR / ALK. AACR 2024 was still about G12C monotherapy resistance; 2026 is about platformizing the whole RAS family.

Revolution Medicines’ RAS(ON) molecules bind RAS and cyclophilin A together and block RAF / PI3K. They do not need the off state, so they skip adaptive GTP loading. The floor also showed dual ON/OFF (VS-7375) and RM-055, built for amplification and daraxonrasib-resistant models.

The second layer of resistance is not another point mutation. Hata’s lab saw drug pressure raise APOBEC3A and push persister cells into a high-plasticity state; Aguirre saw RAS transcription fall in the pancreatic primary but not in liver metastases. Hitting RAS and changing cell identity are two sides of the same event.

Analysis

The floor rewrote the question from “how long can G12C monotherapy last” to “who first makes first-line pancreatic cancer a standard.” daraxonrasib’s Phase 3 is still the only RAS-inhibitor overall-survival win. zoldonrasib and VS-7375 in G12D remain Phase 1 — strong numbers, small samples, not head-to-head.

Chinese groups are not short of G12C molecules (elisrasib’s floor numbers are strong). What they do not have is a multi-selective platform. One of the hotter posts said the number that matters may not be response rate, but whether enough drug actually sits in the tumor.

Still open
  • Whether RASolute 302’s benefit holds across US, European and Asian cohorts.
  • Whether the ON+OFF doublet’s 2026 H2 human data recreates the durable regressions seen preclinically.
  • Whether zoldonrasib’s response rate stays above 40% once the cohort is past 50 patients.
Key readout

daraxonrasib · mOS 13.2 vs 6.7 mo, HR 0.40RASolute 302 · 2L mPDAC · Phase 3

Records

Drug / trialReadoutSetting
daraxonrasibdaraxonrasibmOS 13.2 vs 6.7 mo, HR 0.40RASolute 302 · 2L mPDAC · Phase 3
daraxonrasib ± GnPdaraxonrasib#PANCREATICCANCER#PDACmono ORR 47%; +GnP ORR 55%, DCR 90%1L mPDAC · Phase 1
zoldonrasibzoldonrasibRMC-9805ORR 52%, DCR 93%, mPFS 11.1 mo; ctDNA clearance 73%G12D NSCLC · Phase 1
VS-7375VS-7375NSCLC ORR 68.8% (n=16); pancreas ORR 58%dual ON/OFF G12D · floor disclosure
elisrasibelisrasibG12Ci-naive NSCLC ORR 59.5%; pretreated 32.3%D3S-001 · floor disclosure
RM-055RM-055Active in daraxonrasib-resistant and RAS-amplification modelsPreclinical · the meeting’s hottest single molecule

Floor

Apr 23Thu
  8. 4 more on this topicKRAS · 51
Apr 22Wed
  13. 6 more on this topicdaraxonrasib · 64

Other directions

CBACBA Frontier Insights (CFI)Chinese Biopharmaceutical Association (CBA)www.cba-usa.org