The Pill That Could · The Long Argument

The Long Argument, part 2

* * *

Ciba-Geigy in 1995, when Druker first started lobbying for the compound to be advanced, was distracted.

It had begun to merge with another Swiss pharmaceutical company, Sandoz. The two had been rivals in Basel for most of a century. The deal had been announced in March 1996; it was at the time the largest corporate merger in history, in dollar terms, although it would be quickly surpassed by other late-1990s consolidations. The combined company was called Novartis — a name chosen, in the way these names are chosen, to evoke novelty without committing to anything in particular. The merger was finalized in late 1996. Several thousand jobs were eliminated. Two complete pipelines of research compounds, with their associated programs and people, had to be rationalized into one.

Inside Ciba-Geigy's oncology research, the kinase inhibitor program was small enough that it could plausibly be ended without anyone outside the company noticing. Alex Matter — the head of the program, the one who had hired Lydon and Zimmermann — had been pushing for the compound for years and had kept it alive in the budgets, but he was one voice among many, and the merger threatened to bring in a whole new set of priorities and a whole new layer of management to clear them with. Lydon, the biochemist whose hand had been on the program from the start, was navigating his own future at the merged company.

Druker, on the outside, was watching this and growing more impatient.

* * *

It is worth saying, at this point, that the case against developing CGP 57148 was not stupid.

It was, in fact, the case that any responsible pharmaceutical executive would have made about most molecules in most companies most of the time. Compounds like CGP 57148 — promising in academic studies, owned by the company, advocated by enthusiasts, lacking obvious commercial scale — are common. Most of them, statistically, are not destined to become useful drugs. They fail in animals, or they fail in humans, or they make it through trials but the side effects are unacceptable, or they make it through trials and approval but no one will pay for them, or they cost more to develop than they ever earn back. Pharmaceutical companies that pursue every promising-looking compound from every advocate go bankrupt. Pharmaceutical companies that pursue compounds with care, that demand a clear commercial case before committing development resources, are doing what the system rewards them for doing.

CML was uncommon. The compound was a member of a class — selective kinase inhibitor — that had no approved precedent. The chemistry had real issues. There was a genuine concern, raised internally and at the time entirely reasonable, that even if the drug worked it might work briefly, with cancers evolving around it and patients relapsing within months. There were fears about toxicity. There were fears about resistance. There were fears about the time and money it would take to develop a drug that would, in the end, be used by relatively few people. None of these fears was crazy. Some of them, as we'll see in chapters nine and ten, were even partly right.

The case for developing CGP 57148, in 1996 and 1997, rested on the narrowness and cleanness of what Druker had shown. The molecule did exactly one thing, very well, in cells. The disease was caused by exactly one thing. There was an unusually direct line, by the standards of cancer drug discovery, between what the drug did and what the patient needed. CML, while uncommon, was uniformly fatal; the unmet need was as clean as the molecular case. And if the approach worked, it would establish a paradigm — not just a drug, but a way of finding drugs — that could be applied across cancers.

Both cases were defensible. They were also, when stated by their advocates inside Novartis, mutually unconvincing.

* * *

The new CEO of Novartis, the Swiss-American physician Daniel Vasella, took office in 1996, when the merger was completed. He was forty-three. He had trained as a medical doctor, briefly practiced medicine, and then moved into the pharmaceutical industry through Sandoz, where he had risen rapidly. He had a reputation for being unusually direct and unusually willing to make decisions quickly. He also had, as it would later turn out to matter, the perspective of a doctor; he was not unfamiliar with the experience of having a patient in front of him whom he could not help.

Vasella did not, in 1996 or 1997, have a strong opinion about CGP 57148. He had inherited a pipeline of dozens of compounds, each at different stages of development, each with its own internal champion and its own internal skeptic, each demanding attention. The kinase inhibitor program was a small part of that pipeline. It was being kept alive, but it was not being prioritized.

What Druker was trying to do, from Oregon, was raise the priority.

He did this in several ways. He gave talks. The Nature Medicine paper had been the first formal publication; it was followed by oral presentations at scientific meetings, the most important being the American Society of Hematology annual meeting in December of each year, where the entire blood-cancer research community gathers to share new data. Druker presented his preclinical results to packed rooms. Hematologists understood instantly what they were seeing. Many of them began calling Novartis directly, asking when the trial would start. Some of them wrote letters. The pressure was diffuse, but it accumulated.

All 13 chapters of The Pill That Could · Library · All courses