值得关注的药物研发趋势：CAS Insights 网络研讨会亮点
从微生物组科学到人工智能驱动的建模，药物研发领域正以令人着迷的方式飞速发展。在我们近期的 CAS Insights 网络研讨会上，来自拜耳消费者健康 (Bayer Consumer Health) 和 CAS 的专家共同探讨了正在重塑制药研究的新兴突破。小组成员讨论了精准医疗、早期诊断和新型治疗策略方面的进展。他们的对话为我们提供了一个快速且引人入胜的视角，一窥药物开发的未来。
欲了解更多相关主题，请参阅我们的 CAS Insights 文章，药物研发的最新趋势，其中 CAS 科学家介绍了精准医疗、癌症免疫疗法、神经退行性疾病检测等领域的突破性进展。
微生物组科学迈向系统性健康领域
拜耳消费者健康全球研发前端创新科学创新总监 Rami Ammar 博士首先强调了微生物组研究在药物研发中日益重要的作用。虽然益生菌传统上与消化健康相关，但 Ammar 博士解释说，该领域目前正在探索更广泛的应用，包括心理健康、代谢紊乱和皮肤病学。他将微生物组描述为“隐藏的器官”，并指出了其在遗传学上的主导地位：
“药物研发中最引人注目的趋势之一……是益生菌的应用范围已超越肠道健康，甚至超越了益生菌本身……涵盖了包括益生菌、益生元和后生元在内的广义生物制剂。”
Ammar 博士还指出，药物微生物组学和毒理微生物组学等新兴领域：
“……对于理解我们的肠道微生物组或整体微生物组如何与药物和毒素相互作用，以及如何制定个性化且有效的治疗策略，正变得越来越重要。”
他还列举了针对艰难梭菌感染和抗生素相关腹泻等疾病的获批益生菌疗法案例，强调了基于微生物组的干预措施的临床意义。他最后充满希望地总结道：
“随着我们继续探索微生物组与系统性健康之间复杂的联系，我相信我们正处于突破性发现的前夜，这些发现将有助于彻底改变我们对医学的理解，并为创新治疗方案铺平道路。”
神经退行性疾病的早期诊断
我们的第二位小组成员，CAS 科学分析与洞察经理 Angela Zhao 博士，重点讨论了用于神经退行性疾病早期诊断的生物标志物的重要性，特别是针对阿尔茨海默病和帕金森病。她解释说，虽然生物标志物通常用于监测身体健康（如胆固醇和血糖水平），但在评估大脑健康方面仍存在空白。
“我们目前缺乏一套有效的生物标志物来评估大脑的健康状况。在常规体检中，人们通常不愿进行脊髓液采样或 PET 扫描。但试想一下，如果我们能通过血液或尿液样本获取大脑健康的生物标志物，那将是多么令人惊叹。”
Zhou 博士还指出，近期阿尔茨海默病诊断技术的监管获批，标志着该领域正朝着更易于获取和预防性的方向发展。
PROTACs 在靶向治疗中获得关注
CAS 高级客户成功专家 Gary Gustafson 博士参与了我们的讨论，他拥有超过 25 年的制药行业药物化学家经验，曾是近期获得 FDA 批准的药物 Olutasidenib 开发团队的一员。Gustafson 博士讨论了PROTACs（蛋白降解靶向联合分子）的近期兴起，它们为消除致病蛋白提供了一种新颖的方法。与抑制蛋白质功能的小分子药物不同，PROTACs 通过招募细胞的天然机制来降解蛋白质。正如 Gustafson 博士所言：
“A single PROTAC can actually degrade multiple proteins. They are less hampered by protein mutations or overexpression of the proteins that sometimes can hinder small molecules... “
He pointed out that more than 80 drugs are in clinical development or some kind of clinical evaluation, and that more than one hundred organizations are working in this area.
AI and digital twins: Toward more ethical and translatable research
The panel also explored how artificial intelligence is helping researchers simulate human biology more accurately, reducing reliance on animal models. Dr. Ammar described Bayer’s work with organ-on-chip technologies and physiologically based pharmacokinetic (PBPK) modeling to create digital twins of human systems:
“Reducing the animal use is not only for ethical purposes, but also for transformative and more human translatable results.”
Dr. Zhou added that AI enables more complex biomarker profiling, potentially involving dozens or even hundreds of indicators. Dr. Gustafson emphasized the importance of high-quality data in training AI algorithms and predictive analytics, noting that CAS tools like BioFinder are helping researchers prioritize compounds and explore drug-repurposing opportunities.
Final thoughts: Collaboration, curiosity, and convergence
As the discussion wrapped up, Dr. Zhou encouraged the audience to stay curious and explore adjacent fields to maintain a broad perspective, noting that many great discoveries involve researchers from different fields. It was clear from the wide-ranging, lively discussion that the future of drug discovery will be shaped by the convergence of biology, data science, and engineering—whether it's leveraging the microbiome for systemic health, identifying early biomarkers for neurodegenerative diseases, or designing therapies that eliminate proteins. While tools like AI simulations and organ-on-chip technologies are gaining traction, researchers who work together, with data connected across many parameters, will be better equipped to translate their shared insights into impact.
Watch the full webinar
To hear directly from the experts and explore these trends in more depth, watch the full webinar. The session includes detailed examples, data visualizations, and additional commentary on the future of pharmaceutical research.
Questions from the webinar
We didn’t have time to get to all our audience questions during the hour, so we asked our panelists to provide some thoughts.
Question: Can PROTACS target proteins that were previously considered “undruggable?”
Yes—this is one of the most exciting aspects of PROTAC technology. Unlike traditional small molecules that inhibit protein function by binding to active sites, PROTACs work by tagging proteins for degradation via the ubiquitin–proteasome system. This allows them to target proteins that lack accessible binding pockets or are involved in complex interactions, such as transcription factors and scaffold proteins.
Question: Do you have any insights into radiopharmaceutical conjugates? Which radionuclides do you believe are going to be hot topics going forward?
Radiopharmaceutical conjugates (RDCs) are gaining momentum, especially in oncology, where they combine targeted delivery with potent radiation. Lutetium-177 (^177Lu) remains a leading radionuclide due to its favorable half-life and beta-emission profile, but actinium-225 (^225Ac), an alpha-emitter, is emerging because of its ability to deliver highly localized cytotoxicity. Clinical trials are increasingly exploring ^225Ac-labeled drugs for prostate and neuroendocrine cancers, and major pharma investments suggest alpha therapies will be a key focus going forward. You may be interested in our recent CAS Insights article, “Theranostics: How nuclear medicine is changing precision oncology.”
Question: How can AI and quantum computing help with high-throughput screening of some of the molecular PROTAC targets?
AI is transforming high-throughput screening (HTS) by enabling virtual screening of vast chemical libraries, reducing cost and time while improving hit accuracy. Machine learning models can predict PROTAC activity and optimize compound selection without physical assays. Quantum computing adds another layer by simulating molecular interactions at atomic precision—especially useful for modeling complex PROTAC functions.
Question: CAR-T has been huge for blood cancers, but solid tumors are tougher. What’s the most promising innovation you’ve seen that could finally crack that challenge?
We wrote about recent CAR-T advances last October. Since then, one of the most promising innovations we’ve seen is a new CAR-T cell therapy developed at the Mayo Clinic that targets PD-L1, a protein commonly overexpressed not only on tumor cells but also within the immunosuppressive tumor microenvironment. By designing CAR-T cells to recognize PD-L1, researchers were able to simultaneously attack both the tumor and its protective surroundings, overcoming a major barrier to treatment. In preclinical models of breast cancer, lung cancer, melanoma, and glioblastoma, this approach led to significant tumor reduction.
