Don’t Let Top-Tier Treatments Become the Exclusive Privilege of the Rich
By henry and Yuyang, reporting from Aofeisi Temple
Who would have thought?
The melanoma cancer vaccine Intismeran Autogene, which recently shot to fame after being developed with the help of AI, has already been revealed as potentially becoming a “game for the rich”—
According to estimates by the U.S. investment bank William Blair, the vaccine’s future wholesale acquisition price could reach as high as $475,000 per patient, equivalent to approximately RMB 3.18 million.
And this comes less than a week after Moderna and Merck announced the major breakthrough.
On August 19, the two companies announced that the personalized mRNA cancer vaccine, which they began jointly developing nearly 10 years ago, had achieved positive results in treating melanoma, often referred to as the “king of skin cancers.”
In the Phase 3 trial, Intismeran was used in combination with the PD-1 inhibitor pembrolizumab, commonly known as “Keytruda.” The combination met both the primary endpoint of recurrence-free survival and the key secondary endpoint of distant metastasis-free survival.
Put simply, the treatment may delay cancer recurrence and postpone its spread to distant organs.
With this milestone achievement, Intismeran became the first personalized neoantigen therapy to achieve positive results in a Phase 3 trial, as well as the first mRNA cancer treatment to make it through Phase 3.
Even more notably, AI was involved in almost every stage of the vaccine’s development, from design to manufacturing:
From neoantigen screening and mRNA sequence design to production scheduling, the process transforms each patient’s tumor data into a cancer vaccine made specifically for that patient.
The news not only brought new hope to melanoma patients, but also gave the AI drug-development and innovative-medicine industries a major boost. Moderna’s stock price surged as much as 176.97% at one point.
However, just one week after the breakthrough was announced, the potential treatment price tag of nearly $500,000 began to overshadow the medical achievement itself.
At present, public information offers differing interpretations of the $475,000 figure projected by William Blair: it remains unclear whether the amount refers to a single dose or the complete vaccine treatment process for one patient.
But one thing is certain:
In the Phase 3 trial protocol, Intismeran was not used alone. It had to be combined with Keytruda, which is itself far from inexpensive.
If we use the current market projection of $475,000 and add the list price of up to nine cycles of Keytruda—approximately $221,000—the total drug cost of the combination therapy for one patient could reach nearly $696,000, equivalent to approximately RMB 4.66 million.
In other words, if the actual price after launch ultimately comes close to this level, the cancer-treatment breakthrough that has just made history will immediately face an extremely practical question:
Can ordinary people actually afford it?
How Was the $500,000 Estimate Calculated?
According to currently available public information, Intismeran has not yet been approved, and the pharmaceutical companies have not released a pricing plan. The $475,000 figure comes from a model William Blair used to project future sales revenue.
Although it is unclear exactly which products William Blair used as pricing references, an interesting coincidence is that the figure matches the original price of Kymriah, a CAR-T therapy approved in 2017.
Kymriah is primarily used to treat B-cell acute lymphoblastic leukemia in children and young adults.
Like Intismeran, it requires a patient’s own T cells to be collected, modified, expanded, and then reinfused. In essence, it also has the personalized characteristic of “one manufacturing process per patient.”
When Kymriah launched in 2017, its price for a single treatment was exactly $475,000.
Therefore, although it is too early to claim that Intismeran will actually cost “$500,000 per shot,” the idea is not entirely unfounded.
So why would a cancer vaccine cost so much?
The first reason is personalized production—one drug for one patient.
According to Jiemian News, Zhan Qilin, chief physician of the Hematologic Oncology Department at the Shanghai Public Health Clinical Center, previously said in an interview that the estimated production cost for one patient currently exceeds $100,000. From sample collection to administration of the first dose takes approximately six to nine weeks.
Every patient must undergo an entire process involving tumor sampling through surgery, genetic sequencing, mutation analysis, neoantigen screening, mRNA design and synthesis, lipid nanoparticle encapsulation, quality inspection, and product release.
It is not difficult to see the difference: traditional vaccines can be mass-produced and supplied to large numbers of people, whereas Intismeran’s production line must repeat the process for each individual patient.
The second reason is the enormous R&D investment required for innovative medicines.
According to Moderna CEO Stéphane Bancel, Merck established a partnership with Moderna on personalized mRNA cancer vaccines as early as June 2016, providing an upfront payment of $200 million.
From the outset, the two companies planned to use the vaccine in combination with Keytruda and built small-batch GMP customized manufacturing capabilities for that purpose.
The program then continued for nearly a decade.
In November 2017, mRNA-4157 was administered to its first human participant. Early versions could encode 20 patient-specific neoantigens on a single mRNA strand, a number later expanded to as many as 34.
Recruitment for KEYNOTE-942 began in July 2019. In December 2022, the first positive topline results from the Phase 2b trial were announced. In February 2023, the FDA granted the therapy Breakthrough Therapy Designation. In July of the same year, the Phase 3 INTerpath-001 trial officially began.
In January 2024, the primary analysis of the Phase 2b trial underwent peer review by The Lancet. However, in September of the same year, Moderna disclosed that the FDA did not support pursuing accelerated approval based solely on the existing data and still required a confirmatory Phase 3 trial. This development path was far from smooth.
It was not until June 2026 that five-year Phase 2b data were released. Then, two months later, on August 19, the prespecified interim analysis of the Phase 3 trial finally announced that both the RFS and DMFS endpoints had been met—
In other words, in a randomized, controlled, multicenter study involving a larger patient population, Intismeran demonstrated its efficacy, clearing the most critical hurdle on the path toward regulatory approval.
At this point, mRNA cancer vaccines were no longer merely “promising.” They had taken a crucial step toward real-world clinical application.
Taken together, from the start of the partnership in 2016 to its passage through this key Phase 3 milestone, Intismeran represents a high-investment development program that has lasted nearly a decade.
It also happens to confirm the so-called “rule of tens” often cited in the innovative-drug industry: bringing a new drug from the start of R&D to market often takes more than 10 years and over $1 billion in investment.
Although outsiders have no way of knowing exactly how much Moderna invested in Intismeran, the company’s overall R&D spending offers a sense of just how “cash-burning” this industry is—
In 2025, Moderna reported revenue of $1.944 billion and a net loss of $2.8 billion, while its R&D expenses reached $3.132 billion.
Moreover, as mentioned at the beginning, Intismeran must be used in combination with Keytruda, which is itself far from cheap.
Therefore, even if Intismeran’s eventual price is lower than $475,000, the total drug cost when patients complete the full combination regimen could still rise substantially.
Cancer Vaccines Finally Make It Through Phase 3
Still, a sky-high price is a separate issue from Intismeran’s therapeutic value, which is not diminished by the cost.
For patients, the treatment genuinely offers a new and promising therapeutic option.
To understand why this breakthrough matters, we first need to return to a basic question:
Scientists have been pursuing cancer vaccines for decades. Why was Intismeran the first to make it through Phase 3? In particular, cancer vaccines are hardly a concept that emerged out of nowhere.
For decades, scientists have been trying to do one thing: inform the immune system about the characteristics of cancer cells so that the body can hunt them down itself.
Yet this seemingly straightforward idea had not produced much success in the past.
A 2025 review in Nature Reviews Cancer summarized that early cancer vaccines often targeted “tumor-associated antigens” shared by different patients. They were also frequently used as standalone treatments in patients whose cancers had already reached an advanced stage.
That was where the problem lay.
The targets were not precise enough, while the tumors had already progressed significantly. Patients’ immune systems were also often in a state of prolonged suppression by the tumor.
The vaccine was trying to launch a counterattack, but by then, the battlefield was often already too difficult to fight on.
To address these problems, Intismeran adopted a different strategy.
First, instead of trying to find a single target for all patients, it performs a “genetic fingerprinting” of each individual’s cancer.
The system analyzes the patient’s tumor tissue and normal sample simultaneously, identifies up to 34 neoantigens most likely to activate an immune response from among the many tumor mutations, and then encodes this information into an mRNA strand made specifically for that patient.
In other words, no two people receive exactly the same vaccine.
These neoantigens arise from mutations in the cancer cells themselves and are absent from normal cells. This makes it easier for the immune system to distinguish between the body’s own cells and cancer cells.
Next, the two drugs work together.
Intismeran presents these cancer-cell characteristics to T cells, essentially issuing an “arrest warrant.” Keytruda, used in combination with it, releases the immune system’s brakes, allowing T cells that have recognized the target to attack it.
At the same time, the timing of treatment has changed.
The Phase 3 INTerpath-001 trial recruited a total of 1,137 patients with high-risk stage IIB to stage IV melanoma who had already undergone tumor resection.
In other words, the tumors visible to the naked eye had already been surgically removed.
Intismeran was no longer confronting a large, fully developed advanced tumor. Instead, it was targeting residual cancer cells that might still be hiding in the body after surgery.
Its task was to find those residual cancer cells before the tumor had a chance to return.
In this randomized, double-blind trial, one group of patients received Intismeran plus Keytruda, while the other received placebo plus Keytruda.
Ultimately, the prespecified interim analysis showed that the combination treatment improved both recurrence-free survival and distant metastasis-free survival, with both endpoints achieving statistical significance and clinical relevance.
Moderna and Merck have not yet released the complete Phase 3 data, and overall survival is still being monitored.
But this was not the first time Intismeran had produced positive results.
Previously, five-year follow-up data from the Phase 2b trial showed that, compared with Keytruda alone, Intismeran plus Keytruda reduced the relative risk of recurrence or death by 49% and the relative risk of distant metastasis or death by 59%.
Of course, these two figures are neither a “cure rate” nor a substitute for the final Phase 3 data.
But they at least explain why the entire cancer-vaccine industry has been waiting for this result.
For decades, personalized cancer vaccines remained “highly promising.” Now, for the first time, one has cleared the most critical hurdle in a Phase 3 trial involving more than 1,000 patients.
Cancer vaccines are finally more than a compelling laboratory story.
Not ChatGPT-Style AI Drug Development Yet
Finally, compared with the sensational claims about “AI curing cancer” that have been circulating in recent days, it may be necessary to pour some cold water on the subject.
Although AI was involved in almost every stage of Intismeran’s development, from design to manufacturing, this is not the generative AI centered on large language models that people are more familiar with today.
At Moderna’s second Digital Investor Event in November 2023, the company disclosed that it had been using machine learning to optimize mRNA sequences since 2014. In 2016, it went further by developing specialized drug-design algorithms for personalized neoantigen therapies.
After the patient’s tumor tissue and normal samples have been sequenced, multiple components and AI algorithms in the system process the data in sequence:
They first identify mutations unique to the tumor, then predict which mutations are most likely to form neoantigens capable of triggering an immune response, select up to 34 targets, and finally design the corresponding mRNA sequence.
According to Moderna, the process—from the moment sequencing data enter the system to the generation of the final drug design—can now be completed automatically, without requiring human intervention at each individual step.
AI’s role also extends beyond “designing the vaccine.”
Because each patient receives a separately manufactured vaccine, and because mRNA drugs have extremely demanding production, storage, and transportation requirements, the process also involves a complex supply-chain challenge.
Moderna’s scheduling system arranges manufacturing, quality inspection, transportation, and administration times for every patient.
If any stage is delayed, the system recalculates the subsequent schedule. During the clinical-trial stage, the goal is to transform a patient’s tumor sample into a final, deliverable personalized drug in approximately six weeks.
Strictly speaking, the AI behind Intismeran is more like a specialized system combining bioinformatics, machine learning, and production-optimization algorithms.
It does not write articles or draw pictures like ChatGPT.
What it actually does is something more specific—and more important:
It identifies the targets most worth attacking from among massive numbers of mutations, then transforms a patient’s unique cancer information into a drug that can genuinely be manufactured.
And this may be what makes Intismeran truly worth watching.
Over the past few years, one of the biggest changes AI has brought to ordinary people is that it has begun to make knowledge and capabilities that were once expensive and scarce more accessible—moving toward “intellectual equality.”
But in medicine, the genuinely difficult step is only just beginning.
Once AI has the ability to help design a cancer vaccine specifically for one person, the next question is no longer simply whether we can make it.
The more fundamental question is: Can this highly effective medicine ultimately be made affordable to more people?
AI has opened a breach in cancer’s defenses. The next question is how long it will take to bring down the barriers to access.
References
[1] https://www.modernatx.com/ir-insights-phase-3-intesmeran