Command Palette

Search for a command to run...

BGB-53038, a pan-KRAS inhibitor, shelved at Phase 1 as BeOne pivots to a degrader

OncologySponsorJune 3rd, 2026·5 min read·10.5281/zenodo.20479005

BeOne Medicines terminated the first-in-human study of the pan-KRAS inhibitor BGB-53038 in KRAS-mutant solid tumors, citing lack of promising efficacy, internal prioritization, and a crowded competitive landscape. No efficacy results were posted, and the company is moving toward a KRAS degrader.

Mechanism Risk Score

ComponentPoints
Phase-weighted failure burden4.7 / 40
Archetype severity3.8 / 25
Temporal recency4.3 / 15
Genetic evidence deficit7.1 / 15
Programmatic saturation2.5 / 5

For KRAS in Advanced or metastatic solid tumors with KRAS mutation or amplification, the Mechanism Risk Score is 22/100 (green band). The score is a failure-burden index derived from Claidex post-mortems on this target–disease pair, not a probability of approval.

MRS 22/100 (GREEN). 1 programs across KRAS have been documented for KRAS in Advanced or metastatic solid tumors with KRAS mutation or amplification: 0 Phase 3, 0 Phase 2, 1 Phase 1 — of which 0 were efficacy failures, 0 safety, 0 biomarker, and 0 operational (enrollment, sponsor, or funding). The most informative failure on file is BGB-53038, a pan-KRAS inhibitor, shelved at Phase 1 as BeOne pivots to a degrader. This score quantifies the documented failure burden; the Open Targets association score of 0.52 reflects moderate genetic support, neither rescuing nor compounding the failure record. The MRS is not a prediction of future trial outcomes — it is a structured summary of the empirical record, recomputed live from the Claidex claims table, and intended to flag mechanisms where any new program must explicitly resolve each prior failure mode before pursuit is justified.

This score does not predict whether the next trial will succeed. It flags how heavy the documented mechanistic failure record is before a new program is justified.

Primary figure supporting this claim (BGB-53038 / KRAS / Advanced or metastatic solid tumors with KRAS mutation or amplification): BGB-53038, a pan-KRAS inhibitor, shelved at Phase 1 as BeOne pivots to a degrader

What was tried

BeOne Medicines, the company formerly known as BeiGene, ran a Phase 1a/1b first-in-human study of BGB-53038 in adults with advanced or metastatic solid tumors carrying a KRAS mutation or amplification (NCT06585488). BGB-53038 is an oral pan-KRAS inhibitor. The study was designed in five parts, spanning monotherapy dose escalation, monotherapy safety and dose expansion, and combination escalation and expansion with tislelizumab in non-small cell lung cancer and with cetuximab in colorectal cancer. Enrolled tumor types included non-squamous non-small cell lung cancer, colorectal cancer, pancreatic ductal adenocarcinoma, gastric and gastroesophageal junction cancer, and esophageal adenocarcinoma. The study opened on 13 December 2024 and reached its recorded primary completion on 23 April 2026, with 47 participants enrolled. BeOne terminated it and gave a combined reason: lack of promising efficacy, internal prioritization, and a highly competitive landscape. No results were posted to the registry.

The biological hypothesis

KRAS is among the most frequently mutated oncogenes in human cancer, and it is a genetically and mechanistically strong target. The Open Targets overall association score for KRAS against neoplasm is 0.5244, with high contributions from literature and clinical evidence, and the association against colorectal adenocarcinoma reaches 0.6839, driven by somatic mutation and pathway evidence. The first wave of approved drugs, sotorasib and adagrasib, locked onto the KRAS G12C variant in its inactive state. A pan-KRAS inhibitor aims to cover the broader mutational spectrum, including G12D, G12V, and G13D, which dominate colorectal and pancreatic disease. Preclinical work has shown that a pan-KRAS inhibitor can disable oncogenic signaling and tumor growth across multiple alleles (Nature, 2023, 10.1038/s41586-023-06123-3), and the medicinal chemistry path to broad coverage has been well described (Cancer Discov, 2022, 10.1158/2159-8290.CD-21-1331). The rationale was strong. The execution risk was in dose, selectivity, and timing.

What actually happened

The registry records termination with no posted efficacy data. The stated reasons combine a clinical observation and a portfolio decision. On the clinical side, the sponsor reported a lack of promising efficacy in a 47-patient first-in-human cohort that had not yet reported a formal response analysis. On the portfolio side, BeOne deprioritized BGB-53038 in favor of a preclinical KRAS degrader and shelved several oncology programs in its mid-2026 update (FierceBiotech, 2026). Proof-of-concept data for BGB-53038 had been expected in the second half of 2025, then moved to the first half of 2026, and the program was stopped around that window. So this is a strategic reprioritization, with an early and unquantified efficacy disappointment as the contributing trigger rather than a powered futility result.

Failure mechanism, best guess

The most likely driver is therapeutic index rather than target validity. KRAS itself is not in doubt. The difficulty with broad KRAS inhibition is that wild-type KRAS performs essential functions in normal tissue, so an agent that hits multiple mutant alleles can also engage wild-type protein and narrow the window between anti-tumor activity and on-target toxicity. Marketed KRAS inhibitors already show this tension. In FAERS, sotorasib reports are dominated by gastrointestinal effects and hepatotoxicity, the latter being a recognized class liability. A first-in-human pan-KRAS inhibitor that cannot push exposure high enough to drive deep, durable responses without dose-limiting toxicity will look exactly like this: tolerable, but without the early efficacy that justifies a crowded development path. The competitive context then converts a modest early read into a stop, because the bar to differentiate against approved G12C drugs and a deep pipeline of tri-complex RAS inhibitors is high. The field is also moving toward degraders, which remove the protein rather than occupy it, and BeOne's own pivot reflects that thesis (Science, 2024, 10.1126/science.adm8684; Cancer Cell, 2025, 10.1016/j.ccell.2025.07.006).

How to prevent this next time

The first lever is a Bayesian predictive-probability go-or-stop rule applied to the dose-escalation read, rather than a qualitative judgment after the fact. The predictive probability of eventual success integrates over the posterior for the treatment effect given interim data:

Tying expansion decisions to a pre-specified predictive-probability threshold would have made the go-or-stop point explicit before two combination arms were opened.

The second lever is a competitive-landscape red-team with an expected-value calculation. Entering a target that already has two approved drugs and at least 13 clinical-stage programs raises the commercial threshold, so the probability-weighted value of a me-too pan-KRAS inhibitor must clear that bar before combination expansion spends capital. Historical base rates reinforce the point, since first-in-class oncology assets advance from Phase 1 to approval only about 5 to 10 percent of the time (Wong, Siah, and Lo, Biostatistics, 2019), and a crowded field lowers that further. A translational validation sequence would add a third check: confirm a mutant-selective therapeutic index in the earliest cohorts using paired pharmacodynamic and tolerability readouts before committing to parallel tislelizumab and cetuximab combinations.

The single highest leverage change would have been defining a quantitative go-or-stop on a mutant-selective therapeutic index in the first dose-escalation cohorts, before opening parallel combination arms in an already crowded pan-KRAS field.

What this means for similar programs

KRAS is the opposite of an unvalidated target, so the failure here is not a referendum on the biology. It is a statement about sequencing and timing in a saturated landscape. The Claidex Mechanism Risk Score for KRAS sits at 22, in the green band, the lowest of the four bands. The genetic-deficit component is small because KRAS carries strong somatic and pathway evidence, and the phase-burden component is small because the program stopped in Phase 1, which is the correct shape for a strong target whose problem is sequencing and timing rather than validity. New entrants should expect that occupancy-based inhibition of mutant KRAS faces a tight therapeutic window, that the differentiation bar is now set by approved drugs and tri-complex inhibitors, and that degraders are the direction of travel for broad allele coverage (Cancer Discov, 2025, 10.1158/2159-8290.CD-25-0349).

Open questions

What were the actual response and tolerability numbers in the 47-patient cohort, and will they ever be disclosed? Did monotherapy fail to reach an exposure that engages wild-type-sparing anti-tumor activity, or did toxicity cap the dose first? And does BeOne's degrader carry a real selectivity advantage over the inhibitor it replaces, or is the pivot mainly a way to reset the competitive clock?

Sources

Related failure claims

Linked claims sharing target, indication, or failure mechanism.