The Medarex Bet: A Resignation Written on a Napkin
In 2009 Bristol-Myers Squibb acquired Medarex for roughly 2.4 billion dollars, and the asset it was buying became ipilimumab, later approved as Yervoy and widely credited with opening the modern era of cancer immunotherapy. What is less well known is that the company's own management team argued against the deal, and that the executive pushing it says he wrote out his resignation on a napkin to force it through.
That executive was Jeremy Levin, then a member of the executive committee at Bristol-Myers Squibb and the architect of its String of Pearls acquisition strategy. He later became president and CEO of Teva Pharmaceutical Industries and co-founded Ovid Therapeutics. He told the story on Open Door Salon when asked for the hardest decision of his career.
What was the Medarex acquisition?
Bristol-Myers Squibb announced the agreement on 22 July 2009 at sixteen dollars a share, an aggregate of approximately 2.4 billion dollars, or about 2.1 billion net of Medarex's cash and securities. The lead candidate named in the announcement was ipilimumab, then in Phase III for metastatic melanoma, with studies also running in lung cancer, adjuvant melanoma and prostate cancer. Ipilimumab is an antibody against CTLA-4, a brake on the immune system's response to tumours.
The drug was approved as Yervoy in 2011. It is the reason the deal is now taught as a landmark rather than remembered as expensive.
Why did the management team oppose it?
Levin's account is that opposition was close to unanimous inside the room.
the entire management team, with the exception of the CEO and one person on the team, told me that it was absolutely wrong
He describes a meeting in which several people were shouting at him and others were arguing the company already had everything it needed, while the rest stayed quiet. That is a recognizable shape for a large-company decision: the objection is rarely that the science is wrong, it is that the risk is unnecessary given what is already in hand.
What did he actually do?
One of the company's senior people told him he would have his support if he put his own job behind it. Levin wrote it out.
I, Jeremy Levin, will resign from Bristol-Myers Squibb in the event that the CTLA-4 trial does not succeed.
He is clear that this was not agonizing, which is the detail that makes the story useful rather than merely dramatic.
That decision was done in two seconds.
Why was it the easiest decision, not the hardest?
Because the alternative was unacceptable to him on its own terms.
Because I knew that if they didn't buy it, then I didn't want to be with them anyway.
That reframes what looks like a bet as something closer to a clarification. When a decision reveals whether an organization is the kind you want to work in, the personal stake is not really at risk in the way it appears, because both outcomes resolve the same question. Levin contrasts this with what he calls genuinely difficult decisions, which for him involve people rather than assets.
The most difficult thing is it's when you are working with a colleague that you like, deeply respect, know profoundly and have worked all the time with them, and then you have to fire them.
What happened after the deal closed?
The organization moved, and Levin's reading of why is unsentimental.
So, to their credit, once we did it, they all got behind it because somebody else had taken the risk.
That is a precise description of how conviction functions inside a large company. The dissent was not disloyalty and it did not persist; what it needed was for someone to absorb the downside personally. Once that happened, the same people executed.
It also explains why the objection was never really about the asset. If the team had believed the science was wrong, the acquisition closing would not have changed their view. What changed was the distribution of consequence: with the risk assigned to a named person, backing the decision no longer required anyone else to own an outcome they had not chosen. That is an organizational fact rather than a scientific one, and it is the part that recurs.
What should an operator take from this?
Not that staking your job on a trial result is a repeatable technique. It is not, and a version of this story ending in failure would be told as a cautionary tale about one executive overriding his colleagues. The transferable part is narrower: large organizations default to the option that requires no one to be personally exposed, and that default is not the same thing as a considered judgment about the science.
Knowing which of the two you are facing is the skill. Levin's own advice to founders operating under uncertainty is covered in our piece on where biotech capital has moved, and the broader competitive picture in our piece on China's 2035 biotech strategy. The same judgment made in the other direction, by two operators who decided to stop, runs through the conversation with David Esposito and Pavel Khrimian, where Khrimian reduces the call to three questions about hypothesis, cash and time.
This account is drawn from the recorded, on-the-record conversation on Open Door Salon with Jeremy Levin and Sara Jane Demy, hosted by Lori Ellis. Deal terms verified against the Bristol Myers Squibb announcement of 22 July 2009. Interested in sponsoring Open Door Salon?