⦿ 2021年以来，英矽智能在其专有Pharma.AI平台的帮助下已成功提名22款临床前候选化合物（PCC），仅在今年内完成其中5款候选药物的提名。

⦿ ISM1745的新型骨架以英矽智能Chemistry42的从头设计结果为基础，该生成式AI平台集成40多个生成模型。

⦿ 该候选化合物在多个癌症模型中表现出体内抗肿瘤活性，不仅显示出单药治疗的良好体内药效，还拥有与MAT2A抑制剂等靶向药物、化疗、免疫疗法联合使用的治疗潜力。

马萨诸塞州剑桥市，2024年12月27日 --- 由生成式AI驱动的临床阶段生物科技公司英矽智能宣布提名ISM1745作为临床前候选药物（PCC），一款潜在“同类最佳”MTA协同PRMT5抑制剂。该化合物具有AI驱动的创新分子骨架，用于治疗MTAP缺失癌症。以Chemistry42平台的从头设计结果为基础，英矽智能成功提名今年的第五款PCC，使2021年以来的PCC提名总数达到22个。

蛋白精氨酸甲基转移酶5（PRMT5）参与调解包括RNA剪接、DNA损伤修复和翻译的关键细胞过程。PRMT5表达升高在多种癌症中都有观察到，且与不良预后相关。在MTAP缺失的肿瘤细胞中，积累的甲硫腺苷（MTA）提升细胞对于PRMT5抑制的敏感性，同时与PRMT5结合形成新的MTAP缺失癌症特异性靶点，PRMT5因此作为合成致死潜力靶点得到关注。

英矽智能联合首席执行官兼首席科学官任峰博士表示，“甲硫腺苷磷酸化酶（MTAP）缺失在大约15%的人类癌症中都有发生，在对标准疗法产生耐药的晚期患者中频率更高，突显出巨大的未满足需求。由Chemistry42辅助设计出的ISM1745具有创新分子骨架，使针对MTAP缺失癌症特异性靶点的独特相互作用机制成为可能。基于令人鼓舞的临床前结果，我们期待进一步评估ISM1745这一英矽智能AI驱动的创新项目。"

利用基于结构的生成策略，英矽智能科学家以AI驱动的生成化学引擎Chemistry42的从头分子设计结果为起点，通过 AI辅助的分子评估针对药效团、成药性、可开发性和其他因素进行筛选，最终在多轮优化后，提名ISM1745作为临床前候选化合物。

根据临床前数据，ISM1745在较低剂量下在多个动物模型中展示了强大的体内疗效，不仅作为单药治疗显示出良好的效果，还展现出与化疗、靶向药物（其与MAT2A抑制剂的联用潜力已在临床前体内外试验中获得验证）和免疫疗法联合使用的广阔潜力。此外，该化合物在MTAP-WT细胞系中表现出良好的选择性，药物-药物相互作用（DDI）风险得以降低，同时具有良好的药物代谢和药代动力学（DMPK）特性。

ISM1745已在CDX模型中展现出
与英矽智能MAT2A抑制剂的联用潜力

英矽智能创始人兼首席执行官Alex Zhavoronkov博士表示，“通过今年内的第五款PCC提名，英矽智能AI驱动项目的可复现性再一次得到了验证。自2014年成立以来，我们自有的Pharma.AI平台持续扩展，从论文中的算法发展成为今天的综合性AI解决方案。我们希望确保生成式AI能够加速昂贵且耗时的药物发现和开发过程，让更多人能够活得更长、更健康。”

A Potentially Best-in-class Small Molecule Inhibitor for the Treatment of MTAP-deleted Cancers

⦿ Since 2021, Insilico Medicine has successfully nominated 22 preclinical candidates (PCCs) with the help of its proprietary Pharma.AI platform, among which 5 were nominated just this year.

⦿ The novel scaffold of ISM1745 is based on de novo generation results of Insilico’s Chemistry42, the generative AI platform combining more than 40 generative models.

⦿ With in vivo anti-tumor activity validated in multiple cancer models, the candidate compound showed robust in vivo efficacy as monotherapy as well as combination potential with chemotherapies, targeted agents including MAT2A inhibitor, and immunotherapies.

CAMBRIDGE, Mass., December 27, 2024 — Insilico Medicine (‘Insilico’), a clinical-stage generative artificial intelligence (AI)-driven drug discovery and development company, today announced the nomination of ISM1745, a potentially best-in-class MTA cooperative PRMT5 inhibitor with AI-powered novel scaffold, as preclinical candidate (PCC) for the treatment of MTAP-deleted cancers. Based on de novo design results of Chemistry42, ISM1745 marks the fifth PCC nomination achieved by the Insilico team this year, bringing the total number since 2021 to 22.

Protein arginine methyltransferase 5 (PRMT5) regulates essential cellular processes, including RNA splicing, DNA damage repair, and translation, and elevated PRMT5 expression is observed in various cancers and correlates with poor prognostic outcomes. In MTAP-deleted tumor cells, the accumulating methylthioadenosine (MTA) increases sensitivity to PRMT5 inhibition, while binding with PRMT5 to form a novel MTAP-deleted cancer specific target, showing PRMT5’s potential as a synthetic lethality target.

"MTAP (methylthioadenosine phosphorylase) deletion occurs in about 15% of human cancers, with higher frequency in later line patients suffering from resistance to standard therapies, highlighting great unmet needs," said Feng Ren, PhD, Co-CEO and Chief Scientific Officer of Insilico Medicine. "The novel scaffold of ISM1745 powered by Chemistry42 enables unique interaction with the specific target for MTAP deleted cancers. With promising preclinical results, we look forward to further evaluation of ISM1745 as another Insilico AI-driven program."

Leveraging the structure-based generation strategy, Insilico scientists started with de novo molecule design outputs of Chemistry42, the AI-powered generative chemistry engine. After AI-supported molecule evaluation based on pharmacophores, druglikeness, developability and other factors, ISM1745 was selected as a preclinical candidate following rounds of optimization.

According to the preclinical data, ISM1745 demonstrated robust in vivo efficacy at a low dosage in multiple animal models, with promising potency as monotherapy and broad potential in combination with chemotherapies, targeted agents, and immunotherapies. The combination potential with MAT2A inhibitor has been validated in both in vitro and in vivo tests. Additionally, the compound shows excellent selectivity over MTAP WT cell lines, minimized drug-drug interaction (DDI) risk, and favorable Drug Metabolism and Pharmacokinetics (DMPK) profiles.

"As the fifth PCC nomination solely in this year, the achievement further proves the reproducibility of Insilico's programs," said Alex Zhavoronkov, PhD, Founder and CEO of Insilico Medicine. "Since the foundation of Insilico in 2014, our Pharma.AI platform has been growing nonstop, from algorithms in the papers to the comprehensive AI toolbox today. We want to make sure that generative AI accelerates the costly and lengthy process of drug discovery and development, so more people could live longer and healthier lives."

In 2016, Insilico first described the concept of using generative AI for the design of novel molecules in a peer-reviewed journal, which laid the foundation for the commercially available Pharma.AI platform. Since then, Insilico keeps integrating technical breakthroughs into Pharma.AI platform, which is currently a generative AI-powered solution spanning across biology, chemistry, medicine development and science research. Powered by Pharma.AI, Insilico has nominated 22 preclinical candidates in its comprehensive portfolio of over 30 assets since 2021 and has received IND clearance for 10 molecules.

In early 2024, Insilico published a Nature Biotechnology paper presenting the entire R&D journey from AI algorithms to Phase II clinical trials of ISM001-055, the company's lead drug pipeline with AI-discovered target and AI-designed structure. Following that, Insilico has recently announced positive preliminary results from a Phase IIa trial (NCT05938920), where ISM001-055 showed favorable safety and tolerability across all dose levels, as well as dose-dependent response in forced vital capacity (FVC), after only 12 weeks of dosage.