‘Maybe the Coronavirus Was Lower-Hanging Fruit’

Emerging mRNA technology proved excellent for COVID vaccines. BioNTech’s founders preview what that could mean for cancer and other mysteries.

Illustration of a vaccine needle shaped into a cancer ribbon.
The Atlantic

About the author: Derek Thompson is a staff writer at The Atlantic, where he writes about economics, technology, and the media, and is the author of the Work in Progress newsletter. He is the author of Hit Makers and the host of the podcast Crazy/Genius.

Two years ago, approximately nobody on Earth had ever heard of mRNA vaccines. This was for the very good reason that no country had ever authorized one. As a scientific experiment, synthetic mRNA was more than 40 years old. As a product, it had yet to be born.

Last year, mRNA technology powered the two fastest vaccine developments in history. Moderna famously prepared its COVID-vaccine recipe in about 48 hours. And then there’s BioNTech, a German biotech firm that originally partnered with Pfizer to develop flu therapies but moved quickly to produce its own shot for the new disease. Within 24 hours of the genetic sequencing of the coronavirus, BioNTech had built eight potential vaccine candidates. The company eventually tested more than 20. One of them has now been administered more than 1 billion times around the world—including more than 200 million doses in the United States alone.

Messenger ribonucleic acid—or mRNA—is a tiny molecule that instructs our cells to make proteins that keep us alive. Synthetic-mRNA technology, which powers the COVID-19 vaccines from Moderna and Pfizer-BioNTech, sends specialized instructions to our cells to manufacture specific proteins: in this case, the spike protein that encircles the coronavirus. Our immune system trains itself against these harmless spike proteins so that later, if we confront the real coronavirus, our bodies are primed to destroy it. BioNTech’s founders, the husband-and-wife team of Uğur Şahin and Özlem Türeci, compare this to displaying a Wanted poster of an outlaw to our immune system, so that he can be swiftly eliminated when he shows his face.

The fact that mRNA technology had never delivered an authorized therapy before the coronavirus pandemic could tell us one of two things. Perhaps synthetic mRNA is like a miraculous key that humankind pulled out of our pockets in this pandemic, but it was so perfectly shaped for the coronavirus that we shouldn’t expect it to unlock other scientific mysteries any time soon.

Or perhaps mRNA is merely in the first chapter of a more extraordinary story. This month, BioNTech announced that it had initiated Phase 2 trials of personalized cancer vaccines for patients with colorectal cancer. It is working on other personalized cancer vaccines and exploring possible therapies for malaria using a version of the mRNA technology that had its breakout moment in 2020.

Last week, I spoke with Şahin and Türeci about the history of their COVID vaccine and the promise of mRNA. Our conversation left me feeling optimistic about the future of biotechnology, humbled by the extraordinary challenge of commandeering novel technology to eliminate complex diseases, and deeply fortunate that mRNA tech emerged at the perfect moment in the pandemic. This conversation has been edited for length and clarity.


Derek Thompson: BioNTech was founded in 2008 to focus on cancer therapies. How did you wind your way to mRNA technology?

Uğur Şahin: In the early 1990s, we were both cancer physicians working in parallel on taking care of cancer patients by day and working in labs in the evening. At the time, we could only offer chemotherapy and radiation, and very often we had to tell our patients there were no more options for further treatment. But in the lab, we were seeing the promise of new immunology treatments. There was a clear gap between what we could offer patients, on the one hand, and the emerging science, on the other hand. We wanted to overcome this gap.

Özlem Türeci: In the late 1990s, we became interested in mRNA and its potential for vaccination. You could essentially present to our immune system the blueprint of a wanted foe, in this case the cancer and its specific molecules. Then you could deliver instructions to act upon that wanted poster and destroy the foe.

We thought mRNA had huge potential but also many flaws. In the late 1990s, its potency was very low. There was very little effect on the immune system. So that was our research focus for 30 years. Today’s COVID-19 vaccine uses just 30 micrograms of mRNA. That is a tiny amount of mRNA to activate the immune system of the whole body. It’s almost magic to generate billions of immune cells from such a small amount of mRNA.

Thompson: What did you learn that was so special about mRNA technology?

Türeci: Through years of research, we learned we can treat infectious diseases with mRNA by showing our immune system a wanted poster of a foe—like the spike protein on the coronavirus—and instructing the immune system to target that outlaw for destruction. We’ve also learned that, in addition to showing the wanted poster, we can also modify the message that we send to the body. It’s possible that we can treat autoimmune diseases with mRNA by sending a message that tells our cells to do nothing when they see a certain protein.

Thompson: Moderna has a very famous origin story for its COVID-19 vaccine, which is that it finalized the vaccine recipe in 48 hours. Did you design your vaccine in 48 hours too?

Şahin: Actually, we did it in less than 48 hours! In 24 hours, we generated the genetic sequence of the first eight vaccine candidates.

Thompson: Why make eight different vaccines?

Şahin: When we started the project, this was a new virus without any proven vaccine. So it was not clear what was the best molecule to target. From the beginning, Moderna bet on the full spike protein. But at that time, it wasn’t obvious to us that targeting the spike protein was the best. So we developed one vaccine to target the spike and other vaccines to target other parts of the virus.

We ultimately tested about 20 vaccine candidates on mice. We injected animals and varied the dose to understand which vaccines provided the strongest antibody response and T-cell response and protein-antigen production in the mice. Then we took the four most successful candidates to Phase 1 trials. Those Phase 1 trials told us the single vaccine that worked the best. That’s the final vaccine that showed more than 90 percent efficacy in Phase 3 trials and was authorized by the FDA.

Thompson: Obviously, mRNA’s success was a wonderful surprise. But I think it’s underrated just how surprising it really was. It’s doubly surprising to me not only that this technology worked but that it crushed all these land-speed records for vaccine development with extraordinary effectiveness. That mRNA technology was so well suited for this pandemic seems quite wonderful to the point of being almost miraculous. How do you explain why this technology worked so well for this foe?

Şahin: Nobody has ever asked us the question like that before. I think it may be the mother of all questions. Before corona, there was Ebola, and a different vaccine technology that was viral-vector-based was sufficient. [Editor’s note: Viral-vector vaccines, such as the Johnson & Johnson shot, use a harmless version of an unrelated virus to deliver information that teaches cells to produce an antigen, such as the spike protein, that the immune system learns to neutralize.] Ebola was low-hanging fruit for viral-vector-vaccine technology.

The coronavirus is a very different virus. It has these spike proteins that bind very strongly to the receptors [of our cells]. It turned out that mRNA vaccines were particularly excellent for boosting the immune system’s response. So maybe the coronavirus was lower-hanging fruit for mRNA technology.

Thompson: This is why it’s so important to fund different kinds of vaccine technology. Different tools work for different problems, and there’s no guarantee that mRNA will be the perfect tool for the next epidemic.

I want to ask about the other mRNA vaccines you’re working on. Let’s start with malaria. This year, Yale researchers patented an RNA-based technology to vaccinate against malaria. Reuters reported that you plan to start clinical trials for a malaria vaccine by the end of next year. Why do you think mRNA is a good candidate for malaria?

Şahin: Malaria is a field where scientists have been working for decades. This is a pathogen with many escape mechanisms that have eluded other vaccine technologies. Our strategy is to identify new molecular targets that other scientists have overlooked. We are now testing more than 40 malaria-vaccine candidates in preclinical settings. We believe that mRNA vaccines could, if developed properly, provide a lot of opportunities to prevent infection and disease.

Thompson: I also read that you recently dosed your first patient in a 200-person trial of a new cancer vaccine. How do your cancer vaccines work?

Türeci: We have two types of mRNA vaccines for cancer. First, we have our off-the-shelf vaccines, where we’ve identified molecular features of tumors that are shared by many patients. These are molecules that are broadly present in cancer cells but not in normal cells. By targeting these molecules, you can fight the cancer without getting collateral damage to healthy cells. Second, we have highly personalized vaccines. We identify cancer mutations that are unique to every patient. Every cancer patient has their own mutations, like a fingerprint. We biopsy the tumor, sequence it, and design a unique, individualized vaccine for each patient.

For both types of therapies, we have shown, in early clinical trials, that they are safe and that the tumors shrink. We have moved our vaccine development into Phase 2 trials for melanoma and head and neck cancers. We have also started our treatment for individualized vaccines for high-risk colorectal cancer.

Thompson: So you’re working on two types of cancer vaccines. Are they meant for different kinds of cancers?

Şahin: For personalized vaccines, we’ve come to think that focusing on the stage after surgery might be best. After surgery to remove a tumor, about 60 percent of patients are cured. But 30 to 40 percent see regrowth of that tumor. Certain cancers, like lung and liver cancer, are particularly likely to relapse post-surgery. mRNA vaccines could be perfectly suited to block this recurrence by specifically targeting the molecules associated with regrowth and metastasis.

Thompson: Given that vaccines historically have taken years to be developed, how fast can you reasonably produce an individualized cancer vaccine?

Türeci: mRNA technology is fast enough that we can really accelerate this turnaround time. From tumor biopsy to delivery, we can do this in four to six weeks. Every production of an individualized vaccine is a race against that patient’s tumor.

Thompson: How promising are the data so far?

Şahin: We have hundreds of patients worldwide, and the data in the early trials are convincing. That’s why we’ve moved to Phase 2 trials. But it’s important to say that the Phase 2 stage is where we have to prove that our therapy beats the standard of care that the patient would otherwise get. It would only be scientifically sound to gauge the vaccines after this head-to-head comparison. We think we can make major advances in the next five years, but it really depends on what these Phase 2 studies show us.