Personalised cancer vaccines take a leap forward – can Swiss pharma catch up?
Vaccines tailored to each patient’s tumour look increasingly promising after positive melanoma trial results. But a Roche-backed vaccine stumbled days later, raising questions about where the approach can succeed – and what it means for Swiss pharma.
Select your language
Generated with artificial intelligence.
Listening:
Personalised cancer vaccines take a leap forward – can Swiss pharma catch up?
For decades, the idea of a vaccine to treat cancer promised more than it delivered. That may be changing. This summer, the first successful Phase III trial of an mRNA vaccine designed around each patient’s own tumour was announcedExternal link by US-based Moderna and MSD (known as Merck in the US). It was given to people whose melanoma had been surgically removed, and it made the cancer less likely to return.
“It clearly has the potential to be a breakthrough,” said Olivier Michielin, head of oncology at Geneva University Hospitals. He was already working on experimental therapeutic cancer vaccines over 20 years ago. Back then, scientists could see that the vaccines were activating the immune system, but that wasn’t translating into notable health benefits for patients.
Cancer vaccines fall into two broad categories:
Preventive vaccines, such as those against HPV and hepatitis B, protect against infections that can later cause cancer; researchers are also testing vaccines to prevent cancer in people at high risk.
Therapeutic cancer vaccines are given to people who already have cancer or have been treated for it. They work as a form of immunotherapy, training the immune system to recognise and attack cancer cells.
This article focuses on therapeutic vaccines, and in particular on personalised ones designed around the mutations found in each patient’s tumour, and built with the mRNA technology that’s behind the Covid jabs.
But just nine days after Moderna and MSD’s announcement, Germany-based immunotherapy pioneer BioNTech and Genentech, a subsidiary of Swiss drug giant Roche, External linkstoppedExternal link a failing Phase II trial of their own personalised mRNA vaccine for colorectal cancer. The two companies are still testing the same vaccine in pancreatic cancer.
What do these contrasting results reveal about where personalised vaccines might work best? The answer has different implications for Switzerland’s two pharma giants: Roche has a stake through its long-term partnership with BioNTech, while rival Novartis hasn’t entered the field.
Encouraging results
Moderna and MSD’s vaccine, intismeran autogene, is personalised: each dose is made specifically for one patient, targeting their tumour’s mutations. Given alongside MSD’s immunotherapy pembrolizumab (marketed as Keytruda), the trial showed the vaccine significantly delayed the return or spread of melanoma compared with Keytruda alone, the companies said. They have yet to say whether it helps patients live longer.
Full results of the Phase III trial – the final stage of testing before the vaccine is submitted to regulators for approval – will be presented at the end of October. In an earlier, smaller study, the combination nearly halved the risk of the cancer coming back.
If the full Phase III results confirm the smaller study, it would be “clinically very significant”, said Michielin, especially because Keytruda on its own is already an effective treatment. He pointed out that this trial could validate several ideas at once: mRNA as a cancer treatment, mutations unique to each tumour as targets, and treatments made from scratch for individual patients.
Financial markets took noteExternal link: Moderna shares surged by as much as 60% in pre-market trading after the Phase III announcement. MSD shares also rose, although by a much more modest 6%. The gap is mainly a matter of size, said Stefan Schneider, senior equity analyst for pharma at Swiss investment firm Vontobel and a virologist by training. For Moderna, a successful cancer vaccine could be a major new source of income, while for MSD, one of the world’s biggest drugmakers, it would be a small addition to its business.
Why did one trial succeed and the other fail?
Michielin and Schneider point to several differences between the MSD-Moderna and Roche-BioNTech trials. The first is the cancer itself. Melanoma carries an unusually large number of mutations, giving a personalised vaccine plenty of targets, and it is one of the cancers in which immunotherapy already works best. By contrast, most colorectal cancers are “cold” tumours, which contain few cancer-fighting immune cells and often respond poorly to existing immunotherapies.
The second is how the vaccine was used. Moderna’s was combined with Keytruda, which stops tumours from switching off the immune cells that approach them. The vaccine by Roche and BioNTech, autogene cevumeran, was tested on its own in colorectal cancer.
But in an earlier trial Roche and BioNTech had also tested the same vaccine in melanoma, alongside Keytruda, and it failed to meet its main efficacy goal. Those patients had advanced disease that had spread, or the tumour could not be removed surgically, whereas Moderna’s trial treated patients after surgery.
Roche, however, is cautious about drawing conclusions . In a written response to Swissinfo, the company said its melanoma trial “was not designed to test whether the adjuvant setting is preferable to the metastatic setting” – that is, treating patients after surgery rather than at an advanced stage.
At this stage of research, no single factor can explain the different results – the colorectal patients were also treated after surgery. But together they suggest personalised vaccines may work best early, in cancers the immune system can recognise, said Michielin.
The main open question for Schneider is whether vaccines can also help in cold tumours, where immunotherapy currently does little. That is also where a vaccine could make the biggest difference, which – he suggested – may be why Roche is focusing on them.
Roche and BioNTech are testingExternal link their vaccine in pancreatic cancer, another cold tumour, in a trial that combines it with Roche’s own immunotherapy and chemotherapy after surgery. The outcome could help answer that question, but according to the trial registry, the results are not expected before 2031.
Personalised mRNA vaccines are being developed mainly by two partnerships.
Moderna and MSD are testing an mRNA vaccine (intismeran autogene) in nine mid- and late-stage trials, focusing on melanoma, lung, bladder, and kidney cancer. One large lung cancer trial tests the vaccine on its own.
Roche and BioNTech have tested an mRNA vaccine (autogene cevumeran) in four mid-stage trials. One continues, in pancreatic cancer, where it is combined with Roche’s immunotherapy and chemotherapy after surgery; results are not expected for several years. A colorectal cancer trial was stopped in August 2026 for lack of benefit, and an earlier trial in advanced melanoma failed in 2025. A trial in bladder cancer, combined with a rival company’s immunotherapy, was paused in 2025 after a safety issue and dropped in 2026.
Other personalised approaches, using technologies different from mRNA, are also being tested. French biotech Transgene, for example, is testing a personalised vaccine built on a modified virus, given on its own after surgery for head and neck cancer, now in a mid-stage trial.
Off-the-shelf vaccines, which target features shared by many patients’ cancers rather than being tailored to each individual, are also in late-stage testing. Among these, BioNTech is testing a vaccine in advanced head and neck cancer, and French biotech OSE expects Phase III results for a lung cancer vaccine in 2028. But it’s a highly risky business: Danish firm IO Biotech went bust this year after its melanoma vaccine narrowly missed its goal in a final-stage trial.
Overall, more than 2,000 trials of therapeutic cancer vaccines have been launched since 2000, according to a 2025 studyExternal link.
Two Swiss pharma giants, two different strategies
Big pharma has bet on RNA-based medicines before. Major drugmakers piled in some 25 years ago, then largely retreated around 2010 when progress proved slower than hoped, said Schneider. When they returned in the second half of the 2010s, they mostly did so through partnerships with specialists rather than by buying companies outright. Roche was among them: its US subsidiary Genentech teamed up with BioNTech in 2016 to develop personalised cancer vaccines, sharing the costs and profits.
For Schneider, Roche “has what it takes to be there”. Its diagnostics business can support the tumour testing needed to design personalised treatments; it has long experience matching therapies to the molecular features of a patient’s cancer; and its checkpoint inhibitor, Tecentriq, can be combined with a vaccine. The stakes are modest, though. Schneider does not include the vaccine in his valuation of Roche, and even a success would be a small addition to a company of its size.
Novartis, Switzerland’s other pharma giant, is not working on personalised mRNA cancer-vaccines, at least not publiclyExternal link, despite being active in both RNA technologies and oncology. In a written response to Swissinfo, it stated it “will pursue opportunities where we believe Novartis can make a differentiated contribution for patients”.
Schneider sees this approach as a business calculation: the company is focused on renewing products approaching patent expiry, and personalised vaccines would mean entering a new field, for a product whose profitability is still unclear.
The manufacturing bottleneck
Even if cancer vaccines obtain regulatory approval, there are still significant practical hurdles to overcome.
Each vaccine has to be made for one patient. “You have to take the tumour, sequence it, design the RNA, synthesise it, package it, send it back to the hospital and inject it,” Michielin noted. Today that takes around six weeks – for patients with a fast-moving disease, that wait alone can determine whether a personalised vaccine is a realistic treatment option.
Schneider said there are expectations the timeframe could drop to about four weeks, and Michielin hopes the process will become leaner, for instance by making vaccines on site at university hospitals rather than shipping everything to large factories.
The other issue is price. Most medicines get cheaper as production ramps up, but economies of scale are difficult to achieve for a product made for a single person. Michielin said cost must stay “very, very high among our priorities”. The treatment is likely to be expensive, Schneider noted, but he does not expect costs to sink the business case: cancer drugs already command high prices, and other personalised treatments, such as CAR-T cell therapies, have made it to market.
From trial success to the clinic
No personalised mRNA cancer vaccine has been approved yet, and Moderna and MSD have not yet applied for approval. Michielin advises interested patients to ask their oncologist whether a suitable clinical trial or access programme exists. Even with compelling final data, he cautioned, broader access would still take “many, many months”.
Cancer vaccines are not expected to replace existing treatments. Schneider sees them as “one additional option” for some patients. If they really deliver, Michielin expects vaccines to become an important tool for doctors, but not a cure-all. “In oncology, I think we have come to realise that it’s rarely just one strategy that wins,” he said.
Edited by Nerys Avery/vm/gw
In compliance with the JTI standards
More: SWI swissinfo.ch certified by the Journalism Trust Initiative
You can find an overview of ongoing debates with our journalists here . Please join us!
If you want to start a conversation about a topic raised in this article or want to report factual errors, email us at english@swissinfo.ch.