Part IThe Story
Late one evening in January 2020, Uğur Şahin read a paper in The Lancet about a family that had traveled to Wuhan, and he could not put it down, or rather he put it down and turned to his wife, which in that household amounts to the same thing. The details were clinical and slight. Several family members had been infected with the new coronavirus. Some of them tested positive and showed nothing: no symptoms, no fever. To most readers that week the asymptomatic cases were a curiosity, a footnote to the pneumonia and the overwhelmed hospitals. Şahin was not a virologist. He was an oncologist who had spent nearly thirty years watching disease travel through bodies before anyone noticed it, and he read the footnote the way an oncologist reads a lymph node.
"There were people who had no symptoms, no fever, even though they tested positive for SARS-CoV-2," he said later. "I saw the virus as a cancer that was already metastasising unnoticed." He pictured planes leaving Wuhan full of passengers who looked healthy and would only later fall ill, and he did what he called a very rough calculation, the kind of arithmetic a man who once took a second degree in mathematics by correspondence does in his head: the virus was already everywhere. Özlem Türeci, his wife, his co-founder, the chief medical officer of the company they had built together in Mainz, read it too. Şahin and Türeci argue about papers constantly; it is, by their own description, much of what their marriage is made of. On this one, he recalled, "there was not too much debate."
— Uğur Şahin, on reading the Lancet paper in January 2020I saw the virus as a cancer that was already metastasising unnoticed. Only later do you see what happens. Similarly, I saw planes taking off from Wuhan filled with apparently healthy people who only later became ill and began to infect others.
What they had, at that moment, was a company called BioNTech with roughly a decade of history, no approved product of any kind, a Nasdaq listing barely three months old, and a molecule that most of the pharmaceutical industry had long regarded as a lovely idea and a bad drug. Six months earlier, a reviewer at the U.S. Securities and Exchange Commission had written to "Prof. Ugur Sahin, M.D.," at the company's headquarters on a Mainz street called An der Goldgrube, At the Gold Mine, asking him to "balance your summary disclosure by disclosing that no mRNA immunotherapy has been approved in your new category of therapeutics." The request was routine. It was also, as of January 2020, simply true.
The molecule in question, messenger RNA, is a message. That is not a metaphor; it is the job description. A gene in the nucleus is copied into mRNA, the copy travels to the cell's protein-making machinery, it is read and translated into a protein, and then it falls apart. In a 2006 press release, Mainz's university described the drugs Şahin's team was trying to build as Arbeitskopien zum Einmalgebrauch, single-use working copies of individual genes, which dissolve after they have done their work and leave no permanent change in the genome. For twenty years Şahin had been trying to make that copy survive long enough to be heard.
By the account of Germany's trade-promotion agency, the project he launched in response to the Lancet paper dates from January 24, 2020. It was called Lightspeed. Eleven months later, on a December morning in Coventry, England, a nurse put a needle into the arm of the first person in the world to receive an approved mRNA drug outside a trial.
By the Numbers
The Mainz Message
11 monthsFrom lab and clinical testing to conditional approval of BNT162b2
43,538Participants in the Phase 3 trial whose first readout landed on November 9, 2020
3 billion+Doses BioNTech and Pfizer planned to supply worldwide by end of 2021
~$38BPeak annual Comirnaty sales, reached in 2022
$100B+BioNTech's market capitalization at its August 2021 peak
<3%Typical overlap between two patients' tumors of the same cancer type, per Şahin
500+Patent applications and patents on which Şahin is a co-inventor
A Recommendation for the Hauptschule
The house is still there. In November 2020, when the trial results made him briefly the most discussed scientist on earth, a Turkish news agency sent a photographer to Kurtuluş Mahallesi, a neighborhood of İskenderun on the Mediterranean coast near the Syrian border, to stand outside the home on Şehit Oğuz Yener Caddesi that a four-year-old boy had left fifty-one years earlier. An old neighbor, Sami Uygur, seventy-six, a former head of the local journalists' association, told reporters he had known the family well. The headline the newspaper ran was 51 yıl önce bu evden çıktı: "He left this house 51 years ago." The whole world, it added, had pinned its hopes on him.
He was born there on September 19, 1965. His father had gone ahead to West Germany, to Cologne, where the Ford plants were hiring men from Anatolia and the Mediterranean coast on the explicit understanding, written into the very word the Germans used for them, that they were guests. Gastarbeiter. The arrangement imagined them as labor that would arrive, do its work, and leave without altering the host. In 1969 Şahin and his mother followed.
He grew up in Cologne-Niehl, a district on the Rhine in the shadow of the Ford works, and remembers it with an uncomplicated warmth that is almost disarming in a man so often described as reserved. "I had a very beautiful childhood," he told the Kölner Stadt-Anzeiger in 2021. "It was a time when we children could always be outside — after school, bring the satchel home and then out." They played football on the Rhine meadows, three-on-three, sometimes ten-on-ten. School, he said, did not matter very much in the early years. "Later, of course."
The later is where the story nearly ended. His primary school teacher recommended that he go on to a Hauptschule, the lowest tier of the German secondary system, the track that leads to apprenticeships rather than universities, the track that for a guest worker's son in 1970s Cologne would have looked less like a judgment than a default. A German neighbor intervened. The neighbor's name does not survive in the public record, which is its own small parable: the decisive act was performed by someone who did not need to be remembered for it. Şahin went instead to the Erich-Kästner-Gymnasium in Niehl, named for the author of Emil and the Detectives, took advanced courses in mathematics and chemistry, and graduated in 1984 as the first child at the school with Turkish guest-worker parents.
He borrowed popular science books from the library at his local church. He was, by every account, a football fanatic. A classmate who later worked at the Kölner Stadt-Anzeiger remembered two things about him from their Abitur year: his "insane" mathematical ability and his skill with a ball. Şahin, asked about the second, demurred with the precision of a man who reads data. Compared with the boys who could really play, he said, his talent was "only mediocre."
He stayed in Cologne for medical school, 1984 to 1990, and stayed again for a doctoral thesis in the Clinic I of Internal Medicine under Michael Pfreundschuh, a hematologist working on immune approaches to lymphoma. The thesis title is a mouthful even in German — Bispezifische monoklonale Antikörper zur Aktivierung von Zytostatikavorstufen an Tumorzellen, bispecific monoclonal antibodies for activating chemotherapy precursors directly on tumor cells — and its premise contained the whole of his subsequent career in miniature: do not poison the body; teach a molecule to find the tumor and act only there. It was graded summa cum laude in 1992.
Then he did something that does not appear on the résumés of many oncologists. From 1992 to 1994, while working full-time as a physician, he enrolled in mathematics at the FernUniversität Hagen, Germany's distance-learning university, the institution of night students and correspondence courses. Nobody required it. It was not a credential he would use. It was simply that he wanted to understand things — Ich wollte schon immer die Dinge verstehen, as he put it decades later — and mathematics was a thing.
Day Shift, Night Shift
When Pfreundschuh moved to the university hospital in Homburg, in the Saarland, Şahin went with him. Homburg is a small town near the French border, its medical faculty an outpost of Saarland University, and its oncohematology ward in the early 1990s was a place where young doctors learned, mostly, how people died. Chemotherapy and radiation could buy patients with advanced disease up to a year. Eventually, by Şahin's own estimate, ninety-five percent of them succumbed.
He took care of patients by day and worked in the lab at night. It was on the ward that he met a final-year medical student named Özlem Türeci.
She was born in Siegen in 1967, the daughter of Turkish immigrants of a different kind: her father was an Istanbul-born surgeon at St. Elisabeth-Stift, a small Catholic hospital in Lastrup in rural Lower Saxony, and her mother was a biologist. As a girl she followed her father on his rounds through wards staffed by nuns and decided that what the sisters did, the plain daily labor of helping, was what she wanted to do. She went to gymnasiums in Bad Driburg and Bad Harzburg. She has called herself, with a straight face, a "Prussian Turk." Where Şahin's arrival in Germany had been an industrial transaction, hers was a professional one; where he grew up in the shadow of a car plant, she grew up in the shadow of a crucifix. They found each other on a cancer ward, both haunted, in her word, by "the same dilemma."
"When I made my first experiences with treating cancer patients," Türeci told the New York Times columnist David Gelles in 2021, "I could not believe that this cut-poison-burn type of approach would be the only one we could offer to these patients."
Cut, poison, burn. Surgery, chemotherapy, radiation. The phrase is a clinician's bitter joke and a researcher's indictment, and it became the negative space around which the two of them would spend the next three decades building. Their early work together was the unglamorous labor of target discovery: finding molecules on the surface of tumor cells, of the stomach, pancreas, breast, ovary, prostate and lung, that were absent or hidden on healthy tissue and could therefore serve as addresses for an immune attack. A 2000 paper on tumor-associated antigens in hepatocellular carcinoma lists them both. Şahin completed his habilitation in molecular medicine and immunology at Homburg in 1999; Türeci became a Heisenberg fellow of the German Research Foundation.
Somewhere in those years Şahin went looking for industrial partners, reasoning that if the lab produced something promising, a pharmaceutical company could carry it to patients faster than a university could. The experience curdled into a conviction. A collaboration his university had with a large drug company ended not because the science failed but because the company decided it was getting out of oncology. The project was simply cancelled.
— Uğur Şahin, to Life Science Leader, 2020If you do a project with a large pharmaceutical company, sometimes you find out that the decision to continue the project isn't based on the quality or potential of the research; it is more of a business strategy.
In 2000 he took a year in Zurich, in the Institute for Experimental Immunology run by Hans Hengartner and Rolf Zinkernagel, who four years earlier had shared the Nobel Prize for discovering how T cells recognize infected cells. Zurich was the last stop of his apprenticeship. He came back with a conclusion that he and Türeci had been circling for years. "At that point," he said, "we knew we were going to have to start something ourselves to give projects sufficient time to evolve."
The sentence is worth reading twice. Not to get rich. Not to control the IP. To give projects sufficient time to evolve. The founding motive was about time — the one resource a big company's portfolio review could withdraw overnight — and almost everything Şahin built afterward can be read as an elaborate machine for buying it.
They married in 2002. On the wedding day, by every account, they left the lab, went to the registry office, and went back to the lab.
Single-Use Working Copies
Mainz sits on the Rhine across from the mouth of the Main, a cathedral city famous for two things: carnival, which it takes with a seriousness that alarms visitors, and Johannes Gutenberg, who in the fifteenth century worked out how to make many identical copies of a text from a set of reusable parts. The city's university bears his name. It was there, in 2001, that Şahin and Türeci settled, recruited by Christoph Huber.
Huber ran hematology and oncology at the University Medical Center and had, by the standards of German academic medicine, an unusual appetite for the commercial world. He was older, an Austrian-born immunologist with the mentor's habit of collecting talented people and then getting out of their way. He would co-found both of the companies the couple started and, two decades later, stand beside them at the Werner von Siemens Ring ceremony and in the Mainz town hall as an honorary citizen. In 2001 what he offered was simpler: a lab, a junior research group in a collaborative research center called SFB 432, and the freedom to think about the thing Şahin had started thinking about in 2000, which was whether mRNA could be made into a cancer vaccine.
The industry's verdict on that question was not encouraging. mRNA had been discovered in 1961 and studied exhaustively ever since, but as a drug it had three disqualifying flaws, which Türeci would later list with the weariness of someone who had heard them at every investor meeting for fifteen years: it was "too unstable, too weak in protein production and inflammatory." Injected into the body, the message degraded before it could be read. When it was read, the cell made too little protein, for too short a time. And the immune system, which has evolved to treat stray RNA as evidence of viral invasion, reacted to the drug itself. The field had largely moved on to DNA.
Şahin's group in Mainz began, in 2002, the long work of engineering around each flaw. They redesigned the molecule's structural backbone — the cap at one end, the untranslated regions on either side of the coding sequence, the tail. A 2006 paper in Blood showed that modifying antigen-encoding RNA increased its stability, its translational efficiency and its ability to stimulate T cells. A 2010 paper in Gene Therapy used phosphorothioate cap analogs to make the message last longer in immature dendritic cells, the immune system's scouts. Another, in 2011, worked out that dendritic cells took up naked RNA by a process called macropinocytosis and stopped doing so once they matured — knowledge that would later shape where, and how, you inject the stuff.
In October 2006 the Federal Ministry of Education and Research announced the winners of a new program called GO-Bio, intended to push academic biotechnology toward commercialization. There had been 176 applications. Twenty-eight made the second round. Twelve were funded. Şahin's team received €2 million over three years for "pharmacologically optimized ribonucleic acids as recombinant vaccines and immune adjuvants." The university's press release, written for a lay audience, explained that the optimized RNA chains had induced efficient immune responses in models and had led to the inhibition and regression of tumors. It explained, too, that because such vaccines could be manufactured within a few weeks, they offered the possibility of developing a vaccine tailored to the individual patient. The first clinical test was planned for 2008.
It also contained the sentence about single-use working copies. Read now, the press release is an eerie document: a description, fourteen years early, of the thing that would vaccinate billions of people, written in the modest register of a regional university announcing a ministry grant.
The couple's first company did not, in fact, use mRNA at all. Ganymed Pharmaceuticals, founded in 2001 with Huber and named, with the scholarly whimsy common to German biotech, for the cupbearer of the Greek gods, pursued what it called "ideal monoclonal antibodies" against targets the couple had discovered. Its lead program was an antibody against a splice variant of a protein called claudin-18.2, which in a 2008 paper they described as "a pan-cancer target suitable for therapeutic antibody development": present on gastric and other tumors, essentially hidden on normal tissue. The antibody became zolbetuximab.
Şahin did not run Ganymed. He had no wish to. He served as a medical and scientific officer and later chaired its scientific advisory board; a professional CEO was hired for the day-to-day. Türeci was chief scientific officer from 2001 to 2008 and then, from 2008 until the company was sold, its chief executive. It was she, not he, who first ran a company. In 2016, after a randomized trial showed that zolbetuximab significantly extended overall survival in advanced gastric cancer, Astellas Pharma bought Ganymed in a deal reported at as much as $1.4 billion.
The lesson Şahin drew from Ganymed was not "build one great product." It was the opposite. "At Ganymed, we were focused on one great product; that's what made us successful," he said. "But I wanted to develop multiple products." A single antibody against a single target could help the subset of patients whose tumors carried it. The thing he actually wanted to build was a platform: a way of making many different messages quickly, so that the drug could be fitted to the disease rather than the disease to the drug.
The Twins, the Institute, and the Valley of Death
BioNTech — Biopharmaceutical New Technologies, the capitals preserved in the name like a reminder — was founded in Mainz in 2008, in the middle of the financial crisis, at the precise moment when it was hardest in a generation to raise money for anything. "It was an idea," Şahin said, "that had been incubating in me for almost 15 years."
The money came from a narrow and unusual source. Andreas and Thomas Strüngmann are identical twins from Bavaria who made a fortune in generic pharmaceuticals and spent it, in part, on the kind of long, expensive, unfashionable science that venture funds with ten-year lifespans could not hold. Their family office had already backed Ganymed. Alongside MIG Capital, a Munich venture firm, and Helmut Jeggle, a financier who had managed the Strüngmann money before setting up his own family office and who would become chairman of BioNTech's supervisory board, they put up a seed investment that BioNTech's own history places at €180 million. It is an absurd number for a seed round, which is the point. It bought years. By December 2020 the brothers owned about half of BioNTech, and Bloomberg estimated their combined fortune at more than $24 billion. Patience, in their case, compounded.
The company was only one of the vessels. In 2010 Şahin co-founded TRON — Translational Oncology at the University Medical Center of Johannes Gutenberg University — a non-profit research institute launched with about €30 million and housed in a new building on the medical campus. He served as its scientific director until September 2019. TRON's founding documents describe the problem it was meant to solve with a phrase that has become a cliché in research policy but that in Mainz was meant literally: the gap that had opened between basic and clinical research, "considered 'the valley of death' in translational research." The pharmaceutical industry used to carry discoveries across that valley. Increasingly, it didn't.
"Translation as such is not taught in any university program," Şahin told the Mustafa Prize committee, which honored him in 2019. "It is something that you learn by being involved with interdisciplinary research teams."
The word translation is doing double duty there, and he must have known it. In medicine it means carrying a discovery from the bench to the bedside. In molecular biology it is the precise term for what a ribosome does to an mRNA: reads the message and builds the protein. Şahin had spent his career as a translator in both senses, and in a third, too: the guest worker's son who learned to move between the language of the Ford plant and the language of the gymnasium, between the physician's urgency and the scientist's patience, between the professor's priorities and the investor's.
In 2013 BioNTech hired Katalin Karikó. She was a Hungarian biochemist who had emigrated to the United States in the 1980s and spent decades at the University of Pennsylvania pursuing mRNA when almost no one would fund it, a career that included, as one headline later put it, being demoted and forced to retire. With the immunologist Drew Weissman she had discovered that swapping the nucleoside uridine for a modified version made mRNA far less inflammatory — a solution to the third of the molecule's flaws. She joined BioNTech as a vice president and later became a senior vice president. Her story rhymes with theirs almost too neatly: the immigrant scientist, the dismissed molecule, the decades of stubbornness. In 2014 Şahin, Karikó and Türeci co-wrote a review in Nature Reviews Drug Discovery titled "mRNA-based therapeutics — developing a new class of drugs." It is one of the most cited papers in the field. It was, in effect, a manifesto for a drug class that did not yet exist.
The same year, BioNTech began its first-in-human trial of an individualized mRNA cancer vaccine in solid tumors. The idea was the one sketched in the 2006 press release, now made real with sequencing and software. Biopsy a patient's tumor. Sequence it against the patient's healthy tissue. Identify the mutations that exist only in the cancer. Use algorithms to predict which of those mutated proteins the immune system is most likely to recognize. Write a custom mRNA encoding them. Manufacture it. Inject it. Every patient a different drug. The results, published in Nature in 2017 under the title "Personalized RNA mutanome vaccines mobilize poly-specific therapeutic immunity against cancer," showed that the vaccines generated immune responses against the targeted mutations in melanoma patients.
Şahin explained the logic to the science journalist Kai Kupferschmidt with an arithmetic that is also a philosophy: "When comparing tumors of two patients with the same type of cancer, the similarity of their tumors is less than 3% and more than 97% is unique. The variability and heterogeneity of human cancers is the root cause for the failure of many cancer therapies."
Ninety-seven percent unique. The phrase sounds like a claim about cancer and is also, quietly, a claim about patients — about the impossibility of treating people as categories. It is the closest thing he has to a credo.
The Arrows Were All Halfway
For a decade BioNTech was invisible in the way only a well-funded German company can be. It did deals. In 2016 Genentech paid $310 million upfront to partner on the individualized cancer vaccines. Sanofi signed on for an intratumoral therapy. Genmab, for next-generation immunomodulators. In January 2018 the company closed a $270 million Series A — a Series A, ten years after founding, which tells you something about how long the seed lasted. In August 2018 it signed a multi-year deal with Pfizer to develop mRNA influenza vaccines; under the terms, once BioNTech completed a first-in-human study, Pfizer would take over sole responsibility for development and commercialization. Nobody outside the industry noticed. Inside it, the flu deal was a footnote in Pfizer's vaccines portfolio.
In July 2019 Fidelity led a $325 million Series B. In September the Gates Foundation put in $55 million for HIV and tuberculosis work. And BioNTech filed to go public in the United States.
The SEC's comment letters on the draft registration statement, addressed to Şahin, are among the more quietly revealing documents in the company's history, because they capture, in the deadpan prose of federal regulation, exactly how the company looked from the outside in the summer of 2019. On July 16 the staff asked for "support for your characterization of your immunotherapeutic platforms and product candidates as first-in-class," and asked BioNTech to disclose plainly that no mRNA immunotherapy had ever been approved. It asked why the pipeline table included product candidates identified "merely as 'to be selected' or left blank." On August 20, in a second letter, the staff returned to the pipeline graphic, which used arrows to show how far along each program was. "We note," it wrote, "that all the arrows currently appear to show that all trials are half-way complete."
All the arrows halfway. It is hard to imagine a more exact image of a company that had been working for eleven years on a technology no regulator had ever approved, with a decade of patient capital behind it and a pipeline that was everywhere in progress and nowhere finished.
BioNTech listed on the Nasdaq Global Select Market on October 10, 2019, under the ticker BNTX. Markets were volatile. The deal was cut back. The company raised $150 million, less than it had hoped. Reuters called it "smaller-than-planned."
Three months later, at the J.P. Morgan Healthcare Conference in San Francisco in January 2020 — the same month he would read the Lancet paper — Şahin sat down with Rob Wright of Life Science Leader. The market capitalization had risen by nearly $3 billion between the time the interview was requested and the time it happened. Wright asked how he kept employees from seeing dollar signs. Şahin described the town halls he had held before the IPO to explain to German staff what being public meant, which information had to be shared and which kept confidential, and why. Then he said something about the share price.
— Uğur Şahin, at the J.P. Morgan Healthcare Conference, January 2020I only look at the stock price perhaps once a week. Why worry or get excited if you can't change it?
He meant it. He would also, in the following two years, watch that price carry him onto the list of the world's five hundred richest people.
Lightspeed
"We called it Project Lightspeed," Türeci explained later, "to make clear that the only limitations that will not be questioned are the ones of physics; everything else will be challenged."
The pivot was not total, but it was close. Şahin persuaded the board and investors to redirect a significant share of the company's people and money from cancer to a virus that, in late January, had killed very few people outside China. The company's advantage was the platform itself. Because mRNA is a message rather than a manufactured protein, changing the target means changing the sequence, and changing the sequence is a matter of days. Because BioNTech had spent a decade building variants of the molecule for cancer — uridine-based RNA, nucleoside-modified RNA, self-amplifying RNA, different lipid formulations, different routes — it did not need to choose one design in advance. It could build many and let the data decide.
It built about twenty. Before the disease even had its name, a small group of scientists in Mainz was designing candidate vaccines against different parts of the virus, in different formats, in parallel. By March 2020, BioNTech had several candidates ready to test in humans.
What it lacked was everything else. It had never run a large trial. It had never manufactured at scale. It had never sold a drug. In March it signed two partnerships that divided the world: Fosun Pharma for mainland China, Hong Kong and Macau, and Pfizer, the flu partner, for nearly everywhere else. The carve-outs are telling. BioNTech retained exclusive rights in two countries. One was Germany. The other was Turkey.
Clinical trials began in April, in Germany. Money arrived in quantity and from directions a German biotech had never had to court: a €100 million loan from the European Investment Bank in June, a €250 million investment led by Singapore's Temasek the same month, and €375 million from the German research ministry in September. The candidate that won out, designated BNT162b2, was a lipid-nanoparticle-encapsulated, nucleoside-modified mRNA encoding the virus's spike protein. It combined features from Şahin's and Türeci's earlier work with the uridine modification Karikó had discovered. Pfizer ran the large-scale trial: 43,538 participants.
On Monday, November 9, 2020, the companies announced the first interim readout, based on 94 cases of COVID-19 among participants: the vaccine had prevented more than ninety percent of them. When the couple first saw the efficacy data, the New York Times reported, they celebrated by brewing Turkish tea at home. That was the celebration. Tea. Their daughter, by then a teenager, had spent the year watching her parents work. Şahin, the Guardian noted, still cycled to the headquarters of a company that was suddenly worth $21.9 billion. The Washington Post reported that the couple did not own a car.
His public statement that day had the flattened affect of a man who refuses to let a moment be bigger than its evidence. The result was "a victory for innovation, science and of a global collaborative effort." To another interviewer: "It is the validation of a rigorous scientific approach as well as a validation for the technology. It could open the pharmaceutical field for a new class of molecules." The final analysis, at 170 cases, put efficacy at 95 percent.
On December 2 the United Kingdom granted emergency authorization, the first Western country to approve a COVID-19 vaccine. The first patient was vaccinated in Coventry. The United States followed, then the European Union. On December 3, Bloomberg's index placed Şahin as the 493rd-richest person on the planet, worth $5.1 billion, nearly all of it through a holding company called Medine GmbH, of which he was the sole shareholder, that controlled about 18 percent of BioNTech. Asked for comment, he did not respond.
Something of his temperament survives in the interviews from that winter, which read now less like triumphant press than like ward rounds. When Deutsche Welle asked about the newly discovered British variant, he said that it was not yet known whether the vaccine worked against it, that the company would build a piece of the new variant in the lab within two weeks to test it, and that the odds were good, because "even though this variant has multiple mutations, only 1% of the relevant protein has changed. That means 99% are still the same." When asked whether the ultra-cold storage requirements posed a serious challenge to distribution, he was almost impatient. "These ultra-cold temperatures are nothing new," he said. "Dry ice transportation has been known for more than 50 years. It is not cutting-edge technology. It is just a box with dry ice."
On January 11, 2021, CNBC's Meg Tirrell asked whether he had taken the vaccine himself. "Ah, so we are not yet allowed to take our vaccine," he said, explaining that Germany's prioritization rules put older and high-risk people first, and that the company had found a way to make doses available to its manufacturing teams. He hoped they — "and maybe myself also" — would be vaccinated in the coming weeks. Clips of answers like this, cut down to the sentence about not having been vaccinated, would circulate for a year on social media as supposed proof that the inventor refused his own product. He had in fact been vaccinated in March 2021 and later boosted. The uncut versions had him saying he would "naturally love to be vaccinated."
The book that he and Türeci later produced with Joe Miller of the Financial Times — published in English as The Vaccine: Inside the Race to Conquer the COVID-19 Pandemic and in German as Projekt Lightspeed — appeared on November 2, 2021. One week later, Pfizer's chief executive, Albert Bourla, published his own account, Moonshot: Inside Pfizer's Nine-Month Race to Make the Impossible Possible. Two books, one vaccine, one week apart, each with its own clock: eleven months from Mainz, nine from New York.
Wir sind Impfstoff
In April 2020, before anyone knew whether BioNTech's vaccine would work, a columnist in Berlin's Tagesspiegel wrote about the two scientists under a headline that borrowed the cadence of the Bild front page from 2005, when a German became pope: Wir sind Papst. The new version was Wir sind Impfstoff — "We are vaccine." In October the B.Z. ran the same phrase. The we was doing complicated work. It claimed the couple for Germany. It also, for many of the three million people of Turkish descent in Germany, claimed Germany for them.
The facts of their origins were repeated so often in the autumn of 2020 that they began to function as a kind of national folk tale, recited by German politicians and Turkish newspapers alike. Both children of Turkish immigrants who had come in the late 1960s. Both German citizens, as the Tagesspiegel noted. The first Germans with Turkish roots to enter the country's rich list, at number 93. Vom Gastarbeiterkind zum Weltretter, as one Cologne paper put it: from guest-worker child to savior of the world. "It sounds like a fairy tale," another began. In Turkey the claim was simpler and louder. He was Turkish. He had been born in İskenderun. Reporters went to look at the house.
Şahin has never seemed entirely comfortable with any version of the fairy tale, and he has never repudiated one either. He accepted Germany's Knight Commander's Cross of the Order of Merit from President Frank-Walter Steinmeier in March 2021 and entered his name in the Golden Book of the city of Cologne that September, when the university that had taught him medicine gave him and Türeci honorary doctorates. He told the Kölner Stadt-Anzeiger he would accept the city's invitation gladly. In July 2021 he and Türeci addressed the graduating class of Sabancı University in Istanbul by video, in Turkish. "Be humble, don't lose your curiosity, don't compromise on what you believe in," he told them. "I have always wished this for myself." He predicted that Turkey would return to normal life within three or four months.
In a speech to the Deutsche Nationalstiftung, Germany's national foundation, he reached for an unlikely authority, Helmut Schmidt — the Hamburg Social Democrat, the chancellor whose government presided over the guest-worker era and the halt to recruitment in 1973, the archetypal German pragmatist. "On fundamental questions one must be naive," Şahin quoted him as saying. "I am of the opinion that the problems of the world and of humanity cannot be solved without idealism. Nevertheless, I believe that one should at the same time be realistic and pragmatic." It is a strange and perfect choice. The guest worker's son quoting the guest workers' chancellor on the necessity of naivety.
Some contradictions are simply recorded and left there. On BioNTech's own management board page, beneath his biography, the company lists the attributes it lists for every executive: year of birth, nationality, date of appointment. Under nationality, for the man the German press calls a German citizen and whom Germany has decorated with nearly every honor it has, the page says: Turkish.
He and Türeci have described their own company in terms that mirror the story told about them. BioNTech employs people from some sixty countries. More than half its staff, Deutsche Welle reported in 2020, are women. Türeci credits the vaccine's speed in part to international collaboration, with Pfizer in the United States and Fosun in China. On the question of who would get the vaccine, Şahin said in 2020 that there was "no discussion" about making it available exclusively to individual countries. He has opposed compulsory vaccination and emphasized that it should be voluntary.
The we in Wir sind Impfstoff was generous. The arithmetic, as always, was more complicated.
The Price of a Message
Saving the world turned out to be a business, and the business did not always look like the fairy tale.
In February 2021 a joint investigation by NDR, WDR and the Süddeutsche Zeitung reported that BioNTech and Pfizer's initial offer to the European Union, in June 2020, had been €54.08 per dose. Wolf-Dieter Ludwig, chairman of the drug commission of the German Medical Association, called it "dubious." BioNTech said the figure reflected cost estimates from before mass production and before the dose had even been fixed, and had never been intended as the price for large-scale deliveries; a federal government source described it as "a water-level report." The EU's eventual price was far lower. In August 2021 the price per dose rose to €19.50, which the company justified in part by the need to adapt the vaccine to new variants. Critics said the company was using its dominance.
In May 2021, during a surge of infections and an acute vaccine shortage in Taiwan, President Tsai Ing-wen accused BioNTech of having insisted, during failed negotiations, that the phrase "our country" be deleted from a draft contract. Some read it as Beijing's influence on a German company. Others pointed to a simpler legal problem: under the March 2020 deal, Fosun held exclusive marketing rights to a territory that, by its own definition, included Taiwan. The partnership that had made the vaccine possible in China had also made it, for a time, unavailable to Taiwan.
None of it dented the numbers. In August 2021 BioNTech's market capitalization briefly exceeded $100 billion, making it one of the most valuable companies in Germany. By February 2022 Bloomberg estimated Şahin's net worth at $7.22 billion. Comirnaty sales peaked in 2022 at nearly $38 billion, making it one of the most lucrative pharmaceutical products ever sold. In a prize speech in 2021, addressing "dear Albert" in the audience, Şahin said that the BioNTech and Pfizer teams would produce and supply more than three billion doses worldwide by the end of the year, a figure that "exceeds all expectations." He compared the establishment of mRNA technology to the introduction of recombinant DNA four decades earlier, which had made it possible to manufacture insulin and monoclonal antibodies in fermenters. "Never before in the history of medicine," he said, "has a drug based on a new technology been developed so quickly, so successfully, and made available to hundreds of millions of people worldwide."
The honors arrived in a procession. The Paul Ehrlich and Ludwig Darmstaedter Prize, Germany's most prestigious medical award, received at St. Paul's Church in Frankfurt. The Princess of Asturias Award, shared with Karikó, Weissman, Sarah Gilbert of Oxford and others. The Werner von Siemens Ring, shared with Türeci, Karikó and Huber. The Pour le Mérite. On March 10, 2022, Mainz made Şahin and Türeci honorary citizens.
When the Nobel Prize in Physiology or Medicine for mRNA vaccines came, in October 2023, it went to Karikó and Weissman.
According to one widely repeated account of the Lightspeed year, Şahin, realizing that a company with no commercial experience would suddenly need to scale manufacturing, financing and partnerships globally, read Business Plans for Dummies as a crash course. Whether the story is literally true or merely the kind of thing that gets told about reluctant entrepreneurs, it captures something he has said himself, many times: he never wanted to be one. Asked by a Cologne newspaper whether it was Türeci who had turned the gifted researcher into a man who thought in business terms, he laughed and said, "No." They were both researchers, he said. They both had the goal of helping patients. Türeci, for her part, has described the division between them with more candor: he is the dreamer, in her telling, the one who sees "the beauty of mathematics, science and biology," and believes in an approach everyone else doubts. She is the pragmatist. "It's a symbiosis," she told David Gelles. "I pick up where Uğur leaves things and vice versa."
After the Windfall
Every windfall is also a deadline. The question that hung over BioNTech from 2022 onward was what a company built over a decade for cancer would do once the virus that had made it rich stopped making it money.
The answer, in the founders' minds, had never changed: return to the original thesis. The pandemic had validated the platform and filled the bank. Now the bank would pay for cancer. In January 2023 BioNTech agreed to buy InstaDeep, a British artificial-intelligence company it had worked with since 2019, to industrialize the computational part of its drug design. That year it also signed a partnership with the British government aiming to provide personalized mRNA cancer immunotherapies to as many as 10,000 patients by 2030. In December 2023 it inaugurated its first BioNTainer site in Kigali, Rwanda, a modular mRNA manufacturing facility built in shipping containers, intended to make vaccines where the diseases were. Şahin and Türeci spoke of malaria, tuberculosis and HIV, the "old plagues of humanity," and of a future in which, as a Danish publication summarized their view, companies would be valued by their contribution to society rather than their bottom line.
At the same time, the company quietly stopped being only an mRNA company. It licensed and then, in December 2024, bought the Chinese biotech Biotheus for about $1 billion, securing a bispecific antibody called BNT327, later named pumitamig, which binds both PD-L1 and VEGF-A. It added antibody-drug conjugates. In June 2025 it signed a partnership with Bristol Myers Squibb on pumitamig worth up to $11 billion. The cancer pipeline the company now pitched to investors was a combination strategy — mRNA plus antibodies plus targeted therapies — and its late-stage stars were largely not mRNA.
The same month, BioNTech announced it would buy CureVac, the Tübingen company that had been its rival in the vaccine race, for about $1.25 billion in stock. CureVac's own COVID vaccine had shown low efficacy and been abandoned; in 2022 it had sued BioNTech and Pfizer, claiming Comirnaty infringed its mRNA patents. Buying CureVac ended the litigation, and with it the risk of potentially multibillion-euro damages.
Revenue fell as fast as it had risen. In 2025 BioNTech reported €2.87 billion in revenue, a net loss of €1.14 billion and 7,807 employees; it spent €2.1 billion on research and development. In the first quarter of 2026, Comirnaty revenue fell to €118 million. In May 2026, reporting a quarterly net loss of €532 million, the company announced it would stop making COVID vaccines in Germany and hand manufacturing to Pfizer, and would close sites in Marburg, Idar-Oberstein and Singapore, along with CureVac's operations in Tübingen. About 1,860 jobs were at risk. Tübingen's mayor, Boris Palmer, accused BioNTech of a strategy of "buy first, then kill." The IG BCE union called it "slash-and-burn."
Ingmar Hoerr, CureVac's founder — another German mRNA pioneer of the 1990s, another man who had believed in the molecule when almost no one did, and who had watched his company lose the race to Mainz — told Der Spiegel that the acquisition had been, in effect, misleading to shareholders, that the real motive had been patents, and that cutting staff while buying back shares was "arrogant." He added something more personal. In his view, the BioNTech founders had not demonstrated a sufficiently clear and unique core scientific achievement to warrant a Nobel Prize.
The cancer vaccines themselves were slower and harder than the virus had been. In the late summer of 2026, BioNTech and Genentech stopped a mid-stage trial of the individualized vaccine autogene cevumeran in colorectal cancer after an independent monitoring committee found more deaths in the treatment arm than in the control arm. The shares fell 7.7 percent in a day. Days earlier, they had jumped 22 percent with no company news at all, because Moderna and Merck had reported Phase 3 success for their own personalized melanoma vaccine. The market, it turned out, did not distinguish much between mRNA companies. It traded the category.
In September 2026, Securities and Exchange Commission filings showed six open-market sales of BioNTech shares by Şahin over the course of the month, totaling 182,000 shares and about $17.8 million. He looks at the stock price, he once said, perhaps once a week.
End of Current Appointment
On March 10, 2026, BioNTech announced that Şahin and Türeci would step down as chief executive and chief medical officer by the end of the year and start a new, independent company focused on what it called next-generation mRNA innovations. BioNTech would contribute rights to some of its mRNA technology in exchange for a minority stake, milestones and royalties, but would not fund the new company's operations. It would keep the late-stage cancer portfolio: the antibodies, the conjugates, the fifteen Phase 3 trials it expected to have running by year's end. The founders would take the frontier.
The shares fell by more than 20 percent in early trading. Daina Graybosch, an analyst at Leerink Partners, called the departure "a logical step for a company maturing toward multiple product launches" and said that it also "injects significant uncertainty."
It was, to the day, four years after Mainz made them honorary citizens.
The announcement contains, in compressed form, the shape of everything before it. The men and women who arrived in Cologne in the late 1960s were expected to do a specific job, produce something, and then go, leaving no lasting trace in the host. That is also how the molecule works: a single-use working copy, read and translated, its instructions turned into protein, the message itself dissolving afterward without altering the genome. The founders were supposed to be the message, not the protein. Except that what they produced — a company of nearly eight thousand people, a drug class, a platform that a Swedish committee would one day honor through someone else — stayed. And the messengers did not go home. They went back to the lab, which is the only home either of them has ever consistently described. On their wedding day they returned there within hours.
A boy who was nearly sent to the Hauptschule now holds a W3 professorship, the highest academic rank in Germany, and runs a division at the Helmholtz Institute for Translational Oncology called the mRNA Cancer Immunotherapies Division, where his lab formulates its own lipid nanoparticles and designs individualized neoantigen vaccines for patients whose tumors are ninety-seven percent unique. It is a fifteen-minute walk from the hospital where he started in Mainz in 2001. It is a long way from Şehit Oğuz Yener Caddesi.
Late one evening in January 2020, he read a paper and saw a cancer that was already spreading. In March 2026 the company he founded read his departure as a risk factor. On its website, beneath his photograph, the board page still lists the entries the company lists for every executive, in a typeface designed to be skimmed:
Year of Birth: 1965. Nationality: Turkish. Appointed since: 2008. End of current appointment: 2026.
Part IIThe Playbook
What follows is an attempt to extract, from three decades of Şahin's work, the operating principles that made BioNTech possible — and, in a few cases, the ones whose costs only became visible after the windfall. These are not rules for building a vaccine company. They are patterns about time, capital, partnership and conviction that apply wherever the underlying technology is ahead of its market, and wherever the founder's most valuable asset is the patience to wait for the world to catch up.
The principles
- 1.Found the company to buy time, not to sell a product.
- 2.Build the toolbox before you know the job.
- 3.Read signals from your home discipline into foreign ones.
- 4.Run candidates in parallel and let data choose.
- 5.Rent scale; keep the home ground.
- 6.Match the capital to the clock of the science.
- 7.Split the founder role by temperament, not title.
- 8.Treat the share price as weather.
- 9.Build the institution beside the company.
- 10.Explain mechanisms; refuse miracles.
- 11.A windfall is a loan from the future — budget for repayment.
- 12.Know when the messenger should leave.
M
Three Decades, One Molecule
The decisions that compounded into BioNTech — and the ones that came due afterward.
1984
Graduates from Cologne's Erich-Kästner-Gymnasium, the first child there with Turkish guest-worker parents.
1992
Doctorate summa cum laude on bispecific antibodies against tumor cells; begins correspondence mathematics at FernUniversität Hagen.
2001
Moves to Mainz with Türeci under Christoph Huber; co-founds Ganymed Pharmaceuticals.
2006
Wins one of 12 GO-Bio grants from 176 applicants — €2M to optimize RNA as a cancer vaccine.
2008
Founds BioNTech in the financial crisis with Strüngmann family capital.
2010
Co-founds TRON, a non-profit translational institute, with about €30M.
2013
Hires Katalin Karikó, whose uridine modification will become central to BNT162b2.
2014
First-in-human trial of individualized mRNA cancer vaccines; results published in Nature in 2017.
2019
Nasdaq IPO raises $150M, smaller than planned.
2020
Project Lightspeed: ~20 candidates, Pfizer and Fosun partnerships, approval within 11 months.
2021
Market cap briefly tops $100B; more than 3B doses planned for the year.
2025
Buys CureVac for ~$1.25B; signs BMS deal on pumitamig worth up to $11B.
2026
Founders announce departure to launch a new mRNA company; COVID manufacturing moves to Pfizer.
Principle 1
Found the company to buy time, not to sell a product
Şahin's stated reason for leaving academia was not money, control or even speed. It was duration. He had watched a pharmaceutical partner cancel a promising project because the partner had decided to exit oncology, and he drew a precise conclusion: "we were going to have to start something ourselves to give projects sufficient time to evolve." The company was a vessel for patience, a structure that could keep a decades-long research program alive regardless of whose portfolio strategy shifted.
This framing changes almost every downstream decision. If the purpose of the company is to protect time, you choose investors whose horizon matches the science, you avoid partnerships that hand someone else the off switch, and you tolerate a pipeline whose arrows all look halfway for years. BioNTech's Series A came a full decade after its founding. That is not a failure of fundraising. It is the strategy working.
The reverse lesson is just as important. When outsiders control the timeline, good science dies for reasons unrelated to its quality. Şahin had seen that happen and organized his career to prevent it from happening again.
Tactic: Before founding anything, write down the single resource your current institution keeps withdrawing from you — if the answer is time, design your cap table and partnerships so that nobody else can withdraw it.
Principle 2
Build the toolbox before you know the job
When the virus arrived, BioNTech did not have a coronavirus program. It had something more valuable: two decades of work on the mRNA backbone (caps, untranslated regions, tails), on uridine and nucleoside-modified chemistries, on self-amplifying formats, on lipid formulations and routes of delivery — all developed for cancer. Şahin's team described the result as "different layers" of optimization that combined to produce an exponential improvement in potency and adaptability. The COVID vaccine was not invented in 2020. It was assembled in 2020 from parts made between 2002 and 2019.
The deeper point is that a platform company's real asset is the space of things it can build quickly, not the thing it is currently building. Şahin's frustration at Ganymed — "focused on one great product" — was precisely about that. A single antibody serves a single target. A message-writing machine serves any target whose sequence you can read.
Platforms are expensive and look unfocused to investors. The SEC staff asked why BioNTech listed candidates "to be selected." The honest answer was that optionality was the product.
Tactic: Inventory your organization's reusable capabilities separately from its current products, and fund at least one of them for its own sake, on the assumption that the problem it will solve has not yet appeared.
Principle 3
Read signals from your home discipline into foreign ones
Şahin saw the pandemic early not because he knew more virology than virologists but because he read an infectious-disease paper with an oncologist's pattern library. Asymptomatic carriers looked to him like undetected metastases: the visible disease lags the actual spread. His "very rough" calculation followed from that analogy, and the analogy was what gave him the confidence to redirect the company in January, two months before the World Health Organization declared a pandemic.
Cross-domain reading is not a parlor trick. It is how a specialist sees what generalists in the target field have been trained to discount. Most outbreak watchers in January 2020 were tracking hospitalizations, the visible tumor. Şahin was tracking the people with no symptoms, the micrometastases.
His mathematics degree, taken by correspondence while working as a physician, belongs to the same habit. He collects ways of seeing and applies them outside the fields where they were learned.
Tactic: When a development in an adjacent field crosses your desk, ask explicitly what it would mean if it were happening in your own domain — then check whether that reading predicts something the field's own experts are ignoring.
Principle 4
Run candidates in parallel and let data choose
Lightspeed did not bet on one vaccine design. It built about twenty candidates, varying the antigen and the RNA format, and pushed several into human testing by spring. BNT162b2 was selected because the data favored it, not because it was the plan. This is the opposite of how most organizations behave under time pressure: urgency usually drives premature commitment to a single option, which then has to succeed.
Parallelism is only possible when the marginal cost of each additional option is low. mRNA's defining economic property is that changing the target is a matter of changing a sequence, which Şahin told CNBC could be done "within a few days." He organized the project around that property, converting a technical feature into a decision-making method.
The same logic persisted after the pandemic. "We make the spike protein sequences for potentially relevant variants in the lab every week," he told Der Spiegel in 2022, "and have run this process for hundreds of versions."
Tactic: When the cost of building an extra option is lower than the cost of being wrong, build several in parallel and define in advance the data that will kill all but one.
Principle 5
Rent scale; keep the home ground
In March 2020, a company that had never run a large trial or sold a drug divided the world between two partners: Pfizer for most markets, Fosun for Greater China. It retained exclusive rights in only two countries, Germany and Turkey. The structure acknowledged reality — BioNTech could not manufacture billions of doses or run a 43,538-person trial alone — while keeping the company's identity and its two most symbolically important markets in its own hands.
The Pfizer relationship was not improvised. It grew out of a 2018 influenza deal that had already established working trust. When the emergency came, Şahin called someone he had already partnered with. That is the quiet lesson inside the dramatic one: partnerships that work in a crisis are usually built before the crisis.
There are costs to renting scale. The Fosun territory created the Taiwan dispute in 2021. And in 2026, as COVID revenues collapsed, all vaccine manufacturing went back to the partner. The arrangement that let BioNTech move at lightspeed also defined the shape of its retreat.
Tactic: Sign small, low-stakes partnerships with the large players you might someday need, so that when you need scale urgently you are calling a collaborator rather than negotiating with a stranger.
⇄
Conventional Biotech Wisdom vs. the Mainz Approach
Where Şahin's choices diverged from standard practice — and how they played out.
| Conventional wisdom | Şahin's approach | Outcome |
|---|---|---|
| Focus on a lead asset. | Build a platform with many unnamed candidates. | Validated |
| Raise venture rounds every 18–24 months. | Take a very large seed from patient family capital; Series A a decade later. | Validated |
| Founder-scientist steps aside for a professional CEO. | Hire a CEO at Ganymed; run BioNTech himself. | Mixed |
| Pick the most promising vaccine design early. | Build ~20 candidates in parallel. | Validated |
| Hype the breakthrough. | Frame it as "a box with dry ice" and "99% still the same." | Validated |
| Diversify the windfall through acquisitions. | Buy InstaDeep, Biotheus, CureVac; then close sites. | Contested |
Principle 6
Match the capital to the clock of the science
The Strüngmann twins put up a seed round that BioNTech places at €180 million in 2008, during the financial crisis, for a technology with no approved product anywhere in the world. That capital came from a family office, not a fund with a ten-year life, and its owners had already backed the couple's first company. It was, in the most literal sense, patient money, and it is difficult to imagine BioNTech surviving its first decade on any other kind.
Şahin layered public money on top: a GO-Bio grant in 2006, a federal cluster award in 2012, a €50 million European Investment Bank loan in 2019, a €100 million EIB loan and €375 million from the research ministry in 2020. Each source had its own timeline and its own tolerance for risk. The art was in sequencing them so that no single funder's patience had to cover the whole journey.
When the company finally went to the public markets in 2019, it took $150 million on worse terms than it had wanted. It could afford to, because the IPO was not a lifeline.
Tactic: Estimate honestly how many years your core technology needs before it can prove itself, then refuse capital whose expected holding period is shorter than that number.
Principle 7
Split the founder role by temperament, not title
The partnership between Şahin and Türeci is often romanticized, but its operating logic is precise. By Türeci's own description, he is the dreamer who sees beauty in mathematics and biology and believes in approaches everyone doubts; she is the pragmatist. "I pick up where Uğur leaves things and vice versa." It was Türeci, not Şahin, who first ran a company, as CEO of Ganymed from 2008 to 2016, and Türeci who ran the clinical trials for BNT162b2.
What makes this a playbook rather than a love story is that the division is functional. Each partner covers the other's failure mode. Visionaries without pragmatists build platforms that never reach patients; pragmatists without visionaries ship incremental products. A two-founder structure built around complementary temperaments can do both — and can argue productively, as the couple says they do about nearly every paper, which makes the rare moments of instant agreement, like the January 2020 Lancet paper, unmistakable signals.
Şahin was also comfortable not being the boss when it made sense. At Ganymed he kept to scientific roles and let a hired CEO run operations. Title followed function.
Tactic: Write down your own characteristic failure mode, then choose a co-founder or deputy whose strengths are the inverse — and treat moments of unanimous, immediate agreement between you as unusually strong signals.
"I only look at the stock price perhaps once a week," Şahin told Life Science Leader in January 2020. "Why worry or get excited if you can't change it?" Before the IPO he held town halls to explain to German employees what being public meant, and he steered them toward metrics they could actually influence: patients enrolled, patients responding, papers published.
This discipline mattered most when the price went parabolic. A company whose employees watch their options daily during a pandemic windfall becomes a company that optimizes for the windfall. Şahin's indifference was protective. It kept attention on the work.
The principle has limits, and the later record shows them. In 2026 the stock fell more than 20 percent on the founders' departure news, swung 22 percent on a competitor's trial result and fell 7.7 percent on a halted study. A company that ignores its share price still lives inside it — and its founders, as the 2026 filings showing $17.8 million in sales attest, eventually transact in it too.
Tactic: Pick three operational metrics your team can directly move, report them as prominently as the share price, and deliberately limit how often leadership discusses the stock.
Principle 9
Build the institution beside the company
Şahin did not only found companies. He co-founded TRON in 2010 as a non-profit translational institute, helped establish the Ci3 cluster for individualized immune intervention in 2012, and co-founded the Helmholtz Institute for Translational Oncology. These institutions sat in the "valley of death" between academic discovery and clinical development, where neither universities nor pharma companies wanted to spend money, and they trained the people BioNTech would hire.
The logic is ecosystem-level. A single company cannot produce enough interdisciplinary talent on its own; as Şahin said, "translation as such is not taught in any university program." By building the training ground next door, he created a local labor market that his company could draw on, and a place where long-horizon, non-commercial research could continue regardless of BioNTech's quarterly fortunes.
It also gave him somewhere to go. His HI-TRON lab — formulating its own lipid nanoparticles, building computational models for immunotherapy — is the kind of place where the next company starts.
Tactic: If your company depends on a scarce interdisciplinary skill, fund or co-found a non-profit or academic structure that produces it, and treat it as infrastructure rather than philanthropy.
Principle 10
Explain mechanisms; refuse miracles
Şahin's public communication during the pandemic was conspicuously anti-heroic. Cold-chain logistics: "It is just a box with dry ice." The British variant: "only 1% of the relevant protein has changed. That means 99% are still the same," with a two-week timeline to test it. Adaptation to new variants: technically possible in six weeks, but subject to regulators deciding what evidence they needed. The trial result: "a victory for innovation, science and of a global collaborative effort."
This register built credibility precisely because it refused drama. Each answer gave the listener the mechanism, the uncertainty and the timeline to resolve it. In an environment saturated with both hype and fear, the man who sounded like he was presenting at morning rounds became the most trustworthy voice in the room.
It also inoculated him, partially, against bad-faith editing. When clips of him saying he was "not yet allowed" to take the vaccine circulated out of context, the full interviews, which explained prioritization rules in tedious detail, were there to set the record straight.
Tactic: When announcing anything important, state the mechanism, quantify the remaining uncertainty and give a date by which you will know more — and resist every adjective stronger than the data.
Principle 11
A windfall is a loan from the future — budget for repayment
Comirnaty sales peaked at nearly $38 billion in 2022. By the first quarter of 2026, quarterly COVID revenue had fallen to €118 million. In between, BioNTech did what windfall companies do: it bought capabilities (InstaDeep for AI, Biotheus for bispecific antibodies), eliminated a rival and its litigation (CureVac), signed a major partnership (Bristol Myers Squibb, up to $11 billion) and expanded manufacturing. Some of these moves gave the company a credible late-stage oncology pipeline. Others produced idle capacity, plant closures and roughly 1,860 jobs at risk, along with the charge from Tübingen's mayor of "buy first, then kill."
The lesson is not that the acquisitions were wrong. It is that a windfall changes the cost of mistakes. With tens of billions in the bank, building capacity for a peak demand that could not last looked prudent. Analysts later said the company had built too much. The bill arrived as the founders were leaving.
€
Spending the Windfall
Major post-pandemic moves, categorized by strategic intent.
| Move | Year | Intent | Read |
|---|---|---|---|
| Acquire InstaDeep (AI) | 2023 | Computational drug design | Scan |
| BioNTainer site, Kigali | 2023 | Decentralized mRNA manufacturing | Try |
| Acquire Biotheus (~$1B) | 2024 | Own pumitamig outright | Scale |
| BMS partnership (up to $11B) | 2025 | Accelerate lead oncology asset | Scale |
| Acquire CureVac (~$1.25B) | 2025 | End patent litigation; mRNA assets | Risk |
| Close Marburg, Idar-Oberstein, Singapore, Tübingen | 2026 | Cut overcapacity | Risk |
Tactic: When a single product produces a windfall, model the revenue line falling back to its pre-windfall trend and size every fixed commitment — plants, headcount, acquisitions — to survive that scenario.
Principle 12
Know when the messenger should leave
The March 2026 announcement separated two things that had been fused in BioNTech since 2008: the frontier research program and the commercial company. BioNTech kept the late-stage oncology portfolio and the obligation to launch products. Şahin and Türeci took next-generation mRNA into a new, independent company in which BioNTech would hold a minority stake and receive milestones and royalties, without funding operations.
Founder departures are usually framed as failures or retirements. This one was structured as a split — an acknowledgment that a company preparing multiple oncology launches and a lab chasing the next generation of a technology need different leaders, different capital and different clocks. It mirrors Principle 1 in reverse: the founders left to buy their research program time again, this time from the company they had built to protect it.
The market's reaction, a drop of more than 20 percent, shows the cost. Investors priced the founders as part of the asset. But a founder who stays past the point where the company's needs and their own talents diverge eventually becomes the constraint they once founded the company to escape.
Tactic: Every few years, ask whether the work you are uniquely good at and the work your company most needs are still the same — and if they have diverged, design a structure that lets each go where it belongs rather than forcing one to serve the other.
Free playbook
Get The Business Model Playbook
58 business models, one visual page each: how the money flows, the metrics that matter, and who runs it. Free when you join the Faster Than Normal email.
Free. No spam. Unsubscribe anytime.
Part IIIQuotes and Maxims
In their words
— Uğur Şahin, interviewed by Kai Kupferschmidt for EMBO Molecular Medicine, 2021Our motivation as immunologists has always been to help patients—originally patients with cancer. One of the key challenges here is that each patient's tumor is different. When comparing tumors of two patients with the same type of cancer, the similarity of their tumors is less than 3% and more than 97% is unique.
— Uğur Şahin, to David Gelles at the Atlantic Council, November 2021Definitely not. My research started as a young immunologist asking a question: How can we use the mechanisms of our immune system to fight cancer?
— Uğur Şahin, to Deutsche Welle, December 2020These ultra-cold temperatures are nothing new. Dry ice transportation has been known for more than 50 years. It is not cutting-edge technology. It is just a box with dry ice.
— Uğur Şahin, Empress Theophano Prize speech, 2021Never before in the history of medicine has a drug based on a new technology been developed so quickly, so successfully, and made available to hundreds of millions of people worldwide. This breakthrough was not made possible by a single step. This innovation is based on the accomplishments of many.
— Uğur Şahin, Mustafa Prize profile, 2019Translation as such is not taught in any university program, it is something that you learn by being involved with interdisciplinary research teams.
Maxims
- Found companies to protect time. The scarcest resource in deep science is not money but uninterrupted years, and the right structure exists to guarantee them.
- Build the toolbox before the emergency. The vaccine made in eleven months was assembled from parts built over eighteen years.
- Read across disciplines. An oncologist saw asymptomatic carriers as metastases and acted two months before the world did.
- Let data pick the winner. When options are cheap, build twenty and kill nineteen.
- Partner before you need to. The ally who saves you in a crisis is usually the one you signed a small deal with years earlier.
- Choose money that can wait. Capital whose horizon is shorter than the science's will eventually force the science to lie.
- Divide founder work by temperament. A dreamer and a pragmatist who argue about everything will recognize instantly the rare thing they agree on.
- Look at the stock once a week. Attend to what you can change; the rest is weather.
- Explain the mechanism, not the miracle. "Ninety-nine percent are still the same" persuades better than any superlative.
- Leave when the work and the company diverge. A messenger's job is to deliver the instruction — not to stay in the cell forever.
