HONG KONG, Oct. 10, 2026 /PRNewswire/ — Akeso, Inc. (9926.HK) (“Akeso” or the “Company”) announced that the first patient has been dosed in a Phase III clinical study (COMPASSION-40/AK104-313) evaluating cadonilimab, its first-in-class PD-1/CTLA-4 bispecific antibody, as monotherapy in the neoadjuvant/adjuvant (perioperative) treatment of resectable microsatellite instability-high (MSI-H) or mismatch repair-deficient (dMMR) colon cancer.

This marks a significant new indication for cadonilimab beyond gastric, lung, and cervical cancers, and represents the 13th Phase III/registrational study of the therapy conducted globally. Cadonilimab has moved from later-line to first-line treatment of advanced disease and is now being evaluated in the perioperative setting, further broadening its clinical footprint in immuno-oncology (IO) 2.0.

Colorectal cancer is among the malignancies with the highest incidence and mortality worldwide, with colon cancer accounting for most cases. Perioperative outcomes are a major determinant of long-term prognosis. Patients with MSI-H/dMMR colon cancer derive limited benefit from conventional perioperative chemotherapy, with pathological response rates of only approximately 7%—a persistent and significant unmet medical need.

In metastatic MSI-H/dMMR colorectal cancer, immune checkpoint inhibitors are already established as the standard of care. In localized disease, perioperative treatment holds the potential for curative intent and the prospect of durable clinical cure—an area of substantial clinical importance and patient value. While multiple exploratory studies of PD-(L)1 monoclonal antibodies in this setting have shown encouraging efficacy signals, no immunotherapy has yet been approved anywhere for the perioperative treatment of localized MSI-H/dMMR colon cancer.

MSI-H/dMMR colon cancer is highly responsive to immunotherapy due to its high tumor mutational burden and inflamed immune microenvironment. However, single-target immune checkpoint inhibitors still leave room for improved outcomes. By dual blockade of the PD-1 and CTLA-4 pathways, cadonilimab is designed to more effectively activate and sustain antitumor immune responses in the perioperative setting. This approach has the potential to increase rates of major pathological response (MPR) and pathological complete response (pCR), thereby lowering the risk of postoperative recurrence and improving long-term survival.

Supporting evidence comes from a prior Phase II study of cadonilimab monotherapy as neoadjuvant treatment in patients with MSI-H/dMMR colorectal cancer. Data presented at the 2024 ESMO Immuno-Oncology Congress showed a pathological complete response (pCR) rate of 84.6% and a major pathological response (MPR) rate of 100% among patients who proceeded to surgery, with a manageable safety profile.

As the first approved checkpoint bispecific antibody for cancer, cadonilimab has shown consistent clinical benefit across a broad range of patients, including both PD-L1-positive and PD-L1-negative individuals, in extensive real-world use and multiple Phase III trials. It is increasingly recognized as a foundational therapy in the immuno-oncology (IO) 2.0 era. Cadonilimab is currently being evaluated in 13 Phase III/registrational studies worldwide across major indications such as lung, gastric, cervical, liver and colon cancers, spanning first-line, second-line, perioperative, adjuvant and consolidation settings.

About Cadonilimab

Cadonilimab is a first-in-class PD-1/CTLA-4 bispecific antibody developed by Akeso for cancer immunotherapy. In June 2022, it received initial marketing approval from China’s National Medical Products Administration (NMPA) for the treatment of patients with relapsed or metastatic cervical cancer who had progressed on or after platinum-based chemotherapy. In September 2024, cadonilimab was approved in combination with fluoropyrimidine- and platinum-based chemotherapy for the first-line treatment of patients with locally advanced unresectable or metastatic gastric or gastroesophageal junction (G/GEJ) adenocarcinoma. In May 2025, cadonilimab plus platinum-based chemotherapy, with or without bevacizumab, was approved for the first-line treatment of persistent, recurrent, or metastatic cervical cancer.

By simultaneously targeting PD-1 and CTLA-4, cadonilimab delivers synergistic antitumor activity. It has shown robust efficacy with markedly lower toxicity and improved tolerability compared with conventional combination regimens, demonstrating potent activity across multiple tumor types independent of PD-L1 expression and offering breakthrough clinical value in immunotherapy-resistant and “cold” tumors.

To date, cadonilimab has been evaluated in more than 40 clinical studies spanning over 20 indications, including gastric, lung, liver, cervical, and pancreatic cancers. Thirteen Phase III/registrational studies are ongoing; three have met their primary endpoints; and two global Phase III/registrational trials are advancing.

About Akeso

Akeso (HKEX: 9926.HK) is a leading biopharmaceutical company committed to the research, development, manufacturing and commercialization of the world’s first or best-in-class innovative biological medicines. Founded in 2012, Akeso has built a comprehensive R&D innovation ecosystem anchored by its proprietary Tetrabody antibody technology platform, AI-powered drug R&D platform, Dual-Shield ADC technology platform, Dual-Lock T-cell engager (TCE) technology platform, Tissue-Smart siRNA/mRNA technology platform, and cell therapy technology platforms.

Backed by world-class GMP manufacturing facilities and a highly efficient, integrated commercialization system, Akeso has developed into a globally competitive biopharmaceutical enterprise. Leveraging its fully integrated, multi-functional platform, the Company maintains a robust pipeline of more than 50 innovative assets targeting cancer, autoimmune diseases, inflammation, metabolic disorders, and other major therapeutic areas. Of these, nearly 30 candidates have advanced into clinical trials, including 15 bispecific or multi-specific antibodies and bispecific ADCs. Eight innovative drugs are commercially available, and two additional drugs with three indications are currently under regulatory review for marketing approval.

Akeso is committed to becoming a global leader in biopharmaceuticals through efficient and breakthrough innovation in R&D, developing novel therapies that are first-in-class or best-in-class, and providing better disease solutions for patients around the world.

Forward-Looking Statements

This announcement by Akeso, Inc. (9926.HK, “Akeso”) contains “forward-looking statements”. These statements reflect the current beliefs and expectations of Akeso’s management and are subject to significant risks and uncertainties. These statements are not intended to form the basis of any investment decision or any decision to purchase securities of Akeso. There can be no assurance that the drug candidate(s) indicated in this announcement or Akeso’s other pipeline candidates will obtain the required regulatory approvals or achieve commercial success. If underlying assumptions prove inaccurate or risks or uncertainties materialize, actual results may differ materially from those set forth in the forward-looking statements.

Risks and uncertainties include but are not limited to, general industry conditions and competition; general economic factors, including interest rate and currency exchange rate fluctuations; the impact of pharmaceutical industry regulation and health care legislation in the P.R. China, the United States and internationally; global trends toward health care cost containment; technological advances, new products and patents attained by competitors; challenges inherent in new product development, including obtaining regulatory approval; Akeso’s ability to accurately predict future market conditions; manufacturing difficulties or delays; financial instability of international economies and sovereign risk; dependence on the effectiveness of Akeso’s patents and other protections for innovative products; and the exposure to litigation, including patent litigation, and/or regulatory actions.

Akeso does not undertake any obligation to publicly revise these forward-looking statements to reflect events or circumstances after the date hereof, except as required by law.

QUZHOU, China, Oct. 9, 2026 /PRNewswire/ — This is a report from China.org.cn on the anniversary of Confucius’ birth:

A grand ceremony was held Monday in Quzhou, eastern China’s Zhejiang province, to commemorate the 2,577th anniversary of the birth of Confucius.

Held at the Southern Confucius Ancestral Temple, the event was hosted by the Quzhou municipal government.

Since 2004, Quzhou has developed a contemporary approach to honoring Confucius, with participants wearing modern attire and bowing, while grains, fruits, and vegetables are used as sacrificial offerings. Chanting, string music, and ceremonial music are also incorporated into the rites.

This year, participants sang “Lu Ming” from the “Book of Songs” before a series of traditional rites were performed. Kong Lingli, the 76th-generation lineal descendant of Confucius, later led participants to the Confucius Cultural Park, where representatives recited passages from Confucian classics and all participants bowed before a statue of Confucius.

Nearly 200 representatives of Confucius’ descendants and clan members from 20 provincial-level regions across China, including Hong Kong, Macao, and Taiwan, as well as from overseas, attended the event.

In 2011, the Southern Confucius memorial ceremony was inscribed on China’s national intangible cultural heritage list. In 2024, Quzhou’s ceremony was held on the same day as the Confucius memorial ceremony in Qufu, Shandong province, for the first time.

Monday’s ceremony was also one of the key activities of the 2026 Southern Confucianism Cultural Tourism Season.

Quzhou commemorates 2,577th anniversary of Confucius’ birth
http://www.china.org.cn/2026-09/29/content_118717589.shtml

From process engineers building bespoke “Process Agents” to seamlessly bridging customer demand with manufacturing execution, Quick AI is emerging as UnionTech’s core platform for driving industrial AI applications.

SHANGHAI, Oct. 10, 2026 /PRNewswire/ — When a 3D model of a complex part arrives on a process engineer’s desk, the print job does not begin immediately.

Is the material suitable? How should the part be oriented? Where are supports required? Which areas are prone to deformation? How should print parameters be adjusted? Traditionally, addressing these questions relies on the seasoned judgment of senior engineers, followed by iterative trial-and-error to arrive at a viable approach.

At UnionTech, this process previously took anywhere from several hours to multiple days.

Today, for new projects supported by historical precedent and comparable case data, engineers can leverage self-built Process Agents to generate an initial process plan in just minutes. Engineering teams then review the plan and schedule targeted validations for critical risks.

This paradigm shift is unfolding on UnionTech’s proprietary Enterprise Agent OS—Quick AI.

Allen Yang, Vice General Manager & Head of Digital Center and AI at UnionTech, believes industrial 3D printing is entering a new stage of competition. While hardware is capable of fabricating increasingly complex parts, customers are now demanding consistent, controlled, and continuous small-batch production.

“Hardware determines what we can build, but whether a complex project can be formulated quickly and delivered reliably hinges on engineering expertise and workflow coordination,” said Yang. “Our AI initiatives are designed to better organize, institutionalize, and scale these capabilities.”

Process Experts: Transforming Tribal Knowledge into Process Agents

In industrial manufacturing, crucial domain knowledge often remains locked in personal experience.

The exact same material may require entirely different processing strategies depending on varying wall thicknesses, geometries, and end-use applications. Why a particular parameter adjustment succeeded—or why a specific support strategy failed—is often understood only by the engineers directly involved in the project.

Over the years, this accumulated know-how has formed the core technical barrier of the enterprise. Yet whenever a new project arrived, teams still had to hunt down the engineers with relevant expertise, sift through past records, and re-evaluate from scratch.

Quick AI provides a modern vehicle for this tacit knowledge.

On the platform, process engineers can build their own Process Agents to continuously log process workflows, experimental results, and empirical parameters. Every detail—the selected materials and parameters, challenges encountered, subsequent adjustments, and the final outcome—becomes valuable institutional reference data for future projects.

“In the past, once a project was completed, usually only the final recipe was retained. But what truly matters is the decision-making process along the way,” explained Yang. “Why the part was oriented a certain way, why parameters were tweaked, which methods were attempted and failed—this is the real intellectual property worth preserving.”

When a new project is onboarded, the Process Agent leverages existing logs and similar historical cases to rapidly structure a proposed process plan, helping engineers pinpoint the most promising paths for priority validation.

For projects backed by historical precedent, the preparation phase—which once involved lengthy searches, discussions, and trial runs—is now compressed into minutes. Engineers can focus their energy where it matters most: validating novel structures, qualifying new materials, and mitigating critical risks.

Here, “a few minutes” refers to generating the initial proposal; it does not imply that full process validation is complete. Complex components still demand engineering reviews and necessary physical trials. The transformation lies in giving the team a data-backed starting point from day one, drastically cutting down repetitive trial and error.

“Our goal is to ensure that every project completed by our engineers empowers the team to execute the next project better, faster, and smarter,” said Yang.

Integrating Quick AI into Workflows: Powered by an End-to-End Data Backbone

Process Agents represent only the entry point. Enabling AI to deeply participate in manufacturing requires overcoming a fundamental challenge: bridging siloed business systems.

Customer requirements might be buried in emails, quotation benchmarks are stored in separate business systems, production schedules are tracked on factory shop floors, and quality inspection records are handled by different personnel. Even if AI can parse each individual piece of information, it must accurately map that data to the specific order, part, and design revision.

UnionTech’s decade-long investment in digital transformation laid the groundwork for this integration:

• Unionfab Cloud orchestrates business collaboration across customer engagement, RFQs, quoting, order management, and supply chain logistics.
• Unionfab ONE streamlines digital operations across 3D model preparation, process configuration, and machine control.
• The Auto Product Suite continually drives specialized algorithmic development and intelligent industrial applications.

Quick AI’s role is to act as the collaborative orchestrator across this infrastructure—connecting intelligent agents with enterprise knowledge bases and operational tools, allowing personnel to effortlessly tap into institutional capabilities.

While Cloud and ONE continue to drive business and manufacturing execution, Quick AI serves as the intelligent interface for task interpretation, knowledge retrieval, and tool orchestration. As APIs and workflow automations mature, the operational scope accessible to these agents will expand even further.

“Over the years, UnionTech has built not only a robust suite of software tools, but also an extensive repository of manufacturing expertise and process data. Our systems have accumulated nearly 5 million industrial part models along with their corresponding process parameter datasets,” said Yang. “Quick AI is designed to make these assets intuitive and actionable for front-line teams, ensuring AI directly interfaces with real-world production.”

Automated quotation serves as an active, real-world starting point.

Unionfab has already deployed automated quoting for standardized requirements, pairing 3D model geometry, materials, process rules, and cost algorithms to substantially reduce customer turnaround times. Highly specialized requirements and complex projects are automatically routed to manual engineering review.

These digital foundations empower future Process Agents to assist with requirement interpretation, material selection, and quote preparation. High-stakes decisions involving pricing commitments, critical tolerances, and stringent quality specs continue to be governed by established rules and human sign-off.

AI deployment at UnionTech follows a pragmatic, value-driven trajectory: solve a tangible operational problem, validate it within live workflows, and systematically scale its application.

Unionfab: Exposing AI to Real-World Orders Every Single Day

UnionTech benefits from a vital real-world proving ground: the Unionfab global on-demand manufacturing platform.

Here, customer journeys begin with online engagement and RFQs, move through automated or engineer-assisted quotation, order processing, and supply chain orchestration, and culminate in manufacturing, QA inspection, and global logistics. Incoming platform orders are fulfilled through a hybrid network: partly manufactured by UnionTech’s own fleet of industrial-grade 3D printers, and partly distributed across a partner network of hundreds of 3D printing service providers managed via the Unionfab Cloud ecosystem.

Every stage in this pipeline surfaces concrete operational challenges:

• Why was a particular quote declined?
• Which geometries repeatedly trigger manual engineering revisions?
• Did a specific quality tolerance get accurately transmitted to the shop floor?
• What root cause in an anomaly compromised a delivery window?

These live operational friction points define technical R&D priorities and provide an objective benchmark for whether AI tools deliver actual value.

If a new feature is deployed but engineers still have to conduct extensive manual cross-checks, it requires refinement. If an automated pre-processing step saves digital clicks but fails to enhance physical build stability, it must be re-evaluated.

“Customers don’t lower their standards for quality or lead times just because we use AI,” Yang emphasized. “What matters to them is whether the solution works, whether our response is faster, and whether the parts are delivered strictly to specification.”

Unionfab ensures UnionTech remains in continuous contact with authentic market demands, giving AI R&D immediate, high-fidelity feedback. Recommendations generated by Process Agents are tested against real-world shop floor constraints, cost models, and quality metrics. The fresh insights generated by these projects are then fed back to enrich the enterprise knowledge base.

In targeted production programs, such as tire mold manufacturing, UnionTech has already deployed integrated solutions that combine equipment, advanced materials, proprietary software, and tailored process recipes. These proven manufacturing practices provide empirical benchmarks for further intelligence.

Operating a proprietary digital service platform in tandem with physical production creates a continuous iteration loop: business operations define the challenges, technology provides the tools, and physical manufacturing validates the results.

Manufacturing Outcomes: Giving Experience the Medium to Scale

Industrial AI must answer one fundamental question: How do you verify that the model’s judgment is actually effective?

A 3D model successfully processed in software does not guarantee stable, repeatable physical fabrication. Ultimately, final part quality, total unit cost, and on-time delivery are the only true metrics of whether a process decision was sound.

UnionTech is actively correlating data across geometric features, material and process parameters, real-time machine telemetry, and final quality and delivery outcomes. The objective is to help engineering teams systematically understand why a manufacturing decision succeeded or failed, and which parameters are ripe for standardized reuse.

Governed by strict authorization boundaries and enterprise data security protocols, these historical datasets serve as the fuel for knowledge institutionalization, rule refinement, and algorithm optimization.

The long-term enterprise value of Quick AI is anchored in this closed-loop feedback. The broader the library of reference projects and the more rigorous the validated data, the more effectively Process Agents can guide upcoming manufacturing challenges.

“What we strive to retain above all else is actionable experience that informs the next decision,” said Yang. “Every time a project finishes, our organizational capability must take a step forward.”

This means senior engineers’ tribal knowledge is steadily transformed from individual intuition into shared, accessible organizational capability. Onboarding personnel can ramp up rapidly using historical intelligence, while mature, validated process parameters can be scaled and reproduced across more machines, production facilities, and client applications.

UnionTech’s AI Roadmap: Scaling Reproducible Manufacturing Capabilities

UnionTech has consistently positioned the development of AI capabilities as the core engine of its long-term competitive advantage.

Hardware and materials deliver the physical foundation; software structures process knowledge into executable digital workflows; Unionfab channels a steady stream of customer demand and fulfillment feedback; and Quick AI introduces a unified interface to orchestrate engineering intelligence and business tools.

Whether these synergies unlock real business value is measured by tangible operating outcomes: faster customer adoption, higher engineering capacity per team, superior quality consistency, reliable delivery timelines, and the seamless replication of mature manufacturing recipes across diverse client bases.

UnionTech is executing firmly along this strategic roadmap.

As Quick AI becomes deeply integrated with its mature digital ecosystem, UnionTech aims to create an uninterrupted loop connecting customer demand, process engineering, machine execution, and physical performance feedback—driving the continuous accumulation and scalable replication of industrial manufacturing intelligence.

Competition in industrial additive manufacturing is expanding far beyond standalone hardware specifications. How rapidly an organization interprets requirements, how consistently it delivers parts, and how effectively it reuses accumulated expertise will increasingly determine the quality of long-term business growth.

By embedding AI directly into the manufacturing workflow, UnionTech is explicitly addressing these very challenges.

About UnionTech

UnionTech is a global provider of industrial-grade 3D printing equipment, advanced 3D printing materials, and on-demand digital manufacturing services. Its Quick AI platform serves as an enterprise Agent OS enabling the creation and deployment of specialized Process Agents; Unionfab Cloud drives end-to-end digital collaboration across customer acquisition, order management, and supply chain operations; and Unionfab ONE digitalizes CAD pre-processing, process configuration, and machine-level control. Together, these platforms form a unified digital backbone powering the future of intelligent, automated additive manufacturing.

GUANGZHOU, China, Oct. 10, 2026 /PRNewswire/ — POMDOCTOR LIMITED (“Pomdoctor” or the “Company”) (Nasdaq: POM), a digital healthcare company focused on advancing AI-enabled healthcare solutions and predictive healthcare capabilities, today announced that it received a staff determination letter (the “Letter”) from the Listing Qualifications Department (the “Staff”) of The Nasdaq Stock Market LLC (“Nasdaq”) on October 6, 2026.

The Letter notified the Company that the closing bid price of the Company’s American depositary shares (the “ADSs”) had been below $1.00 per share for 30 consecutive business days, from August 24, 2026 through October 5, 2026. As a result, the Company is not in compliance with the minimum bid price requirement set forth in Nasdaq Listing Rule 5450(a)(1) (the “Minimum Bid Price Requirement”).

Ordinarily, a Nasdaq-listed company would be afforded a 180-calendar-day period to regain compliance with the Minimum Bid Price Requirement. However, pursuant to Nasdaq Listing Rule 5810(c)(3)(A)(iv), the Company is not eligible for any compliance period because the Company effected a 1-for-18 reverse stock split on June 22, 2026, within the prior one-year period.

Accordingly, the Staff determined that the Company’s securities are subject to delisting from The Nasdaq Global Market. Unless the Company timely appeals the Staff’s determination to a Nasdaq Hearings Panel (the “Panel”), the Company’s securities will be suspended from trading at the opening of business on October 15, 2026, and Nasdaq will file a Form 25-NSE with the U.S. Securities and Exchange Commission to remove the Company’s securities from listing and registration on Nasdaq.

The Company intends to timely request a hearing before the Panel to appeal the Staff’s determination and to present a plan to regain compliance with the Minimum Bid Price Requirement. The deadline for the Company to request a hearing is October 13, 2026. A timely hearing request will stay the suspension of the Company’s securities and the filing of the Form 25-NSE pending the Panel’s decision. Accordingly, no suspension or delisting action will take effect while the appeal remains pending.

The Company is evaluating the alternatives available to it to regain compliance with Nasdaq’s continued listing requirements. However, there can be no assurance that the Panel will grant the Company an extension of time to regain compliance, that the Company will regain or maintain compliance with the Minimum Bid Price Requirement or any other Nasdaq continued listing requirement, or that the Company’s appeal will be successful.

About POMDOCTOR LIMITED

POMDOCTOR LIMITED (Nasdaq: POM) is a digital healthcare company focused on advancing AI-enabled healthcare solutions and expanding predictive healthcare capabilities. The Company leverages physician resources, wearable technologies, artificial intelligence, healthcare payment networks and real-world healthcare data to support more personalized, continuous and data-driven healthcare services. For more information, please visit the Company’s website: http://ir.7shiliu.com.

Forward-Looking Statements

This press release contains forward-looking statements. These forward-looking statements include, without limitation, statements concerning the Company’s intention to request a hearing before the Panel, the anticipated effect of a timely hearing request, the Company’s appeal of the Staff’s determination, the Company’s plan and ability to regain and maintain compliance with Nasdaq’s continued listing requirements, the Panel’s consideration of the Company’s appeal and compliance plan, and the continued listing of the Company’s securities on Nasdaq.

Forward-looking statements involve risks and uncertainties, and actual results may differ materially from those expressed or implied by such statements. Factors that could cause actual results to differ materially include, among others, the Panel’s determination regarding the Company’s appeal, the Company’s ability to regain and maintain compliance with the Minimum Bid Price Requirement and other Nasdaq continued listing requirements, fluctuations in the market price of the Company’s ADSs, and the other risks and uncertainties described in the Company’s filings with the U.S. Securities and Exchange Commission.

The Company undertakes no obligation to update any forward-looking statement to reflect events or circumstances occurring after the date of this press release, except as required by applicable law.

Investor Relations Contact

POMDOCTOR LIMITED
Investor Relations Department
Email: ir@7lk.com

Ascent Investor Relations LLC
Tina Xiao
Phone: +1-646-932-7242
Email: investors@ascent-ir.com