返回新闻
产品展示AI Understanding 简报

中国启动人工智能开发的治疗肺纤维化药物rentosertib III期试验

Insilico Medicine 的 AI 设计分子 Rentosertib 在中国进入了一项 320 名患者的 III 期研究,这标志着第一个达到特发性肺纤维化临床测试阶段的 AI 药物。

4 min readRead the linked source
Source-provided image accompanying China starts Phase III trial of AI‑developed drug rentosertib for pulmonary fibrosis
来源参考来源记录
出版商
ua.news
来源链接
ua.newshttps://ua.news/en/farmatsevtika/u-kitayi-rozpochali-iii-fazu-viprobuvan-shi-preparatu-proti-fibrozu-legen-daily-maverick
来源类型
链接来源——主要来源状态尚未确定。
背景60 秒内了解这一点

从这里开始

关键术语

管道
预处理、模型步骤和后处理阶段的有序工作流程。
测试一下自己AI 模型解释测验
Source video from ua.news · shown with attribution.

发生了什么

Insilico Medicine reported that on September 10 the first patient was dosed with rentosertib, an AI‑designed small‑molecule therapy for idiopathic pulmonary fibrosis (IPF). The Phase III trial will enroll 320 participants across 47 Chinese medical centers, with daily dosing for one year. Earlier Phase II data, involving 71 patients, showed that the highest dose of rentosertib improved lung capacity by an average of 98.4 ml over 12 weeks, while the placebo arm showed a decline in the same measures. Daily Maverick highlighted that rentosertib is the first drug discovered through artificial‑intelligence methods to advance to Phase III trials. The article also notes that Insilico has other AI‑generated candidates in development, including a garutadustat molecule for inflammatory bowel disease and the GB‑0895 antibody from Generate:Biomedicines for severe asthma.

On September 10, the first patient in China received rentosertib, an AI‑designed small‑molecule intended to treat idiopathic pulmonary fibrosis (IPF). The Phase III trial is designed to enroll 320 patients at 47 clinical sites, with participants taking the drug daily for a full year.

The earlier Phase II study involved 71 patients and primarily evaluated safety. Participants receiving the highest dose of rentosertib showed a mean increase of 98.4 ml in lung capacity over a 12‑week period, whereas the placebo group experienced a decline in the same metric.

Daily Maverick reported that rentosertib is the first drug discovered through artificial‑intelligence methods to reach Phase III trials, highlighting a milestone for AI‑driven drug discovery. The article also referenced other AI‑generated candidates from Insilico Medicine and Generate:Biomedicines, indicating a broader of AI‑derived therapeutics.

来源详情: ua.news ↗

为什么这很重要

The trial demonstrates that AI can move a drug candidate from computational design to late‑stage human testing within a relatively short timeframe—about 12 months from program start to first‑in‑human dosing for rentosertib. If successful, the therapy could address a major unmet need: IPF is a fatal, progressive disease affecting roughly five million people worldwide, with median survival of three to four years and no existing treatment that restores lung function. A positive outcome would validate AI‑driven discovery as a viable pathway for generating novel therapeutics, potentially reshaping pharmaceutical R&D investment, accelerating pipelines, and encouraging regulators to develop frameworks for AI‑originated medicines. Moreover, the trial underscores China’s growing role in AI‑enabled biotech, signaling competitive pressure on global pharma firms.

The rapid progression from AI‑based design to human testing—approximately 12 months for rentosertib—illustrates how computational methods can compress traditional drug‑discovery timelines, which often span a decade.

IPF remains a high‑mortality disease with limited therapeutic options; existing drugs only slow decline and cannot restore lung function. A successful AI‑derived therapy could dramatically improve patient outcomes and set a precedent for treating other degenerative conditions.

Success would provide concrete evidence to regulators and investors that AI‑generated molecules can meet rigorous safety and efficacy standards, potentially unlocking new funding streams and encouraging collaborations between AI firms and traditional pharma.

Interactive Mechanism

互动机制:它实际上是如何运作的

以交互方式探索这一发展背后的基础技术。

Agent Lifecycle Stage:
1
User Intent & Planning: "Audit customer refund request #4092 and settle payment."
2
Tool Calling: Emits structured JSON call crm_get_transaction(id='4092').
3
Guardrail & Verification:🛡️ Paused: High-value action requires human operator sign-off.
4
Final Settlement: Refund recorded, email receipt dispatched, and audit log stored.
Core takeaway: An AI agent is not just a language model—it is a closed loop of planning, tool invocation, and environment feedback. Production systems require self-healing retries and strict human approval guardrails.
交互式概念检查+10 Points
AI Models Explained Quiz

In AI, what are a model's "parameters"?

接下来看什么

Key indicators to monitor include the Phase III efficacy and safety results, especially any improvement in forced vital capacity versus placebo, and the incidence of adverse events. Regulatory feedback from China’s National Medical Products Administration will be crucial for future approvals and possible international filings. Follow‑on developments from Insilico’s other AI‑generated pipelines—garutadustat and the GB‑0895 antibody—will indicate whether the company can replicate this milestone across disease areas. Finally, industry observers should watch for shifts in funding and partnership activity as investors assess the commercial potential of AI‑derived drugs.

The primary endpoints of the Phase III trial—changes in forced vital capacity and safety outcomes—will determine whether rentosertib can achieve regulatory approval.

Regulatory feedback from China’s health authorities will be pivotal, especially regarding any novel considerations for AI‑originated drug candidates.

Progress on Insilico’s other AI‑derived pipelines, such as garutadustat for inflammatory bowel disease and the GB‑0895 antibody for severe asthma, will reveal whether this milestone can be replicated across therapeutic areas.

Investor and partnership activity in the AI‑driven biotech sector may shift in response to the trial’s results, influencing future R&D strategies.

相关指南和测验

人工智能模型解释AI 的未来AI 伦理测试你所知道的——尝试免费的人工智能测验在我们的词汇表中查找人工智能术语关注 AI 模型发布跟踪器
觉得这有用吗?