Inside Moderna’s Biggest mRNA Test Since COVID

2026年8月，Moderna和Merck公布了针对黑色素瘤的个体化mRNA癌症疫苗的突破性积极三期临床试验结果，标志着癌症治疗领域的重大进展。该试验表明改善了无复发生存期和无远处转移生存期，在过去二十年间有上千次失败临床试验的领域中，这尚属首次。 Moderna首席执行官Stéphane Bancel解释说，成功源于两个关键因素：mRNA技术和个体化。与传统的基于蛋白质或肽的疫苗不同，Moderna的mRNA在注射后会进入淋巴结并进入抗原呈递细胞（APCs）。在这些免疫细胞内，mRNA信息被翻译，*从内部*呈递抗原，这是一种根本不同且更有效的免疫系统教育机制。 第二个关键区别是个性化。癌症是一种DNA疾病。疫苗的生产过程包括对患者肿瘤进行活检，并将其DNA序列逐个核苷酸地与同一患者的健康细胞进行比对。然后，算法会识别出最多34个最相关的突变。这些突变被编织成一个单一的mRNA分子，在大约30天内为该患者量身定制。这种个性化疫苗教导免疫系统识别*该个体*癌细胞的*特定*特征，而这些特征是免疫系统之前所未能识别的。数据显示，大约90%的这些癌症抗原对每个患者来说都是独一无二的，这强调了这种个体化方法的必要性。 这种新疗法与现有免疫疗法（如Keytruda）协同作用。Keytruda是一种领先的检查点抑制剂，它“释放了免疫系统的狗”来攻击癌症，但其对黑色素瘤患者的有效性在五年后仅限于约60%，并且可能导致严重的自身免疫副作用。相比之下，Moderna的mRNA疫苗“非常具体地教导这些狗该寻找什么”，指导T细胞只攻击癌细胞，副作用类似于普通疫苗（例如，暂时性疲劳）。之所以使用“疫苗”这个术语，是因为它能够教育免疫系统，尽管它是一种治疗性疗法，旨在预防已患病患者的癌症复发。 在操作层面，疫苗的个性化特性带来了独特的挑战。从活检到疫苗准备就绪的“针到针”时间目前约为42天，Moderna计划通过自动化和机器人技术进一步缩短这一时间。与需要取出并重编程患者免疫细胞的CAR-T细胞疗法不同，Moderna的方法只需要遗传信息。其生产过程完全是合成的、酶促的和水基的，使其更类似于小分子药物的生产。这使得设备和设施可以小型化，从而显著降低成本并提高产量。当前的算法Intismarine 1.0是有效的，但Moderna计划利用三期患者数据和人工智能开发更先进的2.0版本以改善治疗反应，特别是针对目前没有反应的20%患者。 这种个体化疗法的监管批准将遵循“过程BLA”途径，类似于CAR-T细胞疗法，而非传统的“产品BLA”。FDA将重点关注整个生产过程的稳健性，以确保相同的输入（患者样本）能够始终如一地产生相同的高质量、个性化输出。 展望黑色素瘤之外，Moderna有着宏伟的计划。该疫苗正在与Keytruda联合进行肺癌的三期临床试验以及肾癌和膀胱癌的二期临床试验。他们还在探索其在早期疾病中的应用，例如一期肺癌，在这些情况下，Keytruda的副作用使其不适用。此外，Moderna正在进军现有免疫疗法失败的癌症领域，例如胰腺癌和胃癌，利用其独特的药理作用机制以及与其他新型药物的潜在组合。 除了肿瘤学，Moderna的mRNA平台还应用于其他领域。预计在今年年底前将有一项针对罕见肝脏遗传疾病的关键研究。该公司还将平台指向自身免疫疾病，旨在治疗病根而非仅仅症状，可能通过个体化的自身免疫疗法实现。Moderna认为目前的成功只是一个开始，随着技术和算法的不断改进，未来有望产生更大的影响。

In August 2026, Moderna and Merck announced groundbreaking positive Phase 3 clinical trial results for an individualized mRNA cancer vaccine in melanoma, marking a significant advancement in cancer treatment. This trial demonstrated improved recurrence-free survival and distant metastasis-free survival, a first in a field that has seen over a thousand failed clinical trials over two decades. Stéphane Bancel, Moderna CEO, explained that the success stems from two key factors: mRNA technology and individualization. Unlike traditional protein or peptide-based vaccines, Moderna’s mRNA, when injected, travels to lymph nodes and enters Antigen-Presenting Cells (APCs). Within these immune cells, the mRNA message is translated, presenting the antigen *from within*, a fundamentally different and more effective mechanism for immune system education. The second critical difference is personalization. Cancer is a disease of DNA. The vaccine process involves taking a biopsy of a patient's tumor and comparing its DNA sequence, nucleotide by nucleotide, with healthy cells from the same patient. An algorithm then identifies the up to 34 most relevant mutations. These mutations are stitched into a single mRNA molecule, custom-made for that patient within about 30 days. This personalized vaccine teaches the immune system the *specific* signature of *that individual's* cancer cells, which it had previously missed. Data shows that approximately 90% of these cancer antigens are unique to each patient, underscoring the necessity of this individualized approach. This new treatment works synergistically with existing immunotherapies like Keytruda. Keytruda, a leading checkpoint inhibitor, "unleashes the dogs" of the immune system to attack cancer, but its effectiveness is limited to about 60% of melanoma patients after five years, and it can cause severe autoimmune side effects. Moderna's mRNA vaccine, in contrast, "teaches those dogs what to look for, very specifically," directing T-cells to attack only the cancer cells, with side effects similar to a typical vaccine (e.g., temporary fatigue). The term "vaccine" is used because it teaches the immune system, even though it's a therapeutic treatment to prevent cancer recurrence in patients who already have the disease. Operationally, the personalized nature of the vaccine presents unique challenges. The "needle to needle" time, from biopsy to vaccine readiness, is currently about 42 days, which Moderna aims to reduce further through automation and robotics. Unlike CAR-T cell therapy, which requires taking out and reprogramming a patient's immune cells, Moderna's approach only needs genetic information. The manufacturing process is entirely synthetic, enzymatic, and water-based, making it more akin to small molecule production. This allows for miniaturization of equipment and facilities, significantly reducing costs and increasing throughput. The current algorithm, Intismarine 1.0, is effective, but Moderna plans to use Phase 3 patient data and AI to develop a more advanced 2.0 version to improve responses, particularly for the 20% of patients who currently do not respond. Regulatory approval for this individualized therapy will follow a "process BLA" pathway, similar to CAR-T cell therapies, rather than a traditional product BLA. The FDA focuses on the robustness of the entire manufacturing process to ensure that the same input (patient samples) consistently yields the same high-quality, personalized output. Looking beyond melanoma, Moderna has ambitious plans. The vaccine is being tested in combination with Keytruda in Phase 3 trials for lung cancer and Phase 2 trials for kidney and bladder cancer. They are also exploring its use earlier in diseases, such as Stage 1 lung cancer, where Keytruda's side effects make it unsuitable. Furthermore, Moderna is venturing into cancers where existing immunotherapies have failed, like pancreas and gastric cancer, leveraging its distinct mechanism of action and potential combinations with other novel agents. Beyond oncology, Moderna's mRNA platform is being applied to other areas. A pivotal study for rare genetic diseases of the liver is expected before the end of the year. The company is also pointing its platform towards autoimmune diseases, aiming to treat the root cause rather than just the symptoms, potentially through individualized autoimmune treatments. Moderna views its current successes as just the beginning, with continuous improvements to the technology and algorithms promising even greater impact in the future.