LY3537982

Olomorasib is Eli Lilly's oral KRAS G12C inhibitor. KRAS is a protein that acts like an on/off switch for cell growth, and the G12C mutation jams the switch in the 'on' position, driving about 13% of non-small cell lung cancers. The Phase 3 SUNRAY-01 trial (NCT06119581, enrolling 1,264 patients) tests olomorasib plus pembrolizumab - a checkpoint inhibitor that releases the brakes on the immune system's attack on cancer - with or without chemotherapy, against pembrolizumab plus chemotherapy as first-line treatment for KRAS G12C-mutant advanced NSCLC [1][2]. The bet: a next-generation G12C inhibitor with cleaner safety than first-generation sotorasib (Amgen) and adagrasib (BMS) can capture the first-line market the incumbents failed to win [3][8]. Lilly also runs a parallel Phase 3 (NCT06890598) in resected and unresectable disease [4]. Stakes are high - sotorasib peaked at roughly $340M in 2023 sales after its second-line approval, well below initial bullish projections, and the G12C class needs a true first-line winner to justify the continued investment [11].

Novel compound, not yet approved anywhere. Two Phase 3 programs run in parallel: SUNRAY-01 in first-line advanced NSCLC (NCT06119581) and a perioperative/adjuvant trial in resected or unresectable disease (NCT06890598), launched in 2025 [2][4]. First-in-human results were published in Nature Communications in 2026 (Murciano-Goroff et al.), showing pan-tumor activity across KRAS G12C-mutant solid tumors from the Phase 1a/1b expansion (NCT04956640) [1][5]. No FDA breakthrough therapy or fast track designation has been publicly disclosed for olomorasib - interesting given that the first wave of G12C drugs received accelerated approval pathways. SUNRAY-01 readout most likely lands in 2027-2028 given current recruiting status and the 1,264-patient target; we cannot confirm current enrollment count from public sources, so the readout window has real uncertainty. Lilly's 2025 and 2026 10-K filings both flag olomorasib as a Phase 3 oncology asset alongside imlunestrant and pirtobrutinib expansions [6]. Timing matters strategically: if olomorasib reads positive in 2027, it slots into Lilly's oncology buildout right when post-tirzepatide diversification becomes the investor story.

KRAS is the protein that flips cell growth on and off. The G12C mutation swaps glycine for cysteine at position 12, which locks the switch in the 'on' state and drives uncontrolled division. About 13% of non-small cell lung cancers carry this exact mutation, along with smaller fractions of colorectal and pancreatic cancers [7]. The cysteine creates a chemical handle that small molecules can grab onto covalently, which is what sotorasib (Amgen's Lumakras) and adagrasib (BMS via Mirati, branded Krazati) do [3][8]. Olomorasib is the same class - covalent G12C binder - but engineered with a pharmacokinetic profile intended to improve potency and CNS penetration. Brain penetration matters clinically because roughly 30-40% of advanced NSCLC patients develop brain metastases, and the first-generation G12C drugs have limited intracranial activity. The extent to which olomorasib's CNS exposure translates into intracranial response is not yet fully characterized in published data - the Phase 1a/1b paper reports systemic activity but stops short of definitive intracranial efficacy numbers [1]. The mechanism is genetically airtight: G12C mutations are drivers, not passengers, and two approved drugs prove the target is druggable in humans. What remains unproven is first-line. Sotorasib and adagrasib settled into second-line monotherapy because combination data in 1L was modest and liver enzyme spikes made stacking with checkpoint inhibitors awkward [9]. The Murciano-Goroff Nature Communications paper framed olomorasib's tolerability profile as combination-friendly [1]. Whether that translates to a real win against pembrolizumab plus chemotherapy is the entire SUNRAY-01 thesis.

SUNRAY-01 (NCT06119581) is a randomized Phase 3 in first-line KRAS G12C-mutant advanced or metastatic NSCLC, target enrollment 1,264 patients [2]. Design includes a dose optimization and safety lead-in (Part B), which signals Lilly is taking Project Optimus seriously - the FDA's push to find the lowest effective oncology dose rather than the historical maximum tolerated dose, now an expectation for all new oncology programs. The registrational portion compares olomorasib plus pembrolizumab, with or without chemotherapy, against pembrolizumab plus chemotherapy, the current 1L standard for non-squamous NSCLC without actionable mutations. The primary endpoint in Part B is treatment-emergent adverse events (TEAEs, i.e., side effects appearing or worsening during treatment) for the safety lead-in. Phase 3 efficacy endpoints are most likely progression-free survival (PFS, time until tumor growth or death) and overall survival (OS, time until death from any cause) based on standard 1L NSCLC designs, though this should be confirmed against the registered NCT record before any forecast depends on it. The comparator is the actual standard of care, not placebo, meaning the benefit bar is real. Enrollment target is large for a biomarker-selected population (about 13% of NSCLC patients test G12C positive), so recruiting will take 2-3 years across hundreds of global sites. NCT06890598 (n=700, resected and unresectable disease) extends the franchise into earlier-stage settings with disease-free survival as the primary endpoint [4]. That earlier-line bet is a smart hedge: if first-line wins, adjuvant unlocks a larger and longer-duration market. Design is solid. The risks sit in execution and combination tolerability, not in study architecture.

Our model gives this drug a 37% chance of eventually reaching approval. That starting point comes from the historical approval rate for Phase 3 drugs in this area, which is about 48%. The estimate is nudged up by an unusually large trial enrollment, more secondary endpoints than typical, and an 8-arm study design - but pulled back down by weak earlier-phase results. The remaining factors were close to average for this stage, so they left the number near where it started.

Efficacy and combination tolerability are the central concerns. The CodeBreaK 100/101 sotorasib-plus-pembrolizumab data published by Spira et al. in 2022 showed grade 3-4 hepatotoxicity (ALT/AST elevations) as the dominant treatment-related adverse event, with 88% of grade 3-4 hepatic events occurring outside the DLT window and most requiring corticosteroids or sotorasib discontinuation [9]. If olomorasib has the same on-target liability - plausible since both drugs hit the same cysteine - the SUNRAY-01 combination thesis collapses. Biomarker selection is clean at the headline level (KRAS G12C is a yes/no genetic call), but the population is heterogeneous within G12C: co-mutations in KEAP1, STK11, and TP53 (genes that, when mutated, suppress immune activity and blunt response to checkpoint inhibitors like pembrolizumab) strongly affect immunotherapy efficacy, and SUNRAY-01 does not stratify on those. Safety: G12C inhibitors carry class-wide liver enzyme elevations and gastrointestinal toxicity, and combining with pembrolizumab raises immune-related adverse event risk, particularly hepatitis. Execution risk is real: 1,264 patients in a biomarker-selected population requires global enrollment across 200+ sites and 2-3 years of recruiting. Competitive risk is severe and multi-layered. Divarasib (Roche/Genentech, GDC-6036) is the most direct Phase 3 competitor: a covalent G12C inhibitor with Phase 1 NSCLC monotherapy data showing 53% response rate and 13-month PFS, more potent and reportedly cleaner than first-generation drugs, now in late-stage development [12]. If divarasib reaches market with similar or better efficacy and tolerability, olomorasib's intraclass advantage compresses. One layer further out, Revolution Medicines' daraxonrasib (RMC-6236) is a multi-selective RAS(ON) inhibitor that hits G12C, G12D, G12V, and other oncogenic active-state variants - not pan-KRAS in the wild-type sense - and the RASolute 302 Phase 3 in PDAC reported a median OS of 13.2 months vs 6.7 months for chemotherapy (HR 0.40) [10]. If RMC-6236 extends to NSCLC with similar magnitude, it could leapfrog the entire single-mutation G12C class. Commercial risk: payers may not pay a premium over generic chemotherapy plus pembrolizumab without a clear survival benefit. Sotorasib peaked at roughly $340M in 2023 - a cautionary tale for the entire class [11]. Pricing power requires an OS signal, not just PFS, because payers benchmark new biomarker-targeted oncology drugs against generic backbones and only pay premium prices when survival benefit is unambiguous. Patent/exclusivity timeline for olomorasib is not summarized in public 10-K disclosures at the molecule level; this is a gap worth filling before any long-term commercial model.

Worth watching, conditionally. The signal to watch is SUNRAY-01's safety lead-in (Part B) data, expected late 2026 or 2027. If olomorasib plus pembrolizumab shows manageable hepatotoxicity at a dose that keeps efficacy intact, this becomes a real bet. If the combination forces dose reductions like sotorasib did, the program is in trouble well before Phase 3 readout. The Nature Communications first-in-human paper is the strongest validation yet that olomorasib has the molecular profile to be the next-generation G12C winner [1], but Phase 1 monotherapy data has betrayed plenty of oncology programs that fell apart in combination. Commercial stakes are real but bounded: the US addressable population for KRAS G12C-mutant advanced NSCLC is approximately 25,000 new patients per year, and moving from second-line to first-line more than doubles the eligible pool (because most patients never reach later lines). A first-line label combined with the adjuvant Phase 3 program could plausibly push peak annual sales into the low-single-digit billions if olomorasib displaces the incumbents and survives the divarasib threat - meaningful but a niche asset for Lilly's FY2025 revenue base of roughly $65B per the 2025 10-K [6], dominated by tirzepatide. Check back at the next major oncology conference (ASCO or ESMO) for SUNRAY-01 safety lead-in data. The cleaner read on the entire G12C class will come from Revolution Medicines' daraxonrasib readouts in NSCLC and from divarasib's Phase 3 progression: if multi-RAS works or if divarasib gets there first, single-mutation G12C drugs face obsolescence fast [10][12]. For Lilly's oncology pipeline credibility this matters; for the stock, it does not.