Pan-RAS molecular glue vs KRAS G12C inhibitors compared

How Covalent KRAS G12C Inhibitors Work — and Where They Diverge

Sotorasib (Amgen, FDA-approved May 2021) and adagrasib (Bristol Myers Squibb/Mirati, FDA-approved December 2022) achieve KRAS inhibition by exploiting a structural vulnerability unique to the G12C mutation: the substitution of glycine for cysteine at codon 12. This cysteine’s exceptional nucleophilicity enables irreversible covalent bond formation with both inhibitors at the switch-II pocket (SII-P) region of KRAS. By binding covalently, they trap KRAS in its inactive GDP-bound state, preventing the nucleotide exchange required for GTP loading and downstream effector activation.

Despite sharing the same primary mechanism, sotorasib and adagrasib diverge meaningfully in their isoform selectivity. Sotorasib’s reduced dependence on histidine-95 (H95) for binding stability gives it activity against NRAS G12C and HRAS G12C in addition to KRAS G12C, with 5-fold greater potency against NRAS G12C compared to KRAS G12C. Adagrasib, by contrast, is strictly KRAS-specific: its strong, irreplaceable interaction with H95 — a residue that differs between RAS isoforms — confines its activity to KRAS alone.

This structural nuance has direct clinical consequences. Secondary mutations at H95 confer resistance to adagrasib but not to sotorasib, while mutations at Y96 cause cross-resistance to both agents. The state-selective inhibition strategy also preserves wild-type KRAS signalling, theoretically reducing toxicity in normal tissues where wild-type RAS remains essential for cellular homeostasis — a property that distinguishes both agents from broader RAS-targeting approaches.

The Molecular Glue Mechanism: Pan-RAS Inhibition Explained

Daraxonrasib (RMC-6236, Revolution Medicines) and BI-2865 (Boehringer Ingelheim) inhibit RAS through a fundamentally different principle: instead of covalently modifying a mutation-specific residue, they act as molecular glues that recruit regulatory proteins to RAS, forming abnormal ternary complexes that trap RAS in its active GTP-bound state while simultaneously disrupting critical RAS-effector interactions.

Daraxonrasib targets the RAS-ON (GTP-bound active) state across KRAS, NRAS, and HRAS isoforms, covering G12C, G12D, G12V, G12A, and wild-type RAS. It achieves this through formation of a tri-complex comprising RAS-GTP, the small molecule ligand, and a guanine nucleotide exchange factor (GEF) — a paradoxical mechanism that locks RAS in its active conformation while preventing productive effector binding. This is the mechanistic inverse of covalent G12C inhibitors, which stabilise the inactive state.

BI-2865, developed by Boehringer Ingelheim and currently at preclinical stage, operates through comparable molecular glue principles but with a distinct inhibitory route. Structural studies show that BI-2865 binds KRAS at a site distinct from the classical switch-II pocket and promotes KRAS dimerization as a key inhibitory mechanism, preventing RAF kinase binding and downstream MAPK pathway activation. This dimerization-based route represents an alternative mechanistic path to pan-KRAS inhibition, complementing the tri-complex approach of daraxonrasib.

“Pan-RAS inhibitors target the RAS-ON state across all isoforms and mutation subtypes — the mechanistic inverse of covalent G12C agents, which stabilise the inactive GDP-bound conformation of a single mutation.”

Therapeutic Scope: Which Patients Can Each Approach Reach?

The mutation-specific nature of covalent G12C inhibitors is the most fundamental constraint on their clinical impact. KRAS G12C represents approximately 13% of lung adenocarcinomas in non-Asian populations and only 3% in East Asian populations, while other KRAS mutations — G12D, G12V, G12A, G13D, and Q61 mutations — collectively account for the majority of KRAS-driven cancers. Sotorasib and adagrasib have no activity against these non-G12C subtypes.

The clinical consequence is stark for pancreatic ductal adenocarcinoma (PDAC). G12C mutations represent only approximately 5% of PDAC cases, whereas KRAS mutations at codon 12 collectively account for approximately 90% of PDAC cases. Daraxonrasib’s pan-RAS coverage therefore positions it as a potentially transformative therapy for PDAC — the deadliest solid malignancy — where G12C inhibitors offer no meaningful benefit to the vast majority of patients. According to the National Cancer Institute, pancreatic cancer remains one of the most lethal malignancies with a five-year survival rate below 13%, underscoring the urgency of broader RAS-targeting strategies.

Resistance Mechanisms and the Case for Pan-RAS Breadth

Covalent KRAS G12C inhibitors face two primary resistance categories: on-target resistance through secondary KRAS mutations, and off-target resistance through activation of bypass pathways. Deep mutational scanning has identified extensive secondary mutation landscapes, and critically, these resistance mutations exhibit differential drug-specificity patterns between sotorasib and adagrasib — a distinction with direct sequential therapy implications.

Mutations at codons H95, Q99, and G13D confer strong resistance to adagrasib while remaining sensitive to sotorasib. Conversely, R68S and Y96 mutations cause cross-resistance to both agents, limiting the universality of sequential switching strategies. Resistance emerges rapidly: adaptive ERK rebound occurs often within 72 hours of initial treatment. Off-target resistance mechanisms include KRAS amplification on extrachromosomal DNA, activation of wild-type RAS isoforms through receptor tyrosine kinase (RTK) reactivation, and mutations in downstream effectors including BRAF and MEK1.

Pan-RAS inhibitors demonstrate a structural mechanistic advantage here. Because daraxonrasib and BI-2865 target both mutant and wild-type RAS isoforms, compensatory activation of wild-type RAS represents an ineffective escape strategy — the compensatory pathway is simultaneously suppressed. Additionally, pan-RAS inhibitors retain activity against many of the secondary KRAS mutations that confer resistance to G12C-specific agents. Preclinical and clinical data indicate that RAS-ON multi-selective inhibition overcomes clinically relevant resistance mechanisms to mutation-selective KRAS inhibitors, with compensatory ERK rebound delayed or abrogated compared to G12C inhibitors. As Nature has documented in recent oncology research, the durability of targeted therapy responses is increasingly linked to the breadth of pathway suppression achieved at treatment initiation.

“Because pan-RAS inhibitors target both mutant and wild-type RAS isoforms, compensatory activation of wild-type RAS through EGFR reactivation — the dominant resistance mechanism in CRC — represents an ineffective escape strategy.”

Clinical Efficacy Across Tumor Types: NSCLC, PDAC, and CRC

Clinical efficacy data reveal a consistent pattern: covalent G12C inhibitors perform best in NSCLC, show limited utility in PDAC, and demonstrate dramatically reduced activity in colorectal cancer — a profile that mirrors the biological resistance mechanisms described above.

NSCLC: Approved but Modest

In NSCLC, sotorasib’s CodeBreaK 200 Phase 3 data demonstrated superior progression-free survival versus docetaxel (5.6 vs 4.5 months), and adagrasib similarly improved PFS (5.5 vs 3.8 months). Objective response rates of approximately 37–43% across G12C NSCLC trials represent genuine clinical benefit. However, these improvements are modest, reflecting the limitations of monotherapy in the context of rapid resistance emergence — often within 72 hours as adaptive ERK rebound occurs. According to the FDA, both approvals were based on these pivotal trial data, marking a historic first for direct KRAS inhibition.

PDAC: The Critical Divergence

Pancreatic ductal adenocarcinoma represents the most significant divergence point between the two inhibitor classes. G12C inhibitors offer no meaningful benefit to the approximately 95% of PDAC patients whose tumours carry non-G12C KRAS mutations. Daraxonrasib has received FDA breakthrough therapy designation for PDAC with KRAS mutations at codon 12, with promising response rates and considerably increased patient survival as second- or third-line treatment. Given that KRAS mutations at codon 12 collectively account for approximately 90% of PDAC cases, this positions pan-RAS inhibitors as potentially transformative for a disease where treatment options remain severely limited.

CRC: Where G12C Inhibitors Struggle Most

In colorectal cancer, both sotorasib and adagrasib demonstrate dramatically reduced efficacy compared to NSCLC, with low response rates and rapid resistance emergence. The dominant mechanism is compensatory activation of wild-type RAS through EGFR reactivation — a pathway that G12C inhibitors, by design, cannot suppress. Pan-RAS inhibitors’ simultaneous targeting of wild-type RAS may overcome this limitation, though this hypothesis awaits prospective clinical validation. Research published through ASCO has highlighted CRC as a priority indication for next-generation RAS-targeting strategies.

Combination Strategies and the Future of RAS-Targeted Therapy

Monotherapy with either inhibitor class faces inherent limitations, and the optimal clinical strategy will likely involve combination approaches tailored to tumor mutation profiles and resistance mechanisms. Preclinical and early clinical data support combining G12C inhibitors with MEK inhibitors (trametinib) to suppress MAPK pathway rebound, EGFR inhibitors (afatinib) to prevent compensatory RTK-RAS reactivation, anti-PD-1/PD-L1 immunotherapy to leverage KRAS inhibition-induced immune activation, and CDK4/6 inhibitors (palbociclib) to address cell-cycle pathway activation.

Pan-RAS inhibitors, by simultaneously targeting wild-type RAS, may require different combination partners. Their broader suppression of RAS signalling could achieve superior efficacy through more complete pathway suppression, potentially reducing the compensatory signalling that drives resistance to G12C-specific agents. The field will likely evolve toward personalised combination strategies leveraging both modalities based on tumour mutation profiles — with pan-RAS inhibitors potentially becoming first-line therapy for non-G12C KRAS mutations and combination approaches maximising durability across all RAS-driven cancers.

The development of sequential therapy strategies also warrants attention. Patients developing resistance to adagrasib through H95 mutations might retain sensitivity to sotorasib, and vice versa for other resistance mutations — creating a potential sequential treatment opportunity. However, Y96 mutations cause cross-resistance to both agents, limiting this strategy’s universality. As PatSnap’s life sciences intelligence platform tracks, patent activity around RAS combination therapies has accelerated substantially, reflecting the industry’s recognition that durable responses will require multi-target approaches. The PatSnap Insights blog continues to monitor emerging patent filings across this rapidly evolving therapeutic space.