LG Chem bets on AI-designed antibodies to solve solid tumor toxicity problem

South Korea-based LG Chem (KRX: 051910) has [agreed](https://www.businesswire.com/news/home/20260615992992/en/LabGenius-Therapeutics-and-LG-Chem-Enter-a-Research-Collaboration-Option-and-License-Agreement-to-Develop-an-AIML-Designed-Tumour-Targeting-Antibody) to fund discovery of a tumour-selective multispecific antibody through UK-based LabGenius Therapeutics, in a deal that could deliver triple-digit million dollar milestones if LG Chem exercises its option to license the resulting asset. The collaboration deploys LabGenius's EVA platform — a closed-loop, ML-guided antibody engineering system — against an undisclosed solid tumour antigen expressed across multiple cancer types. The structure, in which LG Chem funds all research and retains an option to in-license, transfers discovery risk to the larger partner while preserving LabGenius's platform economics across a growing roster of collaborators. Financial terms include an undisclosed upfront payment and potential early milestones during the preclinical phase, plus, if the option is exercised, clinical, regulatory, and commercial milestones described as "triple-digit million" in aggregate, along with royalties on net sales. ### Deal context The collaboration is designed to address a recognized limitation of current solid tumour immunotherapies: on-target, off-tumour toxicity, in which antibody drugs bind the intended antigen on healthy tissue as well as malignant cells. LabGenius's EVA platform is described by the company as capable of co-optimizing multiple antibody properties simultaneously — potency, selectivity, and manufacturability — by iterating between machine learning-guided design and high-throughput robotic screening in a closed loop. LabGenius claims discovery cycles of under six weeks. The platform has previously been applied to VHH-based (nanobody) formats, including LabGenius's own lead programme, [LGTX-101](https://labgeniustx.com/press-release-labgenius-therapeutics-announces-poster-presentation-at-the-aacr-annual-meeting-2026/), a trivalent Nectin-4 × CD3 T-cell engager for solid tumours. Under the deal's sequential structure, LabGenius will lead in vitro preclinical work before handing off to LG Chem for in vivo studies. LG Chem then holds the option to formally in-license the asset, at which point the full milestone and royalty package activates. LG Chem funds all activities throughout. No equity investment by LG Chem in LabGenius was disclosed. The target antigen has not been identified publicly, and geographic rights upon option exercise have not been specified. The deal is LabGenius's second disclosed partnership in six months, following a [December 2025 collaboration with Sanofi ](https://labgeniustx.com/sanofi/)covering multiple NANOBODY optimization programmes in inflammation. The repeat Sanofi engagement — itself a follow-on to an earlier undisclosed programme — indicates EVA has delivered measurable preclinical value, lending credibility to the LG Chem arrangement. The LG Chem deal is LabGenius's first disclosed oncology-focused multispecific antibody collaboration, extending the platform into a higher-value therapeutic area. For LG Chem, the deal follows a pattern of oncology pipeline construction through external sourcing. The company [acquired AVEO Oncology](https://www.aveooncology.com/lg-chem-completes-acquisition-of-aveo-oncology/) in January 2023, gaining its first FDA-approved oncology asset, and its first proprietary anti-cancer compound, LGC-D001, entered Phase I clinical study in the US in June 2024. Layering in a discovery-stage multispecific programme while LGC-D001 undergoes early clinical evaluation is consistent with building a sequenced pipeline rather than relying on a single asset. *** This article was generated with AI assistance and reviewed and edited by the AllSci editorial team Explore more at AllSci News: [https://allsci.com/news/](https://allsci.com/news/) --- Spot something wrong? [Report an issue with this article](https://newsgen-prod.reframedata.com/feedback/ai-designed-antibodies-solid-tumors-lg-chem-bets)