The entrepreneur behind the cancer immunotherapy revolution

Nils Lonberg, a scientist at the center of a revolution in cancer therapy, has had a career full of fateful decisions. One of the most crucial: buying an entire bottle of whiskey at a hotel bar.

It was 1998. Lonberg had just been part of a group dinner with Jim Allison, the charismatic, harmonica-playing scientist who would go on to win a Nobel Prize. After the meal, Lonberg took Allison aside and invited him to have a drink; they stayed up drinking, and talking, until the wee small hours of the morning.

Allison had been trying to find a company to turn his discoveries into life-changing medicines. Lonberg, at the time an executive at a small company, wanted to be the one to make it happen.

“He felt like he’d been, I don’t know, toyed with by companies in the past, and they were just kind of moving slowly and not really serious about it,” Lonberg recalled. “I convinced him we would move quickly. So we did move quickly.”

When Lonberg retired to his room, his head still swimming from booze, he sat down and scribbled out several pages of notes documenting their conversation. He wanted to be able to remember exactly what mattered to his new partner.

It would be 20 years before Allison was awarded the Nobel Prize in physiology or medicine for helping to show how the immune system could be harnessed to fight tumors. But what’s often not appreciated is the degree to which Lonberg and his longtime colleague, Alan Korman, helped usher in this transformation in the way cancer is treated. Success required not just keen science, but a kind of management skill: navigating three companies, two mergers, at least six different chief executives, and angry investors.

“It was really Nil’s vision,” said Dr. Drew Pardoll, an oncologist at the School of Medicine at Johns Hopkins University. “Alan’s an extremely smart, really outstanding scientist. But Nils was really the person who interacted with the business side and brought it all together.”

Lonberg was not on the podium in Oslo that day; nor was Korman. But, without them, the class of medicines now known as checkpoint inhibitors — which have changed the way doctors think about tumors and which now are used to treat skin, lung, kidney, and other cancers — might not exist. Nor would the $16 billion in annual sales that these drugs already generate, including $8 billion for their current employer, Bristol-Myers Squibb. In a few years, the market is expected to double. Lonberg is announcing Monday that he is leaving Bristol to advise the venture capital firm Canaan Partners.

This is the story of how Lonberg ended up working inside one of the world’s largest drug companies to prove that the right drugs could turn the body’s own immune system against tumors.

Lonberg is 63 now and living in Northern California. But decades ago he started as a lab rat in a breeding ground for lab rats.

He studied chemistry at Reed College in Portland, Ore., where he published several scientific papers but never graduated due to a missing physical education credit. In 1980, he joined the Harvard lab of Walter Gilbert to earn a Ph.D. That year, Gilbert won a Nobel for helping to invent DNA sequencing, and he was attracting hot talent. As Lonberg pulled all-nighters, his frequent labmate was George Church, later a pioneer of DNA sequencing and synthetic biology, who would work through the night with cardboard blinders affixed to his safety glasses to avoid becoming distracted.

Lonberg went on to Memorial Sloan Kettering, where he began work creating mice with altered genes. He met his wife in the “mouse house,” where the lab animals were kept, and spent hours talking to Dr. Lloyd Old, a Memorial immunologist who believed, ahead of his time, that the immune system could be harnessed against cancer.

But Lonberg wasn’t so much interested in cancer as in mice. Scientists were just beginning to understand how antibodies, a key part of the immune system, could be created in the animals, grown in hamster cells, and used as drugs. But doing so required arduously genetically engineering the mice, so their antibodies would not be rejected by the human body. That was, Lonberg decided, what he wanted to do.

He convinced a Netherlands company, Genpharm, that creating such mice would be a nice side business to its plan of creating genetically engineered livestock. He started work in Mountain View, Calif., in 1989. There he received a government grant that would allow Genpharm to start hiring scientists — a moment that, to this day, he considers one of his proudest.

“It felt really good to be part of building something that was going to employ a bunch of people and help establish … I don’t want to say a community, but good jobs doing good science for a bunch of people,” Lonberg said. “That kind of prospect for growth was kind of really important. It made something abstract — the way science can be abstract — real.”

For nearly a decade, Lonberg would work on turning his mice into little drug factories. About a dozen marketed drugs would be created with its technology, including some that have only recently reached the market, like Darzalex for multiple myeloma (made by Johnson & Johnson and Genmab) and Cosentyx for psoriasis (made by Novartis).

In 1997, however, Genpharm was purchased by Medarex, a New Jersey-based drug company. Medarex wanted to do more than create drugs for other companies. It wanted to create its own medicines, too.

Enter Alan Korman – and Jim Allison.

Korman had known Lonberg since they had worked down the hall from one another at Harvard. Now Korman was working at a Colorado biotech called NexStar to try to turn Allison’s insights into drugs. Just a few years before, Allison, then a professor at the University of California, Berkeley, had published a series of papers showing that tumor cells use a checkpoint protein to evade the white blood cells that would otherwise kill them. He was convinced that blocking this protein, known as CTLA4, for cytotoxic T-lymphocyte-associated protein 4, could allow the body to annihilate some cancers.

But the technology at NeXstar was unsuited to the task. NeXstar was focused on creating drugs out of nucleic acids, the genetic material in RNA and DNA. That wasn’t working to block CTLA4. But Korman and Allison were having better luck in animal models with prototype antibodies made in mice. They wanted to try that approach in people, but NeXstar’s executives were dragging their heels.

“They weren’t sure if they wanted to do it. It just sat, and sat. It was known that I was pretty angry about it,” Allison recalled recently.

Korman wondered if Lonberg might be interested in using the technology at Medarex to create a CTLA4 drug. Lonberg was. And that bottle of whiskey in Boulder, Colo., showed his commitment. He would meet Allison’s challenge to move quickly.

Normally, researchers make a cell line that can be used to manufacture a new medicine in vats before they start clinical trials. Lonberg decided his team at Medarex would move forward with the original cells derived from mice, shaving a year off the process and leaving the cell line for later.

By February 1999, they had created a molecule for a trial. When it was tested in 17 patients with melanoma the following year, it shrank tumors significantly in just three of them. The side effects were harsh, including inflamed lungs and intestines. But in one patient, the tumors seemed to melt away.

It was more than enough to continue clinical trials.

At the same time, NexStar bowed out. While the trials were starting, the company was sold to Gilead Sciences. The cancer project was sold to Medarex, and Korman went with it.

As Medarex released more readouts of its trials, analysts released a steady beat of notes generating excitement for its stock. There was a sense of exuberance.

Bristol-Myers Squibb, in 2004, had signed on to co-develop the anti-CTLA4 drug, paying Medarex $50 million with a promise of $480 million if it proved effective.

Soon, however, there was trouble. Medarex had come to an agreement with the Food and Drug Administration whereby its drug could be approved if 10% of the patients taking it in its clinical trial saw their tumors shrink 30% or more. Except that didn’t happen. In November 2007, Medarex announced that its drug had missed this key hurdle.

“It was a nightmare of a morning dealing with analysts,” said Lonberg. Shares fell 18% to $11, then lost another 24% over the next two years as Pfizer announced that it was discontinuing its own CTLA4 program.

In July 2009, Bristol-Myers Squibb announced plans to purchase Medarex for $2.5 billion, at a 90% premium over its stock price. One person was opposed to the deal: Nils Lonberg. Despite the setback, he saw potential in the drug he was developing; Medarex, he thought, was being undervalued.

Lonberg had been looking at aggregate data for the ongoing study. He thought that, based on the expected life span of a late-stage melanoma patient, the study was likely to be successful.

“I presented it to the board, my model, and they were not persuaded,” Lonberg said. “So they placed a different value on the company than I did.”

Even before the deal was done, Bristol’s executive team asked him to stay on. He decided to do so, to protect his team and work and because he wanted to learn about large drug companies. He thought he’d remain in the job for a year.

He was wrong about his tenure but right about the drug. In 2010, a clinical trial showed it extended the lives of melanoma patients by about four months, to 10 months, the first increase in survival in decades. It was approved, and marketed as Yervoy.

Last year, it had annual sales of $1.3 billion. When Allison won the Nobel, Korman went to Sweden as his guest.

If Medarex had remained independent, it might have become one of the biggest biotechnology companies in the world. The reason? A second drug Lonberg was developing that would become an even bigger success than Yervoy.

A Japanese scientist, Tasuku Honjo, had discovered a checkpoint protein called PD-1 that helped tumor cells evade white blood cells. That raised the possibility that blocking PD-1 would help the immune system to attack cancer.

Korman and Lonberg noticed Honjo’s work, and started working on their own PD-1 antibodies in animal models. After prolonged negotiations and repeated trips to Japan by Lonberg, Medarex licensed the technology from a Japanese drug company, Ono Pharmaceutical.

Medarex almost didn’t move forward with the project. Lonberg remembers a rancorous, all-day meeting in which he fiercely debated another executive who opposed the notion of betting the house on a second cancer immunotherapy. Lonberg’s counterargument: Medarex now had expertise, and the two drugs might prove effective together. Lonberg won. Clinical trials began in 2006, a few years before Medarex was bought by Bristol.

Then, in 2010, the first data about the PD-1 drug emerged. Six of 16 patients in an initial study of the drug, in various types of cancer, saw their tumors shrink significantly. The data were presented at the annual meeting of the American Society of Clinical Oncology, the largest gathering of cancer researchers.

More studies followed. In 2014, Opdivo was approved to treat metastatic melanoma. As Lonberg predicted, it was even more effective against that disease when combined with Yervoy. It has since been approved for two types of lung cancer, kidney cancer, lymphoma, liver cancer, and a particular type of colon cancer. Last year, the drug had annual sales of $6.7 billion.

But Bristol wasn’t alone. When it first presented its Opdivo data at ASCO, it was to a small, uncrowded room. Someone from Merck was there, Lonberg said.

“I don’t think that data would have been shown if BMS was fully aware of the value,” said Lonberg.

Merck had its own PD-1 drug that it had acquired by accident through its $41 billion purchase of Schering-Plough in 2009, which had in turn gotten the medicine when it purchased the Dutch drug firm Organon for $14 billion in 2007.

The drug had originally been developed at Organon by a team led by Andrea van Elsas, a former graduate student of Allison’s. Early on, he considered working with Medarex, but he remembered the company was not in a position to do a deal because of its relationship with Ono. Lonberg remembers turning him down, though they collaborated on another project.

Merck rushed into clinical trials, managing to procure results at a record pace. In September 2014, the company had its own approved checkpoint inhibitor: Keytruda.

Early on, Opdivo had the clear edge against Keytruda. But that changed in the summer of 2016, when a trial of Keytruda succeeded in previously untreated patients with non-small cell lung cancer, and a similar trial of Opdivo failed. One reason: Merck chose to make better use of a diagnostic test that would select which patients would benefit.

“Merck decided to put its full force into it,” van Elsas, who had left the company before the drug was approved, said in an interview. He is now the chief scientific officer at Aduro Biotech, an immunotherapy drug developer.

Some on Wall Street have argued that Keytruda may simply be the better drug. Van Elsas doubts there are “meaningful differences.”

Still, Merck’s strategy of combining Keytruda with chemotherapy has proved at least as effective, if not more effective, than combining Opdivo and Yervoy in lung cancer. Last year, Keytruda sales hit $7.1 billion, almost $400 million more than Opdivo.

Lonberg said the combination “is important and is not done yet.” Richard Evans, an analyst at the consultancy SSR, has argued that, based on patent filings, Bristol’s pipeline likely holds hidden gems — many of them probably developed under Lonberg.

Now Lonberg himself is leaving. He’s decided he wanted to try something new — this time at the venture capital firm.

“I want to see sort of what is out there,” he said recently. “And not what’s out there in terms of an opportunity for me, but what’s out there in terms of technology and new biology that might be applied.”

Members of the team at Bristol’s Redwood City, Calif., offices know life won’t be the same without him. They’ll be naming the new cafeteria in their freshly built laboratories after him. And, as a result of one of Lonberg’s offhand jokes, they plan to name the burger (or “Lonburger”) after him, too.

“I’m not so delusional to think that just because I was lucky enough to be in the right place at the right time that it’s going to happen again,” he said. “But I’d still like to have an impact.”

Correction: An earlier version of this story misstated the date of Keytruda’s approval.